{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.2590711175616836,
  "eval_steps": 3000,
  "global_step": 69000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.294069576263428,
      "epoch": 0.00045355587808418,
      "grad_norm": 1.0,
      "learning_rate": 2e-06,
      "loss": 6.8082,
      "mean_token_accuracy": 0.10324694961309433,
      "num_tokens": 10827.0,
      "step": 5
    },
    {
      "entropy": 7.281773281097412,
      "epoch": 0.00090711175616836,
      "grad_norm": 1.2109375,
      "learning_rate": 4.5e-06,
      "loss": 6.82,
      "mean_token_accuracy": 0.09896889626979828,
      "num_tokens": 21025.0,
      "step": 10
    },
    {
      "entropy": 7.273117637634277,
      "epoch": 0.00136066763425254,
      "grad_norm": 1.1328125,
      "learning_rate": 7e-06,
      "loss": 6.7665,
      "mean_token_accuracy": 0.10093757957220077,
      "num_tokens": 31678.0,
      "step": 15
    },
    {
      "entropy": 7.247605180740356,
      "epoch": 0.00181422351233672,
      "grad_norm": 1.0390625,
      "learning_rate": 9.5e-06,
      "loss": 6.7109,
      "mean_token_accuracy": 0.11424258798360824,
      "num_tokens": 41025.0,
      "step": 20
    },
    {
      "entropy": 7.302770519256592,
      "epoch": 0.0022677793904208998,
      "grad_norm": 0.984375,
      "learning_rate": 1.2e-05,
      "loss": 6.8182,
      "mean_token_accuracy": 0.10542535930871963,
      "num_tokens": 51270.0,
      "step": 25
    },
    {
      "entropy": 7.143268346786499,
      "epoch": 0.00272133526850508,
      "grad_norm": 0.98828125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.7184,
      "mean_token_accuracy": 0.09942657127976418,
      "num_tokens": 62142.0,
      "step": 30
    },
    {
      "entropy": 7.138898038864136,
      "epoch": 0.00317489114658926,
      "grad_norm": 0.921875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.8522,
      "mean_token_accuracy": 0.1027273491024971,
      "num_tokens": 72709.0,
      "step": 35
    },
    {
      "entropy": 7.083880186080933,
      "epoch": 0.00362844702467344,
      "grad_norm": 0.9140625,
      "learning_rate": 1.95e-05,
      "loss": 6.691,
      "mean_token_accuracy": 0.1096431165933609,
      "num_tokens": 82313.0,
      "step": 40
    },
    {
      "entropy": 7.065870761871338,
      "epoch": 0.00408200290275762,
      "grad_norm": 0.9921875,
      "learning_rate": 2.2e-05,
      "loss": 6.709,
      "mean_token_accuracy": 0.11063554510474205,
      "num_tokens": 92450.0,
      "step": 45
    },
    {
      "entropy": 7.089556646347046,
      "epoch": 0.0045355587808417995,
      "grad_norm": 0.98828125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.7247,
      "mean_token_accuracy": 0.09662125557661057,
      "num_tokens": 103978.0,
      "step": 50
    },
    {
      "entropy": 7.111036014556885,
      "epoch": 0.00498911465892598,
      "grad_norm": 0.8046875,
      "learning_rate": 2.7e-05,
      "loss": 6.6995,
      "mean_token_accuracy": 0.11020737215876579,
      "num_tokens": 114134.0,
      "step": 55
    },
    {
      "entropy": 7.1653160572052,
      "epoch": 0.00544267053701016,
      "grad_norm": 0.9375,
      "learning_rate": 2.95e-05,
      "loss": 6.8947,
      "mean_token_accuracy": 0.09984557479619979,
      "num_tokens": 125037.0,
      "step": 60
    },
    {
      "entropy": 7.146482801437378,
      "epoch": 0.00589622641509434,
      "grad_norm": 1.0234375,
      "learning_rate": 3.2e-05,
      "loss": 6.7116,
      "mean_token_accuracy": 0.11133431643247604,
      "num_tokens": 134721.0,
      "step": 65
    },
    {
      "entropy": 7.114493370056152,
      "epoch": 0.00634978229317852,
      "grad_norm": 0.89453125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.6489,
      "mean_token_accuracy": 0.10958910807967186,
      "num_tokens": 144448.0,
      "step": 70
    },
    {
      "entropy": 7.126422929763794,
      "epoch": 0.006803338171262699,
      "grad_norm": 0.921875,
      "learning_rate": 3.7e-05,
      "loss": 6.7058,
      "mean_token_accuracy": 0.10741295143961907,
      "num_tokens": 154343.0,
      "step": 75
    },
    {
      "entropy": 7.110815334320068,
      "epoch": 0.00725689404934688,
      "grad_norm": 0.8671875,
      "learning_rate": 3.95e-05,
      "loss": 6.7713,
      "mean_token_accuracy": 0.10833674892783166,
      "num_tokens": 165650.0,
      "step": 80
    },
    {
      "entropy": 7.209178161621094,
      "epoch": 0.007710449927431059,
      "grad_norm": 1.0,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.9573,
      "mean_token_accuracy": 0.10000077486038209,
      "num_tokens": 177098.0,
      "step": 85
    },
    {
      "entropy": 7.122071313858032,
      "epoch": 0.00816400580551524,
      "grad_norm": 1.0078125,
      "learning_rate": 4.45e-05,
      "loss": 6.6834,
      "mean_token_accuracy": 0.11381161436438561,
      "num_tokens": 187903.0,
      "step": 90
    },
    {
      "entropy": 7.1197929859161375,
      "epoch": 0.00861756168359942,
      "grad_norm": 0.9453125,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 6.7278,
      "mean_token_accuracy": 0.10795715600252151,
      "num_tokens": 198077.0,
      "step": 95
    },
    {
      "entropy": 7.142152261734009,
      "epoch": 0.009071117561683599,
      "grad_norm": 0.87890625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.8489,
      "mean_token_accuracy": 0.10571764856576919,
      "num_tokens": 208641.0,
      "step": 100
    },
    {
      "entropy": 7.104328584671021,
      "epoch": 0.009524673439767779,
      "grad_norm": 0.9609375,
      "learning_rate": 5.2e-05,
      "loss": 6.7348,
      "mean_token_accuracy": 0.114326973259449,
      "num_tokens": 218398.0,
      "step": 105
    },
    {
      "entropy": 7.09408917427063,
      "epoch": 0.00997822931785196,
      "grad_norm": 0.9296875,
      "learning_rate": 5.45e-05,
      "loss": 6.6445,
      "mean_token_accuracy": 0.10643699690699578,
      "num_tokens": 229059.0,
      "step": 110
    },
    {
      "entropy": 7.079065132141113,
      "epoch": 0.01043178519593614,
      "grad_norm": 0.85546875,
      "learning_rate": 5.7e-05,
      "loss": 6.7188,
      "mean_token_accuracy": 0.10808592811226844,
      "num_tokens": 239775.0,
      "step": 115
    },
    {
      "entropy": 7.048356246948242,
      "epoch": 0.01088534107402032,
      "grad_norm": 0.87109375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 6.604,
      "mean_token_accuracy": 0.10746061727404595,
      "num_tokens": 249954.0,
      "step": 120
    },
    {
      "entropy": 7.094142627716065,
      "epoch": 0.011338896952104499,
      "grad_norm": 0.88671875,
      "learning_rate": 6.2e-05,
      "loss": 6.7139,
      "mean_token_accuracy": 0.1050918959081173,
      "num_tokens": 260056.0,
      "step": 125
    },
    {
      "entropy": 7.079626655578613,
      "epoch": 0.01179245283018868,
      "grad_norm": 0.98828125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.6984,
      "mean_token_accuracy": 0.1138601079583168,
      "num_tokens": 269904.0,
      "step": 130
    },
    {
      "entropy": 7.136023092269897,
      "epoch": 0.01224600870827286,
      "grad_norm": 0.96484375,
      "learning_rate": 6.7e-05,
      "loss": 6.6975,
      "mean_token_accuracy": 0.10696214884519577,
      "num_tokens": 280900.0,
      "step": 135
    },
    {
      "entropy": 7.108730602264404,
      "epoch": 0.01269956458635704,
      "grad_norm": 0.984375,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.7283,
      "mean_token_accuracy": 0.1071628138422966,
      "num_tokens": 290844.0,
      "step": 140
    },
    {
      "entropy": 7.061822891235352,
      "epoch": 0.013153120464441219,
      "grad_norm": 0.94140625,
      "learning_rate": 7.2e-05,
      "loss": 6.7121,
      "mean_token_accuracy": 0.10810653567314148,
      "num_tokens": 301345.0,
      "step": 145
    },
    {
      "entropy": 7.080287647247315,
      "epoch": 0.013606676342525399,
      "grad_norm": 0.84765625,
      "learning_rate": 7.45e-05,
      "loss": 6.7084,
      "mean_token_accuracy": 0.10700314119458199,
      "num_tokens": 312681.0,
      "step": 150
    },
    {
      "entropy": 7.125663185119629,
      "epoch": 0.01406023222060958,
      "grad_norm": 0.8984375,
      "learning_rate": 7.7e-05,
      "loss": 6.6748,
      "mean_token_accuracy": 0.10948033705353737,
      "num_tokens": 323492.0,
      "step": 155
    },
    {
      "entropy": 7.086132001876831,
      "epoch": 0.01451378809869376,
      "grad_norm": 0.98828125,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.7303,
      "mean_token_accuracy": 0.10884013846516609,
      "num_tokens": 334521.0,
      "step": 160
    },
    {
      "entropy": 7.097598886489868,
      "epoch": 0.014967343976777939,
      "grad_norm": 0.97265625,
      "learning_rate": 8.2e-05,
      "loss": 6.7547,
      "mean_token_accuracy": 0.11234977021813393,
      "num_tokens": 345468.0,
      "step": 165
    },
    {
      "entropy": 7.158268976211548,
      "epoch": 0.015420899854862119,
      "grad_norm": 1.0546875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.6848,
      "mean_token_accuracy": 0.11324930489063263,
      "num_tokens": 355911.0,
      "step": 170
    },
    {
      "entropy": 7.074994707107544,
      "epoch": 0.0158744557329463,
      "grad_norm": 1.0078125,
      "learning_rate": 8.7e-05,
      "loss": 6.7033,
      "mean_token_accuracy": 0.10836530178785324,
      "num_tokens": 367537.0,
      "step": 175
    },
    {
      "entropy": 7.234744691848755,
      "epoch": 0.01632801161103048,
      "grad_norm": 1.046875,
      "learning_rate": 8.95e-05,
      "loss": 6.8874,
      "mean_token_accuracy": 0.10362846553325653,
      "num_tokens": 377918.0,
      "step": 180
    },
    {
      "entropy": 7.165385055541992,
      "epoch": 0.01678156748911466,
      "grad_norm": 0.91796875,
      "learning_rate": 9.2e-05,
      "loss": 6.7471,
      "mean_token_accuracy": 0.10905997231602668,
      "num_tokens": 388461.0,
      "step": 185
    },
    {
      "entropy": 7.085760498046875,
      "epoch": 0.01723512336719884,
      "grad_norm": 1.078125,
      "learning_rate": 9.45e-05,
      "loss": 6.7294,
      "mean_token_accuracy": 0.1090123176574707,
      "num_tokens": 399771.0,
      "step": 190
    },
    {
      "entropy": 7.09751558303833,
      "epoch": 0.01768867924528302,
      "grad_norm": 0.95703125,
      "learning_rate": 9.7e-05,
      "loss": 6.6457,
      "mean_token_accuracy": 0.11132576763629913,
      "num_tokens": 410254.0,
      "step": 195
    },
    {
      "entropy": 7.104769086837768,
      "epoch": 0.018142235123367198,
      "grad_norm": 1.0,
      "learning_rate": 9.95e-05,
      "loss": 6.6448,
      "mean_token_accuracy": 0.10658138245344162,
      "num_tokens": 419792.0,
      "step": 200
    },
    {
      "entropy": 7.064562034606934,
      "epoch": 0.018595791001451378,
      "grad_norm": 1.015625,
      "learning_rate": 0.000102,
      "loss": 6.6915,
      "mean_token_accuracy": 0.11256536692380906,
      "num_tokens": 429264.0,
      "step": 205
    },
    {
      "entropy": 7.033247804641723,
      "epoch": 0.019049346879535557,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.6445,
      "mean_token_accuracy": 0.11356118023395538,
      "num_tokens": 439624.0,
      "step": 210
    },
    {
      "entropy": 7.0533287525177,
      "epoch": 0.01950290275761974,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000107,
      "loss": 6.6641,
      "mean_token_accuracy": 0.10742003992199897,
      "num_tokens": 449556.0,
      "step": 215
    },
    {
      "entropy": 7.127027368545532,
      "epoch": 0.01995645863570392,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001095,
      "loss": 6.6976,
      "mean_token_accuracy": 0.1103837251663208,
      "num_tokens": 460243.0,
      "step": 220
    },
    {
      "entropy": 7.084058809280395,
      "epoch": 0.0204100145137881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000112,
      "loss": 6.7234,
      "mean_token_accuracy": 0.10724443346261978,
      "num_tokens": 469767.0,
      "step": 225
    },
    {
      "entropy": 7.108950233459472,
      "epoch": 0.02086357039187228,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0001145,
      "loss": 6.6963,
      "mean_token_accuracy": 0.1088191345334053,
      "num_tokens": 480329.0,
      "step": 230
    },
    {
      "entropy": 7.037991762161255,
      "epoch": 0.02131712626995646,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.6476,
      "mean_token_accuracy": 0.10804546251893044,
      "num_tokens": 490096.0,
      "step": 235
    },
    {
      "entropy": 7.215046215057373,
      "epoch": 0.02177068214804064,
      "grad_norm": 1.0625,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.6471,
      "mean_token_accuracy": 0.10496766492724419,
      "num_tokens": 501231.0,
      "step": 240
    },
    {
      "entropy": 6.999786281585694,
      "epoch": 0.022224238026124818,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000122,
      "loss": 6.6219,
      "mean_token_accuracy": 0.11307241767644882,
      "num_tokens": 510722.0,
      "step": 245
    },
    {
      "entropy": 7.085280036926269,
      "epoch": 0.022677793904208998,
      "grad_norm": 1.0,
      "learning_rate": 0.0001245,
      "loss": 6.6711,
      "mean_token_accuracy": 0.10972743034362793,
      "num_tokens": 521452.0,
      "step": 250
    },
    {
      "entropy": 7.075028562545777,
      "epoch": 0.023131349782293177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000127,
      "loss": 6.6852,
      "mean_token_accuracy": 0.10879904478788376,
      "num_tokens": 532397.0,
      "step": 255
    },
    {
      "entropy": 7.080456399917603,
      "epoch": 0.02358490566037736,
      "grad_norm": 0.921875,
      "learning_rate": 0.0001295,
      "loss": 6.6625,
      "mean_token_accuracy": 0.10898046344518661,
      "num_tokens": 542271.0,
      "step": 260
    },
    {
      "entropy": 7.04259991645813,
      "epoch": 0.02403846153846154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000132,
      "loss": 6.6673,
      "mean_token_accuracy": 0.11369868144392967,
      "num_tokens": 552411.0,
      "step": 265
    },
    {
      "entropy": 7.1117438793182375,
      "epoch": 0.02449201741654572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.6571,
      "mean_token_accuracy": 0.11307868435978889,
      "num_tokens": 561518.0,
      "step": 270
    },
    {
      "entropy": 7.074210357666016,
      "epoch": 0.0249455732946299,
      "grad_norm": 1.171875,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.7211,
      "mean_token_accuracy": 0.11085665598511696,
      "num_tokens": 571361.0,
      "step": 275
    },
    {
      "entropy": 7.007254123687744,
      "epoch": 0.02539912917271408,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001395,
      "loss": 6.615,
      "mean_token_accuracy": 0.11170280948281289,
      "num_tokens": 580926.0,
      "step": 280
    },
    {
      "entropy": 7.089535856246949,
      "epoch": 0.025852685050798258,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7083,
      "mean_token_accuracy": 0.10579465702176094,
      "num_tokens": 592282.0,
      "step": 285
    },
    {
      "entropy": 7.124145221710205,
      "epoch": 0.026306240928882438,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001445,
      "loss": 6.7235,
      "mean_token_accuracy": 0.10901360288262367,
      "num_tokens": 602767.0,
      "step": 290
    },
    {
      "entropy": 7.125961971282959,
      "epoch": 0.026759796806966617,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000147,
      "loss": 6.6398,
      "mean_token_accuracy": 0.11005486473441124,
      "num_tokens": 612763.0,
      "step": 295
    },
    {
      "entropy": 6.99384126663208,
      "epoch": 0.027213352685050797,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0001495,
      "loss": 6.6203,
      "mean_token_accuracy": 0.11310217007994652,
      "num_tokens": 622796.0,
      "step": 300
    },
    {
      "entropy": 6.990754795074463,
      "epoch": 0.027666908563134977,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000152,
      "loss": 6.5983,
      "mean_token_accuracy": 0.1153037615120411,
      "num_tokens": 633136.0,
      "step": 305
    },
    {
      "entropy": 7.018349838256836,
      "epoch": 0.02812046444121916,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.6196,
      "mean_token_accuracy": 0.11451629996299743,
      "num_tokens": 643347.0,
      "step": 310
    },
    {
      "entropy": 7.088607692718506,
      "epoch": 0.02857402031930334,
      "grad_norm": 1.25,
      "learning_rate": 0.000157,
      "loss": 6.6067,
      "mean_token_accuracy": 0.1193146750330925,
      "num_tokens": 653096.0,
      "step": 315
    },
    {
      "entropy": 6.883705759048462,
      "epoch": 0.02902757619738752,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001595,
      "loss": 6.6439,
      "mean_token_accuracy": 0.1120058111846447,
      "num_tokens": 663557.0,
      "step": 320
    },
    {
      "entropy": 7.102012681961059,
      "epoch": 0.0294811320754717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000162,
      "loss": 6.7247,
      "mean_token_accuracy": 0.1098653145134449,
      "num_tokens": 674395.0,
      "step": 325
    },
    {
      "entropy": 7.06192626953125,
      "epoch": 0.029934687953555878,
      "grad_norm": 1.109375,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.6559,
      "mean_token_accuracy": 0.1131248153746128,
      "num_tokens": 684776.0,
      "step": 330
    },
    {
      "entropy": 7.036684370040893,
      "epoch": 0.030388243831640058,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.6097,
      "mean_token_accuracy": 0.11125744804739952,
      "num_tokens": 694886.0,
      "step": 335
    },
    {
      "entropy": 7.014414644241333,
      "epoch": 0.030841799709724237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.6881,
      "mean_token_accuracy": 0.11227559223771096,
      "num_tokens": 705426.0,
      "step": 340
    },
    {
      "entropy": 7.103907060623169,
      "epoch": 0.03129535558780842,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.6959,
      "mean_token_accuracy": 0.10788564383983612,
      "num_tokens": 716271.0,
      "step": 345
    },
    {
      "entropy": 7.0990581035614015,
      "epoch": 0.0317489114658926,
      "grad_norm": 0.953125,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.672,
      "mean_token_accuracy": 0.1117681197822094,
      "num_tokens": 726875.0,
      "step": 350
    },
    {
      "entropy": 6.9609973430633545,
      "epoch": 0.032202467343976776,
      "grad_norm": 1.015625,
      "learning_rate": 0.000177,
      "loss": 6.6149,
      "mean_token_accuracy": 0.11049453616142273,
      "num_tokens": 737825.0,
      "step": 355
    },
    {
      "entropy": 7.027775573730469,
      "epoch": 0.03265602322206096,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001795,
      "loss": 6.5764,
      "mean_token_accuracy": 0.11501729264855384,
      "num_tokens": 747424.0,
      "step": 360
    },
    {
      "entropy": 7.032524490356446,
      "epoch": 0.033109579100145135,
      "grad_norm": 1.140625,
      "learning_rate": 0.000182,
      "loss": 6.6177,
      "mean_token_accuracy": 0.11091509833931923,
      "num_tokens": 757649.0,
      "step": 365
    },
    {
      "entropy": 7.0149918556213375,
      "epoch": 0.03356313497822932,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001845,
      "loss": 6.6171,
      "mean_token_accuracy": 0.11376268118619919,
      "num_tokens": 768556.0,
      "step": 370
    },
    {
      "entropy": 7.010801887512207,
      "epoch": 0.034016690856313495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000187,
      "loss": 6.607,
      "mean_token_accuracy": 0.1097978562116623,
      "num_tokens": 778331.0,
      "step": 375
    },
    {
      "entropy": 6.967943716049194,
      "epoch": 0.03447024673439768,
      "grad_norm": 1.125,
      "learning_rate": 0.0001895,
      "loss": 6.6039,
      "mean_token_accuracy": 0.1124206118285656,
      "num_tokens": 788065.0,
      "step": 380
    },
    {
      "entropy": 7.05260066986084,
      "epoch": 0.03492380261248186,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000192,
      "loss": 6.6727,
      "mean_token_accuracy": 0.11531179398298264,
      "num_tokens": 798912.0,
      "step": 385
    },
    {
      "entropy": 7.010505723953247,
      "epoch": 0.03537735849056604,
      "grad_norm": 1.015625,
      "learning_rate": 0.0001945,
      "loss": 6.5932,
      "mean_token_accuracy": 0.11802476570010186,
      "num_tokens": 809580.0,
      "step": 390
    },
    {
      "entropy": 7.041739130020142,
      "epoch": 0.03583091436865022,
      "grad_norm": 1.09375,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.698,
      "mean_token_accuracy": 0.11100108623504638,
      "num_tokens": 820207.0,
      "step": 395
    },
    {
      "entropy": 6.895339822769165,
      "epoch": 0.036284470246734396,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.5155,
      "mean_token_accuracy": 0.11325377970933914,
      "num_tokens": 831038.0,
      "step": 400
    },
    {
      "entropy": 7.01606822013855,
      "epoch": 0.03673802612481858,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000202,
      "loss": 6.619,
      "mean_token_accuracy": 0.11496082544326783,
      "num_tokens": 842144.0,
      "step": 405
    },
    {
      "entropy": 7.006781196594238,
      "epoch": 0.037191582002902755,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.6274,
      "mean_token_accuracy": 0.11621586382389068,
      "num_tokens": 851315.0,
      "step": 410
    },
    {
      "entropy": 6.954495906829834,
      "epoch": 0.03764513788098694,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000207,
      "loss": 6.6407,
      "mean_token_accuracy": 0.11225903108716011,
      "num_tokens": 861233.0,
      "step": 415
    },
    {
      "entropy": 6.989374589920044,
      "epoch": 0.038098693759071114,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002095,
      "loss": 6.6035,
      "mean_token_accuracy": 0.11319806352257729,
      "num_tokens": 871131.0,
      "step": 420
    },
    {
      "entropy": 6.961730718612671,
      "epoch": 0.0385522496371553,
      "grad_norm": 1.109375,
      "learning_rate": 0.000212,
      "loss": 6.6019,
      "mean_token_accuracy": 0.10756484866142273,
      "num_tokens": 882195.0,
      "step": 425
    },
    {
      "entropy": 7.033092260360718,
      "epoch": 0.03900580551523948,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002145,
      "loss": 6.5797,
      "mean_token_accuracy": 0.11503332033753395,
      "num_tokens": 891806.0,
      "step": 430
    },
    {
      "entropy": 6.988789701461792,
      "epoch": 0.03945936139332366,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.6323,
      "mean_token_accuracy": 0.10891624465584755,
      "num_tokens": 901350.0,
      "step": 435
    },
    {
      "entropy": 7.0045853614807125,
      "epoch": 0.03991291727140784,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002195,
      "loss": 6.5539,
      "mean_token_accuracy": 0.11177553907036782,
      "num_tokens": 912061.0,
      "step": 440
    },
    {
      "entropy": 6.900162267684936,
      "epoch": 0.040366473149492016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000222,
      "loss": 6.5201,
      "mean_token_accuracy": 0.11248114556074143,
      "num_tokens": 922220.0,
      "step": 445
    },
    {
      "entropy": 6.956882047653198,
      "epoch": 0.0408200290275762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002245,
      "loss": 6.5593,
      "mean_token_accuracy": 0.11303049996495247,
      "num_tokens": 932319.0,
      "step": 450
    },
    {
      "entropy": 6.911965465545654,
      "epoch": 0.041273584905660375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.58,
      "mean_token_accuracy": 0.11454110518097878,
      "num_tokens": 943254.0,
      "step": 455
    },
    {
      "entropy": 6.996125650405884,
      "epoch": 0.04172714078374456,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.6029,
      "mean_token_accuracy": 0.10846876502037048,
      "num_tokens": 953921.0,
      "step": 460
    },
    {
      "entropy": 6.936468839645386,
      "epoch": 0.042180696661828734,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.526,
      "mean_token_accuracy": 0.11692283749580383,
      "num_tokens": 963832.0,
      "step": 465
    },
    {
      "entropy": 6.945839929580688,
      "epoch": 0.04263425253991292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.6208,
      "mean_token_accuracy": 0.10841431692242623,
      "num_tokens": 974066.0,
      "step": 470
    },
    {
      "entropy": 6.90062198638916,
      "epoch": 0.0430878084179971,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000237,
      "loss": 6.616,
      "mean_token_accuracy": 0.11435005068778992,
      "num_tokens": 984402.0,
      "step": 475
    },
    {
      "entropy": 6.956665706634522,
      "epoch": 0.04354136429608128,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002395,
      "loss": 6.5122,
      "mean_token_accuracy": 0.11360828503966332,
      "num_tokens": 993555.0,
      "step": 480
    },
    {
      "entropy": 6.95721468925476,
      "epoch": 0.04399492017416546,
      "grad_norm": 1.140625,
      "learning_rate": 0.000242,
      "loss": 6.5271,
      "mean_token_accuracy": 0.12218032553791999,
      "num_tokens": 1002920.0,
      "step": 485
    },
    {
      "entropy": 6.885728740692139,
      "epoch": 0.044448476052249636,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002445,
      "loss": 6.608,
      "mean_token_accuracy": 0.10632575079798698,
      "num_tokens": 1013611.0,
      "step": 490
    },
    {
      "entropy": 6.892173099517822,
      "epoch": 0.04490203193033382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000247,
      "loss": 6.4836,
      "mean_token_accuracy": 0.11247801780700684,
      "num_tokens": 1024121.0,
      "step": 495
    },
    {
      "entropy": 7.059810018539428,
      "epoch": 0.045355587808417995,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002495,
      "loss": 6.7215,
      "mean_token_accuracy": 0.1127111479640007,
      "num_tokens": 1033845.0,
      "step": 500
    },
    {
      "entropy": 6.917029905319214,
      "epoch": 0.04580914368650218,
      "grad_norm": 1.015625,
      "learning_rate": 0.000252,
      "loss": 6.5315,
      "mean_token_accuracy": 0.11300075054168701,
      "num_tokens": 1044642.0,
      "step": 505
    },
    {
      "entropy": 6.874821615219116,
      "epoch": 0.046262699564586354,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002545,
      "loss": 6.6032,
      "mean_token_accuracy": 0.10996716171503067,
      "num_tokens": 1054728.0,
      "step": 510
    },
    {
      "entropy": 6.894815063476562,
      "epoch": 0.04671625544267054,
      "grad_norm": 1.03125,
      "learning_rate": 0.000257,
      "loss": 6.558,
      "mean_token_accuracy": 0.10969225391745567,
      "num_tokens": 1065743.0,
      "step": 515
    },
    {
      "entropy": 6.828075170516968,
      "epoch": 0.04716981132075472,
      "grad_norm": 1.125,
      "learning_rate": 0.0002595,
      "loss": 6.4864,
      "mean_token_accuracy": 0.11496817469596862,
      "num_tokens": 1076798.0,
      "step": 520
    },
    {
      "entropy": 6.936053895950318,
      "epoch": 0.047623367198838897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000262,
      "loss": 6.6064,
      "mean_token_accuracy": 0.11198530569672585,
      "num_tokens": 1087445.0,
      "step": 525
    },
    {
      "entropy": 6.88051404953003,
      "epoch": 0.04807692307692308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.5258,
      "mean_token_accuracy": 0.11618097051978112,
      "num_tokens": 1098379.0,
      "step": 530
    },
    {
      "entropy": 6.968464517593384,
      "epoch": 0.048530478955007256,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6118,
      "mean_token_accuracy": 0.11241324171423912,
      "num_tokens": 1108016.0,
      "step": 535
    },
    {
      "entropy": 6.838916206359864,
      "epoch": 0.04898403483309144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.4859,
      "mean_token_accuracy": 0.1130056694149971,
      "num_tokens": 1117887.0,
      "step": 540
    },
    {
      "entropy": 6.830152797698974,
      "epoch": 0.049437590711175615,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.4995,
      "mean_token_accuracy": 0.106851976364851,
      "num_tokens": 1128497.0,
      "step": 545
    },
    {
      "entropy": 6.847278594970703,
      "epoch": 0.0498911465892598,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002745,
      "loss": 6.5224,
      "mean_token_accuracy": 0.11530316844582558,
      "num_tokens": 1138168.0,
      "step": 550
    },
    {
      "entropy": 6.883500242233277,
      "epoch": 0.050344702467343974,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000277,
      "loss": 6.5128,
      "mean_token_accuracy": 0.11332335695624352,
      "num_tokens": 1148357.0,
      "step": 555
    },
    {
      "entropy": 6.851631021499633,
      "epoch": 0.05079825834542816,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002795,
      "loss": 6.5414,
      "mean_token_accuracy": 0.11176767796278,
      "num_tokens": 1159964.0,
      "step": 560
    },
    {
      "entropy": 6.893160057067871,
      "epoch": 0.05125181422351233,
      "grad_norm": 1.125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.5865,
      "mean_token_accuracy": 0.11133491396903991,
      "num_tokens": 1170683.0,
      "step": 565
    },
    {
      "entropy": 6.922141647338867,
      "epoch": 0.051705370101596516,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002845,
      "loss": 6.552,
      "mean_token_accuracy": 0.11003944724798202,
      "num_tokens": 1181680.0,
      "step": 570
    },
    {
      "entropy": 6.8009681224823,
      "epoch": 0.0521589259796807,
      "grad_norm": 1.109375,
      "learning_rate": 0.000287,
      "loss": 6.4299,
      "mean_token_accuracy": 0.11521938815712929,
      "num_tokens": 1191619.0,
      "step": 575
    },
    {
      "entropy": 6.7351405143737795,
      "epoch": 0.052612481857764876,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002895,
      "loss": 6.4436,
      "mean_token_accuracy": 0.1167814739048481,
      "num_tokens": 1202264.0,
      "step": 580
    },
    {
      "entropy": 6.800995492935181,
      "epoch": 0.05306603773584906,
      "grad_norm": 1.171875,
      "learning_rate": 0.000292,
      "loss": 6.4535,
      "mean_token_accuracy": 0.11420593708753586,
      "num_tokens": 1212119.0,
      "step": 585
    },
    {
      "entropy": 6.896333932876587,
      "epoch": 0.053519593613933235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002945,
      "loss": 6.5814,
      "mean_token_accuracy": 0.11330544948577881,
      "num_tokens": 1222317.0,
      "step": 590
    },
    {
      "entropy": 6.882574033737183,
      "epoch": 0.05397314949201742,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000297,
      "loss": 6.5286,
      "mean_token_accuracy": 0.11739532127976418,
      "num_tokens": 1232914.0,
      "step": 595
    },
    {
      "entropy": 6.691092920303345,
      "epoch": 0.054426705370101594,
      "grad_norm": 0.96875,
      "learning_rate": 0.0002995,
      "loss": 6.3838,
      "mean_token_accuracy": 0.11833471432328224,
      "num_tokens": 1244214.0,
      "step": 600
    },
    {
      "entropy": 6.853950929641724,
      "epoch": 0.05488026124818578,
      "grad_norm": 1.171875,
      "learning_rate": 0.000302,
      "loss": 6.4824,
      "mean_token_accuracy": 0.11780562847852707,
      "num_tokens": 1254626.0,
      "step": 605
    },
    {
      "entropy": 6.790469121932984,
      "epoch": 0.05533381712626995,
      "grad_norm": 1.25,
      "learning_rate": 0.0003045,
      "loss": 6.5719,
      "mean_token_accuracy": 0.11432634964585305,
      "num_tokens": 1265461.0,
      "step": 610
    },
    {
      "entropy": 6.8617181301116945,
      "epoch": 0.055787373004354136,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000307,
      "loss": 6.4572,
      "mean_token_accuracy": 0.1158202901482582,
      "num_tokens": 1275810.0,
      "step": 615
    },
    {
      "entropy": 6.72202844619751,
      "epoch": 0.05624092888243832,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003095,
      "loss": 6.4716,
      "mean_token_accuracy": 0.1191383220255375,
      "num_tokens": 1285170.0,
      "step": 620
    },
    {
      "entropy": 6.832190465927124,
      "epoch": 0.056694484760522496,
      "grad_norm": 1.015625,
      "learning_rate": 0.000312,
      "loss": 6.4065,
      "mean_token_accuracy": 0.11602196991443633,
      "num_tokens": 1295460.0,
      "step": 625
    },
    {
      "entropy": 6.798471832275391,
      "epoch": 0.05714804063860668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003145,
      "loss": 6.5401,
      "mean_token_accuracy": 0.11370082646608352,
      "num_tokens": 1306656.0,
      "step": 630
    },
    {
      "entropy": 6.835934972763061,
      "epoch": 0.057601596516690855,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000317,
      "loss": 6.5059,
      "mean_token_accuracy": 0.11711395084857941,
      "num_tokens": 1318019.0,
      "step": 635
    },
    {
      "entropy": 6.653035640716553,
      "epoch": 0.05805515239477504,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003195,
      "loss": 6.4287,
      "mean_token_accuracy": 0.11944135650992393,
      "num_tokens": 1328120.0,
      "step": 640
    },
    {
      "entropy": 6.86923885345459,
      "epoch": 0.058508708272859214,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000322,
      "loss": 6.527,
      "mean_token_accuracy": 0.11324021890759468,
      "num_tokens": 1338354.0,
      "step": 645
    },
    {
      "entropy": 6.790145397186279,
      "epoch": 0.0589622641509434,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.4831,
      "mean_token_accuracy": 0.11808143481612206,
      "num_tokens": 1348666.0,
      "step": 650
    },
    {
      "entropy": 6.80904107093811,
      "epoch": 0.05941582002902757,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5404,
      "mean_token_accuracy": 0.1175168700516224,
      "num_tokens": 1359713.0,
      "step": 655
    },
    {
      "entropy": 6.7852167129516605,
      "epoch": 0.059869375907111756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.4064,
      "mean_token_accuracy": 0.12213529795408248,
      "num_tokens": 1369507.0,
      "step": 660
    },
    {
      "entropy": 6.834307098388672,
      "epoch": 0.06032293178519594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5583,
      "mean_token_accuracy": 0.11135573908686638,
      "num_tokens": 1380726.0,
      "step": 665
    },
    {
      "entropy": 6.805551242828369,
      "epoch": 0.060776487663280115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.4852,
      "mean_token_accuracy": 0.11822790279984474,
      "num_tokens": 1391423.0,
      "step": 670
    },
    {
      "entropy": 6.7903587341308596,
      "epoch": 0.0612300435413643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000337,
      "loss": 6.4651,
      "mean_token_accuracy": 0.1208954781293869,
      "num_tokens": 1402492.0,
      "step": 675
    },
    {
      "entropy": 6.778665447235108,
      "epoch": 0.061683599419448475,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003395,
      "loss": 6.5073,
      "mean_token_accuracy": 0.11749565601348877,
      "num_tokens": 1412497.0,
      "step": 680
    },
    {
      "entropy": 6.856817674636841,
      "epoch": 0.06213715529753266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000342,
      "loss": 6.4803,
      "mean_token_accuracy": 0.1222834125161171,
      "num_tokens": 1423214.0,
      "step": 685
    },
    {
      "entropy": 6.691046953201294,
      "epoch": 0.06259071117561683,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4779,
      "mean_token_accuracy": 0.1166260376572609,
      "num_tokens": 1434044.0,
      "step": 690
    },
    {
      "entropy": 6.79459776878357,
      "epoch": 0.06304426705370102,
      "grad_norm": 1.109375,
      "learning_rate": 0.000347,
      "loss": 6.4429,
      "mean_token_accuracy": 0.11843409463763237,
      "num_tokens": 1444387.0,
      "step": 695
    },
    {
      "entropy": 6.704948234558105,
      "epoch": 0.0634978229317852,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003495,
      "loss": 6.3861,
      "mean_token_accuracy": 0.11568867340683937,
      "num_tokens": 1455157.0,
      "step": 700
    },
    {
      "entropy": 6.735180950164795,
      "epoch": 0.06395137880986937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000352,
      "loss": 6.477,
      "mean_token_accuracy": 0.12313677221536637,
      "num_tokens": 1466145.0,
      "step": 705
    },
    {
      "entropy": 6.7879307746887205,
      "epoch": 0.06440493468795355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003545,
      "loss": 6.4737,
      "mean_token_accuracy": 0.1173794686794281,
      "num_tokens": 1476713.0,
      "step": 710
    },
    {
      "entropy": 6.802341651916504,
      "epoch": 0.06485849056603774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000357,
      "loss": 6.482,
      "mean_token_accuracy": 0.12098045200109482,
      "num_tokens": 1486908.0,
      "step": 715
    },
    {
      "entropy": 6.710763883590698,
      "epoch": 0.06531204644412192,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003595,
      "loss": 6.4084,
      "mean_token_accuracy": 0.1253804959356785,
      "num_tokens": 1498185.0,
      "step": 720
    },
    {
      "entropy": 6.715804719924927,
      "epoch": 0.0657656023222061,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000362,
      "loss": 6.4454,
      "mean_token_accuracy": 0.11805666983127594,
      "num_tokens": 1508376.0,
      "step": 725
    },
    {
      "entropy": 6.765777683258056,
      "epoch": 0.06621915820029027,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003645,
      "loss": 6.5331,
      "mean_token_accuracy": 0.11824854388833046,
      "num_tokens": 1519953.0,
      "step": 730
    },
    {
      "entropy": 6.677792072296143,
      "epoch": 0.06667271407837445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000367,
      "loss": 6.4204,
      "mean_token_accuracy": 0.11954097524285316,
      "num_tokens": 1529793.0,
      "step": 735
    },
    {
      "entropy": 6.716114711761475,
      "epoch": 0.06712626995645864,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003695,
      "loss": 6.4162,
      "mean_token_accuracy": 0.12043775469064713,
      "num_tokens": 1540859.0,
      "step": 740
    },
    {
      "entropy": 6.67792387008667,
      "epoch": 0.06757982583454282,
      "grad_norm": 1.0625,
      "learning_rate": 0.000372,
      "loss": 6.3369,
      "mean_token_accuracy": 0.127581225335598,
      "num_tokens": 1549783.0,
      "step": 745
    },
    {
      "entropy": 6.635021018981933,
      "epoch": 0.06803338171262699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003745,
      "loss": 6.4073,
      "mean_token_accuracy": 0.12199747264385223,
      "num_tokens": 1560522.0,
      "step": 750
    },
    {
      "entropy": 6.677228021621704,
      "epoch": 0.06848693759071117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000377,
      "loss": 6.3805,
      "mean_token_accuracy": 0.12109958082437515,
      "num_tokens": 1570668.0,
      "step": 755
    },
    {
      "entropy": 6.705593681335449,
      "epoch": 0.06894049346879536,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003795,
      "loss": 6.4978,
      "mean_token_accuracy": 0.1242837056517601,
      "num_tokens": 1581523.0,
      "step": 760
    },
    {
      "entropy": 6.739868640899658,
      "epoch": 0.06939404934687954,
      "grad_norm": 1.140625,
      "learning_rate": 0.000382,
      "loss": 6.4013,
      "mean_token_accuracy": 0.1199745386838913,
      "num_tokens": 1591356.0,
      "step": 765
    },
    {
      "entropy": 6.6305756092071535,
      "epoch": 0.06984760522496372,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003845,
      "loss": 6.3821,
      "mean_token_accuracy": 0.12265411987900735,
      "num_tokens": 1601779.0,
      "step": 770
    },
    {
      "entropy": 6.71905837059021,
      "epoch": 0.07030116110304789,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.3932,
      "mean_token_accuracy": 0.12244714498519897,
      "num_tokens": 1611995.0,
      "step": 775
    },
    {
      "entropy": 6.672496271133423,
      "epoch": 0.07075471698113207,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.4211,
      "mean_token_accuracy": 0.1178926706314087,
      "num_tokens": 1621834.0,
      "step": 780
    },
    {
      "entropy": 6.678148460388184,
      "epoch": 0.07120827285921626,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4579,
      "mean_token_accuracy": 0.11523651331663132,
      "num_tokens": 1632881.0,
      "step": 785
    },
    {
      "entropy": 6.615700531005859,
      "epoch": 0.07166182873730044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.4437,
      "mean_token_accuracy": 0.11981857419013978,
      "num_tokens": 1642870.0,
      "step": 790
    },
    {
      "entropy": 6.719856405258179,
      "epoch": 0.07211538461538461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.3202,
      "mean_token_accuracy": 0.12486634775996208,
      "num_tokens": 1653390.0,
      "step": 795
    },
    {
      "entropy": 6.682466459274292,
      "epoch": 0.07256894049346879,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003995,
      "loss": 6.446,
      "mean_token_accuracy": 0.12393867447972298,
      "num_tokens": 1663382.0,
      "step": 800
    },
    {
      "entropy": 6.698669910430908,
      "epoch": 0.07302249637155298,
      "grad_norm": 1.046875,
      "learning_rate": 0.000402,
      "loss": 6.449,
      "mean_token_accuracy": 0.1184344507753849,
      "num_tokens": 1673601.0,
      "step": 805
    },
    {
      "entropy": 6.570560455322266,
      "epoch": 0.07347605224963716,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004045,
      "loss": 6.3681,
      "mean_token_accuracy": 0.126424128562212,
      "num_tokens": 1683565.0,
      "step": 810
    },
    {
      "entropy": 6.632832860946655,
      "epoch": 0.07392960812772134,
      "grad_norm": 1.828125,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.3367,
      "mean_token_accuracy": 0.12308898121118546,
      "num_tokens": 1693636.0,
      "step": 815
    },
    {
      "entropy": 6.7626114845275875,
      "epoch": 0.07438316400580551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004095,
      "loss": 6.4517,
      "mean_token_accuracy": 0.11969177201390266,
      "num_tokens": 1703415.0,
      "step": 820
    },
    {
      "entropy": 6.699670362472534,
      "epoch": 0.0748367198838897,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000412,
      "loss": 6.4305,
      "mean_token_accuracy": 0.12528733313083648,
      "num_tokens": 1713579.0,
      "step": 825
    },
    {
      "entropy": 6.586101007461548,
      "epoch": 0.07529027576197388,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004145,
      "loss": 6.4246,
      "mean_token_accuracy": 0.11557562425732612,
      "num_tokens": 1724583.0,
      "step": 830
    },
    {
      "entropy": 6.649460124969482,
      "epoch": 0.07574383164005806,
      "grad_norm": 1.125,
      "learning_rate": 0.000417,
      "loss": 6.349,
      "mean_token_accuracy": 0.12887958213686942,
      "num_tokens": 1735293.0,
      "step": 835
    },
    {
      "entropy": 6.688171482086181,
      "epoch": 0.07619738751814223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004195,
      "loss": 6.366,
      "mean_token_accuracy": 0.123409453779459,
      "num_tokens": 1745623.0,
      "step": 840
    },
    {
      "entropy": 6.583874368667603,
      "epoch": 0.07665094339622641,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000422,
      "loss": 6.4085,
      "mean_token_accuracy": 0.12038984745740891,
      "num_tokens": 1755713.0,
      "step": 845
    },
    {
      "entropy": 6.691501235961914,
      "epoch": 0.0771044992743106,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004245,
      "loss": 6.3932,
      "mean_token_accuracy": 0.12171499505639076,
      "num_tokens": 1765264.0,
      "step": 850
    },
    {
      "entropy": 6.6751635551452635,
      "epoch": 0.07755805515239478,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000427,
      "loss": 6.4043,
      "mean_token_accuracy": 0.12203977331519127,
      "num_tokens": 1775769.0,
      "step": 855
    },
    {
      "entropy": 6.57877140045166,
      "epoch": 0.07801161103047896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295,
      "loss": 6.3398,
      "mean_token_accuracy": 0.12160504087805749,
      "num_tokens": 1786245.0,
      "step": 860
    },
    {
      "entropy": 6.531408166885376,
      "epoch": 0.07846516690856313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432,
      "loss": 6.3694,
      "mean_token_accuracy": 0.1285658784210682,
      "num_tokens": 1796040.0,
      "step": 865
    },
    {
      "entropy": 6.592545700073242,
      "epoch": 0.07891872278664731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004345,
      "loss": 6.3967,
      "mean_token_accuracy": 0.12193747386336326,
      "num_tokens": 1806007.0,
      "step": 870
    },
    {
      "entropy": 6.7914567470550535,
      "epoch": 0.0793722786647315,
      "grad_norm": 1.15625,
      "learning_rate": 0.000437,
      "loss": 6.4675,
      "mean_token_accuracy": 0.11332392543554307,
      "num_tokens": 1816935.0,
      "step": 875
    },
    {
      "entropy": 6.508628511428833,
      "epoch": 0.07982583454281568,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004395,
      "loss": 6.3239,
      "mean_token_accuracy": 0.1232911117374897,
      "num_tokens": 1827419.0,
      "step": 880
    },
    {
      "entropy": 6.673329305648804,
      "epoch": 0.08027939042089985,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000442,
      "loss": 6.4276,
      "mean_token_accuracy": 0.11844546273350716,
      "num_tokens": 1837933.0,
      "step": 885
    },
    {
      "entropy": 6.630283260345459,
      "epoch": 0.08073294629898403,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004445,
      "loss": 6.4793,
      "mean_token_accuracy": 0.12463728785514831,
      "num_tokens": 1848551.0,
      "step": 890
    },
    {
      "entropy": 6.692006587982178,
      "epoch": 0.08118650217706821,
      "grad_norm": 1.125,
      "learning_rate": 0.000447,
      "loss": 6.3544,
      "mean_token_accuracy": 0.11869508177042007,
      "num_tokens": 1858380.0,
      "step": 895
    },
    {
      "entropy": 6.5042383670806885,
      "epoch": 0.0816400580551524,
      "grad_norm": 1.125,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.2822,
      "mean_token_accuracy": 0.12670984119176865,
      "num_tokens": 1868291.0,
      "step": 900
    },
    {
      "entropy": 6.616396141052246,
      "epoch": 0.08209361393323658,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.386,
      "mean_token_accuracy": 0.12300045192241668,
      "num_tokens": 1879219.0,
      "step": 905
    },
    {
      "entropy": 6.637605619430542,
      "epoch": 0.08254716981132075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3695,
      "mean_token_accuracy": 0.12218734100461007,
      "num_tokens": 1889809.0,
      "step": 910
    },
    {
      "entropy": 6.52483057975769,
      "epoch": 0.08300072568940493,
      "grad_norm": 1.125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3383,
      "mean_token_accuracy": 0.12149963527917862,
      "num_tokens": 1899986.0,
      "step": 915
    },
    {
      "entropy": 6.640835094451904,
      "epoch": 0.08345428156748912,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.3212,
      "mean_token_accuracy": 0.12286782488226891,
      "num_tokens": 1909314.0,
      "step": 920
    },
    {
      "entropy": 6.6047581195831295,
      "epoch": 0.0839078374455733,
      "grad_norm": 1.078125,
      "learning_rate": 0.000462,
      "loss": 6.3726,
      "mean_token_accuracy": 0.12332175076007842,
      "num_tokens": 1918443.0,
      "step": 925
    },
    {
      "entropy": 6.563626766204834,
      "epoch": 0.08436139332365747,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004645,
      "loss": 6.316,
      "mean_token_accuracy": 0.12443615049123764,
      "num_tokens": 1928683.0,
      "step": 930
    },
    {
      "entropy": 6.458373975753784,
      "epoch": 0.08481494920174165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000467,
      "loss": 6.233,
      "mean_token_accuracy": 0.12297066450119018,
      "num_tokens": 1938293.0,
      "step": 935
    },
    {
      "entropy": 6.509970378875733,
      "epoch": 0.08526850507982583,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004695,
      "loss": 6.3464,
      "mean_token_accuracy": 0.12207101732492447,
      "num_tokens": 1948233.0,
      "step": 940
    },
    {
      "entropy": 6.600475120544433,
      "epoch": 0.08572206095791002,
      "grad_norm": 1.125,
      "learning_rate": 0.000472,
      "loss": 6.2553,
      "mean_token_accuracy": 0.12516278997063637,
      "num_tokens": 1958611.0,
      "step": 945
    },
    {
      "entropy": 6.593886280059815,
      "epoch": 0.0861756168359942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004745,
      "loss": 6.4088,
      "mean_token_accuracy": 0.1253935366868973,
      "num_tokens": 1968957.0,
      "step": 950
    },
    {
      "entropy": 6.540861320495606,
      "epoch": 0.08662917271407837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000477,
      "loss": 6.2584,
      "mean_token_accuracy": 0.12826553955674172,
      "num_tokens": 1978503.0,
      "step": 955
    },
    {
      "entropy": 6.460687065124512,
      "epoch": 0.08708272859216255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004795,
      "loss": 6.2557,
      "mean_token_accuracy": 0.12641479223966598,
      "num_tokens": 1988458.0,
      "step": 960
    },
    {
      "entropy": 6.554324579238892,
      "epoch": 0.08753628447024674,
      "grad_norm": 1.09375,
      "learning_rate": 0.000482,
      "loss": 6.3931,
      "mean_token_accuracy": 0.12135342583060264,
      "num_tokens": 1999984.0,
      "step": 965
    },
    {
      "entropy": 6.4496111392974855,
      "epoch": 0.08798984034833092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004845,
      "loss": 6.2431,
      "mean_token_accuracy": 0.13175339475274087,
      "num_tokens": 2010645.0,
      "step": 970
    },
    {
      "entropy": 6.600621032714844,
      "epoch": 0.08844339622641509,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000487,
      "loss": 6.36,
      "mean_token_accuracy": 0.12460424080491066,
      "num_tokens": 2020196.0,
      "step": 975
    },
    {
      "entropy": 6.53763222694397,
      "epoch": 0.08889695210449927,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004895,
      "loss": 6.2527,
      "mean_token_accuracy": 0.12676869034767152,
      "num_tokens": 2030101.0,
      "step": 980
    },
    {
      "entropy": 6.473430156707764,
      "epoch": 0.08935050798258345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 6.3426,
      "mean_token_accuracy": 0.1231530986726284,
      "num_tokens": 2040454.0,
      "step": 985
    },
    {
      "entropy": 6.448872709274292,
      "epoch": 0.08980406386066764,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004945,
      "loss": 6.2774,
      "mean_token_accuracy": 0.12086713090538978,
      "num_tokens": 2050378.0,
      "step": 990
    },
    {
      "entropy": 6.557286500930786,
      "epoch": 0.09025761973875182,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497,
      "loss": 6.3341,
      "mean_token_accuracy": 0.12634406611323357,
      "num_tokens": 2060282.0,
      "step": 995
    },
    {
      "entropy": 6.511825323104858,
      "epoch": 0.09071117561683599,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995,
      "loss": 6.3212,
      "mean_token_accuracy": 0.12492988184094429,
      "num_tokens": 2070563.0,
      "step": 1000
    },
    {
      "entropy": 6.540437269210815,
      "epoch": 0.09116473149492017,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999999985110217,
      "loss": 6.3931,
      "mean_token_accuracy": 0.12742957547307016,
      "num_tokens": 2080153.0,
      "step": 1005
    },
    {
      "entropy": 6.470778942108154,
      "epoch": 0.09161828737300436,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999999924620471,
      "loss": 6.236,
      "mean_token_accuracy": 0.12746582329273223,
      "num_tokens": 2090590.0,
      "step": 1010
    },
    {
      "entropy": 6.5432836532592775,
      "epoch": 0.09207184325108854,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999999817600155,
      "loss": 6.318,
      "mean_token_accuracy": 0.11900549307465554,
      "num_tokens": 2101945.0,
      "step": 1015
    },
    {
      "entropy": 6.5527753829956055,
      "epoch": 0.09252539912917271,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999999664049268,
      "loss": 6.3201,
      "mean_token_accuracy": 0.12768215388059617,
      "num_tokens": 2112514.0,
      "step": 1020
    },
    {
      "entropy": 6.417323160171509,
      "epoch": 0.09297895500725689,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999999463967816,
      "loss": 6.2586,
      "mean_token_accuracy": 0.12710441127419472,
      "num_tokens": 2122296.0,
      "step": 1025
    },
    {
      "entropy": 6.639082574844361,
      "epoch": 0.09343251088534107,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049999992173558,
      "loss": 6.4053,
      "mean_token_accuracy": 0.11305732280015945,
      "num_tokens": 2134512.0,
      "step": 1030
    },
    {
      "entropy": 6.5072770595550535,
      "epoch": 0.09388606676342526,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999998924213228,
      "loss": 6.3261,
      "mean_token_accuracy": 0.12832158356904982,
      "num_tokens": 2143934.0,
      "step": 1035
    },
    {
      "entropy": 6.44057936668396,
      "epoch": 0.09433962264150944,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999998584540105,
      "loss": 6.2249,
      "mean_token_accuracy": 0.12571750357747077,
      "num_tokens": 2155730.0,
      "step": 1040
    },
    {
      "entropy": 6.528116464614868,
      "epoch": 0.09479317851959361,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999998198336437,
      "loss": 6.2518,
      "mean_token_accuracy": 0.12236737087368965,
      "num_tokens": 2166422.0,
      "step": 1045
    },
    {
      "entropy": 6.461041498184204,
      "epoch": 0.09524673439767779,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999997765602233,
      "loss": 6.3449,
      "mean_token_accuracy": 0.12111935913562774,
      "num_tokens": 2177842.0,
      "step": 1050
    },
    {
      "entropy": 6.55608811378479,
      "epoch": 0.09570029027576198,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999997286337502,
      "loss": 6.425,
      "mean_token_accuracy": 0.12338349744677543,
      "num_tokens": 2188614.0,
      "step": 1055
    },
    {
      "entropy": 6.552207946777344,
      "epoch": 0.09615384615384616,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999996760542254,
      "loss": 6.2222,
      "mean_token_accuracy": 0.12883713990449905,
      "num_tokens": 2197886.0,
      "step": 1060
    },
    {
      "entropy": 6.3705360889434814,
      "epoch": 0.09660740203193033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049999961882165,
      "loss": 6.2774,
      "mean_token_accuracy": 0.12635389864444732,
      "num_tokens": 2209930.0,
      "step": 1065
    },
    {
      "entropy": 6.534476470947266,
      "epoch": 0.09706095791001451,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999995569360251,
      "loss": 6.2868,
      "mean_token_accuracy": 0.12291447669267655,
      "num_tokens": 2219767.0,
      "step": 1070
    },
    {
      "entropy": 6.428853178024292,
      "epoch": 0.0975145137880987,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499999490397352,
      "loss": 6.2117,
      "mean_token_accuracy": 0.1293780729174614,
      "num_tokens": 2230082.0,
      "step": 1075
    },
    {
      "entropy": 6.476440477371216,
      "epoch": 0.09796806966618288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999994192056321,
      "loss": 6.2369,
      "mean_token_accuracy": 0.13364363461732864,
      "num_tokens": 2240066.0,
      "step": 1080
    },
    {
      "entropy": 6.405725526809692,
      "epoch": 0.09842162554426705,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999993433608669,
      "loss": 6.2701,
      "mean_token_accuracy": 0.12502679005265235,
      "num_tokens": 2250732.0,
      "step": 1085
    },
    {
      "entropy": 6.442107105255127,
      "epoch": 0.09887518142235123,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999992628630579,
      "loss": 6.1593,
      "mean_token_accuracy": 0.12542710900306703,
      "num_tokens": 2261452.0,
      "step": 1090
    },
    {
      "entropy": 6.452964830398559,
      "epoch": 0.09932873730043541,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999991777122069,
      "loss": 6.2658,
      "mean_token_accuracy": 0.1284552037715912,
      "num_tokens": 2271929.0,
      "step": 1095
    },
    {
      "entropy": 6.414393424987793,
      "epoch": 0.0997822931785196,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999990879083156,
      "loss": 6.2473,
      "mean_token_accuracy": 0.12654093131422997,
      "num_tokens": 2282987.0,
      "step": 1100
    },
    {
      "entropy": 6.367011642456054,
      "epoch": 0.10023584905660378,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999989934513857,
      "loss": 6.2291,
      "mean_token_accuracy": 0.13299606069922448,
      "num_tokens": 2292796.0,
      "step": 1105
    },
    {
      "entropy": 6.531451177597046,
      "epoch": 0.10068940493468795,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999988943414193,
      "loss": 6.2776,
      "mean_token_accuracy": 0.127064548432827,
      "num_tokens": 2303057.0,
      "step": 1110
    },
    {
      "entropy": 6.33807897567749,
      "epoch": 0.10114296081277213,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999987905784184,
      "loss": 6.2576,
      "mean_token_accuracy": 0.1281444862484932,
      "num_tokens": 2314168.0,
      "step": 1115
    },
    {
      "entropy": 6.440323400497436,
      "epoch": 0.10159651669085631,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999986821623853,
      "loss": 6.2274,
      "mean_token_accuracy": 0.12680768072605134,
      "num_tokens": 2325241.0,
      "step": 1120
    },
    {
      "entropy": 6.514708232879639,
      "epoch": 0.1020500725689405,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999985690933219,
      "loss": 6.2553,
      "mean_token_accuracy": 0.12574926167726516,
      "num_tokens": 2334866.0,
      "step": 1125
    },
    {
      "entropy": 6.322805023193359,
      "epoch": 0.10250362844702467,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999984513712311,
      "loss": 6.2558,
      "mean_token_accuracy": 0.12497551888227462,
      "num_tokens": 2345422.0,
      "step": 1130
    },
    {
      "entropy": 6.5371935844421385,
      "epoch": 0.10295718432510885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999983289961146,
      "loss": 6.2772,
      "mean_token_accuracy": 0.12496376484632492,
      "num_tokens": 2356285.0,
      "step": 1135
    },
    {
      "entropy": 6.343741798400879,
      "epoch": 0.10341074020319303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999982019679755,
      "loss": 6.174,
      "mean_token_accuracy": 0.13649375587701798,
      "num_tokens": 2366861.0,
      "step": 1140
    },
    {
      "entropy": 6.431583738327026,
      "epoch": 0.10386429608127722,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999980702868164,
      "loss": 6.2109,
      "mean_token_accuracy": 0.13263476118445397,
      "num_tokens": 2376851.0,
      "step": 1145
    },
    {
      "entropy": 6.428322172164917,
      "epoch": 0.1043178519593614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999979339526396,
      "loss": 6.2835,
      "mean_token_accuracy": 0.13185692876577376,
      "num_tokens": 2386857.0,
      "step": 1150
    },
    {
      "entropy": 6.469877624511719,
      "epoch": 0.10477140783744557,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999977929654484,
      "loss": 6.2088,
      "mean_token_accuracy": 0.13108245208859443,
      "num_tokens": 2397035.0,
      "step": 1155
    },
    {
      "entropy": 6.404506397247315,
      "epoch": 0.10522496371552975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999976473252453,
      "loss": 6.2786,
      "mean_token_accuracy": 0.1281719334423542,
      "num_tokens": 2408177.0,
      "step": 1160
    },
    {
      "entropy": 6.543879461288452,
      "epoch": 0.10567851959361393,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999974970320338,
      "loss": 6.2436,
      "mean_token_accuracy": 0.12375756427645683,
      "num_tokens": 2418445.0,
      "step": 1165
    },
    {
      "entropy": 6.355627489089966,
      "epoch": 0.10613207547169812,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999973420858165,
      "loss": 6.2404,
      "mean_token_accuracy": 0.1271333783864975,
      "num_tokens": 2428990.0,
      "step": 1170
    },
    {
      "entropy": 6.4137852668762205,
      "epoch": 0.10658563134978229,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999971824865968,
      "loss": 6.268,
      "mean_token_accuracy": 0.12645680457353592,
      "num_tokens": 2439643.0,
      "step": 1175
    },
    {
      "entropy": 6.5276916980743405,
      "epoch": 0.10703918722786647,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999970182343782,
      "loss": 6.311,
      "mean_token_accuracy": 0.12705472484230995,
      "num_tokens": 2450691.0,
      "step": 1180
    },
    {
      "entropy": 6.504130935668945,
      "epoch": 0.10749274310595065,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999968493291638,
      "loss": 6.2423,
      "mean_token_accuracy": 0.12769789844751359,
      "num_tokens": 2460706.0,
      "step": 1185
    },
    {
      "entropy": 6.226739454269409,
      "epoch": 0.10794629898403484,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999966757709573,
      "loss": 6.1367,
      "mean_token_accuracy": 0.13683899343013764,
      "num_tokens": 2470947.0,
      "step": 1190
    },
    {
      "entropy": 6.458954811096191,
      "epoch": 0.10839985486211902,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999964975597622,
      "loss": 6.1051,
      "mean_token_accuracy": 0.13537428453564643,
      "num_tokens": 2481132.0,
      "step": 1195
    },
    {
      "entropy": 6.338390970230103,
      "epoch": 0.10885341074020319,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999963146955821,
      "loss": 6.138,
      "mean_token_accuracy": 0.1305425487458706,
      "num_tokens": 2491701.0,
      "step": 1200
    },
    {
      "entropy": 6.297045183181763,
      "epoch": 0.10930696661828737,
      "grad_norm": 1.0,
      "learning_rate": 0.000499996127178421,
      "loss": 6.2305,
      "mean_token_accuracy": 0.12452783957123756,
      "num_tokens": 2502360.0,
      "step": 1205
    },
    {
      "entropy": 6.401448202133179,
      "epoch": 0.10976052249637155,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999959350082825,
      "loss": 6.1111,
      "mean_token_accuracy": 0.1346064828336239,
      "num_tokens": 2512140.0,
      "step": 1210
    },
    {
      "entropy": 6.388574886322021,
      "epoch": 0.11021407837445574,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999957381851709,
      "loss": 6.2384,
      "mean_token_accuracy": 0.1266896404325962,
      "num_tokens": 2522721.0,
      "step": 1215
    },
    {
      "entropy": 6.324961376190186,
      "epoch": 0.1106676342525399,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999955367090901,
      "loss": 6.1038,
      "mean_token_accuracy": 0.13158604130148888,
      "num_tokens": 2531786.0,
      "step": 1220
    },
    {
      "entropy": 6.4173516750335695,
      "epoch": 0.11112119013062409,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999953305800441,
      "loss": 6.2082,
      "mean_token_accuracy": 0.12343020662665367,
      "num_tokens": 2542895.0,
      "step": 1225
    },
    {
      "entropy": 6.308479452133179,
      "epoch": 0.11157474600870827,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999951197980374,
      "loss": 6.2113,
      "mean_token_accuracy": 0.12858483716845512,
      "num_tokens": 2554120.0,
      "step": 1230
    },
    {
      "entropy": 6.494872283935547,
      "epoch": 0.11202830188679246,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999949043630744,
      "loss": 6.2613,
      "mean_token_accuracy": 0.126002337038517,
      "num_tokens": 2563935.0,
      "step": 1235
    },
    {
      "entropy": 6.2588221549987795,
      "epoch": 0.11248185776487664,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999946842751593,
      "loss": 6.1536,
      "mean_token_accuracy": 0.13200209736824037,
      "num_tokens": 2574465.0,
      "step": 1240
    },
    {
      "entropy": 6.344051551818848,
      "epoch": 0.11293541364296081,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999944595342968,
      "loss": 6.1119,
      "mean_token_accuracy": 0.13319186195731164,
      "num_tokens": 2584694.0,
      "step": 1245
    },
    {
      "entropy": 6.293050384521484,
      "epoch": 0.11338896952104499,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999942301404916,
      "loss": 6.0892,
      "mean_token_accuracy": 0.13468018174171448,
      "num_tokens": 2594540.0,
      "step": 1250
    },
    {
      "entropy": 6.27103443145752,
      "epoch": 0.11384252539912917,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999939960937484,
      "loss": 6.0265,
      "mean_token_accuracy": 0.13721646815538407,
      "num_tokens": 2604896.0,
      "step": 1255
    },
    {
      "entropy": 6.413160896301269,
      "epoch": 0.11429608127721336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999937573940721,
      "loss": 6.1974,
      "mean_token_accuracy": 0.14260305911302568,
      "num_tokens": 2614143.0,
      "step": 1260
    },
    {
      "entropy": 6.228365230560303,
      "epoch": 0.11474963715529753,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999935140414674,
      "loss": 6.1018,
      "mean_token_accuracy": 0.13245780095458032,
      "num_tokens": 2623999.0,
      "step": 1265
    },
    {
      "entropy": 6.427315664291382,
      "epoch": 0.11520319303338171,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999932660359395,
      "loss": 6.3045,
      "mean_token_accuracy": 0.1259209007024765,
      "num_tokens": 2636208.0,
      "step": 1270
    },
    {
      "entropy": 6.353607034683227,
      "epoch": 0.11565674891146589,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999930133774936,
      "loss": 6.1813,
      "mean_token_accuracy": 0.13700075224041938,
      "num_tokens": 2645854.0,
      "step": 1275
    },
    {
      "entropy": 6.392856884002685,
      "epoch": 0.11611030478955008,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999927560661348,
      "loss": 6.1641,
      "mean_token_accuracy": 0.13880657628178597,
      "num_tokens": 2655877.0,
      "step": 1280
    },
    {
      "entropy": 6.396235799789428,
      "epoch": 0.11656386066763426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999924941018685,
      "loss": 6.1871,
      "mean_token_accuracy": 0.1347483940422535,
      "num_tokens": 2666157.0,
      "step": 1285
    },
    {
      "entropy": 6.304231357574463,
      "epoch": 0.11701741654571843,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999922274847,
      "loss": 6.1112,
      "mean_token_accuracy": 0.13436754122376443,
      "num_tokens": 2676647.0,
      "step": 1290
    },
    {
      "entropy": 6.259562253952026,
      "epoch": 0.11747097242380261,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499991956214635,
      "loss": 6.1475,
      "mean_token_accuracy": 0.14410557746887206,
      "num_tokens": 2686417.0,
      "step": 1295
    },
    {
      "entropy": 6.3692951679229735,
      "epoch": 0.1179245283018868,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499991680291679,
      "loss": 6.1398,
      "mean_token_accuracy": 0.12797631248831748,
      "num_tokens": 2697540.0,
      "step": 1300
    },
    {
      "entropy": 6.3081450939178465,
      "epoch": 0.11837808417997098,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999913997158378,
      "loss": 6.1925,
      "mean_token_accuracy": 0.12314079701900482,
      "num_tokens": 2709397.0,
      "step": 1305
    },
    {
      "entropy": 6.354034662246704,
      "epoch": 0.11883164005805515,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999911144871169,
      "loss": 6.1452,
      "mean_token_accuracy": 0.12890623807907103,
      "num_tokens": 2720430.0,
      "step": 1310
    },
    {
      "entropy": 6.274476385116577,
      "epoch": 0.11928519593613933,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999908246055226,
      "loss": 6.1047,
      "mean_token_accuracy": 0.128923050314188,
      "num_tokens": 2730648.0,
      "step": 1315
    },
    {
      "entropy": 6.303538846969604,
      "epoch": 0.11973875181422351,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999905300710607,
      "loss": 6.0876,
      "mean_token_accuracy": 0.13420135006308556,
      "num_tokens": 2739979.0,
      "step": 1320
    },
    {
      "entropy": 6.31509165763855,
      "epoch": 0.1201923076923077,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999902308837373,
      "loss": 6.2133,
      "mean_token_accuracy": 0.12678968757390977,
      "num_tokens": 2750329.0,
      "step": 1325
    },
    {
      "entropy": 6.295187377929688,
      "epoch": 0.12064586357039188,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999899270435584,
      "loss": 6.0895,
      "mean_token_accuracy": 0.14072596430778503,
      "num_tokens": 2760535.0,
      "step": 1330
    },
    {
      "entropy": 6.30014271736145,
      "epoch": 0.12109941944847605,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999896185505307,
      "loss": 6.1235,
      "mean_token_accuracy": 0.133646260201931,
      "num_tokens": 2770470.0,
      "step": 1335
    },
    {
      "entropy": 6.295889663696289,
      "epoch": 0.12155297532656023,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999893054046603,
      "loss": 6.1483,
      "mean_token_accuracy": 0.13114681392908095,
      "num_tokens": 2781506.0,
      "step": 1340
    },
    {
      "entropy": 6.219323301315308,
      "epoch": 0.12200653120464441,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999889876059537,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14254361912608146,
      "num_tokens": 2791245.0,
      "step": 1345
    },
    {
      "entropy": 6.264402437210083,
      "epoch": 0.1224600870827286,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999886651544176,
      "loss": 6.0352,
      "mean_token_accuracy": 0.13747782483696938,
      "num_tokens": 2801680.0,
      "step": 1350
    },
    {
      "entropy": 6.328179216384887,
      "epoch": 0.12291364296081277,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999883380500584,
      "loss": 6.1313,
      "mean_token_accuracy": 0.13885134533047677,
      "num_tokens": 2812805.0,
      "step": 1355
    },
    {
      "entropy": 6.274612092971802,
      "epoch": 0.12336719883889695,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999880062928831,
      "loss": 6.1525,
      "mean_token_accuracy": 0.13258814290165902,
      "num_tokens": 2822834.0,
      "step": 1360
    },
    {
      "entropy": 6.32262659072876,
      "epoch": 0.12382075471698113,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999876698828985,
      "loss": 6.093,
      "mean_token_accuracy": 0.1388528026640415,
      "num_tokens": 2832312.0,
      "step": 1365
    },
    {
      "entropy": 6.264431667327881,
      "epoch": 0.12427431059506532,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999873288201116,
      "loss": 6.1074,
      "mean_token_accuracy": 0.14017656743526458,
      "num_tokens": 2842998.0,
      "step": 1370
    },
    {
      "entropy": 6.334399604797364,
      "epoch": 0.1247278664731495,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004999869831045294,
      "loss": 6.1358,
      "mean_token_accuracy": 0.13416288197040557,
      "num_tokens": 2854218.0,
      "step": 1375
    },
    {
      "entropy": 6.257809448242187,
      "epoch": 0.12518142235123367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999866327361589,
      "loss": 6.0218,
      "mean_token_accuracy": 0.14412764012813567,
      "num_tokens": 2864484.0,
      "step": 1380
    },
    {
      "entropy": 6.278401279449463,
      "epoch": 0.12563497822931785,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999862777150077,
      "loss": 6.1335,
      "mean_token_accuracy": 0.13539522439241408,
      "num_tokens": 2873800.0,
      "step": 1385
    },
    {
      "entropy": 6.327728033065796,
      "epoch": 0.12608853410740203,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999859180410829,
      "loss": 6.1425,
      "mean_token_accuracy": 0.13279536589980126,
      "num_tokens": 2884247.0,
      "step": 1390
    },
    {
      "entropy": 6.269402170181275,
      "epoch": 0.12654208998548622,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999855537143919,
      "loss": 6.0613,
      "mean_token_accuracy": 0.13359882831573486,
      "num_tokens": 2894739.0,
      "step": 1395
    },
    {
      "entropy": 6.197747993469238,
      "epoch": 0.1269956458635704,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999851847349425,
      "loss": 6.0982,
      "mean_token_accuracy": 0.13437446355819702,
      "num_tokens": 2904376.0,
      "step": 1400
    },
    {
      "entropy": 6.377307605743408,
      "epoch": 0.12744920174165458,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999848111027419,
      "loss": 6.1661,
      "mean_token_accuracy": 0.12998344823718072,
      "num_tokens": 2914304.0,
      "step": 1405
    },
    {
      "entropy": 6.292350578308105,
      "epoch": 0.12790275761973874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999844328177983,
      "loss": 6.0278,
      "mean_token_accuracy": 0.14133364632725715,
      "num_tokens": 2924400.0,
      "step": 1410
    },
    {
      "entropy": 6.371039295196534,
      "epoch": 0.12835631349782292,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999840498801191,
      "loss": 6.274,
      "mean_token_accuracy": 0.13513783514499664,
      "num_tokens": 2935040.0,
      "step": 1415
    },
    {
      "entropy": 6.21938738822937,
      "epoch": 0.1288098693759071,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999836622897126,
      "loss": 6.0552,
      "mean_token_accuracy": 0.13850467056035995,
      "num_tokens": 2945536.0,
      "step": 1420
    },
    {
      "entropy": 6.202126550674438,
      "epoch": 0.1292634252539913,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999832700465865,
      "loss": 5.9793,
      "mean_token_accuracy": 0.13881825804710388,
      "num_tokens": 2954969.0,
      "step": 1425
    },
    {
      "entropy": 6.261548900604248,
      "epoch": 0.12971698113207547,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999828731507491,
      "loss": 6.1059,
      "mean_token_accuracy": 0.13205914497375487,
      "num_tokens": 2966218.0,
      "step": 1430
    },
    {
      "entropy": 6.315013360977173,
      "epoch": 0.13017053701015965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999824716022085,
      "loss": 6.1289,
      "mean_token_accuracy": 0.12892259508371354,
      "num_tokens": 2976640.0,
      "step": 1435
    },
    {
      "entropy": 6.221342897415161,
      "epoch": 0.13062409288824384,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999820654009731,
      "loss": 6.0923,
      "mean_token_accuracy": 0.13740533366799354,
      "num_tokens": 2985883.0,
      "step": 1440
    },
    {
      "entropy": 6.278382968902588,
      "epoch": 0.13107764876632802,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999816545470513,
      "loss": 6.0839,
      "mean_token_accuracy": 0.12974392548203467,
      "num_tokens": 2996695.0,
      "step": 1445
    },
    {
      "entropy": 6.26499662399292,
      "epoch": 0.1315312046444122,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999812390404514,
      "loss": 6.1001,
      "mean_token_accuracy": 0.13423352390527726,
      "num_tokens": 3007500.0,
      "step": 1450
    },
    {
      "entropy": 6.136911249160766,
      "epoch": 0.13198476052249636,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999808188811823,
      "loss": 5.9549,
      "mean_token_accuracy": 0.13566315323114395,
      "num_tokens": 3017299.0,
      "step": 1455
    },
    {
      "entropy": 6.368114423751831,
      "epoch": 0.13243831640058054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999803940692524,
      "loss": 6.239,
      "mean_token_accuracy": 0.12817587703466415,
      "num_tokens": 3028389.0,
      "step": 1460
    },
    {
      "entropy": 6.223598575592041,
      "epoch": 0.13289187227866472,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999799646046707,
      "loss": 6.0568,
      "mean_token_accuracy": 0.13951559141278266,
      "num_tokens": 3038883.0,
      "step": 1465
    },
    {
      "entropy": 6.345055007934571,
      "epoch": 0.1333454281567489,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999795304874461,
      "loss": 6.1808,
      "mean_token_accuracy": 0.13515384122729301,
      "num_tokens": 3049904.0,
      "step": 1470
    },
    {
      "entropy": 6.184125995635986,
      "epoch": 0.1337989840348331,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999790917175873,
      "loss": 6.0163,
      "mean_token_accuracy": 0.13716932758688927,
      "num_tokens": 3059788.0,
      "step": 1475
    },
    {
      "entropy": 6.27112979888916,
      "epoch": 0.13425253991291727,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999786482951036,
      "loss": 6.0095,
      "mean_token_accuracy": 0.1407728001475334,
      "num_tokens": 3070413.0,
      "step": 1480
    },
    {
      "entropy": 6.198488378524781,
      "epoch": 0.13470609579100146,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999782002200041,
      "loss": 6.1247,
      "mean_token_accuracy": 0.13304170593619347,
      "num_tokens": 3081093.0,
      "step": 1485
    },
    {
      "entropy": 6.299404525756836,
      "epoch": 0.13515965166908564,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999777474922982,
      "loss": 6.0758,
      "mean_token_accuracy": 0.13359509408473969,
      "num_tokens": 3090594.0,
      "step": 1490
    },
    {
      "entropy": 6.272291612625122,
      "epoch": 0.13561320754716982,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999772901119951,
      "loss": 6.1602,
      "mean_token_accuracy": 0.1321714386343956,
      "num_tokens": 3101061.0,
      "step": 1495
    },
    {
      "entropy": 6.250188732147217,
      "epoch": 0.13606676342525398,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999768280791043,
      "loss": 6.0805,
      "mean_token_accuracy": 0.13871695548295976,
      "num_tokens": 3111715.0,
      "step": 1500
    },
    {
      "entropy": 6.187767601013183,
      "epoch": 0.13652031930333816,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999763613936352,
      "loss": 6.0033,
      "mean_token_accuracy": 0.14132409542798996,
      "num_tokens": 3121378.0,
      "step": 1505
    },
    {
      "entropy": 6.173904228210449,
      "epoch": 0.13697387518142234,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999758900555979,
      "loss": 5.9421,
      "mean_token_accuracy": 0.142991504073143,
      "num_tokens": 3131229.0,
      "step": 1510
    },
    {
      "entropy": 6.184138822555542,
      "epoch": 0.13742743105950653,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999754140650017,
      "loss": 6.033,
      "mean_token_accuracy": 0.1338609889149666,
      "num_tokens": 3140859.0,
      "step": 1515
    },
    {
      "entropy": 6.279017591476441,
      "epoch": 0.1378809869375907,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999749334218567,
      "loss": 6.0272,
      "mean_token_accuracy": 0.13679298311471938,
      "num_tokens": 3151692.0,
      "step": 1520
    },
    {
      "entropy": 6.150258111953735,
      "epoch": 0.1383345428156749,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999744481261726,
      "loss": 6.1035,
      "mean_token_accuracy": 0.13335202634334564,
      "num_tokens": 3161879.0,
      "step": 1525
    },
    {
      "entropy": 6.205261278152466,
      "epoch": 0.13878809869375908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999739581779597,
      "loss": 5.9614,
      "mean_token_accuracy": 0.1442497506737709,
      "num_tokens": 3171282.0,
      "step": 1530
    },
    {
      "entropy": 6.204788684844971,
      "epoch": 0.13924165457184326,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499973463577228,
      "loss": 6.0154,
      "mean_token_accuracy": 0.14128000661730766,
      "num_tokens": 3181398.0,
      "step": 1535
    },
    {
      "entropy": 6.210041856765747,
      "epoch": 0.13969521044992744,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999729643239877,
      "loss": 6.1509,
      "mean_token_accuracy": 0.1372975580394268,
      "num_tokens": 3191431.0,
      "step": 1540
    },
    {
      "entropy": 6.246035623550415,
      "epoch": 0.1401487663280116,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999724604182492,
      "loss": 6.1423,
      "mean_token_accuracy": 0.1305283486843109,
      "num_tokens": 3202147.0,
      "step": 1545
    },
    {
      "entropy": 6.317409181594849,
      "epoch": 0.14060232220609578,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999719518600229,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14262037947773934,
      "num_tokens": 3212428.0,
      "step": 1550
    },
    {
      "entropy": 6.198874998092651,
      "epoch": 0.14105587808417996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999714386493192,
      "loss": 6.0638,
      "mean_token_accuracy": 0.1392004132270813,
      "num_tokens": 3222099.0,
      "step": 1555
    },
    {
      "entropy": 6.23433141708374,
      "epoch": 0.14150943396226415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999709207861489,
      "loss": 6.0113,
      "mean_token_accuracy": 0.1394033081829548,
      "num_tokens": 3232079.0,
      "step": 1560
    },
    {
      "entropy": 6.08824315071106,
      "epoch": 0.14196298984034833,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999703982705227,
      "loss": 5.928,
      "mean_token_accuracy": 0.14834348112344742,
      "num_tokens": 3241762.0,
      "step": 1565
    },
    {
      "entropy": 6.164975643157959,
      "epoch": 0.1424165457184325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999698711024513,
      "loss": 6.0104,
      "mean_token_accuracy": 0.1406618483364582,
      "num_tokens": 3252571.0,
      "step": 1570
    },
    {
      "entropy": 6.212284135818481,
      "epoch": 0.1428701015965167,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999693392819456,
      "loss": 6.0603,
      "mean_token_accuracy": 0.13540706485509874,
      "num_tokens": 3263542.0,
      "step": 1575
    },
    {
      "entropy": 6.246869134902954,
      "epoch": 0.14332365747460088,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999688028090166,
      "loss": 6.1386,
      "mean_token_accuracy": 0.13631665483117103,
      "num_tokens": 3273295.0,
      "step": 1580
    },
    {
      "entropy": 6.328453207015992,
      "epoch": 0.14377721335268506,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999682616836755,
      "loss": 6.1427,
      "mean_token_accuracy": 0.13787241727113725,
      "num_tokens": 3283605.0,
      "step": 1585
    },
    {
      "entropy": 6.191803979873657,
      "epoch": 0.14423076923076922,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999677159059334,
      "loss": 5.9952,
      "mean_token_accuracy": 0.14498581141233444,
      "num_tokens": 3293837.0,
      "step": 1590
    },
    {
      "entropy": 6.072175979614258,
      "epoch": 0.1446843251088534,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999671654758016,
      "loss": 5.9983,
      "mean_token_accuracy": 0.14269243702292442,
      "num_tokens": 3304131.0,
      "step": 1595
    },
    {
      "entropy": 6.235024261474609,
      "epoch": 0.14513788098693758,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999666103932915,
      "loss": 6.0234,
      "mean_token_accuracy": 0.13487293496727942,
      "num_tokens": 3313543.0,
      "step": 1600
    },
    {
      "entropy": 6.185160875320435,
      "epoch": 0.14559143686502177,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999660506584145,
      "loss": 6.0201,
      "mean_token_accuracy": 0.13741163462400435,
      "num_tokens": 3324487.0,
      "step": 1605
    },
    {
      "entropy": 6.239445686340332,
      "epoch": 0.14604499274310595,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999654862711823,
      "loss": 6.0262,
      "mean_token_accuracy": 0.137547168135643,
      "num_tokens": 3335094.0,
      "step": 1610
    },
    {
      "entropy": 6.134017467498779,
      "epoch": 0.14649854862119013,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999649172316065,
      "loss": 5.9642,
      "mean_token_accuracy": 0.13992214426398278,
      "num_tokens": 3344252.0,
      "step": 1615
    },
    {
      "entropy": 6.197755432128906,
      "epoch": 0.14695210449927432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499964343539699,
      "loss": 6.0429,
      "mean_token_accuracy": 0.13646679371595383,
      "num_tokens": 3354265.0,
      "step": 1620
    },
    {
      "entropy": 6.2219359397888185,
      "epoch": 0.1474056603773585,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999637651954714,
      "loss": 6.125,
      "mean_token_accuracy": 0.1394253857433796,
      "num_tokens": 3364616.0,
      "step": 1625
    },
    {
      "entropy": 6.27594027519226,
      "epoch": 0.14785921625544268,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999631821989358,
      "loss": 6.0927,
      "mean_token_accuracy": 0.14138601943850518,
      "num_tokens": 3373650.0,
      "step": 1630
    },
    {
      "entropy": 6.101440382003784,
      "epoch": 0.14831277213352684,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999625945501043,
      "loss": 5.9886,
      "mean_token_accuracy": 0.1461173690855503,
      "num_tokens": 3384590.0,
      "step": 1635
    },
    {
      "entropy": 6.26271767616272,
      "epoch": 0.14876632801161102,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499962002248989,
      "loss": 6.0205,
      "mean_token_accuracy": 0.13982762396335602,
      "num_tokens": 3395616.0,
      "step": 1640
    },
    {
      "entropy": 6.153381443023681,
      "epoch": 0.1492198838896952,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999614052956024,
      "loss": 6.0239,
      "mean_token_accuracy": 0.13940254971385002,
      "num_tokens": 3405810.0,
      "step": 1645
    },
    {
      "entropy": 6.254888677597046,
      "epoch": 0.1496734397677794,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999608036899563,
      "loss": 6.0591,
      "mean_token_accuracy": 0.1363350972533226,
      "num_tokens": 3415932.0,
      "step": 1650
    },
    {
      "entropy": 6.151538419723511,
      "epoch": 0.15012699564586357,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999601974320636,
      "loss": 6.0116,
      "mean_token_accuracy": 0.14114982038736343,
      "num_tokens": 3425626.0,
      "step": 1655
    },
    {
      "entropy": 6.121203470230102,
      "epoch": 0.15058055152394775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999595865219367,
      "loss": 6.0208,
      "mean_token_accuracy": 0.13342469930648804,
      "num_tokens": 3434837.0,
      "step": 1660
    },
    {
      "entropy": 6.156308937072754,
      "epoch": 0.15103410740203194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999589709595881,
      "loss": 6.0332,
      "mean_token_accuracy": 0.14258277863264085,
      "num_tokens": 3445238.0,
      "step": 1665
    },
    {
      "entropy": 6.166830587387085,
      "epoch": 0.15148766328011612,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999583507450308,
      "loss": 6.0043,
      "mean_token_accuracy": 0.1305796168744564,
      "num_tokens": 3456738.0,
      "step": 1670
    },
    {
      "entropy": 6.070332002639771,
      "epoch": 0.1519412191582003,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999577258782774,
      "loss": 5.8838,
      "mean_token_accuracy": 0.13995402976870536,
      "num_tokens": 3466296.0,
      "step": 1675
    },
    {
      "entropy": 6.266418838500977,
      "epoch": 0.15239477503628446,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499957096359341,
      "loss": 6.087,
      "mean_token_accuracy": 0.13597871735692024,
      "num_tokens": 3476989.0,
      "step": 1680
    },
    {
      "entropy": 6.111058521270752,
      "epoch": 0.15284833091436864,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999564621882343,
      "loss": 5.9321,
      "mean_token_accuracy": 0.141887167096138,
      "num_tokens": 3487050.0,
      "step": 1685
    },
    {
      "entropy": 6.225695085525513,
      "epoch": 0.15330188679245282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999558233649708,
      "loss": 6.139,
      "mean_token_accuracy": 0.1312915213406086,
      "num_tokens": 3497777.0,
      "step": 1690
    },
    {
      "entropy": 6.17779574394226,
      "epoch": 0.153755442670537,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999551798895636,
      "loss": 6.0475,
      "mean_token_accuracy": 0.13619913384318352,
      "num_tokens": 3509066.0,
      "step": 1695
    },
    {
      "entropy": 6.141854095458984,
      "epoch": 0.1542089985486212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999545317620258,
      "loss": 5.9776,
      "mean_token_accuracy": 0.1392234146595001,
      "num_tokens": 3519567.0,
      "step": 1700
    },
    {
      "entropy": 6.124539947509765,
      "epoch": 0.15466255442670537,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999538789823711,
      "loss": 6.0076,
      "mean_token_accuracy": 0.1419394336640835,
      "num_tokens": 3529402.0,
      "step": 1705
    },
    {
      "entropy": 6.27455096244812,
      "epoch": 0.15511611030478956,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999532215506127,
      "loss": 6.0228,
      "mean_token_accuracy": 0.14713949486613273,
      "num_tokens": 3539805.0,
      "step": 1710
    },
    {
      "entropy": 6.072864151000976,
      "epoch": 0.15556966618287374,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999525594667645,
      "loss": 5.881,
      "mean_token_accuracy": 0.14898346215486527,
      "num_tokens": 3550477.0,
      "step": 1715
    },
    {
      "entropy": 6.0882893085479735,
      "epoch": 0.15602322206095792,
      "grad_norm": 1.0,
      "learning_rate": 0.00049995189273084,
      "loss": 5.9843,
      "mean_token_accuracy": 0.1400413043797016,
      "num_tokens": 3560729.0,
      "step": 1720
    },
    {
      "entropy": 6.1748028755187985,
      "epoch": 0.15647677793904208,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499951221342853,
      "loss": 6.0394,
      "mean_token_accuracy": 0.1325826108455658,
      "num_tokens": 3571538.0,
      "step": 1725
    },
    {
      "entropy": 6.174065685272216,
      "epoch": 0.15693033381712626,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999505453028174,
      "loss": 6.0644,
      "mean_token_accuracy": 0.13689751103520392,
      "num_tokens": 3583128.0,
      "step": 1730
    },
    {
      "entropy": 6.2016395092010494,
      "epoch": 0.15738388969521044,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999498646107472,
      "loss": 5.9871,
      "mean_token_accuracy": 0.1429124280810356,
      "num_tokens": 3594358.0,
      "step": 1735
    },
    {
      "entropy": 6.129647302627563,
      "epoch": 0.15783744557329463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999491792666566,
      "loss": 5.9292,
      "mean_token_accuracy": 0.14557284563779832,
      "num_tokens": 3604345.0,
      "step": 1740
    },
    {
      "entropy": 6.089707612991333,
      "epoch": 0.1582910014513788,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999484892705597,
      "loss": 5.9717,
      "mean_token_accuracy": 0.1409473091363907,
      "num_tokens": 3616129.0,
      "step": 1745
    },
    {
      "entropy": 6.162917709350586,
      "epoch": 0.158744557329463,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999477946224705,
      "loss": 6.0177,
      "mean_token_accuracy": 0.14049824178218842,
      "num_tokens": 3627458.0,
      "step": 1750
    },
    {
      "entropy": 6.149751949310303,
      "epoch": 0.15919811320754718,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999470953224037,
      "loss": 5.9031,
      "mean_token_accuracy": 0.1436824843287468,
      "num_tokens": 3636047.0,
      "step": 1755
    },
    {
      "entropy": 6.0508606910705565,
      "epoch": 0.15965166908563136,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999463913703735,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1433118313550949,
      "num_tokens": 3646930.0,
      "step": 1760
    },
    {
      "entropy": 6.124081325531006,
      "epoch": 0.16010522496371554,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999456827663948,
      "loss": 5.9919,
      "mean_token_accuracy": 0.1459735155105591,
      "num_tokens": 3657279.0,
      "step": 1765
    },
    {
      "entropy": 6.052764081954956,
      "epoch": 0.1605587808417997,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499944969510482,
      "loss": 5.9001,
      "mean_token_accuracy": 0.13991431295871734,
      "num_tokens": 3667070.0,
      "step": 1770
    },
    {
      "entropy": 6.181622266769409,
      "epoch": 0.16101233671988388,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999442516026498,
      "loss": 5.9141,
      "mean_token_accuracy": 0.14285601824522018,
      "num_tokens": 3677560.0,
      "step": 1775
    },
    {
      "entropy": 6.027710103988648,
      "epoch": 0.16146589259796806,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999435290429133,
      "loss": 5.8862,
      "mean_token_accuracy": 0.14947912245988845,
      "num_tokens": 3686701.0,
      "step": 1780
    },
    {
      "entropy": 6.117189884185791,
      "epoch": 0.16191944847605225,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999428018312872,
      "loss": 5.9719,
      "mean_token_accuracy": 0.14543041959404945,
      "num_tokens": 3696338.0,
      "step": 1785
    },
    {
      "entropy": 6.091988801956177,
      "epoch": 0.16237300435413643,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999420699677867,
      "loss": 5.9492,
      "mean_token_accuracy": 0.1432087704539299,
      "num_tokens": 3705752.0,
      "step": 1790
    },
    {
      "entropy": 6.203689908981323,
      "epoch": 0.1628265602322206,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999413334524269,
      "loss": 6.0384,
      "mean_token_accuracy": 0.13643145337700843,
      "num_tokens": 3716225.0,
      "step": 1795
    },
    {
      "entropy": 6.111277675628662,
      "epoch": 0.1632801161103048,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999405922852229,
      "loss": 5.9098,
      "mean_token_accuracy": 0.14482017382979392,
      "num_tokens": 3726593.0,
      "step": 1800
    },
    {
      "entropy": 6.028970766067505,
      "epoch": 0.16373367198838898,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999398464661901,
      "loss": 5.9708,
      "mean_token_accuracy": 0.13964075967669487,
      "num_tokens": 3737563.0,
      "step": 1805
    },
    {
      "entropy": 6.246074914932251,
      "epoch": 0.16418722786647316,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999390959953441,
      "loss": 6.0176,
      "mean_token_accuracy": 0.14332932382822036,
      "num_tokens": 3747752.0,
      "step": 1810
    },
    {
      "entropy": 6.053060626983642,
      "epoch": 0.16464078374455732,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999383408727002,
      "loss": 5.9539,
      "mean_token_accuracy": 0.14173785969614983,
      "num_tokens": 3758795.0,
      "step": 1815
    },
    {
      "entropy": 6.167564058303833,
      "epoch": 0.1650943396226415,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999375810982741,
      "loss": 5.9315,
      "mean_token_accuracy": 0.14446532130241393,
      "num_tokens": 3768690.0,
      "step": 1820
    },
    {
      "entropy": 6.0776159286499025,
      "epoch": 0.16554789550072568,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999368166720815,
      "loss": 6.014,
      "mean_token_accuracy": 0.14304881766438485,
      "num_tokens": 3779018.0,
      "step": 1825
    },
    {
      "entropy": 6.147589683532715,
      "epoch": 0.16600145137880987,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999360475941382,
      "loss": 5.9118,
      "mean_token_accuracy": 0.14721195623278618,
      "num_tokens": 3788985.0,
      "step": 1830
    },
    {
      "entropy": 6.131836366653443,
      "epoch": 0.16645500725689405,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049993527386446,
      "loss": 6.0394,
      "mean_token_accuracy": 0.1322140447795391,
      "num_tokens": 3800787.0,
      "step": 1835
    },
    {
      "entropy": 6.089927768707275,
      "epoch": 0.16690856313497823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999344954830633,
      "loss": 5.9208,
      "mean_token_accuracy": 0.1421339437365532,
      "num_tokens": 3811381.0,
      "step": 1840
    },
    {
      "entropy": 6.0987998962402346,
      "epoch": 0.16736211901306242,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999337124499638,
      "loss": 6.0339,
      "mean_token_accuracy": 0.14644376039505005,
      "num_tokens": 3822055.0,
      "step": 1845
    },
    {
      "entropy": 6.147271633148193,
      "epoch": 0.1678156748911466,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999329247651777,
      "loss": 5.9932,
      "mean_token_accuracy": 0.1378726117312908,
      "num_tokens": 3831900.0,
      "step": 1850
    },
    {
      "entropy": 6.1165355205535885,
      "epoch": 0.16826923076923078,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999321324287213,
      "loss": 5.9383,
      "mean_token_accuracy": 0.14189887940883636,
      "num_tokens": 3842702.0,
      "step": 1855
    },
    {
      "entropy": 6.1751471042633055,
      "epoch": 0.16872278664731494,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999313354406114,
      "loss": 5.9187,
      "mean_token_accuracy": 0.1416791319847107,
      "num_tokens": 3852690.0,
      "step": 1860
    },
    {
      "entropy": 5.9985252857208256,
      "epoch": 0.16917634252539912,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499930533800864,
      "loss": 5.9101,
      "mean_token_accuracy": 0.1507345899939537,
      "num_tokens": 3863040.0,
      "step": 1865
    },
    {
      "entropy": 6.10200867652893,
      "epoch": 0.1696298984034833,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999297275094958,
      "loss": 6.0044,
      "mean_token_accuracy": 0.14036339968442918,
      "num_tokens": 3873613.0,
      "step": 1870
    },
    {
      "entropy": 5.999226474761963,
      "epoch": 0.1700834542815675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999289165665235,
      "loss": 5.8075,
      "mean_token_accuracy": 0.14994008243083953,
      "num_tokens": 3883936.0,
      "step": 1875
    },
    {
      "entropy": 5.995340251922608,
      "epoch": 0.17053701015965167,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999281009719639,
      "loss": 5.7634,
      "mean_token_accuracy": 0.15587486177682877,
      "num_tokens": 3893788.0,
      "step": 1880
    },
    {
      "entropy": 6.001563358306885,
      "epoch": 0.17099056603773585,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999272807258339,
      "loss": 5.8376,
      "mean_token_accuracy": 0.15261975079774856,
      "num_tokens": 3903434.0,
      "step": 1885
    },
    {
      "entropy": 6.05434021949768,
      "epoch": 0.17144412191582004,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999264558281504,
      "loss": 5.8409,
      "mean_token_accuracy": 0.1521665245294571,
      "num_tokens": 3913476.0,
      "step": 1890
    },
    {
      "entropy": 5.971798706054687,
      "epoch": 0.17189767779390422,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999256262789305,
      "loss": 5.8007,
      "mean_token_accuracy": 0.151979461312294,
      "num_tokens": 3923301.0,
      "step": 1895
    },
    {
      "entropy": 6.028662157058716,
      "epoch": 0.1723512336719884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999247920781913,
      "loss": 5.9603,
      "mean_token_accuracy": 0.14160882011055947,
      "num_tokens": 3933335.0,
      "step": 1900
    },
    {
      "entropy": 6.160859155654907,
      "epoch": 0.17280478955007256,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999239532259502,
      "loss": 5.978,
      "mean_token_accuracy": 0.1399615041911602,
      "num_tokens": 3944171.0,
      "step": 1905
    },
    {
      "entropy": 6.11788387298584,
      "epoch": 0.17325834542815674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999231097222244,
      "loss": 6.056,
      "mean_token_accuracy": 0.14640697762370108,
      "num_tokens": 3954160.0,
      "step": 1910
    },
    {
      "entropy": 6.125342941284179,
      "epoch": 0.17371190130624092,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999222615670312,
      "loss": 5.8442,
      "mean_token_accuracy": 0.15440410375595093,
      "num_tokens": 3964110.0,
      "step": 1915
    },
    {
      "entropy": 6.047020721435547,
      "epoch": 0.1741654571843251,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999214087603885,
      "loss": 5.9347,
      "mean_token_accuracy": 0.14503437876701356,
      "num_tokens": 3974293.0,
      "step": 1920
    },
    {
      "entropy": 6.03211555480957,
      "epoch": 0.1746190130624093,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999205513023135,
      "loss": 5.8896,
      "mean_token_accuracy": 0.14775249734520912,
      "num_tokens": 3984346.0,
      "step": 1925
    },
    {
      "entropy": 5.992477941513061,
      "epoch": 0.17507256894049347,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999196891928245,
      "loss": 5.8312,
      "mean_token_accuracy": 0.15544059574604036,
      "num_tokens": 3992897.0,
      "step": 1930
    },
    {
      "entropy": 6.042530679702759,
      "epoch": 0.17552612481857766,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999188224319388,
      "loss": 5.8655,
      "mean_token_accuracy": 0.15182426571846008,
      "num_tokens": 4002575.0,
      "step": 1935
    },
    {
      "entropy": 6.01702356338501,
      "epoch": 0.17597968069666184,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999179510196746,
      "loss": 5.9547,
      "mean_token_accuracy": 0.1407930850982666,
      "num_tokens": 4013752.0,
      "step": 1940
    },
    {
      "entropy": 6.041807699203491,
      "epoch": 0.17643323657474602,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999170749560498,
      "loss": 5.833,
      "mean_token_accuracy": 0.1515166938304901,
      "num_tokens": 4023718.0,
      "step": 1945
    },
    {
      "entropy": 6.038792991638184,
      "epoch": 0.17688679245283018,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999161942410827,
      "loss": 5.9323,
      "mean_token_accuracy": 0.13799159675836564,
      "num_tokens": 4034210.0,
      "step": 1950
    },
    {
      "entropy": 6.134712362289429,
      "epoch": 0.17734034833091436,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999153088747913,
      "loss": 5.9265,
      "mean_token_accuracy": 0.14558120667934418,
      "num_tokens": 4044868.0,
      "step": 1955
    },
    {
      "entropy": 5.951734399795532,
      "epoch": 0.17779390420899854,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499914418857194,
      "loss": 5.9128,
      "mean_token_accuracy": 0.14710975885391236,
      "num_tokens": 4055049.0,
      "step": 1960
    },
    {
      "entropy": 6.109003973007202,
      "epoch": 0.17824746008708273,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999135241883092,
      "loss": 5.9693,
      "mean_token_accuracy": 0.14337438270449637,
      "num_tokens": 4065889.0,
      "step": 1965
    },
    {
      "entropy": 6.076221990585327,
      "epoch": 0.1787010159651669,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999126248681555,
      "loss": 5.9986,
      "mean_token_accuracy": 0.14153247475624084,
      "num_tokens": 4077288.0,
      "step": 1970
    },
    {
      "entropy": 6.059194612503052,
      "epoch": 0.1791545718432511,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999117208967513,
      "loss": 5.9359,
      "mean_token_accuracy": 0.14241613298654557,
      "num_tokens": 4087402.0,
      "step": 1975
    },
    {
      "entropy": 6.074989175796508,
      "epoch": 0.17960812772133528,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999108122741154,
      "loss": 5.9911,
      "mean_token_accuracy": 0.1394512176513672,
      "num_tokens": 4098150.0,
      "step": 1980
    },
    {
      "entropy": 5.997276258468628,
      "epoch": 0.18006168359941946,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999098990002666,
      "loss": 5.8489,
      "mean_token_accuracy": 0.14526972770690919,
      "num_tokens": 4108134.0,
      "step": 1985
    },
    {
      "entropy": 6.0917863845825195,
      "epoch": 0.18051523947750364,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999089810752237,
      "loss": 5.988,
      "mean_token_accuracy": 0.1419826477766037,
      "num_tokens": 4118251.0,
      "step": 1990
    },
    {
      "entropy": 6.037380886077881,
      "epoch": 0.1809687953555878,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499908058499006,
      "loss": 5.8256,
      "mean_token_accuracy": 0.1488537833094597,
      "num_tokens": 4129228.0,
      "step": 1995
    },
    {
      "entropy": 5.989283561706543,
      "epoch": 0.18142235123367198,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999071312716321,
      "loss": 5.8905,
      "mean_token_accuracy": 0.15052457451820372,
      "num_tokens": 4139391.0,
      "step": 2000
    },
    {
      "entropy": 6.038852834701538,
      "epoch": 0.18187590711175616,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999061993931215,
      "loss": 5.8198,
      "mean_token_accuracy": 0.15358539968729018,
      "num_tokens": 4148891.0,
      "step": 2005
    },
    {
      "entropy": 6.235394716262817,
      "epoch": 0.18232946298984035,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999052628634934,
      "loss": 6.1728,
      "mean_token_accuracy": 0.14022321105003357,
      "num_tokens": 4159551.0,
      "step": 2010
    },
    {
      "entropy": 6.079950952529908,
      "epoch": 0.18278301886792453,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999043216827671,
      "loss": 5.9696,
      "mean_token_accuracy": 0.14237845912575722,
      "num_tokens": 4170055.0,
      "step": 2015
    },
    {
      "entropy": 5.993285369873047,
      "epoch": 0.1832365747460087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999033758509621,
      "loss": 5.8201,
      "mean_token_accuracy": 0.14752258732914925,
      "num_tokens": 4180767.0,
      "step": 2020
    },
    {
      "entropy": 6.0369898796081545,
      "epoch": 0.1836901306240929,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499902425368098,
      "loss": 5.9485,
      "mean_token_accuracy": 0.14603547155857086,
      "num_tokens": 4190654.0,
      "step": 2025
    },
    {
      "entropy": 6.0134642124176025,
      "epoch": 0.18414368650217708,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999014702341944,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1434512048959732,
      "num_tokens": 4200753.0,
      "step": 2030
    },
    {
      "entropy": 6.070598554611206,
      "epoch": 0.18459724238026126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999005104492712,
      "loss": 5.9491,
      "mean_token_accuracy": 0.14099996089935302,
      "num_tokens": 4210994.0,
      "step": 2035
    },
    {
      "entropy": 6.0680242538452145,
      "epoch": 0.18505079825834542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998995460133481,
      "loss": 5.8913,
      "mean_token_accuracy": 0.14143144562840462,
      "num_tokens": 4220883.0,
      "step": 2040
    },
    {
      "entropy": 6.077210140228272,
      "epoch": 0.1855043541364296,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998985769264451,
      "loss": 5.9245,
      "mean_token_accuracy": 0.13767858743667602,
      "num_tokens": 4231269.0,
      "step": 2045
    },
    {
      "entropy": 5.974451160430908,
      "epoch": 0.18595791001451378,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998976031885821,
      "loss": 5.93,
      "mean_token_accuracy": 0.1419460207223892,
      "num_tokens": 4241559.0,
      "step": 2050
    },
    {
      "entropy": 6.0086987018585205,
      "epoch": 0.18641146589259797,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998966247997794,
      "loss": 5.8188,
      "mean_token_accuracy": 0.14645475745201111,
      "num_tokens": 4250870.0,
      "step": 2055
    },
    {
      "entropy": 5.967829656600952,
      "epoch": 0.18686502177068215,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998956417600572,
      "loss": 5.7746,
      "mean_token_accuracy": 0.15297217816114425,
      "num_tokens": 4260848.0,
      "step": 2060
    },
    {
      "entropy": 5.986316537857055,
      "epoch": 0.18731857764876633,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998946540694358,
      "loss": 5.9036,
      "mean_token_accuracy": 0.15208268463611602,
      "num_tokens": 4270973.0,
      "step": 2065
    },
    {
      "entropy": 6.132760667800904,
      "epoch": 0.18777213352685052,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998936617279357,
      "loss": 5.9802,
      "mean_token_accuracy": 0.14554873257875442,
      "num_tokens": 4281211.0,
      "step": 2070
    },
    {
      "entropy": 5.941362285614014,
      "epoch": 0.1882256894049347,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998926647355773,
      "loss": 5.8306,
      "mean_token_accuracy": 0.150546196103096,
      "num_tokens": 4291425.0,
      "step": 2075
    },
    {
      "entropy": 6.098439311981201,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998916630923813,
      "loss": 5.9406,
      "mean_token_accuracy": 0.14438463374972343,
      "num_tokens": 4301079.0,
      "step": 2080
    },
    {
      "entropy": 6.045752716064453,
      "epoch": 0.18913280116110304,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998906567983685,
      "loss": 5.9215,
      "mean_token_accuracy": 0.1422693431377411,
      "num_tokens": 4312374.0,
      "step": 2085
    },
    {
      "entropy": 6.030663347244262,
      "epoch": 0.18958635703918722,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998896458535595,
      "loss": 5.8334,
      "mean_token_accuracy": 0.15475437194108962,
      "num_tokens": 4322810.0,
      "step": 2090
    },
    {
      "entropy": 5.984178876876831,
      "epoch": 0.1900399129172714,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998886302579752,
      "loss": 5.8383,
      "mean_token_accuracy": 0.14914876371622085,
      "num_tokens": 4332082.0,
      "step": 2095
    },
    {
      "entropy": 6.012246084213257,
      "epoch": 0.19049346879535559,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499887610011637,
      "loss": 5.9249,
      "mean_token_accuracy": 0.1479165866971016,
      "num_tokens": 4342632.0,
      "step": 2100
    },
    {
      "entropy": 5.986341905593872,
      "epoch": 0.19094702467343977,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998865851145656,
      "loss": 5.8567,
      "mean_token_accuracy": 0.14595670849084855,
      "num_tokens": 4353544.0,
      "step": 2105
    },
    {
      "entropy": 5.9498518943786625,
      "epoch": 0.19140058055152395,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998855555667822,
      "loss": 5.981,
      "mean_token_accuracy": 0.1430116519331932,
      "num_tokens": 4363636.0,
      "step": 2110
    },
    {
      "entropy": 6.076320600509644,
      "epoch": 0.19185413642960814,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998845213683084,
      "loss": 5.831,
      "mean_token_accuracy": 0.1475878320634365,
      "num_tokens": 4373672.0,
      "step": 2115
    },
    {
      "entropy": 5.910988473892212,
      "epoch": 0.19230769230769232,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998834825191652,
      "loss": 5.8096,
      "mean_token_accuracy": 0.14788725823163987,
      "num_tokens": 4383199.0,
      "step": 2120
    },
    {
      "entropy": 6.013854932785034,
      "epoch": 0.1927612481857765,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998824390193743,
      "loss": 5.886,
      "mean_token_accuracy": 0.1449464038014412,
      "num_tokens": 4394446.0,
      "step": 2125
    },
    {
      "entropy": 6.0993109226226805,
      "epoch": 0.19321480406386066,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998813908689573,
      "loss": 5.9237,
      "mean_token_accuracy": 0.14512224718928338,
      "num_tokens": 4405882.0,
      "step": 2130
    },
    {
      "entropy": 5.93023910522461,
      "epoch": 0.19366835994194484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998803380679359,
      "loss": 5.7867,
      "mean_token_accuracy": 0.1542275607585907,
      "num_tokens": 4415280.0,
      "step": 2135
    },
    {
      "entropy": 5.99576849937439,
      "epoch": 0.19412191582002902,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998792806163317,
      "loss": 5.9946,
      "mean_token_accuracy": 0.1401658058166504,
      "num_tokens": 4425755.0,
      "step": 2140
    },
    {
      "entropy": 5.925059509277344,
      "epoch": 0.1945754716981132,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998782185141667,
      "loss": 5.8523,
      "mean_token_accuracy": 0.14792224615812302,
      "num_tokens": 4436259.0,
      "step": 2145
    },
    {
      "entropy": 6.045332574844361,
      "epoch": 0.1950290275761974,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998771517614629,
      "loss": 5.881,
      "mean_token_accuracy": 0.14263954162597656,
      "num_tokens": 4446423.0,
      "step": 2150
    },
    {
      "entropy": 6.00939302444458,
      "epoch": 0.19548258345428157,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998760803582422,
      "loss": 5.8288,
      "mean_token_accuracy": 0.15420894026756288,
      "num_tokens": 4456678.0,
      "step": 2155
    },
    {
      "entropy": 5.945328330993652,
      "epoch": 0.19593613933236576,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998750043045269,
      "loss": 5.8269,
      "mean_token_accuracy": 0.15192546397447587,
      "num_tokens": 4467031.0,
      "step": 2160
    },
    {
      "entropy": 6.024298143386841,
      "epoch": 0.19638969521044994,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998739236003392,
      "loss": 5.9066,
      "mean_token_accuracy": 0.14100065529346467,
      "num_tokens": 4477771.0,
      "step": 2165
    },
    {
      "entropy": 6.0035483837127686,
      "epoch": 0.1968432510885341,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998728382457014,
      "loss": 5.9222,
      "mean_token_accuracy": 0.1434047520160675,
      "num_tokens": 4488426.0,
      "step": 2170
    },
    {
      "entropy": 6.103923177719116,
      "epoch": 0.19729680696661828,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998717482406362,
      "loss": 5.9056,
      "mean_token_accuracy": 0.14213445484638215,
      "num_tokens": 4499735.0,
      "step": 2175
    },
    {
      "entropy": 6.009974479675293,
      "epoch": 0.19775036284470246,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998706535851657,
      "loss": 5.8373,
      "mean_token_accuracy": 0.14943975061178208,
      "num_tokens": 4509500.0,
      "step": 2180
    },
    {
      "entropy": 5.964470767974854,
      "epoch": 0.19820391872278664,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998695542793129,
      "loss": 5.8475,
      "mean_token_accuracy": 0.14806249961256981,
      "num_tokens": 4520431.0,
      "step": 2185
    },
    {
      "entropy": 6.013790941238403,
      "epoch": 0.19865747460087083,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998684503231003,
      "loss": 5.9367,
      "mean_token_accuracy": 0.1468561165034771,
      "num_tokens": 4532294.0,
      "step": 2190
    },
    {
      "entropy": 5.979345083236694,
      "epoch": 0.199111030478955,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499867341716551,
      "loss": 5.8657,
      "mean_token_accuracy": 0.1456448383629322,
      "num_tokens": 4542488.0,
      "step": 2195
    },
    {
      "entropy": 5.9495869159698485,
      "epoch": 0.1995645863570392,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998662284596878,
      "loss": 5.8013,
      "mean_token_accuracy": 0.14156620278954507,
      "num_tokens": 4552933.0,
      "step": 2200
    },
    {
      "entropy": 6.036973190307617,
      "epoch": 0.20001814223512338,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998651105525335,
      "loss": 5.7937,
      "mean_token_accuracy": 0.15192690938711167,
      "num_tokens": 4562373.0,
      "step": 2205
    },
    {
      "entropy": 5.927747917175293,
      "epoch": 0.20047169811320756,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998639879951115,
      "loss": 5.9131,
      "mean_token_accuracy": 0.14802779853343964,
      "num_tokens": 4572715.0,
      "step": 2210
    },
    {
      "entropy": 6.064771509170532,
      "epoch": 0.2009252539912917,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499862860787445,
      "loss": 5.7918,
      "mean_token_accuracy": 0.14533086642622947,
      "num_tokens": 4582782.0,
      "step": 2215
    },
    {
      "entropy": 5.980919456481933,
      "epoch": 0.2013788098693759,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998617289295571,
      "loss": 5.8965,
      "mean_token_accuracy": 0.13982278481125832,
      "num_tokens": 4594108.0,
      "step": 2220
    },
    {
      "entropy": 6.0538468837738035,
      "epoch": 0.20183236574746008,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998605924214714,
      "loss": 5.8628,
      "mean_token_accuracy": 0.1460525505244732,
      "num_tokens": 4604732.0,
      "step": 2225
    },
    {
      "entropy": 6.000491809844971,
      "epoch": 0.20228592162554426,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998594512632114,
      "loss": 5.9955,
      "mean_token_accuracy": 0.13708511143922805,
      "num_tokens": 4614998.0,
      "step": 2230
    },
    {
      "entropy": 5.970520877838135,
      "epoch": 0.20273947750362845,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998583054548006,
      "loss": 5.8136,
      "mean_token_accuracy": 0.1484377056360245,
      "num_tokens": 4624665.0,
      "step": 2235
    },
    {
      "entropy": 6.00954236984253,
      "epoch": 0.20319303338171263,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998571549962628,
      "loss": 5.8101,
      "mean_token_accuracy": 0.15187345445156097,
      "num_tokens": 4633731.0,
      "step": 2240
    },
    {
      "entropy": 5.991653680801392,
      "epoch": 0.2036465892597968,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998559998876216,
      "loss": 5.8384,
      "mean_token_accuracy": 0.14615286365151406,
      "num_tokens": 4643421.0,
      "step": 2245
    },
    {
      "entropy": 5.987816905975341,
      "epoch": 0.204100145137881,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998548401289012,
      "loss": 5.9151,
      "mean_token_accuracy": 0.14586994647979737,
      "num_tokens": 4653795.0,
      "step": 2250
    },
    {
      "entropy": 6.008340549468994,
      "epoch": 0.20455370101596518,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998536757201253,
      "loss": 5.8858,
      "mean_token_accuracy": 0.1444897711277008,
      "num_tokens": 4664840.0,
      "step": 2255
    },
    {
      "entropy": 6.039184617996216,
      "epoch": 0.20500725689404933,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998525066613181,
      "loss": 5.8979,
      "mean_token_accuracy": 0.15061988234519957,
      "num_tokens": 4675508.0,
      "step": 2260
    },
    {
      "entropy": 5.932389974594116,
      "epoch": 0.20546081277213352,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998513329525038,
      "loss": 5.8199,
      "mean_token_accuracy": 0.15319341719150542,
      "num_tokens": 4685995.0,
      "step": 2265
    },
    {
      "entropy": 6.023037481307983,
      "epoch": 0.2059143686502177,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998501545937066,
      "loss": 5.8348,
      "mean_token_accuracy": 0.14633184522390366,
      "num_tokens": 4696659.0,
      "step": 2270
    },
    {
      "entropy": 5.960624837875367,
      "epoch": 0.20636792452830188,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499848971584951,
      "loss": 5.8072,
      "mean_token_accuracy": 0.15713879466056824,
      "num_tokens": 4706438.0,
      "step": 2275
    },
    {
      "entropy": 5.881855773925781,
      "epoch": 0.20682148040638607,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998477839262611,
      "loss": 5.8119,
      "mean_token_accuracy": 0.15000793784856797,
      "num_tokens": 4716402.0,
      "step": 2280
    },
    {
      "entropy": 5.990855503082275,
      "epoch": 0.20727503628447025,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499846591617662,
      "loss": 5.8976,
      "mean_token_accuracy": 0.14913058578968047,
      "num_tokens": 4728488.0,
      "step": 2285
    },
    {
      "entropy": 5.952394151687622,
      "epoch": 0.20772859216255443,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998453946591779,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1516925185918808,
      "num_tokens": 4738693.0,
      "step": 2290
    },
    {
      "entropy": 5.991802835464478,
      "epoch": 0.20818214804063861,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998441930508339,
      "loss": 5.8145,
      "mean_token_accuracy": 0.15066843181848527,
      "num_tokens": 4748895.0,
      "step": 2295
    },
    {
      "entropy": 6.047665977478028,
      "epoch": 0.2086357039187228,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998429867926547,
      "loss": 5.925,
      "mean_token_accuracy": 0.14194059297442435,
      "num_tokens": 4760135.0,
      "step": 2300
    },
    {
      "entropy": 6.035406303405762,
      "epoch": 0.20908925979680695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998417758846651,
      "loss": 5.9523,
      "mean_token_accuracy": 0.15069413632154466,
      "num_tokens": 4769935.0,
      "step": 2305
    },
    {
      "entropy": 6.0572274684906,
      "epoch": 0.20954281567489114,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998405603268903,
      "loss": 5.8976,
      "mean_token_accuracy": 0.14732977300882338,
      "num_tokens": 4779520.0,
      "step": 2310
    },
    {
      "entropy": 5.93422679901123,
      "epoch": 0.20999637155297532,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998393401193554,
      "loss": 5.8067,
      "mean_token_accuracy": 0.1515841841697693,
      "num_tokens": 4791256.0,
      "step": 2315
    },
    {
      "entropy": 5.881869649887085,
      "epoch": 0.2104499274310595,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998381152620857,
      "loss": 5.7539,
      "mean_token_accuracy": 0.15506275743246078,
      "num_tokens": 4801334.0,
      "step": 2320
    },
    {
      "entropy": 5.885318040847778,
      "epoch": 0.21090348330914369,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998368857551067,
      "loss": 5.7321,
      "mean_token_accuracy": 0.1555730938911438,
      "num_tokens": 4811361.0,
      "step": 2325
    },
    {
      "entropy": 5.969886207580567,
      "epoch": 0.21135703918722787,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998356515984433,
      "loss": 5.807,
      "mean_token_accuracy": 0.15263233184814454,
      "num_tokens": 4821190.0,
      "step": 2330
    },
    {
      "entropy": 5.9508056640625,
      "epoch": 0.21181059506531205,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998344127921213,
      "loss": 5.8241,
      "mean_token_accuracy": 0.1541541337966919,
      "num_tokens": 4830694.0,
      "step": 2335
    },
    {
      "entropy": 5.928486108779907,
      "epoch": 0.21226415094339623,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998331693361666,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1508532479405403,
      "num_tokens": 4841344.0,
      "step": 2340
    },
    {
      "entropy": 5.951602220535278,
      "epoch": 0.21271770682148042,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998319212306046,
      "loss": 5.8021,
      "mean_token_accuracy": 0.14895136952400206,
      "num_tokens": 4851161.0,
      "step": 2345
    },
    {
      "entropy": 5.9326536655426025,
      "epoch": 0.21317126269956457,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499830668475461,
      "loss": 5.7841,
      "mean_token_accuracy": 0.14950372502207757,
      "num_tokens": 4860740.0,
      "step": 2350
    },
    {
      "entropy": 5.928380012512207,
      "epoch": 0.21362481857764876,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499829411070762,
      "loss": 5.8168,
      "mean_token_accuracy": 0.15025952607393264,
      "num_tokens": 4871733.0,
      "step": 2355
    },
    {
      "entropy": 5.940114498138428,
      "epoch": 0.21407837445573294,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998281490165335,
      "loss": 5.8779,
      "mean_token_accuracy": 0.14290931671857834,
      "num_tokens": 4883617.0,
      "step": 2360
    },
    {
      "entropy": 6.0039795398712155,
      "epoch": 0.21453193033381712,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998268823128016,
      "loss": 5.8133,
      "mean_token_accuracy": 0.14657811671495438,
      "num_tokens": 4894519.0,
      "step": 2365
    },
    {
      "entropy": 5.913796472549438,
      "epoch": 0.2149854862119013,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998256109595923,
      "loss": 5.7663,
      "mean_token_accuracy": 0.15133122503757476,
      "num_tokens": 4904965.0,
      "step": 2370
    },
    {
      "entropy": 5.905171871185303,
      "epoch": 0.2154390420899855,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998243349569323,
      "loss": 5.6806,
      "mean_token_accuracy": 0.16264960318803787,
      "num_tokens": 4914618.0,
      "step": 2375
    },
    {
      "entropy": 5.793045377731323,
      "epoch": 0.21589259796806967,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998230543048477,
      "loss": 5.7376,
      "mean_token_accuracy": 0.15822856724262238,
      "num_tokens": 4924341.0,
      "step": 2380
    },
    {
      "entropy": 5.860754489898682,
      "epoch": 0.21634615384615385,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499821769003365,
      "loss": 5.7202,
      "mean_token_accuracy": 0.15276679992675782,
      "num_tokens": 4934231.0,
      "step": 2385
    },
    {
      "entropy": 5.90138463973999,
      "epoch": 0.21679970972423804,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998204790525109,
      "loss": 5.7442,
      "mean_token_accuracy": 0.15233908742666244,
      "num_tokens": 4944303.0,
      "step": 2390
    },
    {
      "entropy": 5.892811250686646,
      "epoch": 0.2172532656023222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998191844523119,
      "loss": 5.8289,
      "mean_token_accuracy": 0.15439640134572982,
      "num_tokens": 4954571.0,
      "step": 2395
    },
    {
      "entropy": 5.912092924118042,
      "epoch": 0.21770682148040638,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499817885202795,
      "loss": 5.7943,
      "mean_token_accuracy": 0.15238530337810516,
      "num_tokens": 4964418.0,
      "step": 2400
    },
    {
      "entropy": 5.960055160522461,
      "epoch": 0.21816037735849056,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998165813039869,
      "loss": 5.7833,
      "mean_token_accuracy": 0.1479422241449356,
      "num_tokens": 4975036.0,
      "step": 2405
    },
    {
      "entropy": 5.970535087585449,
      "epoch": 0.21861393323657474,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998152727559145,
      "loss": 5.8868,
      "mean_token_accuracy": 0.14599555134773254,
      "num_tokens": 4985272.0,
      "step": 2410
    },
    {
      "entropy": 6.004713773727417,
      "epoch": 0.21906748911465893,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499813959558605,
      "loss": 5.812,
      "mean_token_accuracy": 0.15417784452438354,
      "num_tokens": 4995187.0,
      "step": 2415
    },
    {
      "entropy": 5.99905571937561,
      "epoch": 0.2195210449927431,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998126417120856,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1530926562845707,
      "num_tokens": 5005013.0,
      "step": 2420
    },
    {
      "entropy": 5.994871044158936,
      "epoch": 0.2199746008708273,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998113192163835,
      "loss": 5.881,
      "mean_token_accuracy": 0.14475401639938354,
      "num_tokens": 5016789.0,
      "step": 2425
    },
    {
      "entropy": 5.921261167526245,
      "epoch": 0.22042815674891147,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499809992071526,
      "loss": 5.828,
      "mean_token_accuracy": 0.1518557384610176,
      "num_tokens": 5026913.0,
      "step": 2430
    },
    {
      "entropy": 5.9377799987792965,
      "epoch": 0.22088171262699566,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998086602775406,
      "loss": 5.9104,
      "mean_token_accuracy": 0.14928861260414122,
      "num_tokens": 5037501.0,
      "step": 2435
    },
    {
      "entropy": 5.983516836166382,
      "epoch": 0.2213352685050798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998073238344547,
      "loss": 5.7602,
      "mean_token_accuracy": 0.15100034326314926,
      "num_tokens": 5048478.0,
      "step": 2440
    },
    {
      "entropy": 5.903777933120727,
      "epoch": 0.221788824383164,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998059827422962,
      "loss": 5.8255,
      "mean_token_accuracy": 0.14812486618757248,
      "num_tokens": 5059118.0,
      "step": 2445
    },
    {
      "entropy": 5.963100814819336,
      "epoch": 0.22224238026124818,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998046370010926,
      "loss": 5.7848,
      "mean_token_accuracy": 0.15494603514671326,
      "num_tokens": 5070180.0,
      "step": 2450
    },
    {
      "entropy": 5.880324077606201,
      "epoch": 0.22269593613933236,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998032866108718,
      "loss": 5.7808,
      "mean_token_accuracy": 0.1510962054133415,
      "num_tokens": 5079912.0,
      "step": 2455
    },
    {
      "entropy": 5.994835710525512,
      "epoch": 0.22314949201741655,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998019315716618,
      "loss": 5.8732,
      "mean_token_accuracy": 0.15168511420488356,
      "num_tokens": 5090549.0,
      "step": 2460
    },
    {
      "entropy": 5.909025812149048,
      "epoch": 0.22360304789550073,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998005718834905,
      "loss": 5.7193,
      "mean_token_accuracy": 0.1588350161910057,
      "num_tokens": 5100880.0,
      "step": 2465
    },
    {
      "entropy": 5.910657835006714,
      "epoch": 0.2240566037735849,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997992075463862,
      "loss": 5.8277,
      "mean_token_accuracy": 0.14884106069803238,
      "num_tokens": 5111796.0,
      "step": 2470
    },
    {
      "entropy": 5.975823974609375,
      "epoch": 0.2245101596516691,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997978385603769,
      "loss": 5.8232,
      "mean_token_accuracy": 0.15248575657606125,
      "num_tokens": 5123468.0,
      "step": 2475
    },
    {
      "entropy": 5.874198150634766,
      "epoch": 0.22496371552975328,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997964649254911,
      "loss": 5.7465,
      "mean_token_accuracy": 0.15882757753133775,
      "num_tokens": 5132925.0,
      "step": 2480
    },
    {
      "entropy": 5.932957172393799,
      "epoch": 0.22541727140783743,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997950866417572,
      "loss": 5.8278,
      "mean_token_accuracy": 0.14813223630189895,
      "num_tokens": 5143445.0,
      "step": 2485
    },
    {
      "entropy": 5.961171960830688,
      "epoch": 0.22587082728592162,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997937037092034,
      "loss": 5.745,
      "mean_token_accuracy": 0.14807531088590623,
      "num_tokens": 5153176.0,
      "step": 2490
    },
    {
      "entropy": 5.9957068920135494,
      "epoch": 0.2263243831640058,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997923161278587,
      "loss": 5.7973,
      "mean_token_accuracy": 0.15407470166683196,
      "num_tokens": 5162765.0,
      "step": 2495
    },
    {
      "entropy": 5.935280895233154,
      "epoch": 0.22677793904208998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997909238977514,
      "loss": 5.8473,
      "mean_token_accuracy": 0.15373046398162843,
      "num_tokens": 5173390.0,
      "step": 2500
    },
    {
      "entropy": 5.851876974105835,
      "epoch": 0.22723149492017417,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997895270189108,
      "loss": 5.7455,
      "mean_token_accuracy": 0.15115862116217613,
      "num_tokens": 5184212.0,
      "step": 2505
    },
    {
      "entropy": 5.847887372970581,
      "epoch": 0.22768505079825835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997881254913653,
      "loss": 5.7465,
      "mean_token_accuracy": 0.15100416243076326,
      "num_tokens": 5194311.0,
      "step": 2510
    },
    {
      "entropy": 5.867375993728638,
      "epoch": 0.22813860667634253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997867193151441,
      "loss": 5.7339,
      "mean_token_accuracy": 0.15452023446559907,
      "num_tokens": 5204708.0,
      "step": 2515
    },
    {
      "entropy": 5.844597864151001,
      "epoch": 0.22859216255442671,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997853084902763,
      "loss": 5.7152,
      "mean_token_accuracy": 0.15443815141916276,
      "num_tokens": 5214566.0,
      "step": 2520
    },
    {
      "entropy": 5.916435194015503,
      "epoch": 0.2290457184325109,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997838930167911,
      "loss": 5.7463,
      "mean_token_accuracy": 0.15494998544454575,
      "num_tokens": 5224762.0,
      "step": 2525
    },
    {
      "entropy": 5.9442017555236815,
      "epoch": 0.22949927431059505,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997824728947176,
      "loss": 5.8129,
      "mean_token_accuracy": 0.15142686367034913,
      "num_tokens": 5235146.0,
      "step": 2530
    },
    {
      "entropy": 5.965577602386475,
      "epoch": 0.22995283018867924,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997810481240853,
      "loss": 5.786,
      "mean_token_accuracy": 0.1514734596014023,
      "num_tokens": 5243867.0,
      "step": 2535
    },
    {
      "entropy": 5.83701639175415,
      "epoch": 0.23040638606676342,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997796187049238,
      "loss": 5.702,
      "mean_token_accuracy": 0.153061743080616,
      "num_tokens": 5254279.0,
      "step": 2540
    },
    {
      "entropy": 6.0492133617401125,
      "epoch": 0.2308599419448476,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997781846372623,
      "loss": 5.7978,
      "mean_token_accuracy": 0.1522805020213127,
      "num_tokens": 5264769.0,
      "step": 2545
    },
    {
      "entropy": 5.893232774734497,
      "epoch": 0.23131349782293179,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997767459211307,
      "loss": 5.7644,
      "mean_token_accuracy": 0.14629466086626053,
      "num_tokens": 5274291.0,
      "step": 2550
    },
    {
      "entropy": 5.868297386169433,
      "epoch": 0.23176705370101597,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997753025565587,
      "loss": 5.695,
      "mean_token_accuracy": 0.15659698024392127,
      "num_tokens": 5284366.0,
      "step": 2555
    },
    {
      "entropy": 5.889286661148072,
      "epoch": 0.23222060957910015,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997738545435763,
      "loss": 5.8173,
      "mean_token_accuracy": 0.15293926745653152,
      "num_tokens": 5294494.0,
      "step": 2560
    },
    {
      "entropy": 5.899365377426148,
      "epoch": 0.23267416545718433,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997724018822133,
      "loss": 5.813,
      "mean_token_accuracy": 0.15529518127441405,
      "num_tokens": 5305606.0,
      "step": 2565
    },
    {
      "entropy": 5.90656361579895,
      "epoch": 0.23312772133526852,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997709445724996,
      "loss": 5.8101,
      "mean_token_accuracy": 0.1502317950129509,
      "num_tokens": 5315677.0,
      "step": 2570
    },
    {
      "entropy": 5.9617547512054445,
      "epoch": 0.23358127721335267,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997694826144656,
      "loss": 5.866,
      "mean_token_accuracy": 0.14772987067699433,
      "num_tokens": 5326453.0,
      "step": 2575
    },
    {
      "entropy": 5.8899938583374025,
      "epoch": 0.23403483309143686,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004997680160081414,
      "loss": 5.7592,
      "mean_token_accuracy": 0.1525429829955101,
      "num_tokens": 5337591.0,
      "step": 2580
    },
    {
      "entropy": 5.914120006561279,
      "epoch": 0.23448838896952104,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997665447535573,
      "loss": 5.7414,
      "mean_token_accuracy": 0.1570722669363022,
      "num_tokens": 5349400.0,
      "step": 2585
    },
    {
      "entropy": 5.9674485206604,
      "epoch": 0.23494194484760522,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997650688507438,
      "loss": 5.7325,
      "mean_token_accuracy": 0.15442833602428435,
      "num_tokens": 5359984.0,
      "step": 2590
    },
    {
      "entropy": 5.855543184280395,
      "epoch": 0.2353955007256894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997635882997314,
      "loss": 5.8149,
      "mean_token_accuracy": 0.16013004034757614,
      "num_tokens": 5369254.0,
      "step": 2595
    },
    {
      "entropy": 5.920786714553833,
      "epoch": 0.2358490566037736,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997621031005507,
      "loss": 5.755,
      "mean_token_accuracy": 0.15076399594545364,
      "num_tokens": 5379648.0,
      "step": 2600
    },
    {
      "entropy": 5.818716669082642,
      "epoch": 0.23630261248185777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997606132532325,
      "loss": 5.7028,
      "mean_token_accuracy": 0.16014982685446738,
      "num_tokens": 5390321.0,
      "step": 2605
    },
    {
      "entropy": 5.850492286682129,
      "epoch": 0.23675616835994195,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997591187578075,
      "loss": 5.7541,
      "mean_token_accuracy": 0.1600041940808296,
      "num_tokens": 5401078.0,
      "step": 2610
    },
    {
      "entropy": 5.8976398468017575,
      "epoch": 0.23720972423802614,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997576196143067,
      "loss": 5.71,
      "mean_token_accuracy": 0.16146167367696762,
      "num_tokens": 5411977.0,
      "step": 2615
    },
    {
      "entropy": 5.871705007553101,
      "epoch": 0.2376632801161103,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499756115822761,
      "loss": 5.7102,
      "mean_token_accuracy": 0.15028589516878127,
      "num_tokens": 5422967.0,
      "step": 2620
    },
    {
      "entropy": 5.950361585617065,
      "epoch": 0.23811683599419448,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997546073832015,
      "loss": 5.7474,
      "mean_token_accuracy": 0.16250978112220765,
      "num_tokens": 5432260.0,
      "step": 2625
    },
    {
      "entropy": 5.842865753173828,
      "epoch": 0.23857039187227866,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997530942956595,
      "loss": 5.7182,
      "mean_token_accuracy": 0.15647428780794143,
      "num_tokens": 5441706.0,
      "step": 2630
    },
    {
      "entropy": 5.885302925109864,
      "epoch": 0.23902394775036284,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997515765601662,
      "loss": 5.7294,
      "mean_token_accuracy": 0.1620200365781784,
      "num_tokens": 5452397.0,
      "step": 2635
    },
    {
      "entropy": 5.802298927307129,
      "epoch": 0.23947750362844702,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997500541767531,
      "loss": 5.6999,
      "mean_token_accuracy": 0.1544307217001915,
      "num_tokens": 5463762.0,
      "step": 2640
    },
    {
      "entropy": 5.895985698699951,
      "epoch": 0.2399310595065312,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997485271454516,
      "loss": 5.7718,
      "mean_token_accuracy": 0.14597637057304383,
      "num_tokens": 5474622.0,
      "step": 2645
    },
    {
      "entropy": 5.872747230529785,
      "epoch": 0.2403846153846154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997469954662933,
      "loss": 5.7431,
      "mean_token_accuracy": 0.15389883369207383,
      "num_tokens": 5484697.0,
      "step": 2650
    },
    {
      "entropy": 5.795325708389282,
      "epoch": 0.24083817126269957,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997454591393097,
      "loss": 5.744,
      "mean_token_accuracy": 0.15505960136651992,
      "num_tokens": 5496433.0,
      "step": 2655
    },
    {
      "entropy": 5.929238319396973,
      "epoch": 0.24129172714078376,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997439181645329,
      "loss": 5.8642,
      "mean_token_accuracy": 0.14926385432481765,
      "num_tokens": 5507917.0,
      "step": 2660
    },
    {
      "entropy": 6.066269302368164,
      "epoch": 0.2417452830188679,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997423725419945,
      "loss": 5.8473,
      "mean_token_accuracy": 0.15168920159339905,
      "num_tokens": 5519159.0,
      "step": 2665
    },
    {
      "entropy": 5.896298265457153,
      "epoch": 0.2421988388969521,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997408222717267,
      "loss": 5.7941,
      "mean_token_accuracy": 0.15869408696889878,
      "num_tokens": 5529180.0,
      "step": 2670
    },
    {
      "entropy": 5.829392433166504,
      "epoch": 0.24265239477503628,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997392673537612,
      "loss": 5.7138,
      "mean_token_accuracy": 0.1638789251446724,
      "num_tokens": 5539363.0,
      "step": 2675
    },
    {
      "entropy": 5.891768503189087,
      "epoch": 0.24310595065312046,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997377077881306,
      "loss": 5.8155,
      "mean_token_accuracy": 0.15023489147424698,
      "num_tokens": 5549841.0,
      "step": 2680
    },
    {
      "entropy": 5.889240789413452,
      "epoch": 0.24355950653120464,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997361435748668,
      "loss": 5.6947,
      "mean_token_accuracy": 0.16268197000026702,
      "num_tokens": 5560299.0,
      "step": 2685
    },
    {
      "entropy": 5.9125152111053465,
      "epoch": 0.24401306240928883,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997345747140024,
      "loss": 5.7689,
      "mean_token_accuracy": 0.15359243452548982,
      "num_tokens": 5571081.0,
      "step": 2690
    },
    {
      "entropy": 5.8776861190795895,
      "epoch": 0.244466618287373,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997330012055695,
      "loss": 5.7108,
      "mean_token_accuracy": 0.15765073597431184,
      "num_tokens": 5581389.0,
      "step": 2695
    },
    {
      "entropy": 5.861866426467896,
      "epoch": 0.2449201741654572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997314230496011,
      "loss": 5.754,
      "mean_token_accuracy": 0.1500275418162346,
      "num_tokens": 5592347.0,
      "step": 2700
    },
    {
      "entropy": 5.839778137207031,
      "epoch": 0.24537373004354138,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997298402461296,
      "loss": 5.7228,
      "mean_token_accuracy": 0.15842826217412947,
      "num_tokens": 5603411.0,
      "step": 2705
    },
    {
      "entropy": 5.9260224342346195,
      "epoch": 0.24582728592162553,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997282527951876,
      "loss": 5.7162,
      "mean_token_accuracy": 0.15986051708459853,
      "num_tokens": 5614615.0,
      "step": 2710
    },
    {
      "entropy": 5.790437936782837,
      "epoch": 0.24628084179970972,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997266606968081,
      "loss": 5.6868,
      "mean_token_accuracy": 0.1561066210269928,
      "num_tokens": 5624582.0,
      "step": 2715
    },
    {
      "entropy": 5.811624383926391,
      "epoch": 0.2467343976777939,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997250639510241,
      "loss": 5.7,
      "mean_token_accuracy": 0.15437885671854018,
      "num_tokens": 5634310.0,
      "step": 2720
    },
    {
      "entropy": 5.889466571807861,
      "epoch": 0.24718795355587808,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004997234625578684,
      "loss": 5.7947,
      "mean_token_accuracy": 0.14918309599161148,
      "num_tokens": 5644018.0,
      "step": 2725
    },
    {
      "entropy": 5.858435821533203,
      "epoch": 0.24764150943396226,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997218565173742,
      "loss": 5.7233,
      "mean_token_accuracy": 0.15628414005041122,
      "num_tokens": 5653592.0,
      "step": 2730
    },
    {
      "entropy": 5.904895401000976,
      "epoch": 0.24809506531204645,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997202458295748,
      "loss": 5.7632,
      "mean_token_accuracy": 0.15781814455986024,
      "num_tokens": 5662812.0,
      "step": 2735
    },
    {
      "entropy": 5.902950763702393,
      "epoch": 0.24854862119013063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997186304945034,
      "loss": 5.7511,
      "mean_token_accuracy": 0.15532326996326445,
      "num_tokens": 5672939.0,
      "step": 2740
    },
    {
      "entropy": 5.827338409423828,
      "epoch": 0.24900217706821481,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997170105121935,
      "loss": 5.7313,
      "mean_token_accuracy": 0.15588268786668777,
      "num_tokens": 5683350.0,
      "step": 2745
    },
    {
      "entropy": 5.919451379776001,
      "epoch": 0.249455732946299,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997153858826786,
      "loss": 5.816,
      "mean_token_accuracy": 0.14666671603918074,
      "num_tokens": 5694516.0,
      "step": 2750
    },
    {
      "entropy": 5.843132162094117,
      "epoch": 0.24990928882438315,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997137566059922,
      "loss": 5.6435,
      "mean_token_accuracy": 0.15954998284578323,
      "num_tokens": 5705443.0,
      "step": 2755
    },
    {
      "entropy": 5.838101053237915,
      "epoch": 0.25036284470246734,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997121226821681,
      "loss": 5.742,
      "mean_token_accuracy": 0.15265777483582496,
      "num_tokens": 5715933.0,
      "step": 2760
    },
    {
      "entropy": 5.910187196731568,
      "epoch": 0.25081640058055155,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049971048411124,
      "loss": 5.8084,
      "mean_token_accuracy": 0.1447537824511528,
      "num_tokens": 5726630.0,
      "step": 2765
    },
    {
      "entropy": 5.9618613719940186,
      "epoch": 0.2512699564586357,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997088408932418,
      "loss": 5.7349,
      "mean_token_accuracy": 0.15529899746179582,
      "num_tokens": 5736890.0,
      "step": 2770
    },
    {
      "entropy": 5.754767274856567,
      "epoch": 0.25172351233671986,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997071930282076,
      "loss": 5.6394,
      "mean_token_accuracy": 0.16765589267015457,
      "num_tokens": 5747042.0,
      "step": 2775
    },
    {
      "entropy": 5.836727952957153,
      "epoch": 0.25217706821480407,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997055405161712,
      "loss": 5.7737,
      "mean_token_accuracy": 0.15657578855752946,
      "num_tokens": 5757367.0,
      "step": 2780
    },
    {
      "entropy": 5.892507696151734,
      "epoch": 0.2526306240928882,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997038833571671,
      "loss": 5.8198,
      "mean_token_accuracy": 0.15040036141872407,
      "num_tokens": 5769557.0,
      "step": 2785
    },
    {
      "entropy": 5.882242345809937,
      "epoch": 0.25308417997097243,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997022215512295,
      "loss": 5.7227,
      "mean_token_accuracy": 0.162361441552639,
      "num_tokens": 5780615.0,
      "step": 2790
    },
    {
      "entropy": 5.901854085922241,
      "epoch": 0.2535377358490566,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997005550983925,
      "loss": 5.7026,
      "mean_token_accuracy": 0.1581372618675232,
      "num_tokens": 5791271.0,
      "step": 2795
    },
    {
      "entropy": 5.816129541397094,
      "epoch": 0.2539912917271408,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499698883998691,
      "loss": 5.6376,
      "mean_token_accuracy": 0.15995124131441116,
      "num_tokens": 5801117.0,
      "step": 2800
    },
    {
      "entropy": 5.778412342071533,
      "epoch": 0.25444484760522496,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996972082521591,
      "loss": 5.6657,
      "mean_token_accuracy": 0.16171006560325624,
      "num_tokens": 5811696.0,
      "step": 2805
    },
    {
      "entropy": 5.8469995021820065,
      "epoch": 0.25489840348330917,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996955278588319,
      "loss": 5.6983,
      "mean_token_accuracy": 0.16263554394245147,
      "num_tokens": 5822207.0,
      "step": 2810
    },
    {
      "entropy": 5.802956390380859,
      "epoch": 0.2553519593613933,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996938428187437,
      "loss": 5.6452,
      "mean_token_accuracy": 0.1667325183749199,
      "num_tokens": 5831927.0,
      "step": 2815
    },
    {
      "entropy": 5.801891422271728,
      "epoch": 0.2558055152394775,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996921531319297,
      "loss": 5.7182,
      "mean_token_accuracy": 0.15510134547948837,
      "num_tokens": 5842093.0,
      "step": 2820
    },
    {
      "entropy": 5.818565034866333,
      "epoch": 0.2562590711175617,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996904587984246,
      "loss": 5.6452,
      "mean_token_accuracy": 0.16608070433139802,
      "num_tokens": 5850551.0,
      "step": 2825
    },
    {
      "entropy": 5.797169351577759,
      "epoch": 0.25671262699564584,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996887598182636,
      "loss": 5.6412,
      "mean_token_accuracy": 0.1570335805416107,
      "num_tokens": 5860651.0,
      "step": 2830
    },
    {
      "entropy": 5.70640926361084,
      "epoch": 0.25716618287373005,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996870561914819,
      "loss": 5.6313,
      "mean_token_accuracy": 0.1594139203429222,
      "num_tokens": 5871500.0,
      "step": 2835
    },
    {
      "entropy": 5.810226488113403,
      "epoch": 0.2576197387518142,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996853479181145,
      "loss": 5.6415,
      "mean_token_accuracy": 0.15245891511440277,
      "num_tokens": 5881523.0,
      "step": 2840
    },
    {
      "entropy": 5.9306385040283205,
      "epoch": 0.2580732946298984,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996836349981969,
      "loss": 5.7752,
      "mean_token_accuracy": 0.1541695550084114,
      "num_tokens": 5891162.0,
      "step": 2845
    },
    {
      "entropy": 5.793090677261352,
      "epoch": 0.2585268505079826,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996819174317645,
      "loss": 5.6667,
      "mean_token_accuracy": 0.15518372505903244,
      "num_tokens": 5901818.0,
      "step": 2850
    },
    {
      "entropy": 5.8290753841400145,
      "epoch": 0.2589804063860668,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996801952188528,
      "loss": 5.7439,
      "mean_token_accuracy": 0.15479833632707596,
      "num_tokens": 5913889.0,
      "step": 2855
    },
    {
      "entropy": 5.848030471801758,
      "epoch": 0.25943396226415094,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004996784683594973,
      "loss": 5.7061,
      "mean_token_accuracy": 0.16026207208633422,
      "num_tokens": 5923429.0,
      "step": 2860
    },
    {
      "entropy": 5.953624963760376,
      "epoch": 0.2598875181422351,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996767368537338,
      "loss": 5.7536,
      "mean_token_accuracy": 0.15464293360710143,
      "num_tokens": 5933847.0,
      "step": 2865
    },
    {
      "entropy": 5.797641706466675,
      "epoch": 0.2603410740203193,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996750007015983,
      "loss": 5.6199,
      "mean_token_accuracy": 0.15784441232681273,
      "num_tokens": 5943874.0,
      "step": 2870
    },
    {
      "entropy": 5.836393356323242,
      "epoch": 0.26079462989840346,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996732599031264,
      "loss": 5.6901,
      "mean_token_accuracy": 0.15797710716724395,
      "num_tokens": 5953653.0,
      "step": 2875
    },
    {
      "entropy": 5.831515836715698,
      "epoch": 0.2612481857764877,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996715144583543,
      "loss": 5.6401,
      "mean_token_accuracy": 0.16178645938634872,
      "num_tokens": 5964039.0,
      "step": 2880
    },
    {
      "entropy": 5.8883569717407225,
      "epoch": 0.26170174165457183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499669764367318,
      "loss": 5.7538,
      "mean_token_accuracy": 0.15600325912237167,
      "num_tokens": 5974083.0,
      "step": 2885
    },
    {
      "entropy": 5.775406360626221,
      "epoch": 0.26215529753265604,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996680096300537,
      "loss": 5.711,
      "mean_token_accuracy": 0.15503483414649963,
      "num_tokens": 5984629.0,
      "step": 2890
    },
    {
      "entropy": 5.7855353355407715,
      "epoch": 0.2626088534107402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996662502465978,
      "loss": 5.6067,
      "mean_token_accuracy": 0.1710730493068695,
      "num_tokens": 5994295.0,
      "step": 2895
    },
    {
      "entropy": 5.8054187297821045,
      "epoch": 0.2630624092888244,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996644862169864,
      "loss": 5.665,
      "mean_token_accuracy": 0.15672381818294526,
      "num_tokens": 6004156.0,
      "step": 2900
    },
    {
      "entropy": 5.815729665756225,
      "epoch": 0.26351596516690856,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996627175412563,
      "loss": 5.7391,
      "mean_token_accuracy": 0.16197800040245056,
      "num_tokens": 6014188.0,
      "step": 2905
    },
    {
      "entropy": 5.904583549499511,
      "epoch": 0.2639695210449927,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499660944219444,
      "loss": 5.8105,
      "mean_token_accuracy": 0.15212338119745256,
      "num_tokens": 6024879.0,
      "step": 2910
    },
    {
      "entropy": 5.926459264755249,
      "epoch": 0.2644230769230769,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996591662515861,
      "loss": 5.8701,
      "mean_token_accuracy": 0.15266856625676156,
      "num_tokens": 6036374.0,
      "step": 2915
    },
    {
      "entropy": 5.832221269607544,
      "epoch": 0.2648766328011611,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996573836377193,
      "loss": 5.7056,
      "mean_token_accuracy": 0.15986677557229995,
      "num_tokens": 6046432.0,
      "step": 2920
    },
    {
      "entropy": 5.8731166362762455,
      "epoch": 0.2653301886792453,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996555963778806,
      "loss": 5.7218,
      "mean_token_accuracy": 0.15033943355083465,
      "num_tokens": 6056864.0,
      "step": 2925
    },
    {
      "entropy": 5.9467894554138185,
      "epoch": 0.26578374455732945,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499653804472107,
      "loss": 5.7396,
      "mean_token_accuracy": 0.1595652773976326,
      "num_tokens": 6067294.0,
      "step": 2930
    },
    {
      "entropy": 5.863904571533203,
      "epoch": 0.26623730043541366,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996520079204353,
      "loss": 5.741,
      "mean_token_accuracy": 0.15215880274772645,
      "num_tokens": 6077587.0,
      "step": 2935
    },
    {
      "entropy": 5.880077457427978,
      "epoch": 0.2666908563134978,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000499650206722903,
      "loss": 5.6892,
      "mean_token_accuracy": 0.1558769851922989,
      "num_tokens": 6089306.0,
      "step": 2940
    },
    {
      "entropy": 5.7567362785339355,
      "epoch": 0.267144412191582,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996484008795471,
      "loss": 5.6433,
      "mean_token_accuracy": 0.15309089198708534,
      "num_tokens": 6099651.0,
      "step": 2945
    },
    {
      "entropy": 5.811124467849732,
      "epoch": 0.2675979680696662,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499646590390405,
      "loss": 5.6503,
      "mean_token_accuracy": 0.1572161540389061,
      "num_tokens": 6110063.0,
      "step": 2950
    },
    {
      "entropy": 5.807859134674072,
      "epoch": 0.26805152394775034,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996447752555142,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15904018133878708,
      "num_tokens": 6120031.0,
      "step": 2955
    },
    {
      "entropy": 5.8083930015563965,
      "epoch": 0.26850507982583455,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996429554749122,
      "loss": 5.7183,
      "mean_token_accuracy": 0.15861095041036605,
      "num_tokens": 6131274.0,
      "step": 2960
    },
    {
      "entropy": 5.888201808929443,
      "epoch": 0.2689586357039187,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996411310486367,
      "loss": 5.7658,
      "mean_token_accuracy": 0.15850751847028732,
      "num_tokens": 6140878.0,
      "step": 2965
    },
    {
      "entropy": 5.859344720840454,
      "epoch": 0.2694121915820029,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996393019767253,
      "loss": 5.7646,
      "mean_token_accuracy": 0.1520871713757515,
      "num_tokens": 6152229.0,
      "step": 2970
    },
    {
      "entropy": 5.737780809402466,
      "epoch": 0.26986574746008707,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996374682592158,
      "loss": 5.6579,
      "mean_token_accuracy": 0.15743956118822097,
      "num_tokens": 6162311.0,
      "step": 2975
    },
    {
      "entropy": 5.8878035068511965,
      "epoch": 0.2703193033381713,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996356298961463,
      "loss": 5.7654,
      "mean_token_accuracy": 0.15596626251935958,
      "num_tokens": 6172806.0,
      "step": 2980
    },
    {
      "entropy": 5.860615348815918,
      "epoch": 0.27077285921625543,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996337868875549,
      "loss": 5.7448,
      "mean_token_accuracy": 0.1540456607937813,
      "num_tokens": 6183479.0,
      "step": 2985
    },
    {
      "entropy": 5.865048408508301,
      "epoch": 0.27122641509433965,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004996319392334792,
      "loss": 5.7043,
      "mean_token_accuracy": 0.15600668787956237,
      "num_tokens": 6194201.0,
      "step": 2990
    },
    {
      "entropy": 5.800745820999145,
      "epoch": 0.2716799709724238,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499630086933958,
      "loss": 5.6423,
      "mean_token_accuracy": 0.1544359117746353,
      "num_tokens": 6204774.0,
      "step": 2995
    },
    {
      "entropy": 5.796499633789063,
      "epoch": 0.27213352685050796,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996282299890292,
      "loss": 5.6475,
      "mean_token_accuracy": 0.15237775593996047,
      "num_tokens": 6215056.0,
      "step": 3000
    },
    {
      "epoch": 0.27213352685050796,
      "eval_entropy": 5.64969411068912,
      "eval_loss": 5.721183776855469,
      "eval_mean_token_accuracy": 0.16399123540223992,
      "eval_num_tokens": 6215056.0,
      "eval_runtime": 20.9259,
      "eval_samples_per_second": 1689.772,
      "eval_steps_per_second": 211.221,
      "step": 3000
    },
    {
      "entropy": 5.928546047210693,
      "epoch": 0.27258708272859217,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996263683987316,
      "loss": 5.754,
      "mean_token_accuracy": 0.15922772735357285,
      "num_tokens": 6226264.0,
      "step": 3005
    },
    {
      "entropy": 5.842231750488281,
      "epoch": 0.2730406386066763,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996245021631033,
      "loss": 5.6621,
      "mean_token_accuracy": 0.1633378177881241,
      "num_tokens": 6235137.0,
      "step": 3010
    },
    {
      "entropy": 5.790783786773682,
      "epoch": 0.27349419448476053,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499622631282183,
      "loss": 5.6825,
      "mean_token_accuracy": 0.16040071249008178,
      "num_tokens": 6245411.0,
      "step": 3015
    },
    {
      "entropy": 5.880749797821045,
      "epoch": 0.2739477503628447,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996207557560095,
      "loss": 5.7469,
      "mean_token_accuracy": 0.15628764629364014,
      "num_tokens": 6255633.0,
      "step": 3020
    },
    {
      "entropy": 5.8452211856842045,
      "epoch": 0.2744013062409289,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996188755846216,
      "loss": 5.7219,
      "mean_token_accuracy": 0.15557182282209397,
      "num_tokens": 6265109.0,
      "step": 3025
    },
    {
      "entropy": 5.8197873592376705,
      "epoch": 0.27485486211901305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499616990768058,
      "loss": 5.7484,
      "mean_token_accuracy": 0.15053788274526597,
      "num_tokens": 6275241.0,
      "step": 3030
    },
    {
      "entropy": 5.869422435760498,
      "epoch": 0.27530841799709727,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996151013063579,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1479799047112465,
      "num_tokens": 6286603.0,
      "step": 3035
    },
    {
      "entropy": 5.895780897140503,
      "epoch": 0.2757619738751814,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996132071995601,
      "loss": 5.6345,
      "mean_token_accuracy": 0.16945138126611708,
      "num_tokens": 6297219.0,
      "step": 3040
    },
    {
      "entropy": 5.763274097442627,
      "epoch": 0.2762155297532656,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996113084477042,
      "loss": 5.6749,
      "mean_token_accuracy": 0.16423184126615525,
      "num_tokens": 6307694.0,
      "step": 3045
    },
    {
      "entropy": 5.769646883010864,
      "epoch": 0.2766690856313498,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996094050508291,
      "loss": 5.604,
      "mean_token_accuracy": 0.16311236768960952,
      "num_tokens": 6317384.0,
      "step": 3050
    },
    {
      "entropy": 5.845623970031738,
      "epoch": 0.27712264150943394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996074970089741,
      "loss": 5.7395,
      "mean_token_accuracy": 0.15895291715860366,
      "num_tokens": 6327558.0,
      "step": 3055
    },
    {
      "entropy": 5.829644441604614,
      "epoch": 0.27757619738751815,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996055843221789,
      "loss": 5.6008,
      "mean_token_accuracy": 0.16788485497236252,
      "num_tokens": 6337828.0,
      "step": 3060
    },
    {
      "entropy": 5.797014570236206,
      "epoch": 0.2780297532656023,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996036669904831,
      "loss": 5.6642,
      "mean_token_accuracy": 0.15934574156999587,
      "num_tokens": 6348678.0,
      "step": 3065
    },
    {
      "entropy": 5.835013437271118,
      "epoch": 0.2784833091436865,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996017450139261,
      "loss": 5.7507,
      "mean_token_accuracy": 0.1531164437532425,
      "num_tokens": 6359670.0,
      "step": 3070
    },
    {
      "entropy": 5.947513818740845,
      "epoch": 0.2789368650217707,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995998183925478,
      "loss": 5.7193,
      "mean_token_accuracy": 0.1559832513332367,
      "num_tokens": 6371034.0,
      "step": 3075
    },
    {
      "entropy": 5.747111892700195,
      "epoch": 0.2793904208998549,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995978871263881,
      "loss": 5.6698,
      "mean_token_accuracy": 0.1631323963403702,
      "num_tokens": 6380672.0,
      "step": 3080
    },
    {
      "entropy": 5.774804639816284,
      "epoch": 0.27984397677793904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995959512154868,
      "loss": 5.7757,
      "mean_token_accuracy": 0.1484546534717083,
      "num_tokens": 6390799.0,
      "step": 3085
    },
    {
      "entropy": 5.823790121078491,
      "epoch": 0.2802975326560232,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499594010659884,
      "loss": 5.6352,
      "mean_token_accuracy": 0.16301891803741456,
      "num_tokens": 6400965.0,
      "step": 3090
    },
    {
      "entropy": 5.842621088027954,
      "epoch": 0.2807510885341074,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995920654596199,
      "loss": 5.6585,
      "mean_token_accuracy": 0.15496362000703812,
      "num_tokens": 6411567.0,
      "step": 3095
    },
    {
      "entropy": 5.816281747817993,
      "epoch": 0.28120464441219156,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995901156147345,
      "loss": 5.7114,
      "mean_token_accuracy": 0.16414457112550734,
      "num_tokens": 6421193.0,
      "step": 3100
    },
    {
      "entropy": 5.827236461639404,
      "epoch": 0.2816582002902758,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995881611252684,
      "loss": 5.7181,
      "mean_token_accuracy": 0.16552949845790862,
      "num_tokens": 6430281.0,
      "step": 3105
    },
    {
      "entropy": 5.8368793487548825,
      "epoch": 0.28211175616835993,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995862019912619,
      "loss": 5.7229,
      "mean_token_accuracy": 0.15318924188613892,
      "num_tokens": 6441055.0,
      "step": 3110
    },
    {
      "entropy": 5.827468013763427,
      "epoch": 0.28256531204644414,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995842382127555,
      "loss": 5.638,
      "mean_token_accuracy": 0.1578447103500366,
      "num_tokens": 6451082.0,
      "step": 3115
    },
    {
      "entropy": 5.795475244522095,
      "epoch": 0.2830188679245283,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995822697897897,
      "loss": 5.7249,
      "mean_token_accuracy": 0.15724158138036728,
      "num_tokens": 6461617.0,
      "step": 3120
    },
    {
      "entropy": 5.884941148757934,
      "epoch": 0.2834724238026125,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995802967224056,
      "loss": 5.7196,
      "mean_token_accuracy": 0.15904805064201355,
      "num_tokens": 6472780.0,
      "step": 3125
    },
    {
      "entropy": 5.719329166412353,
      "epoch": 0.28392597968069666,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995783190106434,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15374166816473006,
      "num_tokens": 6484078.0,
      "step": 3130
    },
    {
      "entropy": 5.816254758834839,
      "epoch": 0.2843795355587808,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995763366545445,
      "loss": 5.7225,
      "mean_token_accuracy": 0.15541643053293228,
      "num_tokens": 6494366.0,
      "step": 3135
    },
    {
      "entropy": 5.751023006439209,
      "epoch": 0.284833091436865,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995743496541497,
      "loss": 5.6544,
      "mean_token_accuracy": 0.1647457867860794,
      "num_tokens": 6505461.0,
      "step": 3140
    },
    {
      "entropy": 5.815382051467895,
      "epoch": 0.2852866473149492,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995723580095001,
      "loss": 5.6527,
      "mean_token_accuracy": 0.16272574812173843,
      "num_tokens": 6516141.0,
      "step": 3145
    },
    {
      "entropy": 5.74614405632019,
      "epoch": 0.2857402031930334,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499570361720637,
      "loss": 5.6605,
      "mean_token_accuracy": 0.15768903791904448,
      "num_tokens": 6525536.0,
      "step": 3150
    },
    {
      "entropy": 5.821631193161011,
      "epoch": 0.28619375907111755,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004995683607876015,
      "loss": 5.7061,
      "mean_token_accuracy": 0.15657218396663666,
      "num_tokens": 6535852.0,
      "step": 3155
    },
    {
      "entropy": 5.84665880203247,
      "epoch": 0.28664731494920176,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995663552104352,
      "loss": 5.6836,
      "mean_token_accuracy": 0.16722792834043504,
      "num_tokens": 6546353.0,
      "step": 3160
    },
    {
      "entropy": 5.704407453536987,
      "epoch": 0.2871008708272859,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995643449891793,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1609612762928009,
      "num_tokens": 6557782.0,
      "step": 3165
    },
    {
      "entropy": 5.824968910217285,
      "epoch": 0.2875544267053701,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995623301238757,
      "loss": 5.6597,
      "mean_token_accuracy": 0.16069735288619996,
      "num_tokens": 6568100.0,
      "step": 3170
    },
    {
      "entropy": 5.805774164199829,
      "epoch": 0.2880079825834543,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995603106145657,
      "loss": 5.7218,
      "mean_token_accuracy": 0.1530776634812355,
      "num_tokens": 6578698.0,
      "step": 3175
    },
    {
      "entropy": 5.830034828186035,
      "epoch": 0.28846153846153844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995582864612913,
      "loss": 5.7009,
      "mean_token_accuracy": 0.15772538930177687,
      "num_tokens": 6588908.0,
      "step": 3180
    },
    {
      "entropy": 5.755585670471191,
      "epoch": 0.28891509433962265,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995562576640943,
      "loss": 5.5465,
      "mean_token_accuracy": 0.16224001944065095,
      "num_tokens": 6599445.0,
      "step": 3185
    },
    {
      "entropy": 5.754920482635498,
      "epoch": 0.2893686502177068,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995542242230167,
      "loss": 5.6333,
      "mean_token_accuracy": 0.16428707391023636,
      "num_tokens": 6610246.0,
      "step": 3190
    },
    {
      "entropy": 5.780771970748901,
      "epoch": 0.289822206095791,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995521861381006,
      "loss": 5.6527,
      "mean_token_accuracy": 0.15799290537834168,
      "num_tokens": 6620501.0,
      "step": 3195
    },
    {
      "entropy": 5.877050399780273,
      "epoch": 0.29027576197387517,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499550143409388,
      "loss": 5.7227,
      "mean_token_accuracy": 0.15273681730031968,
      "num_tokens": 6631206.0,
      "step": 3200
    },
    {
      "entropy": 5.770636701583863,
      "epoch": 0.2907293178519594,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995480960369211,
      "loss": 5.6805,
      "mean_token_accuracy": 0.15422717183828355,
      "num_tokens": 6641600.0,
      "step": 3205
    },
    {
      "entropy": 5.789561319351196,
      "epoch": 0.29118287373004353,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995460440207425,
      "loss": 5.7389,
      "mean_token_accuracy": 0.15444392561912537,
      "num_tokens": 6652319.0,
      "step": 3210
    },
    {
      "entropy": 5.84764404296875,
      "epoch": 0.29163642960812775,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004995439873608945,
      "loss": 5.6796,
      "mean_token_accuracy": 0.15282317996025085,
      "num_tokens": 6663702.0,
      "step": 3215
    },
    {
      "entropy": 5.774284696578979,
      "epoch": 0.2920899854862119,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995419260574196,
      "loss": 5.6669,
      "mean_token_accuracy": 0.1519273892045021,
      "num_tokens": 6673908.0,
      "step": 3220
    },
    {
      "entropy": 5.769118547439575,
      "epoch": 0.29254354136429606,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995398601103603,
      "loss": 5.682,
      "mean_token_accuracy": 0.15996284335851668,
      "num_tokens": 6683237.0,
      "step": 3225
    },
    {
      "entropy": 5.814051771163941,
      "epoch": 0.29299709724238027,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995377895197597,
      "loss": 5.6143,
      "mean_token_accuracy": 0.16433483213186265,
      "num_tokens": 6693451.0,
      "step": 3230
    },
    {
      "entropy": 5.719491100311279,
      "epoch": 0.2934506531204644,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995357142856602,
      "loss": 5.5676,
      "mean_token_accuracy": 0.16208960115909576,
      "num_tokens": 6703479.0,
      "step": 3235
    },
    {
      "entropy": 5.692001056671143,
      "epoch": 0.29390420899854863,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499533634408105,
      "loss": 5.5692,
      "mean_token_accuracy": 0.15959729105234147,
      "num_tokens": 6715321.0,
      "step": 3240
    },
    {
      "entropy": 5.770639801025391,
      "epoch": 0.2943577648766328,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499531549887137,
      "loss": 5.6309,
      "mean_token_accuracy": 0.15734803080558776,
      "num_tokens": 6725541.0,
      "step": 3245
    },
    {
      "entropy": 5.791307258605957,
      "epoch": 0.294811320754717,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995294607227993,
      "loss": 5.6795,
      "mean_token_accuracy": 0.16653743833303453,
      "num_tokens": 6735165.0,
      "step": 3250
    },
    {
      "entropy": 5.812675523757934,
      "epoch": 0.29526487663280115,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995273669151351,
      "loss": 5.6685,
      "mean_token_accuracy": 0.16017006635665892,
      "num_tokens": 6744245.0,
      "step": 3255
    },
    {
      "entropy": 5.793766069412231,
      "epoch": 0.29571843251088537,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499525268464188,
      "loss": 5.6445,
      "mean_token_accuracy": 0.1621537297964096,
      "num_tokens": 6754894.0,
      "step": 3260
    },
    {
      "entropy": 5.82327790260315,
      "epoch": 0.2961719883889695,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995231653700009,
      "loss": 5.7043,
      "mean_token_accuracy": 0.15255022346973418,
      "num_tokens": 6764442.0,
      "step": 3265
    },
    {
      "entropy": 5.78665943145752,
      "epoch": 0.2966255442670537,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004995210576326176,
      "loss": 5.6535,
      "mean_token_accuracy": 0.156018228828907,
      "num_tokens": 6775054.0,
      "step": 3270
    },
    {
      "entropy": 5.80814037322998,
      "epoch": 0.2970791001451379,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995189452520816,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17229300737380981,
      "num_tokens": 6784825.0,
      "step": 3275
    },
    {
      "entropy": 5.80222601890564,
      "epoch": 0.29753265602322204,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995168282284366,
      "loss": 5.6766,
      "mean_token_accuracy": 0.16131959110498428,
      "num_tokens": 6795928.0,
      "step": 3280
    },
    {
      "entropy": 5.726570558547974,
      "epoch": 0.29798621190130625,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995147065617263,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1610921859741211,
      "num_tokens": 6805456.0,
      "step": 3285
    },
    {
      "entropy": 5.799875020980835,
      "epoch": 0.2984397677793904,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995125802519946,
      "loss": 5.6826,
      "mean_token_accuracy": 0.15558315366506575,
      "num_tokens": 6816215.0,
      "step": 3290
    },
    {
      "entropy": 5.817404937744141,
      "epoch": 0.2988933236574746,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995104492992857,
      "loss": 5.7207,
      "mean_token_accuracy": 0.15770665407180787,
      "num_tokens": 6826545.0,
      "step": 3295
    },
    {
      "entropy": 5.704713821411133,
      "epoch": 0.2993468795355588,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995083137036434,
      "loss": 5.6256,
      "mean_token_accuracy": 0.16132950484752656,
      "num_tokens": 6836184.0,
      "step": 3300
    },
    {
      "entropy": 5.845241355895996,
      "epoch": 0.299800435413643,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995061734651121,
      "loss": 5.7128,
      "mean_token_accuracy": 0.15842559039592743,
      "num_tokens": 6846646.0,
      "step": 3305
    },
    {
      "entropy": 5.760690546035766,
      "epoch": 0.30025399129172714,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995040285837358,
      "loss": 5.6189,
      "mean_token_accuracy": 0.16771125495433808,
      "num_tokens": 6857597.0,
      "step": 3310
    },
    {
      "entropy": 5.727470827102661,
      "epoch": 0.3007075471698113,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995018790595589,
      "loss": 5.6962,
      "mean_token_accuracy": 0.1587997391819954,
      "num_tokens": 6867523.0,
      "step": 3315
    },
    {
      "entropy": 5.822723150253296,
      "epoch": 0.3011611030478955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994997248926261,
      "loss": 5.7959,
      "mean_token_accuracy": 0.15733444988727568,
      "num_tokens": 6878190.0,
      "step": 3320
    },
    {
      "entropy": 5.769148635864258,
      "epoch": 0.30161465892597966,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994975660829817,
      "loss": 5.5958,
      "mean_token_accuracy": 0.17123553454875945,
      "num_tokens": 6888483.0,
      "step": 3325
    },
    {
      "entropy": 5.718564510345459,
      "epoch": 0.3020682148040639,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994954026306705,
      "loss": 5.5724,
      "mean_token_accuracy": 0.16064103171229363,
      "num_tokens": 6898337.0,
      "step": 3330
    },
    {
      "entropy": 5.827276992797851,
      "epoch": 0.30252177068214803,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994932345357371,
      "loss": 5.6561,
      "mean_token_accuracy": 0.15858807861804963,
      "num_tokens": 6909131.0,
      "step": 3335
    },
    {
      "entropy": 5.740424537658692,
      "epoch": 0.30297532656023224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994910617982264,
      "loss": 5.6382,
      "mean_token_accuracy": 0.15910502076148986,
      "num_tokens": 6919790.0,
      "step": 3340
    },
    {
      "entropy": 5.753969144821167,
      "epoch": 0.3034288824383164,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994888844181833,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1657829388976097,
      "num_tokens": 6930072.0,
      "step": 3345
    },
    {
      "entropy": 5.6898661136627195,
      "epoch": 0.3038824383164006,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994867023956529,
      "loss": 5.5388,
      "mean_token_accuracy": 0.16034141033887864,
      "num_tokens": 6940646.0,
      "step": 3350
    },
    {
      "entropy": 5.677322959899902,
      "epoch": 0.30433599419448476,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994845157306803,
      "loss": 5.5421,
      "mean_token_accuracy": 0.16668861508369445,
      "num_tokens": 6951645.0,
      "step": 3355
    },
    {
      "entropy": 5.837560987472534,
      "epoch": 0.3047895500725689,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994823244233107,
      "loss": 5.703,
      "mean_token_accuracy": 0.15850953310728072,
      "num_tokens": 6963020.0,
      "step": 3360
    },
    {
      "entropy": 5.787491798400879,
      "epoch": 0.3052431059506531,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994801284735895,
      "loss": 5.6444,
      "mean_token_accuracy": 0.1636459544301033,
      "num_tokens": 6972895.0,
      "step": 3365
    },
    {
      "entropy": 5.849544525146484,
      "epoch": 0.3056966618287373,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994779278815619,
      "loss": 5.7795,
      "mean_token_accuracy": 0.14996607154607772,
      "num_tokens": 6985363.0,
      "step": 3370
    },
    {
      "entropy": 5.838080978393554,
      "epoch": 0.3061502177068215,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994757226472736,
      "loss": 5.7216,
      "mean_token_accuracy": 0.16052751541137694,
      "num_tokens": 6997040.0,
      "step": 3375
    },
    {
      "entropy": 5.793620634078979,
      "epoch": 0.30660377358490565,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994735127707702,
      "loss": 5.6421,
      "mean_token_accuracy": 0.16104835271835327,
      "num_tokens": 7007591.0,
      "step": 3380
    },
    {
      "entropy": 5.819001531600952,
      "epoch": 0.30705732946298986,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994712982520972,
      "loss": 5.6422,
      "mean_token_accuracy": 0.16898961067199708,
      "num_tokens": 7017858.0,
      "step": 3385
    },
    {
      "entropy": 5.713206720352173,
      "epoch": 0.307510885341074,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994690790913008,
      "loss": 5.6155,
      "mean_token_accuracy": 0.16000766903162003,
      "num_tokens": 7027365.0,
      "step": 3390
    },
    {
      "entropy": 5.717849445343018,
      "epoch": 0.3079644412191582,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994668552884265,
      "loss": 5.6084,
      "mean_token_accuracy": 0.16479909121990205,
      "num_tokens": 7038728.0,
      "step": 3395
    },
    {
      "entropy": 5.835485458374023,
      "epoch": 0.3084179970972424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994646268435204,
      "loss": 5.5429,
      "mean_token_accuracy": 0.16616460382938386,
      "num_tokens": 7048736.0,
      "step": 3400
    },
    {
      "entropy": 5.722419404983521,
      "epoch": 0.30887155297532654,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994623937566286,
      "loss": 5.6685,
      "mean_token_accuracy": 0.16291871815919876,
      "num_tokens": 7059205.0,
      "step": 3405
    },
    {
      "entropy": 5.760222864151001,
      "epoch": 0.30932510885341075,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994601560277974,
      "loss": 5.667,
      "mean_token_accuracy": 0.15780892074108124,
      "num_tokens": 7069513.0,
      "step": 3410
    },
    {
      "entropy": 5.734495067596436,
      "epoch": 0.3097786647314949,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499457913657073,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1593753680586815,
      "num_tokens": 7079940.0,
      "step": 3415
    },
    {
      "entropy": 5.781790494918823,
      "epoch": 0.3102322206095791,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994556666445016,
      "loss": 5.5958,
      "mean_token_accuracy": 0.16559863835573196,
      "num_tokens": 7090792.0,
      "step": 3420
    },
    {
      "entropy": 5.704123544692993,
      "epoch": 0.31068577648766327,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994534149901299,
      "loss": 5.5471,
      "mean_token_accuracy": 0.16237700134515762,
      "num_tokens": 7101964.0,
      "step": 3425
    },
    {
      "entropy": 5.715693664550781,
      "epoch": 0.3111393323657475,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994511586940044,
      "loss": 5.6225,
      "mean_token_accuracy": 0.15924402177333832,
      "num_tokens": 7113332.0,
      "step": 3430
    },
    {
      "entropy": 5.885712957382202,
      "epoch": 0.31159288824383163,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994488977561717,
      "loss": 5.764,
      "mean_token_accuracy": 0.15799518972635268,
      "num_tokens": 7123481.0,
      "step": 3435
    },
    {
      "entropy": 5.681813049316406,
      "epoch": 0.31204644412191584,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994466321766786,
      "loss": 5.5308,
      "mean_token_accuracy": 0.16969044506549835,
      "num_tokens": 7133694.0,
      "step": 3440
    },
    {
      "entropy": 5.724149560928344,
      "epoch": 0.3125,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499444361955572,
      "loss": 5.6527,
      "mean_token_accuracy": 0.16183393001556395,
      "num_tokens": 7144491.0,
      "step": 3445
    },
    {
      "entropy": 5.882103443145752,
      "epoch": 0.31295355587808416,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994420870928988,
      "loss": 5.7266,
      "mean_token_accuracy": 0.15855416059494018,
      "num_tokens": 7154716.0,
      "step": 3450
    },
    {
      "entropy": 5.830023527145386,
      "epoch": 0.31340711175616837,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994398075887059,
      "loss": 5.7259,
      "mean_token_accuracy": 0.15234277099370958,
      "num_tokens": 7164258.0,
      "step": 3455
    },
    {
      "entropy": 5.782030057907105,
      "epoch": 0.3138606676342525,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994375234430407,
      "loss": 5.589,
      "mean_token_accuracy": 0.1607864186167717,
      "num_tokens": 7175098.0,
      "step": 3460
    },
    {
      "entropy": 5.742347669601441,
      "epoch": 0.31431422351233673,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004994352346559504,
      "loss": 5.6563,
      "mean_token_accuracy": 0.15974079370498656,
      "num_tokens": 7186480.0,
      "step": 3465
    },
    {
      "entropy": 5.664342737197876,
      "epoch": 0.3147677793904209,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994329412274821,
      "loss": 5.6363,
      "mean_token_accuracy": 0.1637357532978058,
      "num_tokens": 7197056.0,
      "step": 3470
    },
    {
      "entropy": 5.80676236152649,
      "epoch": 0.3152213352685051,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994306431576835,
      "loss": 5.7374,
      "mean_token_accuracy": 0.15914391726255417,
      "num_tokens": 7206040.0,
      "step": 3475
    },
    {
      "entropy": 5.898715686798096,
      "epoch": 0.31567489114658925,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499428340446602,
      "loss": 5.5995,
      "mean_token_accuracy": 0.1631159707903862,
      "num_tokens": 7216344.0,
      "step": 3480
    },
    {
      "entropy": 5.750396013259888,
      "epoch": 0.31612844702467346,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994260330942851,
      "loss": 5.6801,
      "mean_token_accuracy": 0.15558458715677262,
      "num_tokens": 7226729.0,
      "step": 3485
    },
    {
      "entropy": 5.676659345626831,
      "epoch": 0.3165820029027576,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994237211007808,
      "loss": 5.6213,
      "mean_token_accuracy": 0.16082685589790344,
      "num_tokens": 7237616.0,
      "step": 3490
    },
    {
      "entropy": 5.75124158859253,
      "epoch": 0.3170355587808418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994214044661367,
      "loss": 5.5436,
      "mean_token_accuracy": 0.16586074680089952,
      "num_tokens": 7247857.0,
      "step": 3495
    },
    {
      "entropy": 5.63331036567688,
      "epoch": 0.317489114658926,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994190831904007,
      "loss": 5.5215,
      "mean_token_accuracy": 0.16701960563659668,
      "num_tokens": 7257858.0,
      "step": 3500
    },
    {
      "entropy": 5.611927175521851,
      "epoch": 0.31794267053701014,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499416757273621,
      "loss": 5.539,
      "mean_token_accuracy": 0.16479599773883818,
      "num_tokens": 7268440.0,
      "step": 3505
    },
    {
      "entropy": 5.79567551612854,
      "epoch": 0.31839622641509435,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004994144267158454,
      "loss": 5.6125,
      "mean_token_accuracy": 0.16928928196430207,
      "num_tokens": 7278560.0,
      "step": 3510
    },
    {
      "entropy": 5.7546285629272464,
      "epoch": 0.3188497822931785,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994120915171224,
      "loss": 5.6208,
      "mean_token_accuracy": 0.16289258599281312,
      "num_tokens": 7288422.0,
      "step": 3515
    },
    {
      "entropy": 5.8093061447143555,
      "epoch": 0.3193033381712627,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994097516775002,
      "loss": 5.6798,
      "mean_token_accuracy": 0.16264722645282745,
      "num_tokens": 7298613.0,
      "step": 3520
    },
    {
      "entropy": 5.806758069992066,
      "epoch": 0.3197568940493469,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004994074071970271,
      "loss": 5.7208,
      "mean_token_accuracy": 0.15486309304833412,
      "num_tokens": 7309509.0,
      "step": 3525
    },
    {
      "entropy": 5.782179450988769,
      "epoch": 0.3202104499274311,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994050580757515,
      "loss": 5.6159,
      "mean_token_accuracy": 0.16160835027694703,
      "num_tokens": 7319015.0,
      "step": 3530
    },
    {
      "entropy": 5.694505119323731,
      "epoch": 0.32066400580551524,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994027043137223,
      "loss": 5.5828,
      "mean_token_accuracy": 0.16993721425533295,
      "num_tokens": 7329026.0,
      "step": 3535
    },
    {
      "entropy": 5.8108502388000485,
      "epoch": 0.3211175616835994,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994003459109879,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1579549044370651,
      "num_tokens": 7339638.0,
      "step": 3540
    },
    {
      "entropy": 5.797452449798584,
      "epoch": 0.3215711175616836,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993979828675972,
      "loss": 5.5935,
      "mean_token_accuracy": 0.16423679292201995,
      "num_tokens": 7349979.0,
      "step": 3545
    },
    {
      "entropy": 5.655317640304565,
      "epoch": 0.32202467343976776,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499395615183599,
      "loss": 5.5311,
      "mean_token_accuracy": 0.1711718648672104,
      "num_tokens": 7359398.0,
      "step": 3550
    },
    {
      "entropy": 5.670080709457397,
      "epoch": 0.32247822931785197,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993932428590422,
      "loss": 5.5775,
      "mean_token_accuracy": 0.1649343878030777,
      "num_tokens": 7370418.0,
      "step": 3555
    },
    {
      "entropy": 5.7741282939910885,
      "epoch": 0.3229317851959361,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993908658939762,
      "loss": 5.6027,
      "mean_token_accuracy": 0.15476736575365066,
      "num_tokens": 7380461.0,
      "step": 3560
    },
    {
      "entropy": 5.758694982528686,
      "epoch": 0.32338534107402034,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993884842884496,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1681668609380722,
      "num_tokens": 7390646.0,
      "step": 3565
    },
    {
      "entropy": 5.743930101394653,
      "epoch": 0.3238388969521045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993860980425121,
      "loss": 5.7011,
      "mean_token_accuracy": 0.16390874236822128,
      "num_tokens": 7402106.0,
      "step": 3570
    },
    {
      "entropy": 5.784041881561279,
      "epoch": 0.3242924528301887,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993837071562131,
      "loss": 5.6979,
      "mean_token_accuracy": 0.16273627281188965,
      "num_tokens": 7412936.0,
      "step": 3575
    },
    {
      "entropy": 5.752962303161621,
      "epoch": 0.32474600870827286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993813116296017,
      "loss": 5.5556,
      "mean_token_accuracy": 0.16882994323968886,
      "num_tokens": 7422303.0,
      "step": 3580
    },
    {
      "entropy": 5.709475040435791,
      "epoch": 0.325199564586357,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993789114627275,
      "loss": 5.6289,
      "mean_token_accuracy": 0.1605585351586342,
      "num_tokens": 7432829.0,
      "step": 3585
    },
    {
      "entropy": 5.85622272491455,
      "epoch": 0.3256531204644412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993765066556403,
      "loss": 5.6422,
      "mean_token_accuracy": 0.16322125345468522,
      "num_tokens": 7443218.0,
      "step": 3590
    },
    {
      "entropy": 5.7688047885894775,
      "epoch": 0.3261066763425254,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993740972083899,
      "loss": 5.6019,
      "mean_token_accuracy": 0.16230728030204772,
      "num_tokens": 7454531.0,
      "step": 3595
    },
    {
      "entropy": 5.730414962768554,
      "epoch": 0.3265602322206096,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993716831210259,
      "loss": 5.6645,
      "mean_token_accuracy": 0.16234444826841354,
      "num_tokens": 7464993.0,
      "step": 3600
    },
    {
      "entropy": 5.69191312789917,
      "epoch": 0.32701378809869375,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993692643935983,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1614596240222454,
      "num_tokens": 7475755.0,
      "step": 3605
    },
    {
      "entropy": 5.772009658813476,
      "epoch": 0.32746734397677796,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993668410261571,
      "loss": 5.6709,
      "mean_token_accuracy": 0.1652410715818405,
      "num_tokens": 7486778.0,
      "step": 3610
    },
    {
      "entropy": 5.873863935470581,
      "epoch": 0.3279208998548621,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993644130187525,
      "loss": 5.6487,
      "mean_token_accuracy": 0.16203359365463257,
      "num_tokens": 7497439.0,
      "step": 3615
    },
    {
      "entropy": 5.71992974281311,
      "epoch": 0.3283744557329463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993619803714347,
      "loss": 5.6797,
      "mean_token_accuracy": 0.1641794115304947,
      "num_tokens": 7508274.0,
      "step": 3620
    },
    {
      "entropy": 5.724187231063842,
      "epoch": 0.3288280116110305,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993595430842539,
      "loss": 5.6626,
      "mean_token_accuracy": 0.15977796465158461,
      "num_tokens": 7519315.0,
      "step": 3625
    },
    {
      "entropy": 5.81933822631836,
      "epoch": 0.32928156748911463,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993571011572606,
      "loss": 5.6515,
      "mean_token_accuracy": 0.16541533768177033,
      "num_tokens": 7529910.0,
      "step": 3630
    },
    {
      "entropy": 5.755266094207764,
      "epoch": 0.32973512336719885,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993546545905053,
      "loss": 5.5745,
      "mean_token_accuracy": 0.16873921304941178,
      "num_tokens": 7538966.0,
      "step": 3635
    },
    {
      "entropy": 5.755073690414429,
      "epoch": 0.330188679245283,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993522033840386,
      "loss": 5.6146,
      "mean_token_accuracy": 0.158041512966156,
      "num_tokens": 7549683.0,
      "step": 3640
    },
    {
      "entropy": 5.700992918014526,
      "epoch": 0.3306422351233672,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499349747537911,
      "loss": 5.5953,
      "mean_token_accuracy": 0.1624542772769928,
      "num_tokens": 7560629.0,
      "step": 3645
    },
    {
      "entropy": 5.730870103836059,
      "epoch": 0.33109579100145137,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993472870521736,
      "loss": 5.5385,
      "mean_token_accuracy": 0.17402430772781372,
      "num_tokens": 7570613.0,
      "step": 3650
    },
    {
      "entropy": 5.715100336074829,
      "epoch": 0.3315493468795356,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993448219268771,
      "loss": 5.5211,
      "mean_token_accuracy": 0.16446274071931838,
      "num_tokens": 7580640.0,
      "step": 3655
    },
    {
      "entropy": 5.743779134750366,
      "epoch": 0.33200290275761973,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993423521620726,
      "loss": 5.6224,
      "mean_token_accuracy": 0.16669037342071533,
      "num_tokens": 7590164.0,
      "step": 3660
    },
    {
      "entropy": 5.77204122543335,
      "epoch": 0.33245645863570394,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993398777578109,
      "loss": 5.6652,
      "mean_token_accuracy": 0.16001658663153648,
      "num_tokens": 7600173.0,
      "step": 3665
    },
    {
      "entropy": 5.65158109664917,
      "epoch": 0.3329100145137881,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993373987141435,
      "loss": 5.4607,
      "mean_token_accuracy": 0.17389463186264037,
      "num_tokens": 7609958.0,
      "step": 3670
    },
    {
      "entropy": 5.784420680999756,
      "epoch": 0.33336357039187225,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993349150311214,
      "loss": 5.6059,
      "mean_token_accuracy": 0.16194465905427932,
      "num_tokens": 7620237.0,
      "step": 3675
    },
    {
      "entropy": 5.6676253318786625,
      "epoch": 0.33381712626995647,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004993324267087962,
      "loss": 5.5129,
      "mean_token_accuracy": 0.1702829569578171,
      "num_tokens": 7629886.0,
      "step": 3680
    },
    {
      "entropy": 5.706976079940796,
      "epoch": 0.3342706821480406,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993299337472192,
      "loss": 5.5823,
      "mean_token_accuracy": 0.16765174865722657,
      "num_tokens": 7640057.0,
      "step": 3685
    },
    {
      "entropy": 5.674705648422242,
      "epoch": 0.33472423802612483,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499327436146442,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16337296962738038,
      "num_tokens": 7649766.0,
      "step": 3690
    },
    {
      "entropy": 5.715948963165284,
      "epoch": 0.335177793904209,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993249339065162,
      "loss": 5.607,
      "mean_token_accuracy": 0.16630489975214005,
      "num_tokens": 7660890.0,
      "step": 3695
    },
    {
      "entropy": 5.735551595687866,
      "epoch": 0.3356313497822932,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004993224270274937,
      "loss": 5.63,
      "mean_token_accuracy": 0.16649477332830429,
      "num_tokens": 7670737.0,
      "step": 3700
    },
    {
      "entropy": 5.690478563308716,
      "epoch": 0.33608490566037735,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004993199155094263,
      "loss": 5.5932,
      "mean_token_accuracy": 0.16307836025953293,
      "num_tokens": 7681407.0,
      "step": 3705
    },
    {
      "entropy": 5.706917667388916,
      "epoch": 0.33653846153846156,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993173993523658,
      "loss": 5.5156,
      "mean_token_accuracy": 0.17325208634138106,
      "num_tokens": 7691758.0,
      "step": 3710
    },
    {
      "entropy": 5.671240568161011,
      "epoch": 0.3369920174165457,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993148785563643,
      "loss": 5.5515,
      "mean_token_accuracy": 0.16601790338754654,
      "num_tokens": 7702934.0,
      "step": 3715
    },
    {
      "entropy": 5.69557638168335,
      "epoch": 0.3374455732946299,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499312353121474,
      "loss": 5.5392,
      "mean_token_accuracy": 0.1636522337794304,
      "num_tokens": 7713001.0,
      "step": 3720
    },
    {
      "entropy": 5.736032581329345,
      "epoch": 0.3378991291727141,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993098230477471,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1599409118294716,
      "num_tokens": 7724005.0,
      "step": 3725
    },
    {
      "entropy": 5.748565006256103,
      "epoch": 0.33835268505079824,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993072883352358,
      "loss": 5.5614,
      "mean_token_accuracy": 0.1650003969669342,
      "num_tokens": 7733743.0,
      "step": 3730
    },
    {
      "entropy": 5.813916301727295,
      "epoch": 0.33880624092888245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993047489839927,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1575533241033554,
      "num_tokens": 7743762.0,
      "step": 3735
    },
    {
      "entropy": 5.814930248260498,
      "epoch": 0.3392597968069666,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993022049940702,
      "loss": 5.6361,
      "mean_token_accuracy": 0.16483197808265687,
      "num_tokens": 7753758.0,
      "step": 3740
    },
    {
      "entropy": 5.688127851486206,
      "epoch": 0.3397133526850508,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992996563655209,
      "loss": 5.5699,
      "mean_token_accuracy": 0.165471251308918,
      "num_tokens": 7764692.0,
      "step": 3745
    },
    {
      "entropy": 5.683252620697021,
      "epoch": 0.340166908563135,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992971030983975,
      "loss": 5.6076,
      "mean_token_accuracy": 0.15619706213474274,
      "num_tokens": 7776892.0,
      "step": 3750
    },
    {
      "entropy": 5.818699216842651,
      "epoch": 0.3406204644412192,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992945451927531,
      "loss": 5.6558,
      "mean_token_accuracy": 0.16157331615686416,
      "num_tokens": 7787143.0,
      "step": 3755
    },
    {
      "entropy": 5.6922383308410645,
      "epoch": 0.34107402031930334,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992919826486402,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16458574086427688,
      "num_tokens": 7798537.0,
      "step": 3760
    },
    {
      "entropy": 5.70722393989563,
      "epoch": 0.3415275761973875,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992894154661118,
      "loss": 5.5466,
      "mean_token_accuracy": 0.16931652128696442,
      "num_tokens": 7808434.0,
      "step": 3765
    },
    {
      "entropy": 5.727777719497681,
      "epoch": 0.3419811320754717,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992868436452213,
      "loss": 5.6646,
      "mean_token_accuracy": 0.16223400980234146,
      "num_tokens": 7819592.0,
      "step": 3770
    },
    {
      "entropy": 5.751184320449829,
      "epoch": 0.34243468795355586,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004992842671860217,
      "loss": 5.5501,
      "mean_token_accuracy": 0.16743091493844986,
      "num_tokens": 7830602.0,
      "step": 3775
    },
    {
      "entropy": 5.776014566421509,
      "epoch": 0.34288824383164007,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992816860885662,
      "loss": 5.6971,
      "mean_token_accuracy": 0.16009064316749572,
      "num_tokens": 7840850.0,
      "step": 3780
    },
    {
      "entropy": 5.83824052810669,
      "epoch": 0.3433417997097242,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992791003529084,
      "loss": 5.6309,
      "mean_token_accuracy": 0.1558419495820999,
      "num_tokens": 7852371.0,
      "step": 3785
    },
    {
      "entropy": 5.6923058986663815,
      "epoch": 0.34379535558780844,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992765099791017,
      "loss": 5.6581,
      "mean_token_accuracy": 0.15808210968971254,
      "num_tokens": 7863290.0,
      "step": 3790
    },
    {
      "entropy": 5.6762631893157955,
      "epoch": 0.3442489114658926,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992739149671995,
      "loss": 5.4947,
      "mean_token_accuracy": 0.168349090218544,
      "num_tokens": 7873927.0,
      "step": 3795
    },
    {
      "entropy": 5.698384523391724,
      "epoch": 0.3447024673439768,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992713153172556,
      "loss": 5.5329,
      "mean_token_accuracy": 0.16461328268051148,
      "num_tokens": 7884977.0,
      "step": 3800
    },
    {
      "entropy": 5.671432590484619,
      "epoch": 0.34515602322206096,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992687110293238,
      "loss": 5.6109,
      "mean_token_accuracy": 0.16780710071325303,
      "num_tokens": 7894620.0,
      "step": 3805
    },
    {
      "entropy": 5.681638097763061,
      "epoch": 0.3456095791001451,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992661021034579,
      "loss": 5.5377,
      "mean_token_accuracy": 0.17044852972030639,
      "num_tokens": 7904541.0,
      "step": 3810
    },
    {
      "entropy": 5.677857828140259,
      "epoch": 0.3460631349782293,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004992634885397118,
      "loss": 5.5551,
      "mean_token_accuracy": 0.16660392731428147,
      "num_tokens": 7915435.0,
      "step": 3815
    },
    {
      "entropy": 5.675379896163941,
      "epoch": 0.3465166908563135,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992608703381396,
      "loss": 5.54,
      "mean_token_accuracy": 0.1684734731912613,
      "num_tokens": 7925870.0,
      "step": 3820
    },
    {
      "entropy": 5.698918628692627,
      "epoch": 0.3469702467343977,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004992582474987955,
      "loss": 5.541,
      "mean_token_accuracy": 0.1659692794084549,
      "num_tokens": 7937046.0,
      "step": 3825
    },
    {
      "entropy": 5.677004718780518,
      "epoch": 0.34742380261248185,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992556200217337,
      "loss": 5.4992,
      "mean_token_accuracy": 0.1643098548054695,
      "num_tokens": 7947150.0,
      "step": 3830
    },
    {
      "entropy": 5.739584589004517,
      "epoch": 0.34787735849056606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004992529879070085,
      "loss": 5.6016,
      "mean_token_accuracy": 0.1661156862974167,
      "num_tokens": 7956815.0,
      "step": 3835
    },
    {
      "entropy": 5.73980484008789,
      "epoch": 0.3483309143686502,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992503511546745,
      "loss": 5.6341,
      "mean_token_accuracy": 0.16282225549221038,
      "num_tokens": 7968114.0,
      "step": 3840
    },
    {
      "entropy": 5.691490125656128,
      "epoch": 0.3487844702467344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992477097647859,
      "loss": 5.5408,
      "mean_token_accuracy": 0.16325437575578688,
      "num_tokens": 7977840.0,
      "step": 3845
    },
    {
      "entropy": 5.76722149848938,
      "epoch": 0.3492380261248186,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004992450637373976,
      "loss": 5.6678,
      "mean_token_accuracy": 0.1663844257593155,
      "num_tokens": 7988663.0,
      "step": 3850
    },
    {
      "entropy": 5.655972480773926,
      "epoch": 0.34969158200290273,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992424130725642,
      "loss": 5.48,
      "mean_token_accuracy": 0.1633950486779213,
      "num_tokens": 7999693.0,
      "step": 3855
    },
    {
      "entropy": 5.750917959213257,
      "epoch": 0.35014513788098695,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992397577703406,
      "loss": 5.6568,
      "mean_token_accuracy": 0.15703431218862535,
      "num_tokens": 8010154.0,
      "step": 3860
    },
    {
      "entropy": 5.7925513744354244,
      "epoch": 0.3505986937590711,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004992370978307816,
      "loss": 5.6324,
      "mean_token_accuracy": 0.16514518857002258,
      "num_tokens": 8020921.0,
      "step": 3865
    },
    {
      "entropy": 5.6815894603729244,
      "epoch": 0.3510522496371553,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992344332539424,
      "loss": 5.5597,
      "mean_token_accuracy": 0.1687205910682678,
      "num_tokens": 8031404.0,
      "step": 3870
    },
    {
      "entropy": 5.654696989059448,
      "epoch": 0.35150580551523947,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992317640398779,
      "loss": 5.57,
      "mean_token_accuracy": 0.16497284919023514,
      "num_tokens": 8041310.0,
      "step": 3875
    },
    {
      "entropy": 5.681327152252197,
      "epoch": 0.3519593613933237,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992290901886434,
      "loss": 5.559,
      "mean_token_accuracy": 0.16811044812202453,
      "num_tokens": 8051054.0,
      "step": 3880
    },
    {
      "entropy": 5.717649364471436,
      "epoch": 0.35241291727140783,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499226411700294,
      "loss": 5.5503,
      "mean_token_accuracy": 0.16839763820171355,
      "num_tokens": 8061063.0,
      "step": 3885
    },
    {
      "entropy": 5.685747528076172,
      "epoch": 0.35286647314949204,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004992237285748855,
      "loss": 5.5605,
      "mean_token_accuracy": 0.1663712427020073,
      "num_tokens": 8071715.0,
      "step": 3890
    },
    {
      "entropy": 5.768758487701416,
      "epoch": 0.3533200290275762,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004992210408124731,
      "loss": 5.6295,
      "mean_token_accuracy": 0.156692710518837,
      "num_tokens": 8082679.0,
      "step": 3895
    },
    {
      "entropy": 5.718517923355103,
      "epoch": 0.35377358490566035,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004992183484131123,
      "loss": 5.6508,
      "mean_token_accuracy": 0.1636080786585808,
      "num_tokens": 8093785.0,
      "step": 3900
    },
    {
      "entropy": 5.654246377944946,
      "epoch": 0.35422714078374457,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992156513768591,
      "loss": 5.5534,
      "mean_token_accuracy": 0.16666727364063263,
      "num_tokens": 8105822.0,
      "step": 3905
    },
    {
      "entropy": 5.649802350997925,
      "epoch": 0.3546806966618287,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499212949703769,
      "loss": 5.4627,
      "mean_token_accuracy": 0.17031579464673996,
      "num_tokens": 8115521.0,
      "step": 3910
    },
    {
      "entropy": 5.711194086074829,
      "epoch": 0.35513425253991293,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004992102433938981,
      "loss": 5.5528,
      "mean_token_accuracy": 0.1639755219221115,
      "num_tokens": 8125524.0,
      "step": 3915
    },
    {
      "entropy": 5.670215368270874,
      "epoch": 0.3555878084179971,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992075324473021,
      "loss": 5.4705,
      "mean_token_accuracy": 0.17687036842107773,
      "num_tokens": 8135552.0,
      "step": 3920
    },
    {
      "entropy": 5.7498579025268555,
      "epoch": 0.3560413642960813,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004992048168640371,
      "loss": 5.5912,
      "mean_token_accuracy": 0.16968106478452682,
      "num_tokens": 8145617.0,
      "step": 3925
    },
    {
      "entropy": 5.6834397315979,
      "epoch": 0.35649492017416545,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004992020966441595,
      "loss": 5.6008,
      "mean_token_accuracy": 0.1607098937034607,
      "num_tokens": 8156729.0,
      "step": 3930
    },
    {
      "entropy": 5.697573709487915,
      "epoch": 0.35694847605224966,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991993717877254,
      "loss": 5.5865,
      "mean_token_accuracy": 0.16268351525068284,
      "num_tokens": 8167637.0,
      "step": 3935
    },
    {
      "entropy": 5.648408317565918,
      "epoch": 0.3574020319303338,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991966422947911,
      "loss": 5.5628,
      "mean_token_accuracy": 0.170132377743721,
      "num_tokens": 8177354.0,
      "step": 3940
    },
    {
      "entropy": 5.63679690361023,
      "epoch": 0.357855587808418,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991939081654132,
      "loss": 5.5381,
      "mean_token_accuracy": 0.1754870280623436,
      "num_tokens": 8187580.0,
      "step": 3945
    },
    {
      "entropy": 5.660647439956665,
      "epoch": 0.3583091436865022,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499191169399648,
      "loss": 5.6353,
      "mean_token_accuracy": 0.15820996463298798,
      "num_tokens": 8199382.0,
      "step": 3950
    },
    {
      "entropy": 5.687941408157348,
      "epoch": 0.35876269956458634,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004991884259975525,
      "loss": 5.5178,
      "mean_token_accuracy": 0.16903046071529387,
      "num_tokens": 8208484.0,
      "step": 3955
    },
    {
      "entropy": 5.708724451065064,
      "epoch": 0.35921625544267055,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499185677959183,
      "loss": 5.5645,
      "mean_token_accuracy": 0.16452753990888597,
      "num_tokens": 8219056.0,
      "step": 3960
    },
    {
      "entropy": 5.7164100170135494,
      "epoch": 0.3596698113207547,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991829252845966,
      "loss": 5.5797,
      "mean_token_accuracy": 0.1641779735684395,
      "num_tokens": 8230149.0,
      "step": 3965
    },
    {
      "entropy": 5.706824064254761,
      "epoch": 0.3601233671988389,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991801679738502,
      "loss": 5.6529,
      "mean_token_accuracy": 0.15674541145563126,
      "num_tokens": 8240803.0,
      "step": 3970
    },
    {
      "entropy": 5.706471729278564,
      "epoch": 0.3605769230769231,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991774060270008,
      "loss": 5.5946,
      "mean_token_accuracy": 0.16189683228731155,
      "num_tokens": 8251563.0,
      "step": 3975
    },
    {
      "entropy": 5.671058511734008,
      "epoch": 0.3610304789550073,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004991746394441055,
      "loss": 5.5315,
      "mean_token_accuracy": 0.1629596546292305,
      "num_tokens": 8261342.0,
      "step": 3980
    },
    {
      "entropy": 5.712836742401123,
      "epoch": 0.36148403483309144,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991718682252216,
      "loss": 5.6412,
      "mean_token_accuracy": 0.1584403395652771,
      "num_tokens": 8271540.0,
      "step": 3985
    },
    {
      "entropy": 5.757377099990845,
      "epoch": 0.3619375907111756,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991690923704063,
      "loss": 5.5587,
      "mean_token_accuracy": 0.16804730147123337,
      "num_tokens": 8280593.0,
      "step": 3990
    },
    {
      "entropy": 5.687913990020752,
      "epoch": 0.3623911465892598,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991663118797171,
      "loss": 5.5785,
      "mean_token_accuracy": 0.16617974787950515,
      "num_tokens": 8290126.0,
      "step": 3995
    },
    {
      "entropy": 5.699881505966187,
      "epoch": 0.36284470246734396,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004991635267532113,
      "loss": 5.5149,
      "mean_token_accuracy": 0.1671944186091423,
      "num_tokens": 8301416.0,
      "step": 4000
    },
    {
      "entropy": 5.742449140548706,
      "epoch": 0.36329825834542817,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991607369909468,
      "loss": 5.59,
      "mean_token_accuracy": 0.16540921181440355,
      "num_tokens": 8311034.0,
      "step": 4005
    },
    {
      "entropy": 5.688007831573486,
      "epoch": 0.3637518142235123,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991579425929811,
      "loss": 5.6005,
      "mean_token_accuracy": 0.16405065953731537,
      "num_tokens": 8322203.0,
      "step": 4010
    },
    {
      "entropy": 5.6758204936981205,
      "epoch": 0.36420537010159654,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499155143559372,
      "loss": 5.4789,
      "mean_token_accuracy": 0.17067255973815917,
      "num_tokens": 8330981.0,
      "step": 4015
    },
    {
      "entropy": 5.636501121520996,
      "epoch": 0.3646589259796807,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991523398901773,
      "loss": 5.5188,
      "mean_token_accuracy": 0.16740261018276215,
      "num_tokens": 8341294.0,
      "step": 4020
    },
    {
      "entropy": 5.790857219696045,
      "epoch": 0.3651124818577649,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004991495315854553,
      "loss": 5.6633,
      "mean_token_accuracy": 0.1621541276574135,
      "num_tokens": 8351824.0,
      "step": 4025
    },
    {
      "entropy": 5.688350629806519,
      "epoch": 0.36556603773584906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991467186452637,
      "loss": 5.5408,
      "mean_token_accuracy": 0.167211352288723,
      "num_tokens": 8361975.0,
      "step": 4030
    },
    {
      "entropy": 5.64109468460083,
      "epoch": 0.3660195936139332,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991439010696609,
      "loss": 5.482,
      "mean_token_accuracy": 0.16923303604125978,
      "num_tokens": 8372145.0,
      "step": 4035
    },
    {
      "entropy": 5.672000598907471,
      "epoch": 0.3664731494920174,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991410788587051,
      "loss": 5.5725,
      "mean_token_accuracy": 0.16295102685689927,
      "num_tokens": 8382542.0,
      "step": 4040
    },
    {
      "entropy": 5.721254110336304,
      "epoch": 0.3669267053701016,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991382520124547,
      "loss": 5.572,
      "mean_token_accuracy": 0.16192779392004014,
      "num_tokens": 8393450.0,
      "step": 4045
    },
    {
      "entropy": 5.723563814163208,
      "epoch": 0.3673802612481858,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991354205309681,
      "loss": 5.5075,
      "mean_token_accuracy": 0.16757548451423646,
      "num_tokens": 8404929.0,
      "step": 4050
    },
    {
      "entropy": 5.734829044342041,
      "epoch": 0.36783381712626995,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499132584414304,
      "loss": 5.5422,
      "mean_token_accuracy": 0.16685086190700532,
      "num_tokens": 8415703.0,
      "step": 4055
    },
    {
      "entropy": 5.728091859817505,
      "epoch": 0.36828737300435416,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991297436625207,
      "loss": 5.6072,
      "mean_token_accuracy": 0.16412289291620255,
      "num_tokens": 8426220.0,
      "step": 4060
    },
    {
      "entropy": 5.648223352432251,
      "epoch": 0.3687409288824383,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991268982756774,
      "loss": 5.4911,
      "mean_token_accuracy": 0.16989713162183762,
      "num_tokens": 8436220.0,
      "step": 4065
    },
    {
      "entropy": 5.782150840759277,
      "epoch": 0.3691944847605225,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991240482538327,
      "loss": 5.6445,
      "mean_token_accuracy": 0.15544479340314865,
      "num_tokens": 8446724.0,
      "step": 4070
    },
    {
      "entropy": 5.689100742340088,
      "epoch": 0.3696480406386067,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004991211935970455,
      "loss": 5.5028,
      "mean_token_accuracy": 0.16541081219911574,
      "num_tokens": 8457270.0,
      "step": 4075
    },
    {
      "entropy": 5.654791688919067,
      "epoch": 0.37010159651669083,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991183343053748,
      "loss": 5.5721,
      "mean_token_accuracy": 0.15914703905582428,
      "num_tokens": 8468137.0,
      "step": 4080
    },
    {
      "entropy": 5.730112028121948,
      "epoch": 0.37055515239477504,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00049911547037888,
      "loss": 5.5037,
      "mean_token_accuracy": 0.17106323689222336,
      "num_tokens": 8477574.0,
      "step": 4085
    },
    {
      "entropy": 5.67306604385376,
      "epoch": 0.3710087082728592,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004991126018176202,
      "loss": 5.5959,
      "mean_token_accuracy": 0.16015791594982148,
      "num_tokens": 8487268.0,
      "step": 4090
    },
    {
      "entropy": 5.792642593383789,
      "epoch": 0.3714622641509434,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991097286216545,
      "loss": 5.6546,
      "mean_token_accuracy": 0.16796909868717194,
      "num_tokens": 8497785.0,
      "step": 4095
    },
    {
      "entropy": 5.751044225692749,
      "epoch": 0.37191582002902757,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991068507910427,
      "loss": 5.5268,
      "mean_token_accuracy": 0.1661210760474205,
      "num_tokens": 8508169.0,
      "step": 4100
    },
    {
      "entropy": 5.6846167087554935,
      "epoch": 0.3723693759071118,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499103968325844,
      "loss": 5.4959,
      "mean_token_accuracy": 0.16699425280094146,
      "num_tokens": 8519201.0,
      "step": 4105
    },
    {
      "entropy": 5.7432060718536375,
      "epoch": 0.37282293178519593,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004991010812261181,
      "loss": 5.6047,
      "mean_token_accuracy": 0.1590347096323967,
      "num_tokens": 8531093.0,
      "step": 4110
    },
    {
      "entropy": 5.601396274566651,
      "epoch": 0.37327648766328014,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990981894919248,
      "loss": 5.474,
      "mean_token_accuracy": 0.17184109836816788,
      "num_tokens": 8541730.0,
      "step": 4115
    },
    {
      "entropy": 5.6851277351379395,
      "epoch": 0.3737300435413643,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990952931233238,
      "loss": 5.5322,
      "mean_token_accuracy": 0.16669057607650756,
      "num_tokens": 8552570.0,
      "step": 4120
    },
    {
      "entropy": 5.711943817138672,
      "epoch": 0.37418359941944845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499092392120375,
      "loss": 5.4496,
      "mean_token_accuracy": 0.16830459237098694,
      "num_tokens": 8562340.0,
      "step": 4125
    },
    {
      "entropy": 5.690545225143433,
      "epoch": 0.37463715529753266,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004990894864831385,
      "loss": 5.6457,
      "mean_token_accuracy": 0.16141000092029573,
      "num_tokens": 8573240.0,
      "step": 4130
    },
    {
      "entropy": 5.74811692237854,
      "epoch": 0.3750907111756168,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990865762116743,
      "loss": 5.5008,
      "mean_token_accuracy": 0.17035448253154756,
      "num_tokens": 8583194.0,
      "step": 4135
    },
    {
      "entropy": 5.673021078109741,
      "epoch": 0.37554426705370103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990836613060426,
      "loss": 5.5799,
      "mean_token_accuracy": 0.1630268007516861,
      "num_tokens": 8593603.0,
      "step": 4140
    },
    {
      "entropy": 5.630602216720581,
      "epoch": 0.3759978229317852,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990807417663036,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1646374523639679,
      "num_tokens": 8603176.0,
      "step": 4145
    },
    {
      "entropy": 5.777580118179321,
      "epoch": 0.3764513788098694,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004990778175925179,
      "loss": 5.596,
      "mean_token_accuracy": 0.1675265073776245,
      "num_tokens": 8614801.0,
      "step": 4150
    },
    {
      "entropy": 5.7292474746704105,
      "epoch": 0.37690493468795355,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004990748887847457,
      "loss": 5.6252,
      "mean_token_accuracy": 0.16409169584512712,
      "num_tokens": 8624999.0,
      "step": 4155
    },
    {
      "entropy": 5.659530401229858,
      "epoch": 0.37735849056603776,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990719553430477,
      "loss": 5.4784,
      "mean_token_accuracy": 0.17621796131134032,
      "num_tokens": 8635358.0,
      "step": 4160
    },
    {
      "entropy": 5.682961511611938,
      "epoch": 0.3778120464441219,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990690172674847,
      "loss": 5.5547,
      "mean_token_accuracy": 0.1646416664123535,
      "num_tokens": 8645430.0,
      "step": 4165
    },
    {
      "entropy": 5.717040538787842,
      "epoch": 0.3782656023222061,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990660745581172,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1789042666554451,
      "num_tokens": 8655115.0,
      "step": 4170
    },
    {
      "entropy": 5.636549854278565,
      "epoch": 0.3787191582002903,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990631272150062,
      "loss": 5.5041,
      "mean_token_accuracy": 0.16583775132894515,
      "num_tokens": 8665391.0,
      "step": 4175
    },
    {
      "entropy": 5.702080345153808,
      "epoch": 0.37917271407837444,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004990601752382127,
      "loss": 5.5887,
      "mean_token_accuracy": 0.16532650887966155,
      "num_tokens": 8676819.0,
      "step": 4180
    },
    {
      "entropy": 5.778425836563111,
      "epoch": 0.37962626995645865,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990572186277975,
      "loss": 5.5227,
      "mean_token_accuracy": 0.171260467171669,
      "num_tokens": 8687052.0,
      "step": 4185
    },
    {
      "entropy": 5.689051246643066,
      "epoch": 0.3800798258345428,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990542573838221,
      "loss": 5.5942,
      "mean_token_accuracy": 0.1582357779145241,
      "num_tokens": 8697756.0,
      "step": 4190
    },
    {
      "entropy": 5.596316719055176,
      "epoch": 0.380533381712627,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990512915063475,
      "loss": 5.5045,
      "mean_token_accuracy": 0.1715959295630455,
      "num_tokens": 8708388.0,
      "step": 4195
    },
    {
      "entropy": 5.747104120254517,
      "epoch": 0.38098693759071117,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990483209954352,
      "loss": 5.5915,
      "mean_token_accuracy": 0.1606937661767006,
      "num_tokens": 8719475.0,
      "step": 4200
    },
    {
      "entropy": 5.6669543266296385,
      "epoch": 0.3814404934687954,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004990453458511464,
      "loss": 5.454,
      "mean_token_accuracy": 0.17628490179777145,
      "num_tokens": 8729395.0,
      "step": 4205
    },
    {
      "entropy": 5.690093278884888,
      "epoch": 0.38189404934687954,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004990423660735428,
      "loss": 5.5132,
      "mean_token_accuracy": 0.17369182407855988,
      "num_tokens": 8740351.0,
      "step": 4210
    },
    {
      "entropy": 5.721720170974732,
      "epoch": 0.3823476052249637,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499039381662686,
      "loss": 5.5275,
      "mean_token_accuracy": 0.1712048977613449,
      "num_tokens": 8749260.0,
      "step": 4215
    },
    {
      "entropy": 5.719943189620972,
      "epoch": 0.3828011611030479,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004990363926186377,
      "loss": 5.6179,
      "mean_token_accuracy": 0.15975076854228973,
      "num_tokens": 8759604.0,
      "step": 4220
    },
    {
      "entropy": 5.750780868530273,
      "epoch": 0.38325471698113206,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990333989414597,
      "loss": 5.6349,
      "mean_token_accuracy": 0.15848522633314133,
      "num_tokens": 8769815.0,
      "step": 4225
    },
    {
      "entropy": 5.646167278289795,
      "epoch": 0.38370827285921627,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990304006312138,
      "loss": 5.5614,
      "mean_token_accuracy": 0.16491910070180893,
      "num_tokens": 8780557.0,
      "step": 4230
    },
    {
      "entropy": 5.693453121185303,
      "epoch": 0.3841618287373004,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990273976879622,
      "loss": 5.5145,
      "mean_token_accuracy": 0.17479519844055175,
      "num_tokens": 8790993.0,
      "step": 4235
    },
    {
      "entropy": 5.70437707901001,
      "epoch": 0.38461538461538464,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990243901117669,
      "loss": 5.5828,
      "mean_token_accuracy": 0.15901805460453033,
      "num_tokens": 8800657.0,
      "step": 4240
    },
    {
      "entropy": 5.6222734451293945,
      "epoch": 0.3850689404934688,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990213779026903,
      "loss": 5.5851,
      "mean_token_accuracy": 0.1631401777267456,
      "num_tokens": 8812983.0,
      "step": 4245
    },
    {
      "entropy": 5.734277868270874,
      "epoch": 0.385522496371553,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990183610607944,
      "loss": 5.543,
      "mean_token_accuracy": 0.16615747958421706,
      "num_tokens": 8822980.0,
      "step": 4250
    },
    {
      "entropy": 5.662422704696655,
      "epoch": 0.38597605224963716,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990153395861415,
      "loss": 5.5319,
      "mean_token_accuracy": 0.1691455915570259,
      "num_tokens": 8833821.0,
      "step": 4255
    },
    {
      "entropy": 5.62187876701355,
      "epoch": 0.3864296081277213,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990123134787946,
      "loss": 5.4695,
      "mean_token_accuracy": 0.170258067548275,
      "num_tokens": 8843929.0,
      "step": 4260
    },
    {
      "entropy": 5.683696556091308,
      "epoch": 0.3868831640058055,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004990092827388159,
      "loss": 5.4591,
      "mean_token_accuracy": 0.17503607273101807,
      "num_tokens": 8854180.0,
      "step": 4265
    },
    {
      "entropy": 5.589134645462036,
      "epoch": 0.3873367198838897,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004990062473662681,
      "loss": 5.463,
      "mean_token_accuracy": 0.1750032678246498,
      "num_tokens": 8865018.0,
      "step": 4270
    },
    {
      "entropy": 5.705612516403198,
      "epoch": 0.3877902757619739,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499003207361214,
      "loss": 5.6611,
      "mean_token_accuracy": 0.16093496978282928,
      "num_tokens": 8875423.0,
      "step": 4275
    },
    {
      "entropy": 5.781583690643311,
      "epoch": 0.38824383164005805,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004990001627237165,
      "loss": 5.5854,
      "mean_token_accuracy": 0.1674569606781006,
      "num_tokens": 8887850.0,
      "step": 4280
    },
    {
      "entropy": 5.6495767593383786,
      "epoch": 0.38869738751814226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989971134538386,
      "loss": 5.4845,
      "mean_token_accuracy": 0.17036610543727876,
      "num_tokens": 8897429.0,
      "step": 4285
    },
    {
      "entropy": 5.62213659286499,
      "epoch": 0.3891509433962264,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989940595516432,
      "loss": 5.5079,
      "mean_token_accuracy": 0.16780147105455398,
      "num_tokens": 8907048.0,
      "step": 4290
    },
    {
      "entropy": 5.6579169750213625,
      "epoch": 0.3896044992743106,
      "grad_norm": 1.0,
      "learning_rate": 0.0004989910010171938,
      "loss": 5.5632,
      "mean_token_accuracy": 0.16277251839637757,
      "num_tokens": 8917987.0,
      "step": 4295
    },
    {
      "entropy": 5.714656162261963,
      "epoch": 0.3900580551523948,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004989879378505532,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16808317601680756,
      "num_tokens": 8929339.0,
      "step": 4300
    },
    {
      "entropy": 5.576473188400269,
      "epoch": 0.39051161103047893,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989848700517849,
      "loss": 5.4732,
      "mean_token_accuracy": 0.16945151686668397,
      "num_tokens": 8939473.0,
      "step": 4305
    },
    {
      "entropy": 5.632296085357666,
      "epoch": 0.39096516690856314,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989817976209526,
      "loss": 5.5056,
      "mean_token_accuracy": 0.16843720525503159,
      "num_tokens": 8949173.0,
      "step": 4310
    },
    {
      "entropy": 5.724737358093262,
      "epoch": 0.3914187227866473,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989787205581196,
      "loss": 5.4705,
      "mean_token_accuracy": 0.17448844164609909,
      "num_tokens": 8959879.0,
      "step": 4315
    },
    {
      "entropy": 5.661236667633057,
      "epoch": 0.3918722786647315,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989756388633496,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1667757049202919,
      "num_tokens": 8970643.0,
      "step": 4320
    },
    {
      "entropy": 5.635674095153808,
      "epoch": 0.39232583454281567,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989725525367063,
      "loss": 5.4405,
      "mean_token_accuracy": 0.17421406060457229,
      "num_tokens": 8979882.0,
      "step": 4325
    },
    {
      "entropy": 5.694489431381226,
      "epoch": 0.3927793904208999,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989694615782534,
      "loss": 5.5696,
      "mean_token_accuracy": 0.1630749970674515,
      "num_tokens": 8990925.0,
      "step": 4330
    },
    {
      "entropy": 5.679839992523194,
      "epoch": 0.39323294629898403,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989663659880552,
      "loss": 5.5923,
      "mean_token_accuracy": 0.1655128762125969,
      "num_tokens": 9001365.0,
      "step": 4335
    },
    {
      "entropy": 5.7446623802185055,
      "epoch": 0.3936865021770682,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989632657661754,
      "loss": 5.5877,
      "mean_token_accuracy": 0.16630492359399796,
      "num_tokens": 9011942.0,
      "step": 4340
    },
    {
      "entropy": 5.694921827316284,
      "epoch": 0.3941400580551524,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498960160912678,
      "loss": 5.6212,
      "mean_token_accuracy": 0.16706815212965012,
      "num_tokens": 9022610.0,
      "step": 4345
    },
    {
      "entropy": 5.657366991043091,
      "epoch": 0.39459361393323655,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004989570514276276,
      "loss": 5.5033,
      "mean_token_accuracy": 0.1658786192536354,
      "num_tokens": 9033034.0,
      "step": 4350
    },
    {
      "entropy": 5.693386363983154,
      "epoch": 0.39504716981132076,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989539373110883,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1648570790886879,
      "num_tokens": 9043418.0,
      "step": 4355
    },
    {
      "entropy": 5.730675506591797,
      "epoch": 0.3955007256894049,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989508185631245,
      "loss": 5.6473,
      "mean_token_accuracy": 0.16303568482398986,
      "num_tokens": 9052708.0,
      "step": 4360
    },
    {
      "entropy": 5.610522842407226,
      "epoch": 0.39595428156748913,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989476951838006,
      "loss": 5.503,
      "mean_token_accuracy": 0.16568560302257537,
      "num_tokens": 9062883.0,
      "step": 4365
    },
    {
      "entropy": 5.690759563446045,
      "epoch": 0.3964078374455733,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989445671731814,
      "loss": 5.4655,
      "mean_token_accuracy": 0.1712013304233551,
      "num_tokens": 9072830.0,
      "step": 4370
    },
    {
      "entropy": 5.583675479888916,
      "epoch": 0.3968613933236575,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989414345313315,
      "loss": 5.417,
      "mean_token_accuracy": 0.16587775200605392,
      "num_tokens": 9082749.0,
      "step": 4375
    },
    {
      "entropy": 5.6703362464904785,
      "epoch": 0.39731494920174165,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989382972583156,
      "loss": 5.5146,
      "mean_token_accuracy": 0.1637563332915306,
      "num_tokens": 9094188.0,
      "step": 4380
    },
    {
      "entropy": 5.6142324924469,
      "epoch": 0.3977685050798258,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989351553541986,
      "loss": 5.5483,
      "mean_token_accuracy": 0.1618421718478203,
      "num_tokens": 9105023.0,
      "step": 4385
    },
    {
      "entropy": 5.60534520149231,
      "epoch": 0.39822206095791,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989320088190456,
      "loss": 5.4782,
      "mean_token_accuracy": 0.17100915908813477,
      "num_tokens": 9115563.0,
      "step": 4390
    },
    {
      "entropy": 5.668853235244751,
      "epoch": 0.3986756168359942,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989288576529216,
      "loss": 5.509,
      "mean_token_accuracy": 0.16283219754695893,
      "num_tokens": 9125394.0,
      "step": 4395
    },
    {
      "entropy": 5.6934130668640135,
      "epoch": 0.3991291727140784,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989257018558917,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16816064417362214,
      "num_tokens": 9137061.0,
      "step": 4400
    },
    {
      "entropy": 5.669379186630249,
      "epoch": 0.39958272859216254,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004989225414280213,
      "loss": 5.476,
      "mean_token_accuracy": 0.17045701593160628,
      "num_tokens": 9148196.0,
      "step": 4405
    },
    {
      "entropy": 5.5835472583770756,
      "epoch": 0.40003628447024675,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004989193763693756,
      "loss": 5.494,
      "mean_token_accuracy": 0.17453179508447647,
      "num_tokens": 9158013.0,
      "step": 4410
    },
    {
      "entropy": 5.637920618057251,
      "epoch": 0.4004898403483309,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989162066800202,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1738430827856064,
      "num_tokens": 9167813.0,
      "step": 4415
    },
    {
      "entropy": 5.620611953735351,
      "epoch": 0.4009433962264151,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989130323600205,
      "loss": 5.4628,
      "mean_token_accuracy": 0.17226067781448365,
      "num_tokens": 9177783.0,
      "step": 4420
    },
    {
      "entropy": 5.668359851837158,
      "epoch": 0.40139695210449927,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004989098534094424,
      "loss": 5.5088,
      "mean_token_accuracy": 0.17067906260490417,
      "num_tokens": 9187556.0,
      "step": 4425
    },
    {
      "entropy": 5.655431270599365,
      "epoch": 0.4018505079825834,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989066698283512,
      "loss": 5.5866,
      "mean_token_accuracy": 0.16263027787208556,
      "num_tokens": 9198659.0,
      "step": 4430
    },
    {
      "entropy": 5.669175672531128,
      "epoch": 0.40230406386066764,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989034816168132,
      "loss": 5.5561,
      "mean_token_accuracy": 0.16660318672657012,
      "num_tokens": 9209546.0,
      "step": 4435
    },
    {
      "entropy": 5.627757024765015,
      "epoch": 0.4027576197387518,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989002887748941,
      "loss": 5.4648,
      "mean_token_accuracy": 0.1715000495314598,
      "num_tokens": 9219339.0,
      "step": 4440
    },
    {
      "entropy": 5.656140756607056,
      "epoch": 0.403211175616836,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988970913026601,
      "loss": 5.6487,
      "mean_token_accuracy": 0.16045075058937072,
      "num_tokens": 9230364.0,
      "step": 4445
    },
    {
      "entropy": 5.673274517059326,
      "epoch": 0.40366473149492016,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988938892001771,
      "loss": 5.4897,
      "mean_token_accuracy": 0.16654825061559678,
      "num_tokens": 9240732.0,
      "step": 4450
    },
    {
      "entropy": 5.599252319335937,
      "epoch": 0.40411828737300437,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004988906824675113,
      "loss": 5.4359,
      "mean_token_accuracy": 0.17193734049797058,
      "num_tokens": 9251049.0,
      "step": 4455
    },
    {
      "entropy": 5.626266956329346,
      "epoch": 0.4045718432510885,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004988874711047293,
      "loss": 5.4841,
      "mean_token_accuracy": 0.17143381983041764,
      "num_tokens": 9261660.0,
      "step": 4460
    },
    {
      "entropy": 5.650390625,
      "epoch": 0.40502539912917274,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988842551118972,
      "loss": 5.5465,
      "mean_token_accuracy": 0.170607990026474,
      "num_tokens": 9272114.0,
      "step": 4465
    },
    {
      "entropy": 5.645572662353516,
      "epoch": 0.4054789550072569,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988810344890818,
      "loss": 5.5135,
      "mean_token_accuracy": 0.17275023311376572,
      "num_tokens": 9282125.0,
      "step": 4470
    },
    {
      "entropy": 5.7307713508605955,
      "epoch": 0.40593251088534105,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988778092363494,
      "loss": 5.5466,
      "mean_token_accuracy": 0.16705104261636733,
      "num_tokens": 9294078.0,
      "step": 4475
    },
    {
      "entropy": 5.658994054794311,
      "epoch": 0.40638606676342526,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988745793537671,
      "loss": 5.5095,
      "mean_token_accuracy": 0.16595903933048248,
      "num_tokens": 9304026.0,
      "step": 4480
    },
    {
      "entropy": 5.674858808517456,
      "epoch": 0.4068396226415094,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988713448414011,
      "loss": 5.5545,
      "mean_token_accuracy": 0.1764197275042534,
      "num_tokens": 9313933.0,
      "step": 4485
    },
    {
      "entropy": 5.660690259933472,
      "epoch": 0.4072931785195936,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000498868105699319,
      "loss": 5.5765,
      "mean_token_accuracy": 0.16426668763160707,
      "num_tokens": 9323932.0,
      "step": 4490
    },
    {
      "entropy": 5.5835301876068115,
      "epoch": 0.4077467343976778,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988648619275872,
      "loss": 5.4754,
      "mean_token_accuracy": 0.17924543023109435,
      "num_tokens": 9333467.0,
      "step": 4495
    },
    {
      "entropy": 5.666203594207763,
      "epoch": 0.408200290275762,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988616135262729,
      "loss": 5.5319,
      "mean_token_accuracy": 0.16330583691596984,
      "num_tokens": 9344909.0,
      "step": 4500
    },
    {
      "entropy": 5.749293851852417,
      "epoch": 0.40865384615384615,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004988583604954436,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1672707676887512,
      "num_tokens": 9355630.0,
      "step": 4505
    },
    {
      "entropy": 5.601924467086792,
      "epoch": 0.40910740203193036,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004988551028351663,
      "loss": 5.5262,
      "mean_token_accuracy": 0.165559022128582,
      "num_tokens": 9365700.0,
      "step": 4510
    },
    {
      "entropy": 5.630667209625244,
      "epoch": 0.4095609579100145,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988518405455084,
      "loss": 5.6004,
      "mean_token_accuracy": 0.16178341209888458,
      "num_tokens": 9377088.0,
      "step": 4515
    },
    {
      "entropy": 5.690512752532959,
      "epoch": 0.41001451378809867,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988485736265374,
      "loss": 5.3891,
      "mean_token_accuracy": 0.17818945944309234,
      "num_tokens": 9386855.0,
      "step": 4520
    },
    {
      "entropy": 5.622279644012451,
      "epoch": 0.4104680696661829,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000498845302078321,
      "loss": 5.5203,
      "mean_token_accuracy": 0.16728461235761644,
      "num_tokens": 9398661.0,
      "step": 4525
    },
    {
      "entropy": 5.662732839584351,
      "epoch": 0.41092162554426703,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988420259009266,
      "loss": 5.5199,
      "mean_token_accuracy": 0.17487220466136932,
      "num_tokens": 9408670.0,
      "step": 4530
    },
    {
      "entropy": 5.680340766906738,
      "epoch": 0.41137518142235124,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000498838745094422,
      "loss": 5.531,
      "mean_token_accuracy": 0.1707653671503067,
      "num_tokens": 9418740.0,
      "step": 4535
    },
    {
      "entropy": 5.592594003677368,
      "epoch": 0.4118287373004354,
      "grad_norm": 1.0,
      "learning_rate": 0.0004988354596588751,
      "loss": 5.5133,
      "mean_token_accuracy": 0.1718428000807762,
      "num_tokens": 9429339.0,
      "step": 4540
    },
    {
      "entropy": 5.652766942977905,
      "epoch": 0.4122822931785196,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988321695943539,
      "loss": 5.4573,
      "mean_token_accuracy": 0.17410281002521516,
      "num_tokens": 9438996.0,
      "step": 4545
    },
    {
      "entropy": 5.635617685317993,
      "epoch": 0.41273584905660377,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988288749009264,
      "loss": 5.4764,
      "mean_token_accuracy": 0.17122164964675904,
      "num_tokens": 9448669.0,
      "step": 4550
    },
    {
      "entropy": 5.588901138305664,
      "epoch": 0.413189404934688,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004988255755786608,
      "loss": 5.4564,
      "mean_token_accuracy": 0.17379783987998962,
      "num_tokens": 9458857.0,
      "step": 4555
    },
    {
      "entropy": 5.644811010360717,
      "epoch": 0.41364296081277213,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988222716276251,
      "loss": 5.4576,
      "mean_token_accuracy": 0.17216896563768386,
      "num_tokens": 9470241.0,
      "step": 4560
    },
    {
      "entropy": 5.688581943511963,
      "epoch": 0.4140965166908563,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000498818963047888,
      "loss": 5.6176,
      "mean_token_accuracy": 0.16512774527072907,
      "num_tokens": 9480528.0,
      "step": 4565
    },
    {
      "entropy": 5.707815265655517,
      "epoch": 0.4145500725689405,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988156498395176,
      "loss": 5.5291,
      "mean_token_accuracy": 0.16812336444854736,
      "num_tokens": 9491000.0,
      "step": 4570
    },
    {
      "entropy": 5.693633604049682,
      "epoch": 0.41500362844702465,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988123320025825,
      "loss": 5.5084,
      "mean_token_accuracy": 0.1725713789463043,
      "num_tokens": 9501612.0,
      "step": 4575
    },
    {
      "entropy": 5.645354413986206,
      "epoch": 0.41545718432510886,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988090095371514,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1642218790948391,
      "num_tokens": 9513999.0,
      "step": 4580
    },
    {
      "entropy": 5.672644186019897,
      "epoch": 0.415910740203193,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004988056824432928,
      "loss": 5.4907,
      "mean_token_accuracy": 0.16187755018472672,
      "num_tokens": 9524392.0,
      "step": 4585
    },
    {
      "entropy": 5.691357946395874,
      "epoch": 0.41636429608127723,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004988023507210758,
      "loss": 5.4872,
      "mean_token_accuracy": 0.1723600745201111,
      "num_tokens": 9534227.0,
      "step": 4590
    },
    {
      "entropy": 5.637865591049194,
      "epoch": 0.4168178519593614,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987990143705692,
      "loss": 5.4877,
      "mean_token_accuracy": 0.17282977551221848,
      "num_tokens": 9544242.0,
      "step": 4595
    },
    {
      "entropy": 5.601320028305054,
      "epoch": 0.4172714078374456,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004987956733918418,
      "loss": 5.4146,
      "mean_token_accuracy": 0.16889237761497497,
      "num_tokens": 9553921.0,
      "step": 4600
    },
    {
      "entropy": 5.529018545150757,
      "epoch": 0.41772496371552975,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987923277849629,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1648541048169136,
      "num_tokens": 9564768.0,
      "step": 4605
    },
    {
      "entropy": 5.6799657344818115,
      "epoch": 0.4181785195936139,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987889775500017,
      "loss": 5.5097,
      "mean_token_accuracy": 0.16558238118886948,
      "num_tokens": 9576360.0,
      "step": 4610
    },
    {
      "entropy": 5.645033693313598,
      "epoch": 0.4186320754716981,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004987856226870272,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16887864768505095,
      "num_tokens": 9587623.0,
      "step": 4615
    },
    {
      "entropy": 5.6239402294158936,
      "epoch": 0.4190856313497823,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987822631961092,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1658405616879463,
      "num_tokens": 9597837.0,
      "step": 4620
    },
    {
      "entropy": 5.668896579742432,
      "epoch": 0.4195391872278665,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498778899077317,
      "loss": 5.4938,
      "mean_token_accuracy": 0.1671911209821701,
      "num_tokens": 9609509.0,
      "step": 4625
    },
    {
      "entropy": 5.637753105163574,
      "epoch": 0.41999274310595064,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049877553033072,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1714413955807686,
      "num_tokens": 9619954.0,
      "step": 4630
    },
    {
      "entropy": 5.5957275390625,
      "epoch": 0.42044629898403485,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987721569563882,
      "loss": 5.4746,
      "mean_token_accuracy": 0.1704351544380188,
      "num_tokens": 9629742.0,
      "step": 4635
    },
    {
      "entropy": 5.6290453433990475,
      "epoch": 0.420899854862119,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000498768778954391,
      "loss": 5.4947,
      "mean_token_accuracy": 0.17307782173156738,
      "num_tokens": 9641378.0,
      "step": 4640
    },
    {
      "entropy": 5.7364380836486815,
      "epoch": 0.4213534107402032,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004987653963247985,
      "loss": 5.5701,
      "mean_token_accuracy": 0.1703578621149063,
      "num_tokens": 9652279.0,
      "step": 4645
    },
    {
      "entropy": 5.666488122940064,
      "epoch": 0.42180696661828737,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004987620090676805,
      "loss": 5.5464,
      "mean_token_accuracy": 0.1691744804382324,
      "num_tokens": 9663416.0,
      "step": 4650
    },
    {
      "entropy": 5.635804986953735,
      "epoch": 0.4222605224963715,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987586171831072,
      "loss": 5.5084,
      "mean_token_accuracy": 0.17069009840488433,
      "num_tokens": 9674647.0,
      "step": 4655
    },
    {
      "entropy": 5.627348804473877,
      "epoch": 0.42271407837445574,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004987552206711487,
      "loss": 5.3729,
      "mean_token_accuracy": 0.17568349838256836,
      "num_tokens": 9685084.0,
      "step": 4660
    },
    {
      "entropy": 5.546395874023437,
      "epoch": 0.4231676342525399,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987518195318752,
      "loss": 5.4367,
      "mean_token_accuracy": 0.1702194795012474,
      "num_tokens": 9695661.0,
      "step": 4665
    },
    {
      "entropy": 5.675781869888306,
      "epoch": 0.4236211901306241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000498748413765357,
      "loss": 5.4906,
      "mean_token_accuracy": 0.17196833938360215,
      "num_tokens": 9705773.0,
      "step": 4670
    },
    {
      "entropy": 5.642674303054809,
      "epoch": 0.42407474600870826,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987450033716646,
      "loss": 5.5318,
      "mean_token_accuracy": 0.16780486702919006,
      "num_tokens": 9715799.0,
      "step": 4675
    },
    {
      "entropy": 5.678176498413086,
      "epoch": 0.42452830188679247,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987415883508686,
      "loss": 5.4922,
      "mean_token_accuracy": 0.1741717353463173,
      "num_tokens": 9727012.0,
      "step": 4680
    },
    {
      "entropy": 5.624595403671265,
      "epoch": 0.4249818577648766,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987381687030394,
      "loss": 5.5047,
      "mean_token_accuracy": 0.1686483457684517,
      "num_tokens": 9737044.0,
      "step": 4685
    },
    {
      "entropy": 5.73467435836792,
      "epoch": 0.42543541364296084,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987347444282479,
      "loss": 5.5885,
      "mean_token_accuracy": 0.15938138514757155,
      "num_tokens": 9747648.0,
      "step": 4690
    },
    {
      "entropy": 5.68295259475708,
      "epoch": 0.425888969521045,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987313155265649,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16666015088558198,
      "num_tokens": 9757812.0,
      "step": 4695
    },
    {
      "entropy": 5.677083969116211,
      "epoch": 0.42634252539912915,
      "grad_norm": 0.875,
      "learning_rate": 0.0004987278819980613,
      "loss": 5.5077,
      "mean_token_accuracy": 0.175352643430233,
      "num_tokens": 9768337.0,
      "step": 4700
    },
    {
      "entropy": 5.682326030731201,
      "epoch": 0.42679608127721336,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987244438428081,
      "loss": 5.548,
      "mean_token_accuracy": 0.16384970843791963,
      "num_tokens": 9779433.0,
      "step": 4705
    },
    {
      "entropy": 5.598498392105102,
      "epoch": 0.4272496371552975,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987210010608764,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1729959085583687,
      "num_tokens": 9790069.0,
      "step": 4710
    },
    {
      "entropy": 5.612740516662598,
      "epoch": 0.4277031930333817,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987175536523373,
      "loss": 5.413,
      "mean_token_accuracy": 0.17409739047288894,
      "num_tokens": 9799669.0,
      "step": 4715
    },
    {
      "entropy": 5.602735471725464,
      "epoch": 0.4281567489114659,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987141016172622,
      "loss": 5.4565,
      "mean_token_accuracy": 0.1733306437730789,
      "num_tokens": 9809615.0,
      "step": 4720
    },
    {
      "entropy": 5.598521518707275,
      "epoch": 0.4286103047895501,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004987106449557225,
      "loss": 5.4288,
      "mean_token_accuracy": 0.17376641631126405,
      "num_tokens": 9819666.0,
      "step": 4725
    },
    {
      "entropy": 5.727816772460938,
      "epoch": 0.42906386066763424,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987071836677896,
      "loss": 5.5897,
      "mean_token_accuracy": 0.16895099133253097,
      "num_tokens": 9831015.0,
      "step": 4730
    },
    {
      "entropy": 5.61959285736084,
      "epoch": 0.42951741654571846,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987037177535353,
      "loss": 5.445,
      "mean_token_accuracy": 0.164634008705616,
      "num_tokens": 9841354.0,
      "step": 4735
    },
    {
      "entropy": 5.564237451553344,
      "epoch": 0.4299709724238026,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987002472130309,
      "loss": 5.4452,
      "mean_token_accuracy": 0.16783540546894074,
      "num_tokens": 9850467.0,
      "step": 4740
    },
    {
      "entropy": 5.710805034637451,
      "epoch": 0.43042452830188677,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986967720463485,
      "loss": 5.5211,
      "mean_token_accuracy": 0.16725909262895583,
      "num_tokens": 9861115.0,
      "step": 4745
    },
    {
      "entropy": 5.713682174682617,
      "epoch": 0.430878084179971,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986932922535598,
      "loss": 5.5218,
      "mean_token_accuracy": 0.168181936442852,
      "num_tokens": 9871172.0,
      "step": 4750
    },
    {
      "entropy": 5.633159351348877,
      "epoch": 0.43133164005805513,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986898078347368,
      "loss": 5.4289,
      "mean_token_accuracy": 0.1761467546224594,
      "num_tokens": 9881997.0,
      "step": 4755
    },
    {
      "entropy": 5.609486722946167,
      "epoch": 0.43178519593613934,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986863187899516,
      "loss": 5.4029,
      "mean_token_accuracy": 0.17335348427295685,
      "num_tokens": 9892310.0,
      "step": 4760
    },
    {
      "entropy": 5.551136064529419,
      "epoch": 0.4322387518142235,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986828251192764,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1719319611787796,
      "num_tokens": 9902664.0,
      "step": 4765
    },
    {
      "entropy": 5.614825916290283,
      "epoch": 0.4326923076923077,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986793268227833,
      "loss": 5.446,
      "mean_token_accuracy": 0.17410656809806824,
      "num_tokens": 9912491.0,
      "step": 4770
    },
    {
      "entropy": 5.6420478343963625,
      "epoch": 0.43314586357039186,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986758239005448,
      "loss": 5.4114,
      "mean_token_accuracy": 0.17530306428670883,
      "num_tokens": 9922057.0,
      "step": 4775
    },
    {
      "entropy": 5.603491544723511,
      "epoch": 0.4335994194484761,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986723163526331,
      "loss": 5.4703,
      "mean_token_accuracy": 0.16902925074100494,
      "num_tokens": 9933661.0,
      "step": 4780
    },
    {
      "entropy": 5.545865392684936,
      "epoch": 0.43405297532656023,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986688041791209,
      "loss": 5.4331,
      "mean_token_accuracy": 0.16436014771461488,
      "num_tokens": 9945177.0,
      "step": 4785
    },
    {
      "entropy": 5.630179929733276,
      "epoch": 0.4345065312046444,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498665287380081,
      "loss": 5.4649,
      "mean_token_accuracy": 0.16838951110839845,
      "num_tokens": 9956226.0,
      "step": 4790
    },
    {
      "entropy": 5.670184659957886,
      "epoch": 0.4349600870827286,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986617659555859,
      "loss": 5.5281,
      "mean_token_accuracy": 0.1700868308544159,
      "num_tokens": 9965877.0,
      "step": 4795
    },
    {
      "entropy": 5.601760721206665,
      "epoch": 0.43541364296081275,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986582399057085,
      "loss": 5.4742,
      "mean_token_accuracy": 0.17551955133676528,
      "num_tokens": 9975592.0,
      "step": 4800
    },
    {
      "entropy": 5.680169630050659,
      "epoch": 0.43586719883889696,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004986547092305216,
      "loss": 5.5056,
      "mean_token_accuracy": 0.16957487165927887,
      "num_tokens": 9986094.0,
      "step": 4805
    },
    {
      "entropy": 5.717319345474243,
      "epoch": 0.4363207547169811,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986511739300984,
      "loss": 5.5406,
      "mean_token_accuracy": 0.16883666813373566,
      "num_tokens": 9997241.0,
      "step": 4810
    },
    {
      "entropy": 5.607332468032837,
      "epoch": 0.43677431059506533,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986476340045119,
      "loss": 5.4581,
      "mean_token_accuracy": 0.17112519294023515,
      "num_tokens": 10008091.0,
      "step": 4815
    },
    {
      "entropy": 5.587670564651489,
      "epoch": 0.4372278664731495,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004986440894538354,
      "loss": 5.5149,
      "mean_token_accuracy": 0.16817084699869156,
      "num_tokens": 10018794.0,
      "step": 4820
    },
    {
      "entropy": 5.617847633361817,
      "epoch": 0.4376814223512337,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004986405402781421,
      "loss": 5.4295,
      "mean_token_accuracy": 0.17280754894018174,
      "num_tokens": 10028414.0,
      "step": 4825
    },
    {
      "entropy": 5.634881925582886,
      "epoch": 0.43813497822931785,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986369864775054,
      "loss": 5.5077,
      "mean_token_accuracy": 0.16782476902008056,
      "num_tokens": 10038749.0,
      "step": 4830
    },
    {
      "entropy": 5.6245917797088625,
      "epoch": 0.438588534107402,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986334280519989,
      "loss": 5.4539,
      "mean_token_accuracy": 0.17029527723789215,
      "num_tokens": 10049866.0,
      "step": 4835
    },
    {
      "entropy": 5.6066179275512695,
      "epoch": 0.4390420899854862,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498629865001696,
      "loss": 5.4886,
      "mean_token_accuracy": 0.17579846531152726,
      "num_tokens": 10060680.0,
      "step": 4840
    },
    {
      "entropy": 5.681512403488159,
      "epoch": 0.43949564586357037,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004986262973266706,
      "loss": 5.5728,
      "mean_token_accuracy": 0.16835084557533264,
      "num_tokens": 10070931.0,
      "step": 4845
    },
    {
      "entropy": 5.582299041748047,
      "epoch": 0.4399492017416546,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986227250269964,
      "loss": 5.5073,
      "mean_token_accuracy": 0.1723349377512932,
      "num_tokens": 10080189.0,
      "step": 4850
    },
    {
      "entropy": 5.640298318862915,
      "epoch": 0.44040275761973874,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986191481027472,
      "loss": 5.4149,
      "mean_token_accuracy": 0.17179947793483735,
      "num_tokens": 10088813.0,
      "step": 4855
    },
    {
      "entropy": 5.5767192363739015,
      "epoch": 0.44085631349782295,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986155665539971,
      "loss": 5.4386,
      "mean_token_accuracy": 0.17210467159748077,
      "num_tokens": 10098125.0,
      "step": 4860
    },
    {
      "entropy": 5.666972684860229,
      "epoch": 0.4413098693759071,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049861198038082,
      "loss": 5.4766,
      "mean_token_accuracy": 0.16837365478277205,
      "num_tokens": 10109125.0,
      "step": 4865
    },
    {
      "entropy": 5.607843923568725,
      "epoch": 0.4417634252539913,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004986083895832903,
      "loss": 5.4597,
      "mean_token_accuracy": 0.17088931798934937,
      "num_tokens": 10120182.0,
      "step": 4870
    },
    {
      "entropy": 5.563991737365723,
      "epoch": 0.44221698113207547,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004986047941614821,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1657083287835121,
      "num_tokens": 10130941.0,
      "step": 4875
    },
    {
      "entropy": 5.683834409713745,
      "epoch": 0.4426705370101596,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004986011941154696,
      "loss": 5.4817,
      "mean_token_accuracy": 0.16850205659866332,
      "num_tokens": 10141708.0,
      "step": 4880
    },
    {
      "entropy": 5.627482271194458,
      "epoch": 0.44312409288824384,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985975894453275,
      "loss": 5.4216,
      "mean_token_accuracy": 0.16762909889221192,
      "num_tokens": 10151816.0,
      "step": 4885
    },
    {
      "entropy": 5.666119718551636,
      "epoch": 0.443577648766328,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004985939801511304,
      "loss": 5.5404,
      "mean_token_accuracy": 0.16791291683912277,
      "num_tokens": 10162470.0,
      "step": 4890
    },
    {
      "entropy": 5.635271310806274,
      "epoch": 0.4440312046444122,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004985903662329527,
      "loss": 5.4278,
      "mean_token_accuracy": 0.17861440777778625,
      "num_tokens": 10173171.0,
      "step": 4895
    },
    {
      "entropy": 5.6425152778625485,
      "epoch": 0.44448476052249636,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004985867476908693,
      "loss": 5.4696,
      "mean_token_accuracy": 0.17133731096982957,
      "num_tokens": 10184185.0,
      "step": 4900
    },
    {
      "entropy": 5.674123239517212,
      "epoch": 0.44493831640058057,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985831245249549,
      "loss": 5.5259,
      "mean_token_accuracy": 0.17461849972605706,
      "num_tokens": 10195516.0,
      "step": 4905
    },
    {
      "entropy": 5.618943500518799,
      "epoch": 0.4453918722786647,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985794967352846,
      "loss": 5.4537,
      "mean_token_accuracy": 0.17528860867023469,
      "num_tokens": 10204936.0,
      "step": 4910
    },
    {
      "entropy": 5.646315479278565,
      "epoch": 0.44584542815674894,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985758643219334,
      "loss": 5.4884,
      "mean_token_accuracy": 0.1720237150788307,
      "num_tokens": 10216218.0,
      "step": 4915
    },
    {
      "entropy": 5.66897406578064,
      "epoch": 0.4462989840348331,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004985722272849762,
      "loss": 5.4856,
      "mean_token_accuracy": 0.16680216640233994,
      "num_tokens": 10226370.0,
      "step": 4920
    },
    {
      "entropy": 5.541078567504883,
      "epoch": 0.44675253991291725,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985685856244884,
      "loss": 5.418,
      "mean_token_accuracy": 0.18198757171630858,
      "num_tokens": 10236170.0,
      "step": 4925
    },
    {
      "entropy": 5.612875270843506,
      "epoch": 0.44720609579100146,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985649393405453,
      "loss": 5.4902,
      "mean_token_accuracy": 0.16974830478429795,
      "num_tokens": 10246119.0,
      "step": 4930
    },
    {
      "entropy": 5.630044221878052,
      "epoch": 0.4476596516690856,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985612884332223,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16304934322834014,
      "num_tokens": 10256762.0,
      "step": 4935
    },
    {
      "entropy": 5.5931806564331055,
      "epoch": 0.4481132075471698,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985576329025949,
      "loss": 5.4349,
      "mean_token_accuracy": 0.17295026481151582,
      "num_tokens": 10267317.0,
      "step": 4940
    },
    {
      "entropy": 5.6881492137908936,
      "epoch": 0.448566763425254,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004985539727487385,
      "loss": 5.5798,
      "mean_token_accuracy": 0.15689071714878083,
      "num_tokens": 10279111.0,
      "step": 4945
    },
    {
      "entropy": 5.61900143623352,
      "epoch": 0.4490203193033382,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004985503079717292,
      "loss": 5.4445,
      "mean_token_accuracy": 0.17076381295919418,
      "num_tokens": 10289383.0,
      "step": 4950
    },
    {
      "entropy": 5.611308670043945,
      "epoch": 0.44947387518142234,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985466385716425,
      "loss": 5.4578,
      "mean_token_accuracy": 0.1737099438905716,
      "num_tokens": 10299806.0,
      "step": 4955
    },
    {
      "entropy": 5.5558473587036135,
      "epoch": 0.44992743105950656,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004985429645485543,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16585210114717483,
      "num_tokens": 10310348.0,
      "step": 4960
    },
    {
      "entropy": 5.625494909286499,
      "epoch": 0.4503809869375907,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004985392859025407,
      "loss": 5.4737,
      "mean_token_accuracy": 0.17041534036397935,
      "num_tokens": 10320354.0,
      "step": 4965
    },
    {
      "entropy": 5.601232576370239,
      "epoch": 0.45083454281567487,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985356026336776,
      "loss": 5.3997,
      "mean_token_accuracy": 0.17218953967094422,
      "num_tokens": 10330292.0,
      "step": 4970
    },
    {
      "entropy": 5.576334428787232,
      "epoch": 0.4512880986937591,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985319147420414,
      "loss": 5.366,
      "mean_token_accuracy": 0.18071993440389633,
      "num_tokens": 10340449.0,
      "step": 4975
    },
    {
      "entropy": 5.603624248504639,
      "epoch": 0.45174165457184323,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985282222277082,
      "loss": 5.442,
      "mean_token_accuracy": 0.17101425379514695,
      "num_tokens": 10350637.0,
      "step": 4980
    },
    {
      "entropy": 5.54790735244751,
      "epoch": 0.45219521044992744,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985245250907544,
      "loss": 5.3815,
      "mean_token_accuracy": 0.17386381030082704,
      "num_tokens": 10361548.0,
      "step": 4985
    },
    {
      "entropy": 5.566483211517334,
      "epoch": 0.4526487663280116,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985208233312564,
      "loss": 5.3907,
      "mean_token_accuracy": 0.1698409304022789,
      "num_tokens": 10371765.0,
      "step": 4990
    },
    {
      "entropy": 5.661066150665283,
      "epoch": 0.4531023222060958,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985171169492908,
      "loss": 5.4967,
      "mean_token_accuracy": 0.1759715974330902,
      "num_tokens": 10382108.0,
      "step": 4995
    },
    {
      "entropy": 5.606108808517456,
      "epoch": 0.45355587808417996,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004985134059449344,
      "loss": 5.4415,
      "mean_token_accuracy": 0.1780141994357109,
      "num_tokens": 10392390.0,
      "step": 5000
    },
    {
      "entropy": 5.593348503112793,
      "epoch": 0.4540094339622642,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004985096903182637,
      "loss": 5.5074,
      "mean_token_accuracy": 0.17851048856973648,
      "num_tokens": 10402094.0,
      "step": 5005
    },
    {
      "entropy": 5.660210275650025,
      "epoch": 0.45446298984034833,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004985059700693557,
      "loss": 5.4447,
      "mean_token_accuracy": 0.17331443428993226,
      "num_tokens": 10413350.0,
      "step": 5010
    },
    {
      "entropy": 5.674293041229248,
      "epoch": 0.4549165457184325,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985022451982872,
      "loss": 5.5776,
      "mean_token_accuracy": 0.16241093277931212,
      "num_tokens": 10424292.0,
      "step": 5015
    },
    {
      "entropy": 5.69236478805542,
      "epoch": 0.4553701015965167,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004984985157051354,
      "loss": 5.4254,
      "mean_token_accuracy": 0.17173273414373397,
      "num_tokens": 10433724.0,
      "step": 5020
    },
    {
      "entropy": 5.644928026199341,
      "epoch": 0.45582365747460085,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984947815899774,
      "loss": 5.5155,
      "mean_token_accuracy": 0.16911423206329346,
      "num_tokens": 10444724.0,
      "step": 5025
    },
    {
      "entropy": 5.585797357559204,
      "epoch": 0.45627721335268506,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984910428528902,
      "loss": 5.4332,
      "mean_token_accuracy": 0.17657725960016252,
      "num_tokens": 10455086.0,
      "step": 5030
    },
    {
      "entropy": 5.552824258804321,
      "epoch": 0.4567307692307692,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984872994939514,
      "loss": 5.3995,
      "mean_token_accuracy": 0.16780003756284714,
      "num_tokens": 10465076.0,
      "step": 5035
    },
    {
      "entropy": 5.628185558319092,
      "epoch": 0.45718432510885343,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984835515132382,
      "loss": 5.4097,
      "mean_token_accuracy": 0.17855704724788665,
      "num_tokens": 10474817.0,
      "step": 5040
    },
    {
      "entropy": 5.590842294692993,
      "epoch": 0.4576378809869376,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004984797989108282,
      "loss": 5.4381,
      "mean_token_accuracy": 0.1680952191352844,
      "num_tokens": 10485735.0,
      "step": 5045
    },
    {
      "entropy": 5.577192163467407,
      "epoch": 0.4580914368650218,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984760416867991,
      "loss": 5.5278,
      "mean_token_accuracy": 0.16953626424074172,
      "num_tokens": 10496731.0,
      "step": 5050
    },
    {
      "entropy": 5.570709562301635,
      "epoch": 0.45854499274310595,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984722798412286,
      "loss": 5.3987,
      "mean_token_accuracy": 0.1745694950222969,
      "num_tokens": 10506787.0,
      "step": 5055
    },
    {
      "entropy": 5.653943300247192,
      "epoch": 0.4589985486211901,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004984685133741941,
      "loss": 5.4581,
      "mean_token_accuracy": 0.1692812368273735,
      "num_tokens": 10517170.0,
      "step": 5060
    },
    {
      "entropy": 5.604233932495117,
      "epoch": 0.4594521044992743,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498464742285774,
      "loss": 5.46,
      "mean_token_accuracy": 0.1756867364048958,
      "num_tokens": 10527342.0,
      "step": 5065
    },
    {
      "entropy": 5.5937105178833,
      "epoch": 0.45990566037735847,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498460966576046,
      "loss": 5.433,
      "mean_token_accuracy": 0.1744078904390335,
      "num_tokens": 10537253.0,
      "step": 5070
    },
    {
      "entropy": 5.653906154632568,
      "epoch": 0.4603592162554427,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984571862450884,
      "loss": 5.4459,
      "mean_token_accuracy": 0.16896500289440156,
      "num_tokens": 10547249.0,
      "step": 5075
    },
    {
      "entropy": 5.689875221252441,
      "epoch": 0.46081277213352684,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984534012929791,
      "loss": 5.4873,
      "mean_token_accuracy": 0.17034367769956588,
      "num_tokens": 10558174.0,
      "step": 5080
    },
    {
      "entropy": 5.67174391746521,
      "epoch": 0.46126632801161105,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004984496117197967,
      "loss": 5.4359,
      "mean_token_accuracy": 0.17626606971025466,
      "num_tokens": 10567610.0,
      "step": 5085
    },
    {
      "entropy": 5.512884426116943,
      "epoch": 0.4617198838896952,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004984458175256193,
      "loss": 5.4515,
      "mean_token_accuracy": 0.17041234970092772,
      "num_tokens": 10578494.0,
      "step": 5090
    },
    {
      "entropy": 5.667339754104614,
      "epoch": 0.4621734397677794,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984420187105253,
      "loss": 5.5543,
      "mean_token_accuracy": 0.16734294444322587,
      "num_tokens": 10589482.0,
      "step": 5095
    },
    {
      "entropy": 5.714143896102906,
      "epoch": 0.46262699564586357,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004984382152745936,
      "loss": 5.5033,
      "mean_token_accuracy": 0.16372307240962983,
      "num_tokens": 10599965.0,
      "step": 5100
    },
    {
      "entropy": 5.607032728195191,
      "epoch": 0.4630805515239477,
      "grad_norm": 0.875,
      "learning_rate": 0.0004984344072179026,
      "loss": 5.5384,
      "mean_token_accuracy": 0.17114560455083846,
      "num_tokens": 10610500.0,
      "step": 5105
    },
    {
      "entropy": 5.703943061828613,
      "epoch": 0.46353410740203194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984305945405312,
      "loss": 5.4792,
      "mean_token_accuracy": 0.17351720333099366,
      "num_tokens": 10619843.0,
      "step": 5110
    },
    {
      "entropy": 5.620040988922119,
      "epoch": 0.4639876632801161,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000498426777242558,
      "loss": 5.4189,
      "mean_token_accuracy": 0.17909690588712693,
      "num_tokens": 10631026.0,
      "step": 5115
    },
    {
      "entropy": 5.440194177627563,
      "epoch": 0.4644412191582003,
      "grad_norm": 1.0,
      "learning_rate": 0.0004984229553240623,
      "loss": 5.4071,
      "mean_token_accuracy": 0.17747279554605483,
      "num_tokens": 10641130.0,
      "step": 5120
    },
    {
      "entropy": 5.710953187942505,
      "epoch": 0.46489477503628446,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984191287851228,
      "loss": 5.4911,
      "mean_token_accuracy": 0.17253502756357192,
      "num_tokens": 10652305.0,
      "step": 5125
    },
    {
      "entropy": 5.679422616958618,
      "epoch": 0.46534833091436867,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984152976258188,
      "loss": 5.5138,
      "mean_token_accuracy": 0.17212531566619874,
      "num_tokens": 10663083.0,
      "step": 5130
    },
    {
      "entropy": 5.602247858047486,
      "epoch": 0.4658018867924528,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984114618462296,
      "loss": 5.5342,
      "mean_token_accuracy": 0.1666450873017311,
      "num_tokens": 10673602.0,
      "step": 5135
    },
    {
      "entropy": 5.576183080673218,
      "epoch": 0.46625544267053703,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004984076214464343,
      "loss": 5.3597,
      "mean_token_accuracy": 0.18366137892007828,
      "num_tokens": 10684622.0,
      "step": 5140
    },
    {
      "entropy": 5.608708000183105,
      "epoch": 0.4667089985486212,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004984037764265126,
      "loss": 5.4792,
      "mean_token_accuracy": 0.171570348739624,
      "num_tokens": 10695129.0,
      "step": 5145
    },
    {
      "entropy": 5.566263818740845,
      "epoch": 0.46716255442670535,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983999267865438,
      "loss": 5.3269,
      "mean_token_accuracy": 0.1831385925412178,
      "num_tokens": 10705236.0,
      "step": 5150
    },
    {
      "entropy": 5.559924459457397,
      "epoch": 0.46761611030478956,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983960725266078,
      "loss": 5.4354,
      "mean_token_accuracy": 0.17425220608711242,
      "num_tokens": 10715748.0,
      "step": 5155
    },
    {
      "entropy": 5.570201396942139,
      "epoch": 0.4680696661828737,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983922136467838,
      "loss": 5.3498,
      "mean_token_accuracy": 0.18027273714542388,
      "num_tokens": 10725799.0,
      "step": 5160
    },
    {
      "entropy": 5.638973999023437,
      "epoch": 0.4685232220609579,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498388350147152,
      "loss": 5.4945,
      "mean_token_accuracy": 0.16939412355422973,
      "num_tokens": 10735668.0,
      "step": 5165
    },
    {
      "entropy": 5.662532472610474,
      "epoch": 0.4689767779390421,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983844820277922,
      "loss": 5.5932,
      "mean_token_accuracy": 0.1683347389101982,
      "num_tokens": 10746466.0,
      "step": 5170
    },
    {
      "entropy": 5.7226646900177,
      "epoch": 0.4694303338171263,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004983806092887843,
      "loss": 5.4903,
      "mean_token_accuracy": 0.170631568133831,
      "num_tokens": 10756724.0,
      "step": 5175
    },
    {
      "entropy": 5.611787223815918,
      "epoch": 0.46988388969521044,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004983767319302086,
      "loss": 5.5413,
      "mean_token_accuracy": 0.1692361503839493,
      "num_tokens": 10766919.0,
      "step": 5180
    },
    {
      "entropy": 5.565976285934449,
      "epoch": 0.47033744557329465,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983728499521451,
      "loss": 5.3111,
      "mean_token_accuracy": 0.1810767963528633,
      "num_tokens": 10777550.0,
      "step": 5185
    },
    {
      "entropy": 5.609437608718872,
      "epoch": 0.4707910014513788,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983689633546741,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17074895054101943,
      "num_tokens": 10788456.0,
      "step": 5190
    },
    {
      "entropy": 5.579924440383911,
      "epoch": 0.47124455732946297,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983650721378761,
      "loss": 5.4118,
      "mean_token_accuracy": 0.1763128623366356,
      "num_tokens": 10800049.0,
      "step": 5195
    },
    {
      "entropy": 5.6112401485443115,
      "epoch": 0.4716981132075472,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983611763018314,
      "loss": 5.4538,
      "mean_token_accuracy": 0.1694786950945854,
      "num_tokens": 10810743.0,
      "step": 5200
    },
    {
      "entropy": 5.722457981109619,
      "epoch": 0.47215166908563133,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004983572758466208,
      "loss": 5.6501,
      "mean_token_accuracy": 0.16584568619728088,
      "num_tokens": 10821897.0,
      "step": 5205
    },
    {
      "entropy": 5.692024803161621,
      "epoch": 0.47260522496371554,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004983533707723247,
      "loss": 5.5064,
      "mean_token_accuracy": 0.17261601835489274,
      "num_tokens": 10831589.0,
      "step": 5210
    },
    {
      "entropy": 5.609469699859619,
      "epoch": 0.4730587808417997,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983494610790241,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16788049787282944,
      "num_tokens": 10841404.0,
      "step": 5215
    },
    {
      "entropy": 5.577998208999634,
      "epoch": 0.4735123367198839,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983455467667996,
      "loss": 5.4547,
      "mean_token_accuracy": 0.16858301162719727,
      "num_tokens": 10851577.0,
      "step": 5220
    },
    {
      "entropy": 5.585168886184692,
      "epoch": 0.47396589259796806,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983416278357322,
      "loss": 5.3676,
      "mean_token_accuracy": 0.17800675630569457,
      "num_tokens": 10861403.0,
      "step": 5225
    },
    {
      "entropy": 5.595915460586548,
      "epoch": 0.4744194484760523,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983377042859032,
      "loss": 5.416,
      "mean_token_accuracy": 0.16920799762010574,
      "num_tokens": 10871933.0,
      "step": 5230
    },
    {
      "entropy": 5.5529121398925785,
      "epoch": 0.47487300435413643,
      "grad_norm": 1.0,
      "learning_rate": 0.0004983337761173936,
      "loss": 5.4844,
      "mean_token_accuracy": 0.17355937510728836,
      "num_tokens": 10881800.0,
      "step": 5235
    },
    {
      "entropy": 5.67032265663147,
      "epoch": 0.4753265602322206,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983298433302843,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1732437029480934,
      "num_tokens": 10891830.0,
      "step": 5240
    },
    {
      "entropy": 5.5511308193206785,
      "epoch": 0.4757801161103048,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983259059246572,
      "loss": 5.3629,
      "mean_token_accuracy": 0.1802822083234787,
      "num_tokens": 10901557.0,
      "step": 5245
    },
    {
      "entropy": 5.5265007495880125,
      "epoch": 0.47623367198838895,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983219639005934,
      "loss": 5.4896,
      "mean_token_accuracy": 0.16791147738695145,
      "num_tokens": 10912155.0,
      "step": 5250
    },
    {
      "entropy": 5.609215116500854,
      "epoch": 0.47668722786647316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983180172581745,
      "loss": 5.3888,
      "mean_token_accuracy": 0.17777832001447677,
      "num_tokens": 10922291.0,
      "step": 5255
    },
    {
      "entropy": 5.660725116729736,
      "epoch": 0.4771407837445573,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004983140659974819,
      "loss": 5.4694,
      "mean_token_accuracy": 0.17444853037595748,
      "num_tokens": 10933176.0,
      "step": 5260
    },
    {
      "entropy": 5.595969295501709,
      "epoch": 0.47759433962264153,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004983101101185977,
      "loss": 5.5461,
      "mean_token_accuracy": 0.165020552277565,
      "num_tokens": 10944049.0,
      "step": 5265
    },
    {
      "entropy": 5.5567363739013675,
      "epoch": 0.4780478955007257,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004983061496216035,
      "loss": 5.3823,
      "mean_token_accuracy": 0.1773837149143219,
      "num_tokens": 10954347.0,
      "step": 5270
    },
    {
      "entropy": 5.603483200073242,
      "epoch": 0.4785014513788099,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983021845065812,
      "loss": 5.4351,
      "mean_token_accuracy": 0.17545117288827897,
      "num_tokens": 10965123.0,
      "step": 5275
    },
    {
      "entropy": 5.5426102638244625,
      "epoch": 0.47895500725689405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004982982147736128,
      "loss": 5.429,
      "mean_token_accuracy": 0.17363492995500565,
      "num_tokens": 10976395.0,
      "step": 5280
    },
    {
      "entropy": 5.617046165466308,
      "epoch": 0.4794085631349782,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004982942404227805,
      "loss": 5.4184,
      "mean_token_accuracy": 0.17311217486858368,
      "num_tokens": 10986797.0,
      "step": 5285
    },
    {
      "entropy": 5.655211639404297,
      "epoch": 0.4798621190130624,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982902614541662,
      "loss": 5.4858,
      "mean_token_accuracy": 0.1699137955904007,
      "num_tokens": 10997229.0,
      "step": 5290
    },
    {
      "entropy": 5.557010078430176,
      "epoch": 0.48031567489114657,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982862778678526,
      "loss": 5.4061,
      "mean_token_accuracy": 0.18039322644472122,
      "num_tokens": 11008180.0,
      "step": 5295
    },
    {
      "entropy": 5.612529039382935,
      "epoch": 0.4807692307692308,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982822896639218,
      "loss": 5.4279,
      "mean_token_accuracy": 0.179520645737648,
      "num_tokens": 11019248.0,
      "step": 5300
    },
    {
      "entropy": 5.558038425445557,
      "epoch": 0.48122278664731494,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982782968424563,
      "loss": 5.352,
      "mean_token_accuracy": 0.1724263459444046,
      "num_tokens": 11029760.0,
      "step": 5305
    },
    {
      "entropy": 5.563699245452881,
      "epoch": 0.48167634252539915,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982742994035388,
      "loss": 5.4403,
      "mean_token_accuracy": 0.1763938307762146,
      "num_tokens": 11039211.0,
      "step": 5310
    },
    {
      "entropy": 5.577710294723511,
      "epoch": 0.4821298984034833,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982702973472518,
      "loss": 5.4115,
      "mean_token_accuracy": 0.17324711233377457,
      "num_tokens": 11049892.0,
      "step": 5315
    },
    {
      "entropy": 5.665327739715576,
      "epoch": 0.4825834542815675,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982662906736783,
      "loss": 5.4634,
      "mean_token_accuracy": 0.17439174950122832,
      "num_tokens": 11060823.0,
      "step": 5320
    },
    {
      "entropy": 5.543615531921387,
      "epoch": 0.48303701015965167,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004982622793829009,
      "loss": 5.4282,
      "mean_token_accuracy": 0.17477934807538986,
      "num_tokens": 11072353.0,
      "step": 5325
    },
    {
      "entropy": 5.606219053268433,
      "epoch": 0.4834905660377358,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982582634750027,
      "loss": 5.4479,
      "mean_token_accuracy": 0.1762908637523651,
      "num_tokens": 11082655.0,
      "step": 5330
    },
    {
      "entropy": 5.619220972061157,
      "epoch": 0.48394412191582004,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004982542429500667,
      "loss": 5.4757,
      "mean_token_accuracy": 0.1755259931087494,
      "num_tokens": 11093315.0,
      "step": 5335
    },
    {
      "entropy": 5.478820943832398,
      "epoch": 0.4843976777939042,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982502178081761,
      "loss": 5.3626,
      "mean_token_accuracy": 0.1801849588751793,
      "num_tokens": 11103074.0,
      "step": 5340
    },
    {
      "entropy": 5.587391710281372,
      "epoch": 0.4848512336719884,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982461880494141,
      "loss": 5.3863,
      "mean_token_accuracy": 0.1792117804288864,
      "num_tokens": 11113636.0,
      "step": 5345
    },
    {
      "entropy": 5.571300601959228,
      "epoch": 0.48530478955007256,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000498242153673864,
      "loss": 5.367,
      "mean_token_accuracy": 0.17792698591947556,
      "num_tokens": 11123509.0,
      "step": 5350
    },
    {
      "entropy": 5.555966091156006,
      "epoch": 0.48575834542815677,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982381146816094,
      "loss": 5.413,
      "mean_token_accuracy": 0.17205502539873124,
      "num_tokens": 11133405.0,
      "step": 5355
    },
    {
      "entropy": 5.676434469223023,
      "epoch": 0.4862119013062409,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004982340710727336,
      "loss": 5.4994,
      "mean_token_accuracy": 0.1640792816877365,
      "num_tokens": 11144725.0,
      "step": 5360
    },
    {
      "entropy": 5.577190208435058,
      "epoch": 0.48666545718432513,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004982300228473203,
      "loss": 5.378,
      "mean_token_accuracy": 0.17402091771364211,
      "num_tokens": 11155446.0,
      "step": 5365
    },
    {
      "entropy": 5.566140747070312,
      "epoch": 0.4871190130624093,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982259700054533,
      "loss": 5.4423,
      "mean_token_accuracy": 0.1782988503575325,
      "num_tokens": 11165380.0,
      "step": 5370
    },
    {
      "entropy": 5.599629878997803,
      "epoch": 0.48757256894049344,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982219125472163,
      "loss": 5.4825,
      "mean_token_accuracy": 0.17474941462278365,
      "num_tokens": 11175641.0,
      "step": 5375
    },
    {
      "entropy": 5.750193023681641,
      "epoch": 0.48802612481857766,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004982178504726933,
      "loss": 5.6549,
      "mean_token_accuracy": 0.17236959785223008,
      "num_tokens": 11186257.0,
      "step": 5380
    },
    {
      "entropy": 5.677424907684326,
      "epoch": 0.4884796806966618,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004982137837819682,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16541945934295654,
      "num_tokens": 11197284.0,
      "step": 5385
    },
    {
      "entropy": 5.629317426681519,
      "epoch": 0.488933236574746,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982097124751252,
      "loss": 5.4406,
      "mean_token_accuracy": 0.17467922419309617,
      "num_tokens": 11208413.0,
      "step": 5390
    },
    {
      "entropy": 5.571043968200684,
      "epoch": 0.4893867924528302,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004982056365522486,
      "loss": 5.438,
      "mean_token_accuracy": 0.17819569408893585,
      "num_tokens": 11219016.0,
      "step": 5395
    },
    {
      "entropy": 5.607671308517456,
      "epoch": 0.4898403483309144,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982015560134225,
      "loss": 5.405,
      "mean_token_accuracy": 0.17524216771125795,
      "num_tokens": 11228934.0,
      "step": 5400
    },
    {
      "entropy": 5.603353881835938,
      "epoch": 0.49029390420899854,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981974708587312,
      "loss": 5.3277,
      "mean_token_accuracy": 0.1815441757440567,
      "num_tokens": 11239596.0,
      "step": 5405
    },
    {
      "entropy": 5.519843006134034,
      "epoch": 0.49074746008708275,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004981933810882594,
      "loss": 5.3232,
      "mean_token_accuracy": 0.17441688776016234,
      "num_tokens": 11249560.0,
      "step": 5410
    },
    {
      "entropy": 5.6501236915588375,
      "epoch": 0.4912010159651669,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004981892867020917,
      "loss": 5.5281,
      "mean_token_accuracy": 0.17009060010313987,
      "num_tokens": 11259787.0,
      "step": 5415
    },
    {
      "entropy": 5.669698715209961,
      "epoch": 0.49165457184325106,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004981851877003125,
      "loss": 5.4462,
      "mean_token_accuracy": 0.17556439489126205,
      "num_tokens": 11269717.0,
      "step": 5420
    },
    {
      "entropy": 5.540651464462281,
      "epoch": 0.4921081277213353,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981810840830068,
      "loss": 5.4329,
      "mean_token_accuracy": 0.18177998811006546,
      "num_tokens": 11279863.0,
      "step": 5425
    },
    {
      "entropy": 5.613092374801636,
      "epoch": 0.49256168359941943,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981769758502596,
      "loss": 5.555,
      "mean_token_accuracy": 0.16333883106708527,
      "num_tokens": 11291161.0,
      "step": 5430
    },
    {
      "entropy": 5.5472633838653564,
      "epoch": 0.49301523947750364,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981728630021555,
      "loss": 5.3762,
      "mean_token_accuracy": 0.17442505955696105,
      "num_tokens": 11301151.0,
      "step": 5435
    },
    {
      "entropy": 5.524767255783081,
      "epoch": 0.4934687953555878,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981687455387797,
      "loss": 5.3818,
      "mean_token_accuracy": 0.1735568791627884,
      "num_tokens": 11311466.0,
      "step": 5440
    },
    {
      "entropy": 5.597518110275269,
      "epoch": 0.493922351233672,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981646234602173,
      "loss": 5.3424,
      "mean_token_accuracy": 0.17718082964420317,
      "num_tokens": 11321522.0,
      "step": 5445
    },
    {
      "entropy": 5.68077597618103,
      "epoch": 0.49437590711175616,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981604967665539,
      "loss": 5.4617,
      "mean_token_accuracy": 0.17782650887966156,
      "num_tokens": 11332317.0,
      "step": 5450
    },
    {
      "entropy": 5.593668365478516,
      "epoch": 0.4948294629898404,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004981563654578743,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1720297485589981,
      "num_tokens": 11343024.0,
      "step": 5455
    },
    {
      "entropy": 5.524274826049805,
      "epoch": 0.49528301886792453,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981522295342643,
      "loss": 5.365,
      "mean_token_accuracy": 0.17542414665222167,
      "num_tokens": 11353114.0,
      "step": 5460
    },
    {
      "entropy": 5.566524648666382,
      "epoch": 0.4957365747460087,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981480889958092,
      "loss": 5.411,
      "mean_token_accuracy": 0.17415190041065215,
      "num_tokens": 11363446.0,
      "step": 5465
    },
    {
      "entropy": 5.521554613113404,
      "epoch": 0.4961901306240929,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981439438425949,
      "loss": 5.4142,
      "mean_token_accuracy": 0.18057068884372712,
      "num_tokens": 11373413.0,
      "step": 5470
    },
    {
      "entropy": 5.528734445571899,
      "epoch": 0.49664368650217705,
      "grad_norm": 1.0,
      "learning_rate": 0.000498139794074707,
      "loss": 5.2985,
      "mean_token_accuracy": 0.17956365793943405,
      "num_tokens": 11383142.0,
      "step": 5475
    },
    {
      "entropy": 5.62391300201416,
      "epoch": 0.49709724238026126,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981356396922313,
      "loss": 5.4436,
      "mean_token_accuracy": 0.17436255961656572,
      "num_tokens": 11394635.0,
      "step": 5480
    },
    {
      "entropy": 5.548166561126709,
      "epoch": 0.4975507982583454,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004981314806952535,
      "loss": 5.3495,
      "mean_token_accuracy": 0.17908620834350586,
      "num_tokens": 11404442.0,
      "step": 5485
    },
    {
      "entropy": 5.488615989685059,
      "epoch": 0.49800435413642963,
      "grad_norm": 1.0,
      "learning_rate": 0.00049812731708386,
      "loss": 5.3055,
      "mean_token_accuracy": 0.1778433859348297,
      "num_tokens": 11414217.0,
      "step": 5490
    },
    {
      "entropy": 5.632096195220948,
      "epoch": 0.4984579100145138,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004981231488581367,
      "loss": 5.435,
      "mean_token_accuracy": 0.1736169531941414,
      "num_tokens": 11423834.0,
      "step": 5495
    },
    {
      "entropy": 5.467576503753662,
      "epoch": 0.498911465892598,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981189760181699,
      "loss": 5.3164,
      "mean_token_accuracy": 0.17622301876544952,
      "num_tokens": 11433613.0,
      "step": 5500
    },
    {
      "entropy": 5.507274103164673,
      "epoch": 0.49936502177068215,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004981147985640456,
      "loss": 5.3959,
      "mean_token_accuracy": 0.1766986832022667,
      "num_tokens": 11443915.0,
      "step": 5505
    },
    {
      "entropy": 5.668768692016601,
      "epoch": 0.4998185776487663,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981106164958506,
      "loss": 5.4205,
      "mean_token_accuracy": 0.17393814772367477,
      "num_tokens": 11454588.0,
      "step": 5510
    },
    {
      "entropy": 5.569959020614624,
      "epoch": 0.5002721335268505,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981064298136712,
      "loss": 5.4766,
      "mean_token_accuracy": 0.17282384485006333,
      "num_tokens": 11465533.0,
      "step": 5515
    },
    {
      "entropy": 5.569711923599243,
      "epoch": 0.5007256894049347,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004981022385175939,
      "loss": 5.4259,
      "mean_token_accuracy": 0.1630531370639801,
      "num_tokens": 11475776.0,
      "step": 5520
    },
    {
      "entropy": 5.567735767364502,
      "epoch": 0.5011792452830188,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004980980426077054,
      "loss": 5.4036,
      "mean_token_accuracy": 0.18170717507600784,
      "num_tokens": 11486200.0,
      "step": 5525
    },
    {
      "entropy": 5.587890720367431,
      "epoch": 0.5016328011611031,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004980938420840926,
      "loss": 5.4315,
      "mean_token_accuracy": 0.1731704905629158,
      "num_tokens": 11496768.0,
      "step": 5530
    },
    {
      "entropy": 5.611096572875977,
      "epoch": 0.5020863570391872,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004980896369468423,
      "loss": 5.3881,
      "mean_token_accuracy": 0.17920028865337373,
      "num_tokens": 11507096.0,
      "step": 5535
    },
    {
      "entropy": 5.5796503067016605,
      "epoch": 0.5025399129172714,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980854271960415,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1763727992773056,
      "num_tokens": 11517301.0,
      "step": 5540
    },
    {
      "entropy": 5.62597918510437,
      "epoch": 0.5029934687953556,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004980812128317772,
      "loss": 5.5175,
      "mean_token_accuracy": 0.17588469237089158,
      "num_tokens": 11528616.0,
      "step": 5545
    },
    {
      "entropy": 5.653271436691284,
      "epoch": 0.5034470246734397,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980769938541364,
      "loss": 5.4301,
      "mean_token_accuracy": 0.18069574236869812,
      "num_tokens": 11539344.0,
      "step": 5550
    },
    {
      "entropy": 5.5967125415802,
      "epoch": 0.503900580551524,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980727702632066,
      "loss": 5.4819,
      "mean_token_accuracy": 0.17302369624376296,
      "num_tokens": 11549511.0,
      "step": 5555
    },
    {
      "entropy": 5.592402076721191,
      "epoch": 0.5043541364296081,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004980685420590752,
      "loss": 5.4184,
      "mean_token_accuracy": 0.17681699395179748,
      "num_tokens": 11559425.0,
      "step": 5560
    },
    {
      "entropy": 5.674917411804199,
      "epoch": 0.5048076923076923,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980643092418294,
      "loss": 5.5278,
      "mean_token_accuracy": 0.16729987114667894,
      "num_tokens": 11570406.0,
      "step": 5565
    },
    {
      "entropy": 5.604274845123291,
      "epoch": 0.5052612481857764,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980600718115568,
      "loss": 5.4232,
      "mean_token_accuracy": 0.17531515061855316,
      "num_tokens": 11580509.0,
      "step": 5570
    },
    {
      "entropy": 5.593520736694336,
      "epoch": 0.5057148040638607,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004980558297683452,
      "loss": 5.431,
      "mean_token_accuracy": 0.1766576960682869,
      "num_tokens": 11590860.0,
      "step": 5575
    },
    {
      "entropy": 5.602160930633545,
      "epoch": 0.5061683599419449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980515831122822,
      "loss": 5.4209,
      "mean_token_accuracy": 0.17564485520124434,
      "num_tokens": 11601954.0,
      "step": 5580
    },
    {
      "entropy": 5.609522008895874,
      "epoch": 0.506621915820029,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004980473318434555,
      "loss": 5.4391,
      "mean_token_accuracy": 0.17501976639032363,
      "num_tokens": 11612581.0,
      "step": 5585
    },
    {
      "entropy": 5.5128148078918455,
      "epoch": 0.5070754716981132,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004980430759619532,
      "loss": 5.3458,
      "mean_token_accuracy": 0.17789707481861114,
      "num_tokens": 11622599.0,
      "step": 5590
    },
    {
      "entropy": 5.4920727729797365,
      "epoch": 0.5075290275761973,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980388154678633,
      "loss": 5.3789,
      "mean_token_accuracy": 0.17600786685943604,
      "num_tokens": 11633611.0,
      "step": 5595
    },
    {
      "entropy": 5.540155076980591,
      "epoch": 0.5079825834542816,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980345503612738,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17380252927541734,
      "num_tokens": 11643159.0,
      "step": 5600
    },
    {
      "entropy": 5.554949855804443,
      "epoch": 0.5084361393323658,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980302806422729,
      "loss": 5.3301,
      "mean_token_accuracy": 0.18010354191064834,
      "num_tokens": 11653174.0,
      "step": 5605
    },
    {
      "entropy": 5.570931339263916,
      "epoch": 0.5088896952104499,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980260063109491,
      "loss": 5.3818,
      "mean_token_accuracy": 0.17837395519018173,
      "num_tokens": 11662621.0,
      "step": 5610
    },
    {
      "entropy": 5.487890195846558,
      "epoch": 0.5093432510885341,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004980217273673906,
      "loss": 5.3409,
      "mean_token_accuracy": 0.1755935877561569,
      "num_tokens": 11673538.0,
      "step": 5615
    },
    {
      "entropy": 5.584263563156128,
      "epoch": 0.5097968069666183,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980174438116859,
      "loss": 5.355,
      "mean_token_accuracy": 0.18188655823469163,
      "num_tokens": 11683203.0,
      "step": 5620
    },
    {
      "entropy": 5.511223268508911,
      "epoch": 0.5102503628447025,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980131556439236,
      "loss": 5.3434,
      "mean_token_accuracy": 0.18352581113576888,
      "num_tokens": 11693313.0,
      "step": 5625
    },
    {
      "entropy": 5.590123558044434,
      "epoch": 0.5107039187227866,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004980088628641925,
      "loss": 5.4414,
      "mean_token_accuracy": 0.17572334557771682,
      "num_tokens": 11704194.0,
      "step": 5630
    },
    {
      "entropy": 5.572856807708741,
      "epoch": 0.5111574746008708,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004980045654725812,
      "loss": 5.3004,
      "mean_token_accuracy": 0.18134156614542007,
      "num_tokens": 11715109.0,
      "step": 5635
    },
    {
      "entropy": 5.505990886688233,
      "epoch": 0.511611030478955,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980002634691787,
      "loss": 5.3408,
      "mean_token_accuracy": 0.18385408073663712,
      "num_tokens": 11724961.0,
      "step": 5640
    },
    {
      "entropy": 5.480311536788941,
      "epoch": 0.5120645863570392,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497995956854074,
      "loss": 5.3565,
      "mean_token_accuracy": 0.1773085355758667,
      "num_tokens": 11735334.0,
      "step": 5645
    },
    {
      "entropy": 5.5139413356781,
      "epoch": 0.5125181422351234,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979916456273561,
      "loss": 5.3665,
      "mean_token_accuracy": 0.17331959158182145,
      "num_tokens": 11746858.0,
      "step": 5650
    },
    {
      "entropy": 5.6394970417022705,
      "epoch": 0.5129716981132075,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979873297891141,
      "loss": 5.4643,
      "mean_token_accuracy": 0.17096892297267913,
      "num_tokens": 11757819.0,
      "step": 5655
    },
    {
      "entropy": 5.578557634353638,
      "epoch": 0.5134252539912917,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979830093394372,
      "loss": 5.3963,
      "mean_token_accuracy": 0.17578381448984146,
      "num_tokens": 11767470.0,
      "step": 5660
    },
    {
      "entropy": 5.511473894119263,
      "epoch": 0.513878809869376,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979786842784149,
      "loss": 5.3225,
      "mean_token_accuracy": 0.18720675110816956,
      "num_tokens": 11777060.0,
      "step": 5665
    },
    {
      "entropy": 5.539907693862915,
      "epoch": 0.5143323657474601,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004979743546061366,
      "loss": 5.3779,
      "mean_token_accuracy": 0.179094859957695,
      "num_tokens": 11787176.0,
      "step": 5670
    },
    {
      "entropy": 5.490522623062134,
      "epoch": 0.5147859216255443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979700203226918,
      "loss": 5.2427,
      "mean_token_accuracy": 0.1881948873400688,
      "num_tokens": 11796419.0,
      "step": 5675
    },
    {
      "entropy": 5.593269300460816,
      "epoch": 0.5152394775036284,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979656814281702,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1741935670375824,
      "num_tokens": 11807279.0,
      "step": 5680
    },
    {
      "entropy": 5.544387626647949,
      "epoch": 0.5156930333817126,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004979613379226615,
      "loss": 5.3819,
      "mean_token_accuracy": 0.1720898911356926,
      "num_tokens": 11817669.0,
      "step": 5685
    },
    {
      "entropy": 5.501872730255127,
      "epoch": 0.5161465892597968,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979569898062554,
      "loss": 5.3753,
      "mean_token_accuracy": 0.18298592865467073,
      "num_tokens": 11827247.0,
      "step": 5690
    },
    {
      "entropy": 5.57254548072815,
      "epoch": 0.516600145137881,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497952637079042,
      "loss": 5.362,
      "mean_token_accuracy": 0.17560323923826218,
      "num_tokens": 11838201.0,
      "step": 5695
    },
    {
      "entropy": 5.53571834564209,
      "epoch": 0.5170537010159652,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004979482797411114,
      "loss": 5.3133,
      "mean_token_accuracy": 0.1851049169898033,
      "num_tokens": 11847561.0,
      "step": 5700
    },
    {
      "entropy": 5.582940626144409,
      "epoch": 0.5175072568940493,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979439177925534,
      "loss": 5.4319,
      "mean_token_accuracy": 0.1830618754029274,
      "num_tokens": 11858272.0,
      "step": 5705
    },
    {
      "entropy": 5.607063817977905,
      "epoch": 0.5179608127721336,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004979395512334584,
      "loss": 5.4662,
      "mean_token_accuracy": 0.17159156501293182,
      "num_tokens": 11869660.0,
      "step": 5710
    },
    {
      "entropy": 5.618724775314331,
      "epoch": 0.5184143686502177,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979351800639166,
      "loss": 5.4987,
      "mean_token_accuracy": 0.171364863216877,
      "num_tokens": 11880239.0,
      "step": 5715
    },
    {
      "entropy": 5.676260328292846,
      "epoch": 0.5188679245283019,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979308042840186,
      "loss": 5.4668,
      "mean_token_accuracy": 0.1713451027870178,
      "num_tokens": 11890630.0,
      "step": 5720
    },
    {
      "entropy": 5.644466686248779,
      "epoch": 0.519321480406386,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004979264238938547,
      "loss": 5.4871,
      "mean_token_accuracy": 0.16992147564888,
      "num_tokens": 11900948.0,
      "step": 5725
    },
    {
      "entropy": 5.512283897399902,
      "epoch": 0.5197750362844702,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004979220388935155,
      "loss": 5.3824,
      "mean_token_accuracy": 0.17406109869480133,
      "num_tokens": 11912528.0,
      "step": 5730
    },
    {
      "entropy": 5.519187307357788,
      "epoch": 0.5202285921625545,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979176492830918,
      "loss": 5.3426,
      "mean_token_accuracy": 0.17976761609315872,
      "num_tokens": 11922211.0,
      "step": 5735
    },
    {
      "entropy": 5.571788311004639,
      "epoch": 0.5206821480406386,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004979132550626742,
      "loss": 5.377,
      "mean_token_accuracy": 0.17560189813375474,
      "num_tokens": 11932520.0,
      "step": 5740
    },
    {
      "entropy": 5.545016241073609,
      "epoch": 0.5211357039187228,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004979088562323538,
      "loss": 5.293,
      "mean_token_accuracy": 0.18467068821191787,
      "num_tokens": 11943312.0,
      "step": 5745
    },
    {
      "entropy": 5.4827127933502195,
      "epoch": 0.5215892597968069,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979044527922214,
      "loss": 5.4606,
      "mean_token_accuracy": 0.17181172966957092,
      "num_tokens": 11954455.0,
      "step": 5750
    },
    {
      "entropy": 5.57046070098877,
      "epoch": 0.5220428156748912,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979000447423681,
      "loss": 5.3524,
      "mean_token_accuracy": 0.1794617787003517,
      "num_tokens": 11964650.0,
      "step": 5755
    },
    {
      "entropy": 5.524151277542114,
      "epoch": 0.5224963715529753,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004978956320828851,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17838129699230193,
      "num_tokens": 11975134.0,
      "step": 5760
    },
    {
      "entropy": 5.5696474552154545,
      "epoch": 0.5229499274310595,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978912148138637,
      "loss": 5.4274,
      "mean_token_accuracy": 0.17480608224868774,
      "num_tokens": 11986486.0,
      "step": 5765
    },
    {
      "entropy": 5.633633422851562,
      "epoch": 0.5234034833091437,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497886792935395,
      "loss": 5.4444,
      "mean_token_accuracy": 0.1734721377491951,
      "num_tokens": 11998045.0,
      "step": 5770
    },
    {
      "entropy": 5.634159660339355,
      "epoch": 0.5238570391872278,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004978823664475707,
      "loss": 5.4267,
      "mean_token_accuracy": 0.17873497009277345,
      "num_tokens": 12008130.0,
      "step": 5775
    },
    {
      "entropy": 5.544890308380127,
      "epoch": 0.5243105950653121,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978779353504824,
      "loss": 5.4998,
      "mean_token_accuracy": 0.16745634078979493,
      "num_tokens": 12018753.0,
      "step": 5780
    },
    {
      "entropy": 5.579509639739991,
      "epoch": 0.5247641509433962,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978734996442214,
      "loss": 5.3882,
      "mean_token_accuracy": 0.17942756414413452,
      "num_tokens": 12028202.0,
      "step": 5785
    },
    {
      "entropy": 5.576986360549927,
      "epoch": 0.5252177068214804,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978690593288798,
      "loss": 5.4305,
      "mean_token_accuracy": 0.18245516568422318,
      "num_tokens": 12039652.0,
      "step": 5790
    },
    {
      "entropy": 5.620332813262939,
      "epoch": 0.5256712626995645,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004978646144045491,
      "loss": 5.3697,
      "mean_token_accuracy": 0.17105551362037658,
      "num_tokens": 12051269.0,
      "step": 5795
    },
    {
      "entropy": 5.445338344573974,
      "epoch": 0.5261248185776488,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004978601648713215,
      "loss": 5.2544,
      "mean_token_accuracy": 0.17919528782367705,
      "num_tokens": 12060647.0,
      "step": 5800
    },
    {
      "entropy": 5.515891742706299,
      "epoch": 0.526578374455733,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978557107292889,
      "loss": 5.403,
      "mean_token_accuracy": 0.1814267262816429,
      "num_tokens": 12071669.0,
      "step": 5805
    },
    {
      "entropy": 5.515309429168701,
      "epoch": 0.5270319303338171,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978512519785433,
      "loss": 5.2696,
      "mean_token_accuracy": 0.18379224985837936,
      "num_tokens": 12081971.0,
      "step": 5810
    },
    {
      "entropy": 5.512398910522461,
      "epoch": 0.5274854862119013,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497846788619177,
      "loss": 5.4025,
      "mean_token_accuracy": 0.18137486279010773,
      "num_tokens": 12092465.0,
      "step": 5815
    },
    {
      "entropy": 5.5811525821685795,
      "epoch": 0.5279390420899854,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978423206512825,
      "loss": 5.351,
      "mean_token_accuracy": 0.1765703558921814,
      "num_tokens": 12101752.0,
      "step": 5820
    },
    {
      "entropy": 5.611680221557617,
      "epoch": 0.5283925979680697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497837848074952,
      "loss": 5.4266,
      "mean_token_accuracy": 0.1774984195828438,
      "num_tokens": 12111805.0,
      "step": 5825
    },
    {
      "entropy": 5.605977153778076,
      "epoch": 0.5288461538461539,
      "grad_norm": 0.9375,
      "learning_rate": 0.000497833370890278,
      "loss": 5.4652,
      "mean_token_accuracy": 0.17292639762163162,
      "num_tokens": 12122511.0,
      "step": 5830
    },
    {
      "entropy": 5.5387050151824955,
      "epoch": 0.529299709724238,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978288890973529,
      "loss": 5.3391,
      "mean_token_accuracy": 0.17948324829339982,
      "num_tokens": 12132510.0,
      "step": 5835
    },
    {
      "entropy": 5.529839754104614,
      "epoch": 0.5297532656023222,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004978244026962698,
      "loss": 5.3596,
      "mean_token_accuracy": 0.17620361894369124,
      "num_tokens": 12142419.0,
      "step": 5840
    },
    {
      "entropy": 5.548328828811646,
      "epoch": 0.5302068214804064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004978199116871212,
      "loss": 5.3942,
      "mean_token_accuracy": 0.17589981257915496,
      "num_tokens": 12152759.0,
      "step": 5845
    },
    {
      "entropy": 5.519029521942139,
      "epoch": 0.5306603773584906,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00049781541607,
      "loss": 5.352,
      "mean_token_accuracy": 0.17770048826932908,
      "num_tokens": 12162758.0,
      "step": 5850
    },
    {
      "entropy": 5.528452920913696,
      "epoch": 0.5311139332365747,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978109158449993,
      "loss": 5.4023,
      "mean_token_accuracy": 0.17557057440280915,
      "num_tokens": 12172664.0,
      "step": 5855
    },
    {
      "entropy": 5.534512901306153,
      "epoch": 0.5315674891146589,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004978064110122121,
      "loss": 5.3733,
      "mean_token_accuracy": 0.17872561067342757,
      "num_tokens": 12182028.0,
      "step": 5860
    },
    {
      "entropy": 5.5767758846282955,
      "epoch": 0.532021044992743,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004978019015717314,
      "loss": 5.4254,
      "mean_token_accuracy": 0.17671894729137422,
      "num_tokens": 12192168.0,
      "step": 5865
    },
    {
      "entropy": 5.480035495758057,
      "epoch": 0.5324746008708273,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977973875236508,
      "loss": 5.3214,
      "mean_token_accuracy": 0.17603713423013687,
      "num_tokens": 12202889.0,
      "step": 5870
    },
    {
      "entropy": 5.484272623062134,
      "epoch": 0.5329281567489115,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004977928688680634,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1813318744301796,
      "num_tokens": 12212514.0,
      "step": 5875
    },
    {
      "entropy": 5.513952398300171,
      "epoch": 0.5333817126269956,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004977883456050626,
      "loss": 5.2224,
      "mean_token_accuracy": 0.18456854969263076,
      "num_tokens": 12222353.0,
      "step": 5880
    },
    {
      "entropy": 5.614924669265747,
      "epoch": 0.5338352685050798,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497783817734742,
      "loss": 5.5128,
      "mean_token_accuracy": 0.17102179080247878,
      "num_tokens": 12232845.0,
      "step": 5885
    },
    {
      "entropy": 5.568334293365479,
      "epoch": 0.534288824383164,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004977792852571954,
      "loss": 5.4261,
      "mean_token_accuracy": 0.17077252715826036,
      "num_tokens": 12242950.0,
      "step": 5890
    },
    {
      "entropy": 5.563507461547852,
      "epoch": 0.5347423802612482,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977747481725164,
      "loss": 5.406,
      "mean_token_accuracy": 0.17852654159069062,
      "num_tokens": 12253132.0,
      "step": 5895
    },
    {
      "entropy": 5.545976257324218,
      "epoch": 0.5351959361393324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977702064807987,
      "loss": 5.3649,
      "mean_token_accuracy": 0.17776333838701247,
      "num_tokens": 12263214.0,
      "step": 5900
    },
    {
      "entropy": 5.4558085918426515,
      "epoch": 0.5356494920174165,
      "grad_norm": 1.0,
      "learning_rate": 0.0004977656601821366,
      "loss": 5.2902,
      "mean_token_accuracy": 0.18908421099185943,
      "num_tokens": 12273250.0,
      "step": 5905
    },
    {
      "entropy": 5.684365606307983,
      "epoch": 0.5361030478955007,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004977611092766238,
      "loss": 5.4839,
      "mean_token_accuracy": 0.18275137096643448,
      "num_tokens": 12283767.0,
      "step": 5910
    },
    {
      "entropy": 5.496643781661987,
      "epoch": 0.5365566037735849,
      "grad_norm": 0.875,
      "learning_rate": 0.0004977565537643544,
      "loss": 5.3221,
      "mean_token_accuracy": 0.17724623829126357,
      "num_tokens": 12294612.0,
      "step": 5915
    },
    {
      "entropy": 5.497534513473511,
      "epoch": 0.5370101596516691,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004977519936454227,
      "loss": 5.4078,
      "mean_token_accuracy": 0.17884652316570282,
      "num_tokens": 12305614.0,
      "step": 5920
    },
    {
      "entropy": 5.499866771697998,
      "epoch": 0.5374637155297532,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977474289199231,
      "loss": 5.2892,
      "mean_token_accuracy": 0.17646802216768265,
      "num_tokens": 12316152.0,
      "step": 5925
    },
    {
      "entropy": 5.460436773300171,
      "epoch": 0.5379172714078374,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977428595879499,
      "loss": 5.36,
      "mean_token_accuracy": 0.18156999796628953,
      "num_tokens": 12325940.0,
      "step": 5930
    },
    {
      "entropy": 5.518941020965576,
      "epoch": 0.5383708272859217,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977382856495975,
      "loss": 5.349,
      "mean_token_accuracy": 0.18231007754802703,
      "num_tokens": 12335482.0,
      "step": 5935
    },
    {
      "entropy": 5.581254148483277,
      "epoch": 0.5388243831640058,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004977337071049606,
      "loss": 5.436,
      "mean_token_accuracy": 0.17495866268873214,
      "num_tokens": 12346386.0,
      "step": 5940
    },
    {
      "entropy": 5.62946310043335,
      "epoch": 0.53927793904209,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497729123954134,
      "loss": 5.4003,
      "mean_token_accuracy": 0.17766263484954833,
      "num_tokens": 12356282.0,
      "step": 5945
    },
    {
      "entropy": 5.5623939037323,
      "epoch": 0.5397314949201741,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004977245361972123,
      "loss": 5.3752,
      "mean_token_accuracy": 0.1799081191420555,
      "num_tokens": 12366994.0,
      "step": 5950
    },
    {
      "entropy": 5.515581130981445,
      "epoch": 0.5401850507982583,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004977199438342904,
      "loss": 5.2576,
      "mean_token_accuracy": 0.18349101692438125,
      "num_tokens": 12377524.0,
      "step": 5955
    },
    {
      "entropy": 5.505928182601929,
      "epoch": 0.5406386066763426,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004977153468654634,
      "loss": 5.4312,
      "mean_token_accuracy": 0.1794543370604515,
      "num_tokens": 12387593.0,
      "step": 5960
    },
    {
      "entropy": 5.605174350738525,
      "epoch": 0.5410921625544267,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977107452908262,
      "loss": 5.3838,
      "mean_token_accuracy": 0.1782233878970146,
      "num_tokens": 12398732.0,
      "step": 5965
    },
    {
      "entropy": 5.553425455093384,
      "epoch": 0.5415457184325109,
      "grad_norm": 1.0,
      "learning_rate": 0.000497706139110474,
      "loss": 5.393,
      "mean_token_accuracy": 0.18314189165830613,
      "num_tokens": 12409821.0,
      "step": 5970
    },
    {
      "entropy": 5.556648111343383,
      "epoch": 0.541999274310595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004977015283245021,
      "loss": 5.4116,
      "mean_token_accuracy": 0.17372148483991623,
      "num_tokens": 12420044.0,
      "step": 5975
    },
    {
      "entropy": 5.4501725196838375,
      "epoch": 0.5424528301886793,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976969129330059,
      "loss": 5.32,
      "mean_token_accuracy": 0.1746181458234787,
      "num_tokens": 12430980.0,
      "step": 5980
    },
    {
      "entropy": 5.516680097579956,
      "epoch": 0.5429063860667634,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004976922929360808,
      "loss": 5.3777,
      "mean_token_accuracy": 0.1816636934876442,
      "num_tokens": 12441999.0,
      "step": 5985
    },
    {
      "entropy": 5.596748208999633,
      "epoch": 0.5433599419448476,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004976876683338222,
      "loss": 5.4189,
      "mean_token_accuracy": 0.17574200332164763,
      "num_tokens": 12452837.0,
      "step": 5990
    },
    {
      "entropy": 5.547367906570434,
      "epoch": 0.5438134978229318,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497683039126326,
      "loss": 5.3353,
      "mean_token_accuracy": 0.18248194307088852,
      "num_tokens": 12462142.0,
      "step": 5995
    },
    {
      "entropy": 5.537907600402832,
      "epoch": 0.5442670537010159,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976784053136878,
      "loss": 5.3846,
      "mean_token_accuracy": 0.17923251092433928,
      "num_tokens": 12472756.0,
      "step": 6000
    },
    {
      "epoch": 0.5442670537010159,
      "eval_entropy": 5.300756282288564,
      "eval_loss": 5.398468971252441,
      "eval_mean_token_accuracy": 0.1842530439409735,
      "eval_num_tokens": 12472756.0,
      "eval_runtime": 20.7729,
      "eval_samples_per_second": 1702.214,
      "eval_steps_per_second": 212.777,
      "step": 6000
    },
    {
      "entropy": 5.589378356933594,
      "epoch": 0.5447206095791002,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004976737668960034,
      "loss": 5.5023,
      "mean_token_accuracy": 0.16921191662549973,
      "num_tokens": 12483115.0,
      "step": 6005
    },
    {
      "entropy": 5.606891059875489,
      "epoch": 0.5451741654571843,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976691238733688,
      "loss": 5.3434,
      "mean_token_accuracy": 0.17560307085514068,
      "num_tokens": 12493280.0,
      "step": 6010
    },
    {
      "entropy": 5.555288028717041,
      "epoch": 0.5456277213352685,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049766447624588,
      "loss": 5.4899,
      "mean_token_accuracy": 0.1748093470931053,
      "num_tokens": 12503665.0,
      "step": 6015
    },
    {
      "entropy": 5.522700214385987,
      "epoch": 0.5460812772133526,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497659824013633,
      "loss": 5.3842,
      "mean_token_accuracy": 0.17242820858955382,
      "num_tokens": 12513381.0,
      "step": 6020
    },
    {
      "entropy": 5.633601522445678,
      "epoch": 0.5465348330914369,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004976551671767241,
      "loss": 5.4278,
      "mean_token_accuracy": 0.1727713316679001,
      "num_tokens": 12524121.0,
      "step": 6025
    },
    {
      "entropy": 5.546627807617187,
      "epoch": 0.5469883889695211,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004976505057352496,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17922226190567017,
      "num_tokens": 12533972.0,
      "step": 6030
    },
    {
      "entropy": 5.598764562606812,
      "epoch": 0.5474419448476052,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004976458396893059,
      "loss": 5.5091,
      "mean_token_accuracy": 0.16582815796136857,
      "num_tokens": 12544110.0,
      "step": 6035
    },
    {
      "entropy": 5.587916040420533,
      "epoch": 0.5478955007256894,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976411690389896,
      "loss": 5.3301,
      "mean_token_accuracy": 0.17198276072740554,
      "num_tokens": 12554273.0,
      "step": 6040
    },
    {
      "entropy": 5.588861322402954,
      "epoch": 0.5483490566037735,
      "grad_norm": 0.984375,
      "learning_rate": 0.000497636493784397,
      "loss": 5.4118,
      "mean_token_accuracy": 0.1747738778591156,
      "num_tokens": 12564633.0,
      "step": 6045
    },
    {
      "entropy": 5.5353899002075195,
      "epoch": 0.5488026124818578,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004976318139256251,
      "loss": 5.3269,
      "mean_token_accuracy": 0.18243416100740434,
      "num_tokens": 12573790.0,
      "step": 6050
    },
    {
      "entropy": 5.546478748321533,
      "epoch": 0.549256168359942,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976271294627706,
      "loss": 5.4166,
      "mean_token_accuracy": 0.17173689603805542,
      "num_tokens": 12584265.0,
      "step": 6055
    },
    {
      "entropy": 5.532885789871216,
      "epoch": 0.5497097242380261,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004976224403959303,
      "loss": 5.2774,
      "mean_token_accuracy": 0.17905397415161134,
      "num_tokens": 12594413.0,
      "step": 6060
    },
    {
      "entropy": 5.52966341972351,
      "epoch": 0.5501632801161103,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976177467252012,
      "loss": 5.3433,
      "mean_token_accuracy": 0.18766953349113463,
      "num_tokens": 12604637.0,
      "step": 6065
    },
    {
      "entropy": 5.580493021011352,
      "epoch": 0.5506168359941945,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004976130484506804,
      "loss": 5.4231,
      "mean_token_accuracy": 0.18023614436388016,
      "num_tokens": 12614873.0,
      "step": 6070
    },
    {
      "entropy": 5.555962562561035,
      "epoch": 0.5510703918722787,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004976083455724649,
      "loss": 5.3565,
      "mean_token_accuracy": 0.17727595716714858,
      "num_tokens": 12626182.0,
      "step": 6075
    },
    {
      "entropy": 5.501303195953369,
      "epoch": 0.5515239477503628,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976036380906523,
      "loss": 5.2818,
      "mean_token_accuracy": 0.18620312511920928,
      "num_tokens": 12635418.0,
      "step": 6080
    },
    {
      "entropy": 5.5395348072052,
      "epoch": 0.551977503628447,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975989260053395,
      "loss": 5.4414,
      "mean_token_accuracy": 0.16959528923034667,
      "num_tokens": 12645942.0,
      "step": 6085
    },
    {
      "entropy": 5.557878732681274,
      "epoch": 0.5524310595065312,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004975942093166242,
      "loss": 5.3681,
      "mean_token_accuracy": 0.17604986876249312,
      "num_tokens": 12656460.0,
      "step": 6090
    },
    {
      "entropy": 5.562954711914062,
      "epoch": 0.5528846153846154,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975894880246041,
      "loss": 5.3042,
      "mean_token_accuracy": 0.17875290811061859,
      "num_tokens": 12666916.0,
      "step": 6095
    },
    {
      "entropy": 5.491450262069702,
      "epoch": 0.5533381712626996,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004975847621293766,
      "loss": 5.3514,
      "mean_token_accuracy": 0.18175258487462997,
      "num_tokens": 12677562.0,
      "step": 6100
    },
    {
      "entropy": 5.489293241500855,
      "epoch": 0.5537917271407837,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004975800316310394,
      "loss": 5.3207,
      "mean_token_accuracy": 0.18005092144012452,
      "num_tokens": 12687349.0,
      "step": 6105
    },
    {
      "entropy": 5.5710242748260494,
      "epoch": 0.5542452830188679,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004975752965296906,
      "loss": 5.3769,
      "mean_token_accuracy": 0.17738821506500244,
      "num_tokens": 12697850.0,
      "step": 6110
    },
    {
      "entropy": 5.493455505371093,
      "epoch": 0.5546988388969522,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975705568254278,
      "loss": 5.2428,
      "mean_token_accuracy": 0.17960745394229888,
      "num_tokens": 12708283.0,
      "step": 6115
    },
    {
      "entropy": 5.540375185012818,
      "epoch": 0.5551523947750363,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975658125183493,
      "loss": 5.3476,
      "mean_token_accuracy": 0.1787947803735733,
      "num_tokens": 12718201.0,
      "step": 6120
    },
    {
      "entropy": 5.479549789428711,
      "epoch": 0.5556059506531205,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000497561063608553,
      "loss": 5.292,
      "mean_token_accuracy": 0.18358754068613053,
      "num_tokens": 12728860.0,
      "step": 6125
    },
    {
      "entropy": 5.527003002166748,
      "epoch": 0.5560595065312046,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004975563100961372,
      "loss": 5.354,
      "mean_token_accuracy": 0.17611497938632964,
      "num_tokens": 12739007.0,
      "step": 6130
    },
    {
      "entropy": 5.524245548248291,
      "epoch": 0.5565130624092888,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004975515519812002,
      "loss": 5.2078,
      "mean_token_accuracy": 0.18419712632894517,
      "num_tokens": 12749004.0,
      "step": 6135
    },
    {
      "entropy": 5.496869516372681,
      "epoch": 0.556966618287373,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975467892638402,
      "loss": 5.3626,
      "mean_token_accuracy": 0.18162387162446975,
      "num_tokens": 12760085.0,
      "step": 6140
    },
    {
      "entropy": 5.532464933395386,
      "epoch": 0.5574201741654572,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004975420219441561,
      "loss": 5.3144,
      "mean_token_accuracy": 0.18599002957344055,
      "num_tokens": 12770303.0,
      "step": 6145
    },
    {
      "entropy": 5.5668566703796385,
      "epoch": 0.5578737300435413,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004975372500222462,
      "loss": 5.3268,
      "mean_token_accuracy": 0.18526234328746796,
      "num_tokens": 12780334.0,
      "step": 6150
    },
    {
      "entropy": 5.521327352523803,
      "epoch": 0.5583272859216255,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004975324734982094,
      "loss": 5.3897,
      "mean_token_accuracy": 0.17213321775197982,
      "num_tokens": 12790980.0,
      "step": 6155
    },
    {
      "entropy": 5.559855127334595,
      "epoch": 0.5587808417997098,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975276923721442,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17847093641757966,
      "num_tokens": 12801713.0,
      "step": 6160
    },
    {
      "entropy": 5.581979513168335,
      "epoch": 0.5592343976777939,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975229066441496,
      "loss": 5.4391,
      "mean_token_accuracy": 0.17114087343215942,
      "num_tokens": 12813040.0,
      "step": 6165
    },
    {
      "entropy": 5.521907711029053,
      "epoch": 0.5596879535558781,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004975181163143247,
      "loss": 5.373,
      "mean_token_accuracy": 0.1722074955701828,
      "num_tokens": 12824048.0,
      "step": 6170
    },
    {
      "entropy": 5.556381750106811,
      "epoch": 0.5601415094339622,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975133213827684,
      "loss": 5.3349,
      "mean_token_accuracy": 0.1731349363923073,
      "num_tokens": 12834297.0,
      "step": 6175
    },
    {
      "entropy": 5.628904151916504,
      "epoch": 0.5605950653120464,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975085218495798,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17205490320920944,
      "num_tokens": 12844574.0,
      "step": 6180
    },
    {
      "entropy": 5.476094388961792,
      "epoch": 0.5610486211901307,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975037177148585,
      "loss": 5.3343,
      "mean_token_accuracy": 0.1756583869457245,
      "num_tokens": 12854670.0,
      "step": 6185
    },
    {
      "entropy": 5.502668714523315,
      "epoch": 0.5615021770682148,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004974989089787035,
      "loss": 5.3474,
      "mean_token_accuracy": 0.18209386765956878,
      "num_tokens": 12865281.0,
      "step": 6190
    },
    {
      "entropy": 5.6341406345367435,
      "epoch": 0.561955732946299,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974940956412143,
      "loss": 5.3597,
      "mean_token_accuracy": 0.1820847824215889,
      "num_tokens": 12875222.0,
      "step": 6195
    },
    {
      "entropy": 5.65907883644104,
      "epoch": 0.5624092888243831,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004974892777024905,
      "loss": 5.4245,
      "mean_token_accuracy": 0.16868771463632584,
      "num_tokens": 12885079.0,
      "step": 6200
    },
    {
      "entropy": 5.475528192520142,
      "epoch": 0.5628628447024674,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004974844551626318,
      "loss": 5.3626,
      "mean_token_accuracy": 0.18181288093328477,
      "num_tokens": 12896402.0,
      "step": 6205
    },
    {
      "entropy": 5.545027017593384,
      "epoch": 0.5633164005805515,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004974796280217379,
      "loss": 5.3348,
      "mean_token_accuracy": 0.1814814731478691,
      "num_tokens": 12906403.0,
      "step": 6210
    },
    {
      "entropy": 5.571792650222778,
      "epoch": 0.5637699564586357,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974747962799086,
      "loss": 5.3618,
      "mean_token_accuracy": 0.17638490051031114,
      "num_tokens": 12917061.0,
      "step": 6215
    },
    {
      "entropy": 5.523290824890137,
      "epoch": 0.5642235123367199,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004974699599372437,
      "loss": 5.378,
      "mean_token_accuracy": 0.18156666308641434,
      "num_tokens": 12927182.0,
      "step": 6220
    },
    {
      "entropy": 5.494165849685669,
      "epoch": 0.564677068214804,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004974651189938434,
      "loss": 5.3496,
      "mean_token_accuracy": 0.1800533875823021,
      "num_tokens": 12937371.0,
      "step": 6225
    },
    {
      "entropy": 5.538199424743652,
      "epoch": 0.5651306240928883,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004974602734498079,
      "loss": 5.4027,
      "mean_token_accuracy": 0.17287339568138121,
      "num_tokens": 12948569.0,
      "step": 6230
    },
    {
      "entropy": 5.569438743591308,
      "epoch": 0.5655841799709724,
      "grad_norm": 0.90625,
      "learning_rate": 0.000497455423305237,
      "loss": 5.3436,
      "mean_token_accuracy": 0.17129313349723815,
      "num_tokens": 12958503.0,
      "step": 6235
    },
    {
      "entropy": 5.575509071350098,
      "epoch": 0.5660377358490566,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004974505685602314,
      "loss": 5.368,
      "mean_token_accuracy": 0.1803085759282112,
      "num_tokens": 12968714.0,
      "step": 6240
    },
    {
      "entropy": 5.497352027893067,
      "epoch": 0.5664912917271407,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004974457092148912,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17698344886302947,
      "num_tokens": 12978857.0,
      "step": 6245
    },
    {
      "entropy": 5.623220825195313,
      "epoch": 0.566944847605225,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974408452693172,
      "loss": 5.4974,
      "mean_token_accuracy": 0.17146317958831786,
      "num_tokens": 12989863.0,
      "step": 6250
    },
    {
      "entropy": 5.5716962814331055,
      "epoch": 0.5673984034833092,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974359767236099,
      "loss": 5.3044,
      "mean_token_accuracy": 0.1790088012814522,
      "num_tokens": 13000307.0,
      "step": 6255
    },
    {
      "entropy": 5.550614261627198,
      "epoch": 0.5678519593613933,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974311035778698,
      "loss": 5.3787,
      "mean_token_accuracy": 0.1838524878025055,
      "num_tokens": 13009053.0,
      "step": 6260
    },
    {
      "entropy": 5.540517282485962,
      "epoch": 0.5683055152394775,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974262258321979,
      "loss": 5.406,
      "mean_token_accuracy": 0.17350224703550338,
      "num_tokens": 13019136.0,
      "step": 6265
    },
    {
      "entropy": 5.594241189956665,
      "epoch": 0.5687590711175616,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004974213434866949,
      "loss": 5.446,
      "mean_token_accuracy": 0.1712695136666298,
      "num_tokens": 13030356.0,
      "step": 6270
    },
    {
      "entropy": 5.602076768875122,
      "epoch": 0.5692126269956459,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004974164565414617,
      "loss": 5.4559,
      "mean_token_accuracy": 0.17462238520383835,
      "num_tokens": 13040477.0,
      "step": 6275
    },
    {
      "entropy": 5.55633134841919,
      "epoch": 0.56966618287373,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004974115649965998,
      "loss": 5.3222,
      "mean_token_accuracy": 0.18193186223506927,
      "num_tokens": 13051755.0,
      "step": 6280
    },
    {
      "entropy": 5.541445636749268,
      "epoch": 0.5701197387518142,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974066688522098,
      "loss": 5.3493,
      "mean_token_accuracy": 0.18550432324409485,
      "num_tokens": 13061670.0,
      "step": 6285
    },
    {
      "entropy": 5.4989865779876705,
      "epoch": 0.5705732946298984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974017681083934,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1777109295129776,
      "num_tokens": 13070917.0,
      "step": 6290
    },
    {
      "entropy": 5.532304143905639,
      "epoch": 0.5710268505079826,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004973968627652516,
      "loss": 5.3885,
      "mean_token_accuracy": 0.17264553159475327,
      "num_tokens": 13080577.0,
      "step": 6295
    },
    {
      "entropy": 5.580759334564209,
      "epoch": 0.5714804063860668,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000497391952822886,
      "loss": 5.4129,
      "mean_token_accuracy": 0.17470873892307281,
      "num_tokens": 13091446.0,
      "step": 6300
    },
    {
      "entropy": 5.586512184143066,
      "epoch": 0.5719339622641509,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004973870382813983,
      "loss": 5.4488,
      "mean_token_accuracy": 0.17591034173965453,
      "num_tokens": 13102752.0,
      "step": 6305
    },
    {
      "entropy": 5.6238028049469,
      "epoch": 0.5723875181422351,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973821191408899,
      "loss": 5.4457,
      "mean_token_accuracy": 0.17629251033067703,
      "num_tokens": 13113944.0,
      "step": 6310
    },
    {
      "entropy": 5.575624752044678,
      "epoch": 0.5728410740203193,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973771954014626,
      "loss": 5.3887,
      "mean_token_accuracy": 0.17858594805002212,
      "num_tokens": 13124316.0,
      "step": 6315
    },
    {
      "entropy": 5.624091053009034,
      "epoch": 0.5732946298984035,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004973722670632182,
      "loss": 5.4551,
      "mean_token_accuracy": 0.1725321814417839,
      "num_tokens": 13133567.0,
      "step": 6320
    },
    {
      "entropy": 5.511906528472901,
      "epoch": 0.5737481857764877,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973673341262587,
      "loss": 5.3886,
      "mean_token_accuracy": 0.17628957331180573,
      "num_tokens": 13144279.0,
      "step": 6325
    },
    {
      "entropy": 5.572101020812989,
      "epoch": 0.5742017416545718,
      "grad_norm": 1.015625,
      "learning_rate": 0.000497362396590686,
      "loss": 5.4501,
      "mean_token_accuracy": 0.17332976907491685,
      "num_tokens": 13154463.0,
      "step": 6330
    },
    {
      "entropy": 5.607514715194702,
      "epoch": 0.574655297532656,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004973574544566024,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16702012121677398,
      "num_tokens": 13165565.0,
      "step": 6335
    },
    {
      "entropy": 5.5187538146972654,
      "epoch": 0.5751088534107403,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004973525077241098,
      "loss": 5.3295,
      "mean_token_accuracy": 0.18137111365795136,
      "num_tokens": 13175839.0,
      "step": 6340
    },
    {
      "entropy": 5.462864255905151,
      "epoch": 0.5755624092888244,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973475563933109,
      "loss": 5.3085,
      "mean_token_accuracy": 0.1800212725996971,
      "num_tokens": 13186372.0,
      "step": 6345
    },
    {
      "entropy": 5.509666633605957,
      "epoch": 0.5760159651669086,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004973426004643077,
      "loss": 5.37,
      "mean_token_accuracy": 0.18036460280418395,
      "num_tokens": 13196259.0,
      "step": 6350
    },
    {
      "entropy": 5.572441911697387,
      "epoch": 0.5764695210449927,
      "grad_norm": 1.0,
      "learning_rate": 0.0004973376399372029,
      "loss": 5.344,
      "mean_token_accuracy": 0.1819022089242935,
      "num_tokens": 13206497.0,
      "step": 6355
    },
    {
      "entropy": 5.549651670455932,
      "epoch": 0.5769230769230769,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973326748120991,
      "loss": 5.3967,
      "mean_token_accuracy": 0.17948508709669114,
      "num_tokens": 13217188.0,
      "step": 6360
    },
    {
      "entropy": 5.524637079238891,
      "epoch": 0.5773766328011611,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973277050890988,
      "loss": 5.3873,
      "mean_token_accuracy": 0.17857213616371154,
      "num_tokens": 13226544.0,
      "step": 6365
    },
    {
      "entropy": 5.600632667541504,
      "epoch": 0.5778301886792453,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004973227307683049,
      "loss": 5.4246,
      "mean_token_accuracy": 0.17547977715730667,
      "num_tokens": 13236572.0,
      "step": 6370
    },
    {
      "entropy": 5.669041633605957,
      "epoch": 0.5782837445573294,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973177518498204,
      "loss": 5.4281,
      "mean_token_accuracy": 0.17712328135967254,
      "num_tokens": 13247958.0,
      "step": 6375
    },
    {
      "entropy": 5.547800731658936,
      "epoch": 0.5787373004354136,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973127683337482,
      "loss": 5.3319,
      "mean_token_accuracy": 0.18092712610960007,
      "num_tokens": 13258255.0,
      "step": 6380
    },
    {
      "entropy": 5.465287494659424,
      "epoch": 0.5791908563134979,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973077802201911,
      "loss": 5.3557,
      "mean_token_accuracy": 0.17363798320293428,
      "num_tokens": 13268416.0,
      "step": 6385
    },
    {
      "entropy": 5.503123950958252,
      "epoch": 0.579644412191582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004973027875092526,
      "loss": 5.3293,
      "mean_token_accuracy": 0.17840101420879365,
      "num_tokens": 13278087.0,
      "step": 6390
    },
    {
      "entropy": 5.524763393402099,
      "epoch": 0.5800979680696662,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004972977902010357,
      "loss": 5.2447,
      "mean_token_accuracy": 0.1844359055161476,
      "num_tokens": 13288121.0,
      "step": 6395
    },
    {
      "entropy": 5.6309037685394285,
      "epoch": 0.5805515239477503,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972927882956439,
      "loss": 5.4657,
      "mean_token_accuracy": 0.16494472771883012,
      "num_tokens": 13298620.0,
      "step": 6400
    },
    {
      "entropy": 5.517560768127441,
      "epoch": 0.5810050798258345,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004972877817931807,
      "loss": 5.2753,
      "mean_token_accuracy": 0.17818261831998825,
      "num_tokens": 13308413.0,
      "step": 6405
    },
    {
      "entropy": 5.477993535995483,
      "epoch": 0.5814586357039188,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972827706937493,
      "loss": 5.3562,
      "mean_token_accuracy": 0.18004698753356935,
      "num_tokens": 13319387.0,
      "step": 6410
    },
    {
      "entropy": 5.563785362243652,
      "epoch": 0.5819121915820029,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004972777549974537,
      "loss": 5.3143,
      "mean_token_accuracy": 0.17967743724584578,
      "num_tokens": 13330494.0,
      "step": 6415
    },
    {
      "entropy": 5.541955375671387,
      "epoch": 0.5823657474600871,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972727347043975,
      "loss": 5.3155,
      "mean_token_accuracy": 0.1786987692117691,
      "num_tokens": 13341009.0,
      "step": 6420
    },
    {
      "entropy": 5.490311861038208,
      "epoch": 0.5828193033381712,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972677098146845,
      "loss": 5.2766,
      "mean_token_accuracy": 0.1829589769244194,
      "num_tokens": 13351269.0,
      "step": 6425
    },
    {
      "entropy": 5.4687377452850345,
      "epoch": 0.5832728592162555,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972626803284185,
      "loss": 5.3916,
      "mean_token_accuracy": 0.1788855239748955,
      "num_tokens": 13362160.0,
      "step": 6430
    },
    {
      "entropy": 5.5932941913604735,
      "epoch": 0.5837264150943396,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972576462457038,
      "loss": 5.4421,
      "mean_token_accuracy": 0.18139465898275375,
      "num_tokens": 13372497.0,
      "step": 6435
    },
    {
      "entropy": 5.603217458724975,
      "epoch": 0.5841799709724238,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004972526075666442,
      "loss": 5.3849,
      "mean_token_accuracy": 0.17963762879371642,
      "num_tokens": 13381665.0,
      "step": 6440
    },
    {
      "entropy": 5.481933689117431,
      "epoch": 0.584633526850508,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972475642913442,
      "loss": 5.2862,
      "mean_token_accuracy": 0.18237100541591644,
      "num_tokens": 13392507.0,
      "step": 6445
    },
    {
      "entropy": 5.484732341766358,
      "epoch": 0.5850870827285921,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972425164199078,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18799172341823578,
      "num_tokens": 13402171.0,
      "step": 6450
    },
    {
      "entropy": 5.608668804168701,
      "epoch": 0.5855406386066764,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004972374639524396,
      "loss": 5.4976,
      "mean_token_accuracy": 0.1756341800093651,
      "num_tokens": 13412933.0,
      "step": 6455
    },
    {
      "entropy": 5.547793292999268,
      "epoch": 0.5859941944847605,
      "grad_norm": 0.859375,
      "learning_rate": 0.000497232406889044,
      "loss": 5.3367,
      "mean_token_accuracy": 0.17190575897693633,
      "num_tokens": 13423563.0,
      "step": 6460
    },
    {
      "entropy": 5.600666761398315,
      "epoch": 0.5864477503628447,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004972273452298255,
      "loss": 5.3459,
      "mean_token_accuracy": 0.18437232375144957,
      "num_tokens": 13433998.0,
      "step": 6465
    },
    {
      "entropy": 5.516443872451783,
      "epoch": 0.5869013062409288,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972222789748888,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18250495493412017,
      "num_tokens": 13443878.0,
      "step": 6470
    },
    {
      "entropy": 5.495645189285279,
      "epoch": 0.5873548621190131,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004972172081243389,
      "loss": 5.3899,
      "mean_token_accuracy": 0.17166997939348222,
      "num_tokens": 13456051.0,
      "step": 6475
    },
    {
      "entropy": 5.5309515476226805,
      "epoch": 0.5878084179970973,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004972121326782806,
      "loss": 5.3665,
      "mean_token_accuracy": 0.18319807052612305,
      "num_tokens": 13467276.0,
      "step": 6480
    },
    {
      "entropy": 5.55036153793335,
      "epoch": 0.5882619738751814,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004972070526368186,
      "loss": 5.3465,
      "mean_token_accuracy": 0.17851588428020476,
      "num_tokens": 13478331.0,
      "step": 6485
    },
    {
      "entropy": 5.58794755935669,
      "epoch": 0.5887155297532656,
      "grad_norm": 1.0,
      "learning_rate": 0.0004972019680000582,
      "loss": 5.2955,
      "mean_token_accuracy": 0.17752917855978012,
      "num_tokens": 13489879.0,
      "step": 6490
    },
    {
      "entropy": 5.493243551254272,
      "epoch": 0.5891690856313497,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004971968787681044,
      "loss": 5.4016,
      "mean_token_accuracy": 0.181935915350914,
      "num_tokens": 13500879.0,
      "step": 6495
    },
    {
      "entropy": 5.616210842132569,
      "epoch": 0.589622641509434,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971917849410627,
      "loss": 5.3875,
      "mean_token_accuracy": 0.17739370465278625,
      "num_tokens": 13511719.0,
      "step": 6500
    },
    {
      "entropy": 5.50895209312439,
      "epoch": 0.5900761973875182,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971866865190381,
      "loss": 5.318,
      "mean_token_accuracy": 0.17822761982679367,
      "num_tokens": 13521850.0,
      "step": 6505
    },
    {
      "entropy": 5.563834190368652,
      "epoch": 0.5905297532656023,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004971815835021362,
      "loss": 5.3519,
      "mean_token_accuracy": 0.1825216919183731,
      "num_tokens": 13532095.0,
      "step": 6510
    },
    {
      "entropy": 5.515305042266846,
      "epoch": 0.5909833091436865,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971764758904626,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17511919140815735,
      "num_tokens": 13543133.0,
      "step": 6515
    },
    {
      "entropy": 5.525648593902588,
      "epoch": 0.5914368650217707,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004971713636841229,
      "loss": 5.402,
      "mean_token_accuracy": 0.17349975258111955,
      "num_tokens": 13553771.0,
      "step": 6520
    },
    {
      "entropy": 5.501309537887574,
      "epoch": 0.5918904208998549,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004971662468832228,
      "loss": 5.3479,
      "mean_token_accuracy": 0.18141622096300125,
      "num_tokens": 13564548.0,
      "step": 6525
    },
    {
      "entropy": 5.475589799880981,
      "epoch": 0.592343976777939,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497161125487868,
      "loss": 5.3231,
      "mean_token_accuracy": 0.1807534858584404,
      "num_tokens": 13574567.0,
      "step": 6530
    },
    {
      "entropy": 5.496382856369019,
      "epoch": 0.5927975326560232,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004971559994981645,
      "loss": 5.3686,
      "mean_token_accuracy": 0.1793726459145546,
      "num_tokens": 13585302.0,
      "step": 6535
    },
    {
      "entropy": 5.572533655166626,
      "epoch": 0.5932510885341074,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971508689142184,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17545087039470672,
      "num_tokens": 13594754.0,
      "step": 6540
    },
    {
      "entropy": 5.486157846450806,
      "epoch": 0.5937046444121916,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971457337361356,
      "loss": 5.3268,
      "mean_token_accuracy": 0.177515971660614,
      "num_tokens": 13604691.0,
      "step": 6545
    },
    {
      "entropy": 5.4501392364501955,
      "epoch": 0.5941582002902758,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004971405939640225,
      "loss": 5.2214,
      "mean_token_accuracy": 0.18570286631584168,
      "num_tokens": 13613911.0,
      "step": 6550
    },
    {
      "entropy": 5.636072492599487,
      "epoch": 0.5946117561683599,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971354495979854,
      "loss": 5.4398,
      "mean_token_accuracy": 0.17701357752084732,
      "num_tokens": 13624285.0,
      "step": 6555
    },
    {
      "entropy": 5.521263599395752,
      "epoch": 0.5950653120464441,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971303006381305,
      "loss": 5.2916,
      "mean_token_accuracy": 0.18147354274988176,
      "num_tokens": 13635434.0,
      "step": 6560
    },
    {
      "entropy": 5.4838104248046875,
      "epoch": 0.5955188679245284,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004971251470845643,
      "loss": 5.345,
      "mean_token_accuracy": 0.1768563508987427,
      "num_tokens": 13645582.0,
      "step": 6565
    },
    {
      "entropy": 5.5116071701049805,
      "epoch": 0.5959724238026125,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004971199889373935,
      "loss": 5.3055,
      "mean_token_accuracy": 0.18206693828105927,
      "num_tokens": 13655029.0,
      "step": 6570
    },
    {
      "entropy": 5.480277967453003,
      "epoch": 0.5964259796806967,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004971148261967249,
      "loss": 5.294,
      "mean_token_accuracy": 0.18128707706928254,
      "num_tokens": 13665204.0,
      "step": 6575
    },
    {
      "entropy": 5.5318831443786625,
      "epoch": 0.5968795355587808,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971096588626649,
      "loss": 5.3953,
      "mean_token_accuracy": 0.17587229311466218,
      "num_tokens": 13676281.0,
      "step": 6580
    },
    {
      "entropy": 5.528276824951172,
      "epoch": 0.597333091436865,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004971044869353206,
      "loss": 5.353,
      "mean_token_accuracy": 0.17965753376483917,
      "num_tokens": 13687144.0,
      "step": 6585
    },
    {
      "entropy": 5.529869556427002,
      "epoch": 0.5977866473149492,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000497099310414799,
      "loss": 5.3345,
      "mean_token_accuracy": 0.18199270814657212,
      "num_tokens": 13696371.0,
      "step": 6590
    },
    {
      "entropy": 5.573552417755127,
      "epoch": 0.5982402031930334,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970941293012069,
      "loss": 5.2835,
      "mean_token_accuracy": 0.17838052809238433,
      "num_tokens": 13706900.0,
      "step": 6595
    },
    {
      "entropy": 5.5088542938232425,
      "epoch": 0.5986937590711175,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004970889435946518,
      "loss": 5.3598,
      "mean_token_accuracy": 0.1811433807015419,
      "num_tokens": 13717874.0,
      "step": 6600
    },
    {
      "entropy": 5.550598001480102,
      "epoch": 0.5991473149492017,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970837532952406,
      "loss": 5.3263,
      "mean_token_accuracy": 0.1788433775305748,
      "num_tokens": 13728020.0,
      "step": 6605
    },
    {
      "entropy": 5.562851047515869,
      "epoch": 0.599600870827286,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004970785584030808,
      "loss": 5.372,
      "mean_token_accuracy": 0.1790483370423317,
      "num_tokens": 13738007.0,
      "step": 6610
    },
    {
      "entropy": 5.53517484664917,
      "epoch": 0.6000544267053701,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004970733589182799,
      "loss": 5.3681,
      "mean_token_accuracy": 0.17741776257753372,
      "num_tokens": 13749429.0,
      "step": 6615
    },
    {
      "entropy": 5.50953049659729,
      "epoch": 0.6005079825834543,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004970681548409452,
      "loss": 5.2357,
      "mean_token_accuracy": 0.18484993875026703,
      "num_tokens": 13760279.0,
      "step": 6620
    },
    {
      "entropy": 5.569452714920044,
      "epoch": 0.6009615384615384,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970629461711846,
      "loss": 5.4547,
      "mean_token_accuracy": 0.1746961295604706,
      "num_tokens": 13770575.0,
      "step": 6625
    },
    {
      "entropy": 5.4972024917602536,
      "epoch": 0.6014150943396226,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004970577329091057,
      "loss": 5.2939,
      "mean_token_accuracy": 0.18609136939048768,
      "num_tokens": 13781068.0,
      "step": 6630
    },
    {
      "entropy": 5.549247598648071,
      "epoch": 0.6018686502177069,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004970525150548162,
      "loss": 5.3836,
      "mean_token_accuracy": 0.18321933150291442,
      "num_tokens": 13790745.0,
      "step": 6635
    },
    {
      "entropy": 5.521773195266723,
      "epoch": 0.602322206095791,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970472926084242,
      "loss": 5.3162,
      "mean_token_accuracy": 0.1791524738073349,
      "num_tokens": 13800196.0,
      "step": 6640
    },
    {
      "entropy": 5.490709257125855,
      "epoch": 0.6027757619738752,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970420655700375,
      "loss": 5.2291,
      "mean_token_accuracy": 0.18693480491638184,
      "num_tokens": 13809954.0,
      "step": 6645
    },
    {
      "entropy": 5.512080097198487,
      "epoch": 0.6032293178519593,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004970368339397643,
      "loss": 5.4023,
      "mean_token_accuracy": 0.1746376097202301,
      "num_tokens": 13820281.0,
      "step": 6650
    },
    {
      "entropy": 5.586239385604858,
      "epoch": 0.6036828737300436,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004970315977177129,
      "loss": 5.3464,
      "mean_token_accuracy": 0.18440013825893403,
      "num_tokens": 13830614.0,
      "step": 6655
    },
    {
      "entropy": 5.500546884536743,
      "epoch": 0.6041364296081277,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004970263569039914,
      "loss": 5.2936,
      "mean_token_accuracy": 0.181098048388958,
      "num_tokens": 13840551.0,
      "step": 6660
    },
    {
      "entropy": 5.625216722488403,
      "epoch": 0.6045899854862119,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004970211114987083,
      "loss": 5.478,
      "mean_token_accuracy": 0.17690049409866332,
      "num_tokens": 13851830.0,
      "step": 6665
    },
    {
      "entropy": 5.556969070434571,
      "epoch": 0.6050435413642961,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970158615019719,
      "loss": 5.286,
      "mean_token_accuracy": 0.18250710368156434,
      "num_tokens": 13862299.0,
      "step": 6670
    },
    {
      "entropy": 5.575377559661865,
      "epoch": 0.6054970972423802,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004970106069138912,
      "loss": 5.4059,
      "mean_token_accuracy": 0.1716219589114189,
      "num_tokens": 13872852.0,
      "step": 6675
    },
    {
      "entropy": 5.499694538116455,
      "epoch": 0.6059506531204645,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004970053477345743,
      "loss": 5.3559,
      "mean_token_accuracy": 0.18448759168386458,
      "num_tokens": 13883601.0,
      "step": 6680
    },
    {
      "entropy": 5.550814008712768,
      "epoch": 0.6064042089985486,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004970000839641304,
      "loss": 5.3464,
      "mean_token_accuracy": 0.18434916138648988,
      "num_tokens": 13894246.0,
      "step": 6685
    },
    {
      "entropy": 5.4789330005645756,
      "epoch": 0.6068577648766328,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004969948156026681,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17830681800842285,
      "num_tokens": 13904823.0,
      "step": 6690
    },
    {
      "entropy": 5.598898124694824,
      "epoch": 0.6073113207547169,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969895426502965,
      "loss": 5.3862,
      "mean_token_accuracy": 0.17962592840194702,
      "num_tokens": 13915338.0,
      "step": 6695
    },
    {
      "entropy": 5.507212829589844,
      "epoch": 0.6077648766328012,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004969842651071246,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17749970853328706,
      "num_tokens": 13925838.0,
      "step": 6700
    },
    {
      "entropy": 5.5650975704193115,
      "epoch": 0.6082184325108854,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969789829732615,
      "loss": 5.3388,
      "mean_token_accuracy": 0.17786233574151994,
      "num_tokens": 13936341.0,
      "step": 6705
    },
    {
      "entropy": 5.559619951248169,
      "epoch": 0.6086719883889695,
      "grad_norm": 1.0,
      "learning_rate": 0.0004969736962488164,
      "loss": 5.3525,
      "mean_token_accuracy": 0.1801693171262741,
      "num_tokens": 13945438.0,
      "step": 6710
    },
    {
      "entropy": 5.509589719772339,
      "epoch": 0.6091255442670537,
      "grad_norm": 0.875,
      "learning_rate": 0.0004969684049338987,
      "loss": 5.2386,
      "mean_token_accuracy": 0.180333910882473,
      "num_tokens": 13956029.0,
      "step": 6715
    },
    {
      "entropy": 5.5770862102508545,
      "epoch": 0.6095791001451378,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496963109028618,
      "loss": 5.3515,
      "mean_token_accuracy": 0.17293535470962523,
      "num_tokens": 13966498.0,
      "step": 6720
    },
    {
      "entropy": 5.5024824142456055,
      "epoch": 0.6100326560232221,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969578085330834,
      "loss": 5.2984,
      "mean_token_accuracy": 0.1841035008430481,
      "num_tokens": 13976997.0,
      "step": 6725
    },
    {
      "entropy": 5.555294322967529,
      "epoch": 0.6104862119013063,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004969525034474049,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18184979408979415,
      "num_tokens": 13986664.0,
      "step": 6730
    },
    {
      "entropy": 5.456352567672729,
      "epoch": 0.6109397677793904,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000496947193771692,
      "loss": 5.3061,
      "mean_token_accuracy": 0.18491313904523848,
      "num_tokens": 13997828.0,
      "step": 6735
    },
    {
      "entropy": 5.480406856536865,
      "epoch": 0.6113933236574746,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969418795060548,
      "loss": 5.2586,
      "mean_token_accuracy": 0.18284006267786027,
      "num_tokens": 14008293.0,
      "step": 6740
    },
    {
      "entropy": 5.478730154037476,
      "epoch": 0.6118468795355588,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004969365606506027,
      "loss": 5.2436,
      "mean_token_accuracy": 0.186459219455719,
      "num_tokens": 14018031.0,
      "step": 6745
    },
    {
      "entropy": 5.46801495552063,
      "epoch": 0.612300435413643,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004969312372054461,
      "loss": 5.2332,
      "mean_token_accuracy": 0.18784987032413483,
      "num_tokens": 14028013.0,
      "step": 6750
    },
    {
      "entropy": 5.491974496841431,
      "epoch": 0.6127539912917271,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004969259091706948,
      "loss": 5.3637,
      "mean_token_accuracy": 0.1775025248527527,
      "num_tokens": 14039787.0,
      "step": 6755
    },
    {
      "entropy": 5.5373763084411625,
      "epoch": 0.6132075471698113,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969205765464593,
      "loss": 5.2606,
      "mean_token_accuracy": 0.18175734281539918,
      "num_tokens": 14049223.0,
      "step": 6760
    },
    {
      "entropy": 5.5544661521911625,
      "epoch": 0.6136611030478955,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004969152393328497,
      "loss": 5.2985,
      "mean_token_accuracy": 0.18559947311878205,
      "num_tokens": 14059004.0,
      "step": 6765
    },
    {
      "entropy": 5.536505842208863,
      "epoch": 0.6141146589259797,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004969098975299764,
      "loss": 5.4062,
      "mean_token_accuracy": 0.17162830382585526,
      "num_tokens": 14069615.0,
      "step": 6770
    },
    {
      "entropy": 5.525615072250366,
      "epoch": 0.6145682148040639,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004969045511379499,
      "loss": 5.3252,
      "mean_token_accuracy": 0.18242162317037583,
      "num_tokens": 14078886.0,
      "step": 6775
    },
    {
      "entropy": 5.610492515563965,
      "epoch": 0.615021770682148,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004968992001568806,
      "loss": 5.4129,
      "mean_token_accuracy": 0.17199260890483856,
      "num_tokens": 14089021.0,
      "step": 6780
    },
    {
      "entropy": 5.4925415992736815,
      "epoch": 0.6154753265602322,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968938445868795,
      "loss": 5.2152,
      "mean_token_accuracy": 0.1890949234366417,
      "num_tokens": 14098316.0,
      "step": 6785
    },
    {
      "entropy": 5.463628149032592,
      "epoch": 0.6159288824383164,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004968884844280569,
      "loss": 5.3117,
      "mean_token_accuracy": 0.1768708735704422,
      "num_tokens": 14108874.0,
      "step": 6790
    },
    {
      "entropy": 5.520146799087525,
      "epoch": 0.6163824383164006,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968831196805242,
      "loss": 5.3576,
      "mean_token_accuracy": 0.1788746789097786,
      "num_tokens": 14119613.0,
      "step": 6795
    },
    {
      "entropy": 5.559129047393799,
      "epoch": 0.6168359941944848,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968777503443918,
      "loss": 5.3215,
      "mean_token_accuracy": 0.17656567692756653,
      "num_tokens": 14129085.0,
      "step": 6800
    },
    {
      "entropy": 5.440955114364624,
      "epoch": 0.6172895500725689,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004968723764197711,
      "loss": 5.2268,
      "mean_token_accuracy": 0.1918613076210022,
      "num_tokens": 14138461.0,
      "step": 6805
    },
    {
      "entropy": 5.47119174003601,
      "epoch": 0.6177431059506531,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496866997906773,
      "loss": 5.2868,
      "mean_token_accuracy": 0.18615336567163468,
      "num_tokens": 14149126.0,
      "step": 6810
    },
    {
      "entropy": 5.4483057975769045,
      "epoch": 0.6181966618287373,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004968616148055089,
      "loss": 5.2881,
      "mean_token_accuracy": 0.18620866686105728,
      "num_tokens": 14160068.0,
      "step": 6815
    },
    {
      "entropy": 5.509595775604248,
      "epoch": 0.6186502177068215,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968562271160901,
      "loss": 5.3862,
      "mean_token_accuracy": 0.18133724480867386,
      "num_tokens": 14170288.0,
      "step": 6820
    },
    {
      "entropy": 5.591276216506958,
      "epoch": 0.6191037735849056,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000496850834838628,
      "loss": 5.3755,
      "mean_token_accuracy": 0.17940264344215393,
      "num_tokens": 14181231.0,
      "step": 6825
    },
    {
      "entropy": 5.559400033950806,
      "epoch": 0.6195573294629898,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004968454379732342,
      "loss": 5.2794,
      "mean_token_accuracy": 0.1832197815179825,
      "num_tokens": 14191715.0,
      "step": 6830
    },
    {
      "entropy": 5.463568258285522,
      "epoch": 0.6200108853410741,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00049684003652002,
      "loss": 5.2953,
      "mean_token_accuracy": 0.17444696128368378,
      "num_tokens": 14203416.0,
      "step": 6835
    },
    {
      "entropy": 5.4512610912323,
      "epoch": 0.6204644412191582,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004968346304790973,
      "loss": 5.2323,
      "mean_token_accuracy": 0.19206695556640624,
      "num_tokens": 14213658.0,
      "step": 6840
    },
    {
      "entropy": 5.544649982452393,
      "epoch": 0.6209179970972424,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004968292198505781,
      "loss": 5.4084,
      "mean_token_accuracy": 0.17560362964868545,
      "num_tokens": 14223956.0,
      "step": 6845
    },
    {
      "entropy": 5.562683057785034,
      "epoch": 0.6213715529753265,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000496823804634574,
      "loss": 5.3982,
      "mean_token_accuracy": 0.17950718104839325,
      "num_tokens": 14236801.0,
      "step": 6850
    },
    {
      "entropy": 5.515814352035522,
      "epoch": 0.6218251088534107,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004968183848311972,
      "loss": 5.2099,
      "mean_token_accuracy": 0.18720452338457108,
      "num_tokens": 14247382.0,
      "step": 6855
    },
    {
      "entropy": 5.555305004119873,
      "epoch": 0.622278664731495,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004968129604405596,
      "loss": 5.2996,
      "mean_token_accuracy": 0.1855861246585846,
      "num_tokens": 14258627.0,
      "step": 6860
    },
    {
      "entropy": 5.4478870868682865,
      "epoch": 0.6227322206095791,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004968075314627734,
      "loss": 5.3,
      "mean_token_accuracy": 0.18473073691129685,
      "num_tokens": 14269013.0,
      "step": 6865
    },
    {
      "entropy": 5.474765825271606,
      "epoch": 0.6231857764876633,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000496802097897951,
      "loss": 5.2832,
      "mean_token_accuracy": 0.18246868699789048,
      "num_tokens": 14279620.0,
      "step": 6870
    },
    {
      "entropy": 5.5203248977661135,
      "epoch": 0.6236393323657474,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004967966597462046,
      "loss": 5.2874,
      "mean_token_accuracy": 0.1900581583380699,
      "num_tokens": 14289448.0,
      "step": 6875
    },
    {
      "entropy": 5.53533444404602,
      "epoch": 0.6240928882438317,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967912170076469,
      "loss": 5.291,
      "mean_token_accuracy": 0.17730221003293992,
      "num_tokens": 14299999.0,
      "step": 6880
    },
    {
      "entropy": 5.460947799682617,
      "epoch": 0.6245464441219158,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004967857696823901,
      "loss": 5.2673,
      "mean_token_accuracy": 0.1848415583372116,
      "num_tokens": 14310384.0,
      "step": 6885
    },
    {
      "entropy": 5.509242057800293,
      "epoch": 0.625,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967803177705472,
      "loss": 5.3818,
      "mean_token_accuracy": 0.17576736062765122,
      "num_tokens": 14320644.0,
      "step": 6890
    },
    {
      "entropy": 5.552798414230347,
      "epoch": 0.6254535558780842,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967748612722308,
      "loss": 5.3416,
      "mean_token_accuracy": 0.1810254707932472,
      "num_tokens": 14330077.0,
      "step": 6895
    },
    {
      "entropy": 5.5750954151153564,
      "epoch": 0.6259071117561683,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967694001875538,
      "loss": 5.4386,
      "mean_token_accuracy": 0.17333186119794847,
      "num_tokens": 14341549.0,
      "step": 6900
    },
    {
      "entropy": 5.564305830001831,
      "epoch": 0.6263606676342526,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496763934516629,
      "loss": 5.2874,
      "mean_token_accuracy": 0.18876885771751403,
      "num_tokens": 14352146.0,
      "step": 6905
    },
    {
      "entropy": 5.471659088134766,
      "epoch": 0.6268142235123367,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004967584642595696,
      "loss": 5.325,
      "mean_token_accuracy": 0.17631787657737732,
      "num_tokens": 14362521.0,
      "step": 6910
    },
    {
      "entropy": 5.541382551193237,
      "epoch": 0.6272677793904209,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004967529894164887,
      "loss": 5.3537,
      "mean_token_accuracy": 0.17927271872758865,
      "num_tokens": 14373216.0,
      "step": 6915
    },
    {
      "entropy": 5.488159418106079,
      "epoch": 0.627721335268505,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967475099874992,
      "loss": 5.2567,
      "mean_token_accuracy": 0.1907469481229782,
      "num_tokens": 14383104.0,
      "step": 6920
    },
    {
      "entropy": 5.4569638729095455,
      "epoch": 0.6281748911465893,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000496742025972715,
      "loss": 5.2899,
      "mean_token_accuracy": 0.19110392779111862,
      "num_tokens": 14393854.0,
      "step": 6925
    },
    {
      "entropy": 5.516732025146484,
      "epoch": 0.6286284470246735,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496736537372249,
      "loss": 5.2274,
      "mean_token_accuracy": 0.1899970456957817,
      "num_tokens": 14403213.0,
      "step": 6930
    },
    {
      "entropy": 5.603625202178955,
      "epoch": 0.6290820029027576,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004967310441862151,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1735685408115387,
      "num_tokens": 14412792.0,
      "step": 6935
    },
    {
      "entropy": 5.546102237701416,
      "epoch": 0.6295355587808418,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004967255464147265,
      "loss": 5.3143,
      "mean_token_accuracy": 0.18926624953746796,
      "num_tokens": 14423061.0,
      "step": 6940
    },
    {
      "entropy": 5.512146854400635,
      "epoch": 0.6299891146589259,
      "grad_norm": 1.0,
      "learning_rate": 0.0004967200440578972,
      "loss": 5.2589,
      "mean_token_accuracy": 0.19065099507570266,
      "num_tokens": 14433338.0,
      "step": 6945
    },
    {
      "entropy": 5.45038628578186,
      "epoch": 0.6304426705370102,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004967145371158408,
      "loss": 5.2484,
      "mean_token_accuracy": 0.19122012555599213,
      "num_tokens": 14444124.0,
      "step": 6950
    },
    {
      "entropy": 5.501307201385498,
      "epoch": 0.6308962264150944,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004967090255886714,
      "loss": 5.3908,
      "mean_token_accuracy": 0.17991872131824493,
      "num_tokens": 14453868.0,
      "step": 6955
    },
    {
      "entropy": 5.517397832870484,
      "epoch": 0.6313497822931785,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004967035094765028,
      "loss": 5.2692,
      "mean_token_accuracy": 0.18014791905879973,
      "num_tokens": 14465015.0,
      "step": 6960
    },
    {
      "entropy": 5.467123174667359,
      "epoch": 0.6318033381712627,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004966979887794491,
      "loss": 5.2254,
      "mean_token_accuracy": 0.19137182533740998,
      "num_tokens": 14475523.0,
      "step": 6965
    },
    {
      "entropy": 5.466613149642944,
      "epoch": 0.6322568940493469,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966924634976245,
      "loss": 5.2968,
      "mean_token_accuracy": 0.1851485162973404,
      "num_tokens": 14485523.0,
      "step": 6970
    },
    {
      "entropy": 5.538769245147705,
      "epoch": 0.6327104499274311,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004966869336311433,
      "loss": 5.3623,
      "mean_token_accuracy": 0.18265851140022277,
      "num_tokens": 14495237.0,
      "step": 6975
    },
    {
      "entropy": 5.564757013320923,
      "epoch": 0.6331640058055152,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966813991801197,
      "loss": 5.4848,
      "mean_token_accuracy": 0.1746712803840637,
      "num_tokens": 14506431.0,
      "step": 6980
    },
    {
      "entropy": 5.602010774612427,
      "epoch": 0.6336175616835994,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004966758601446684,
      "loss": 5.4194,
      "mean_token_accuracy": 0.17271297425031662,
      "num_tokens": 14517445.0,
      "step": 6985
    },
    {
      "entropy": 5.554762315750122,
      "epoch": 0.6340711175616836,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966703165249038,
      "loss": 5.3482,
      "mean_token_accuracy": 0.17258450388908386,
      "num_tokens": 14527818.0,
      "step": 6990
    },
    {
      "entropy": 5.549514579772949,
      "epoch": 0.6345246734397678,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966647683209406,
      "loss": 5.4332,
      "mean_token_accuracy": 0.16915443241596223,
      "num_tokens": 14538235.0,
      "step": 6995
    },
    {
      "entropy": 5.6472475051879885,
      "epoch": 0.634978229317852,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966592155328935,
      "loss": 5.4463,
      "mean_token_accuracy": 0.1713977947831154,
      "num_tokens": 14549784.0,
      "step": 7000
    },
    {
      "entropy": 5.499601411819458,
      "epoch": 0.6354317851959361,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004966536581608771,
      "loss": 5.4273,
      "mean_token_accuracy": 0.17051296085119247,
      "num_tokens": 14561057.0,
      "step": 7005
    },
    {
      "entropy": 5.546831130981445,
      "epoch": 0.6358853410740203,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004966480962050069,
      "loss": 5.3432,
      "mean_token_accuracy": 0.1802305594086647,
      "num_tokens": 14570859.0,
      "step": 7010
    },
    {
      "entropy": 5.5499834537506105,
      "epoch": 0.6363388969521045,
      "grad_norm": 1.0,
      "learning_rate": 0.0004966425296653974,
      "loss": 5.3498,
      "mean_token_accuracy": 0.18582831472158431,
      "num_tokens": 14581066.0,
      "step": 7015
    },
    {
      "entropy": 5.500417184829712,
      "epoch": 0.6367924528301887,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000496636958542164,
      "loss": 5.3133,
      "mean_token_accuracy": 0.18046621680259706,
      "num_tokens": 14590709.0,
      "step": 7020
    },
    {
      "entropy": 5.511330270767212,
      "epoch": 0.6372460087082729,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004966313828354218,
      "loss": 5.2974,
      "mean_token_accuracy": 0.18662593364715577,
      "num_tokens": 14600849.0,
      "step": 7025
    },
    {
      "entropy": 5.455791187286377,
      "epoch": 0.637699564586357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000496625802545286,
      "loss": 5.1919,
      "mean_token_accuracy": 0.18537684082984923,
      "num_tokens": 14611427.0,
      "step": 7030
    },
    {
      "entropy": 5.508231830596924,
      "epoch": 0.6381531204644412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004966202176718723,
      "loss": 5.3486,
      "mean_token_accuracy": 0.18688730001449586,
      "num_tokens": 14623860.0,
      "step": 7035
    },
    {
      "entropy": 5.537319087982178,
      "epoch": 0.6386066763425254,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966146282152959,
      "loss": 5.2776,
      "mean_token_accuracy": 0.18067865520715715,
      "num_tokens": 14633481.0,
      "step": 7040
    },
    {
      "entropy": 5.535480594635009,
      "epoch": 0.6390602322206096,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966090341756726,
      "loss": 5.2586,
      "mean_token_accuracy": 0.19394663125276565,
      "num_tokens": 14643519.0,
      "step": 7045
    },
    {
      "entropy": 5.390807437896728,
      "epoch": 0.6395137880986937,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496603435553118,
      "loss": 5.2387,
      "mean_token_accuracy": 0.19258788973093033,
      "num_tokens": 14654570.0,
      "step": 7050
    },
    {
      "entropy": 5.439500379562378,
      "epoch": 0.6399673439767779,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965978323477478,
      "loss": 5.2894,
      "mean_token_accuracy": 0.17954834699630737,
      "num_tokens": 14665596.0,
      "step": 7055
    },
    {
      "entropy": 5.537118530273437,
      "epoch": 0.6404208998548622,
      "grad_norm": 1.03125,
      "learning_rate": 0.000496592224559678,
      "loss": 5.2384,
      "mean_token_accuracy": 0.18761214017868041,
      "num_tokens": 14675660.0,
      "step": 7060
    },
    {
      "entropy": 5.550399494171143,
      "epoch": 0.6408744557329463,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965866121890245,
      "loss": 5.3923,
      "mean_token_accuracy": 0.1781720206141472,
      "num_tokens": 14685229.0,
      "step": 7065
    },
    {
      "entropy": 5.52332239151001,
      "epoch": 0.6413280116110305,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004965809952359034,
      "loss": 5.2985,
      "mean_token_accuracy": 0.17515805810689927,
      "num_tokens": 14695672.0,
      "step": 7070
    },
    {
      "entropy": 5.501980543136597,
      "epoch": 0.6417815674891146,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004965753737004309,
      "loss": 5.2568,
      "mean_token_accuracy": 0.1830699175596237,
      "num_tokens": 14705160.0,
      "step": 7075
    },
    {
      "entropy": 5.48081636428833,
      "epoch": 0.6422351233671988,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004965697475827231,
      "loss": 5.2957,
      "mean_token_accuracy": 0.18334358930587769,
      "num_tokens": 14715487.0,
      "step": 7080
    },
    {
      "entropy": 5.545005559921265,
      "epoch": 0.6426886792452831,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965641168828965,
      "loss": 5.3329,
      "mean_token_accuracy": 0.18440192192792892,
      "num_tokens": 14726987.0,
      "step": 7085
    },
    {
      "entropy": 5.4650185108184814,
      "epoch": 0.6431422351233672,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004965584816010676,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1867426112294197,
      "num_tokens": 14737810.0,
      "step": 7090
    },
    {
      "entropy": 5.546742534637451,
      "epoch": 0.6435957910014514,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965528417373528,
      "loss": 5.3085,
      "mean_token_accuracy": 0.18062587231397628,
      "num_tokens": 14748997.0,
      "step": 7095
    },
    {
      "entropy": 5.515628051757813,
      "epoch": 0.6440493468795355,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004965471972918688,
      "loss": 5.3136,
      "mean_token_accuracy": 0.17992234230041504,
      "num_tokens": 14760439.0,
      "step": 7100
    },
    {
      "entropy": 5.42392373085022,
      "epoch": 0.6445029027576198,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004965415482647322,
      "loss": 5.2268,
      "mean_token_accuracy": 0.1856522426009178,
      "num_tokens": 14770828.0,
      "step": 7105
    },
    {
      "entropy": 5.4523519515991214,
      "epoch": 0.6449564586357039,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00049653589465606,
      "loss": 5.3692,
      "mean_token_accuracy": 0.17964595556259155,
      "num_tokens": 14781753.0,
      "step": 7110
    },
    {
      "entropy": 5.59224910736084,
      "epoch": 0.6454100145137881,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496530236465969,
      "loss": 5.3884,
      "mean_token_accuracy": 0.17452527433633805,
      "num_tokens": 14792241.0,
      "step": 7115
    },
    {
      "entropy": 5.466744709014892,
      "epoch": 0.6458635703918723,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004965245736945763,
      "loss": 5.1985,
      "mean_token_accuracy": 0.19304679036140443,
      "num_tokens": 14802580.0,
      "step": 7120
    },
    {
      "entropy": 5.487527847290039,
      "epoch": 0.6463171262699564,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004965189063419989,
      "loss": 5.4117,
      "mean_token_accuracy": 0.17699665278196336,
      "num_tokens": 14813103.0,
      "step": 7125
    },
    {
      "entropy": 5.527288436889648,
      "epoch": 0.6467706821480407,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004965132344083541,
      "loss": 5.2994,
      "mean_token_accuracy": 0.17992949336767197,
      "num_tokens": 14823334.0,
      "step": 7130
    },
    {
      "entropy": 5.533066177368164,
      "epoch": 0.6472242380261248,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004965075578937593,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17671651989221573,
      "num_tokens": 14834850.0,
      "step": 7135
    },
    {
      "entropy": 5.469491338729858,
      "epoch": 0.647677793904209,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004965018767983316,
      "loss": 5.3381,
      "mean_token_accuracy": 0.1797954857349396,
      "num_tokens": 14844276.0,
      "step": 7140
    },
    {
      "entropy": 5.385136699676513,
      "epoch": 0.6481313497822931,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004964961911221886,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19689569771289825,
      "num_tokens": 14854352.0,
      "step": 7145
    },
    {
      "entropy": 5.52975492477417,
      "epoch": 0.6485849056603774,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496490500865448,
      "loss": 5.3462,
      "mean_token_accuracy": 0.1827696070075035,
      "num_tokens": 14864395.0,
      "step": 7150
    },
    {
      "entropy": 5.498007678985596,
      "epoch": 0.6490384615384616,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004964848060282275,
      "loss": 5.2283,
      "mean_token_accuracy": 0.18727268576622008,
      "num_tokens": 14874373.0,
      "step": 7155
    },
    {
      "entropy": 5.491959953308106,
      "epoch": 0.6494920174165457,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964791066106447,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18243176639080047,
      "num_tokens": 14884932.0,
      "step": 7160
    },
    {
      "entropy": 5.5285765647888185,
      "epoch": 0.6499455732946299,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964734026128175,
      "loss": 5.3897,
      "mean_token_accuracy": 0.17832595705986024,
      "num_tokens": 14896254.0,
      "step": 7165
    },
    {
      "entropy": 5.572258043289184,
      "epoch": 0.650399129172714,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964676940348638,
      "loss": 5.3231,
      "mean_token_accuracy": 0.1801915466785431,
      "num_tokens": 14907392.0,
      "step": 7170
    },
    {
      "entropy": 5.564219951629639,
      "epoch": 0.6508526850507983,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496461980876902,
      "loss": 5.4305,
      "mean_token_accuracy": 0.17841965556144715,
      "num_tokens": 14917676.0,
      "step": 7175
    },
    {
      "entropy": 5.515146255493164,
      "epoch": 0.6513062409288825,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964562631390498,
      "loss": 5.2854,
      "mean_token_accuracy": 0.1819108560681343,
      "num_tokens": 14928500.0,
      "step": 7180
    },
    {
      "entropy": 5.483141279220581,
      "epoch": 0.6517597968069666,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004964505408214257,
      "loss": 5.243,
      "mean_token_accuracy": 0.18780999034643173,
      "num_tokens": 14938970.0,
      "step": 7185
    },
    {
      "entropy": 5.482848167419434,
      "epoch": 0.6522133526850508,
      "grad_norm": 1.0,
      "learning_rate": 0.000496444813924148,
      "loss": 5.3065,
      "mean_token_accuracy": 0.1825691819190979,
      "num_tokens": 14949017.0,
      "step": 7190
    },
    {
      "entropy": 5.521585893630982,
      "epoch": 0.652666908563135,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964390824473351,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18902090042829514,
      "num_tokens": 14958529.0,
      "step": 7195
    },
    {
      "entropy": 5.471276807785034,
      "epoch": 0.6531204644412192,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004964333463911056,
      "loss": 5.3325,
      "mean_token_accuracy": 0.18068593144416809,
      "num_tokens": 14968044.0,
      "step": 7200
    },
    {
      "entropy": 5.519154214859009,
      "epoch": 0.6535740203193033,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004964276057555779,
      "loss": 5.3461,
      "mean_token_accuracy": 0.18163236528635024,
      "num_tokens": 14977867.0,
      "step": 7205
    },
    {
      "entropy": 5.576827430725098,
      "epoch": 0.6540275761973875,
      "grad_norm": 1.046875,
      "learning_rate": 0.000496421860540871,
      "loss": 5.3499,
      "mean_token_accuracy": 0.18708072751760482,
      "num_tokens": 14986936.0,
      "step": 7210
    },
    {
      "entropy": 5.499208927154541,
      "epoch": 0.6544811320754716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964161107471036,
      "loss": 5.3146,
      "mean_token_accuracy": 0.1859233096241951,
      "num_tokens": 14996681.0,
      "step": 7215
    },
    {
      "entropy": 5.50371241569519,
      "epoch": 0.6549346879535559,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004964103563743946,
      "loss": 5.3101,
      "mean_token_accuracy": 0.18214258253574372,
      "num_tokens": 15007100.0,
      "step": 7220
    },
    {
      "entropy": 5.530852222442627,
      "epoch": 0.6553882438316401,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000496404597422863,
      "loss": 5.3575,
      "mean_token_accuracy": 0.1881464421749115,
      "num_tokens": 15017452.0,
      "step": 7225
    },
    {
      "entropy": 5.495759153366089,
      "epoch": 0.6558417997097242,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000496398833892628,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18454673141241074,
      "num_tokens": 15028362.0,
      "step": 7230
    },
    {
      "entropy": 5.53149824142456,
      "epoch": 0.6562953555878084,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963930657838086,
      "loss": 5.3204,
      "mean_token_accuracy": 0.1792814999818802,
      "num_tokens": 15038691.0,
      "step": 7235
    },
    {
      "entropy": 5.503436279296875,
      "epoch": 0.6567489114658926,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963872930965242,
      "loss": 5.2949,
      "mean_token_accuracy": 0.18664567470550536,
      "num_tokens": 15050037.0,
      "step": 7240
    },
    {
      "entropy": 5.5270062446594235,
      "epoch": 0.6572024673439768,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004963815158308941,
      "loss": 5.2687,
      "mean_token_accuracy": 0.18515150398015975,
      "num_tokens": 15059793.0,
      "step": 7245
    },
    {
      "entropy": 5.48774676322937,
      "epoch": 0.657656023222061,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004963757339870379,
      "loss": 5.3611,
      "mean_token_accuracy": 0.1739514410495758,
      "num_tokens": 15070500.0,
      "step": 7250
    },
    {
      "entropy": 5.557533264160156,
      "epoch": 0.6581095791001451,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004963699475650751,
      "loss": 5.4125,
      "mean_token_accuracy": 0.17817330807447435,
      "num_tokens": 15081109.0,
      "step": 7255
    },
    {
      "entropy": 5.464095830917358,
      "epoch": 0.6585631349782293,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004963641565651255,
      "loss": 5.295,
      "mean_token_accuracy": 0.18745609223842621,
      "num_tokens": 15092315.0,
      "step": 7260
    },
    {
      "entropy": 5.526384782791138,
      "epoch": 0.6590166908563135,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004963583609873084,
      "loss": 5.3606,
      "mean_token_accuracy": 0.18208035826683044,
      "num_tokens": 15103957.0,
      "step": 7265
    },
    {
      "entropy": 5.556434249877929,
      "epoch": 0.6594702467343977,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004963525608317443,
      "loss": 5.3472,
      "mean_token_accuracy": 0.17901953905820847,
      "num_tokens": 15114996.0,
      "step": 7270
    },
    {
      "entropy": 5.518425130844117,
      "epoch": 0.6599238026124818,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004963467560985527,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17808068990707399,
      "num_tokens": 15125098.0,
      "step": 7275
    },
    {
      "entropy": 5.448391628265381,
      "epoch": 0.660377358490566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004963409467878538,
      "loss": 5.2763,
      "mean_token_accuracy": 0.1811485469341278,
      "num_tokens": 15134746.0,
      "step": 7280
    },
    {
      "entropy": 5.566460371017456,
      "epoch": 0.6608309143686503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963351328997678,
      "loss": 5.3115,
      "mean_token_accuracy": 0.17681878209114074,
      "num_tokens": 15145698.0,
      "step": 7285
    },
    {
      "entropy": 5.540733051300049,
      "epoch": 0.6612844702467344,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004963293144344148,
      "loss": 5.3639,
      "mean_token_accuracy": 0.1819513663649559,
      "num_tokens": 15156026.0,
      "step": 7290
    },
    {
      "entropy": 5.5315940380096436,
      "epoch": 0.6617380261248186,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004963234913919151,
      "loss": 5.2934,
      "mean_token_accuracy": 0.17740343809127807,
      "num_tokens": 15165900.0,
      "step": 7295
    },
    {
      "entropy": 5.499019193649292,
      "epoch": 0.6621915820029027,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963176637723893,
      "loss": 5.3654,
      "mean_token_accuracy": 0.177557073533535,
      "num_tokens": 15176693.0,
      "step": 7300
    },
    {
      "entropy": 5.536378002166748,
      "epoch": 0.6626451378809869,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004963118315759578,
      "loss": 5.3817,
      "mean_token_accuracy": 0.17143151313066482,
      "num_tokens": 15186535.0,
      "step": 7305
    },
    {
      "entropy": 5.5319818496704105,
      "epoch": 0.6630986937590712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004963059948027411,
      "loss": 5.2914,
      "mean_token_accuracy": 0.17779401242733,
      "num_tokens": 15196019.0,
      "step": 7310
    },
    {
      "entropy": 5.423022127151489,
      "epoch": 0.6635522496371553,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004963001534528601,
      "loss": 5.1571,
      "mean_token_accuracy": 0.19192803800106048,
      "num_tokens": 15205273.0,
      "step": 7315
    },
    {
      "entropy": 5.527570486068726,
      "epoch": 0.6640058055152395,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962943075264355,
      "loss": 5.3163,
      "mean_token_accuracy": 0.17965196520090104,
      "num_tokens": 15216716.0,
      "step": 7320
    },
    {
      "entropy": 5.532290601730347,
      "epoch": 0.6644593613933236,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004962884570235884,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1758708655834198,
      "num_tokens": 15227306.0,
      "step": 7325
    },
    {
      "entropy": 5.456675958633423,
      "epoch": 0.6649129172714079,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004962826019444394,
      "loss": 5.2574,
      "mean_token_accuracy": 0.18692494183778763,
      "num_tokens": 15237302.0,
      "step": 7330
    },
    {
      "entropy": 5.492032623291015,
      "epoch": 0.665366473149492,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004962767422891099,
      "loss": 5.2547,
      "mean_token_accuracy": 0.1889828249812126,
      "num_tokens": 15247064.0,
      "step": 7335
    },
    {
      "entropy": 5.5133216857910154,
      "epoch": 0.6658200290275762,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000496270878057721,
      "loss": 5.3161,
      "mean_token_accuracy": 0.18626102805137634,
      "num_tokens": 15257006.0,
      "step": 7340
    },
    {
      "entropy": 5.5310060501098635,
      "epoch": 0.6662735849056604,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004962650092503939,
      "loss": 5.3115,
      "mean_token_accuracy": 0.19045017808675765,
      "num_tokens": 15266720.0,
      "step": 7345
    },
    {
      "entropy": 5.508117914199829,
      "epoch": 0.6667271407837445,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004962591358672502,
      "loss": 5.3201,
      "mean_token_accuracy": 0.18087735772132874,
      "num_tokens": 15277330.0,
      "step": 7350
    },
    {
      "entropy": 5.490307807922363,
      "epoch": 0.6671806966618288,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496253257908411,
      "loss": 5.2604,
      "mean_token_accuracy": 0.18703862726688386,
      "num_tokens": 15287401.0,
      "step": 7355
    },
    {
      "entropy": 5.460469102859497,
      "epoch": 0.6676342525399129,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004962473753739981,
      "loss": 5.2144,
      "mean_token_accuracy": 0.18987584263086318,
      "num_tokens": 15297426.0,
      "step": 7360
    },
    {
      "entropy": 5.512601900100708,
      "epoch": 0.6680878084179971,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000496241488264133,
      "loss": 5.2122,
      "mean_token_accuracy": 0.17890442311763763,
      "num_tokens": 15308412.0,
      "step": 7365
    },
    {
      "entropy": 5.494791793823242,
      "epoch": 0.6685413642960812,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962355965789377,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18206070810556413,
      "num_tokens": 15318821.0,
      "step": 7370
    },
    {
      "entropy": 5.428279018402099,
      "epoch": 0.6689949201741655,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004962297003185338,
      "loss": 5.284,
      "mean_token_accuracy": 0.18319466859102249,
      "num_tokens": 15328737.0,
      "step": 7375
    },
    {
      "entropy": 5.52919020652771,
      "epoch": 0.6694484760522497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004962237994830433,
      "loss": 5.3792,
      "mean_token_accuracy": 0.17462361603975296,
      "num_tokens": 15339695.0,
      "step": 7380
    },
    {
      "entropy": 5.563119554519654,
      "epoch": 0.6699020319303338,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004962178940725883,
      "loss": 5.3496,
      "mean_token_accuracy": 0.18139780610799788,
      "num_tokens": 15350270.0,
      "step": 7385
    },
    {
      "entropy": 5.520648050308227,
      "epoch": 0.670355587808418,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004962119840872908,
      "loss": 5.3272,
      "mean_token_accuracy": 0.1830776020884514,
      "num_tokens": 15360986.0,
      "step": 7390
    },
    {
      "entropy": 5.482888650894165,
      "epoch": 0.6708091436865021,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004962060695272733,
      "loss": 5.3153,
      "mean_token_accuracy": 0.18322502970695495,
      "num_tokens": 15371786.0,
      "step": 7395
    },
    {
      "entropy": 5.507193279266358,
      "epoch": 0.6712626995645864,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004962001503926577,
      "loss": 5.2871,
      "mean_token_accuracy": 0.19009895771741867,
      "num_tokens": 15382960.0,
      "step": 7400
    },
    {
      "entropy": 5.521780300140381,
      "epoch": 0.6717162554426706,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004961942266835667,
      "loss": 5.2713,
      "mean_token_accuracy": 0.18810440748929977,
      "num_tokens": 15392807.0,
      "step": 7405
    },
    {
      "entropy": 5.498525714874267,
      "epoch": 0.6721698113207547,
      "grad_norm": 1.0,
      "learning_rate": 0.0004961882984001227,
      "loss": 5.3172,
      "mean_token_accuracy": 0.17630916088819504,
      "num_tokens": 15403712.0,
      "step": 7410
    },
    {
      "entropy": 5.498222351074219,
      "epoch": 0.6726233671988389,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004961823655424484,
      "loss": 5.3845,
      "mean_token_accuracy": 0.17348767071962357,
      "num_tokens": 15414839.0,
      "step": 7415
    },
    {
      "entropy": 5.530002164840698,
      "epoch": 0.6730769230769231,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961764281106664,
      "loss": 5.2852,
      "mean_token_accuracy": 0.1857857435941696,
      "num_tokens": 15425320.0,
      "step": 7420
    },
    {
      "entropy": 5.50027289390564,
      "epoch": 0.6735304789550073,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004961704861048995,
      "loss": 5.263,
      "mean_token_accuracy": 0.18382281810045242,
      "num_tokens": 15435476.0,
      "step": 7425
    },
    {
      "entropy": 5.478959846496582,
      "epoch": 0.6739840348330914,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004961645395252705,
      "loss": 5.3773,
      "mean_token_accuracy": 0.1767744243144989,
      "num_tokens": 15445786.0,
      "step": 7430
    },
    {
      "entropy": 5.524657535552978,
      "epoch": 0.6744375907111756,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961585883719025,
      "loss": 5.3577,
      "mean_token_accuracy": 0.18489955067634584,
      "num_tokens": 15455915.0,
      "step": 7435
    },
    {
      "entropy": 5.54436674118042,
      "epoch": 0.6748911465892597,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004961526326449186,
      "loss": 5.3439,
      "mean_token_accuracy": 0.1796664759516716,
      "num_tokens": 15466523.0,
      "step": 7440
    },
    {
      "entropy": 5.51652889251709,
      "epoch": 0.675344702467344,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004961466723444419,
      "loss": 5.3379,
      "mean_token_accuracy": 0.1811014547944069,
      "num_tokens": 15476821.0,
      "step": 7445
    },
    {
      "entropy": 5.483820390701294,
      "epoch": 0.6757982583454282,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004961407074705955,
      "loss": 5.2628,
      "mean_token_accuracy": 0.18104134052991866,
      "num_tokens": 15487182.0,
      "step": 7450
    },
    {
      "entropy": 5.501073026657105,
      "epoch": 0.6762518142235123,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004961347380235031,
      "loss": 5.2571,
      "mean_token_accuracy": 0.18460666239261628,
      "num_tokens": 15497600.0,
      "step": 7455
    },
    {
      "entropy": 5.524596357345581,
      "epoch": 0.6767053701015965,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961287640032879,
      "loss": 5.3944,
      "mean_token_accuracy": 0.17238223403692246,
      "num_tokens": 15508278.0,
      "step": 7460
    },
    {
      "entropy": 5.57806396484375,
      "epoch": 0.6771589259796807,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961227854100734,
      "loss": 5.4086,
      "mean_token_accuracy": 0.17855887562036515,
      "num_tokens": 15518458.0,
      "step": 7465
    },
    {
      "entropy": 5.475199174880982,
      "epoch": 0.6776124818577649,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961168022439835,
      "loss": 5.2303,
      "mean_token_accuracy": 0.18490078896284104,
      "num_tokens": 15528738.0,
      "step": 7470
    },
    {
      "entropy": 5.493577146530152,
      "epoch": 0.6780660377358491,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961108145051417,
      "loss": 5.2666,
      "mean_token_accuracy": 0.18824383467435837,
      "num_tokens": 15538656.0,
      "step": 7475
    },
    {
      "entropy": 5.446999168395996,
      "epoch": 0.6785195936139332,
      "grad_norm": 0.921875,
      "learning_rate": 0.000496104822193672,
      "loss": 5.2699,
      "mean_token_accuracy": 0.18458060622215272,
      "num_tokens": 15549285.0,
      "step": 7480
    },
    {
      "entropy": 5.559575128555298,
      "epoch": 0.6789731494920174,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004960988253096982,
      "loss": 5.3514,
      "mean_token_accuracy": 0.18502087444067,
      "num_tokens": 15559766.0,
      "step": 7485
    },
    {
      "entropy": 5.494141435623169,
      "epoch": 0.6794267053701016,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004960928238533443,
      "loss": 5.218,
      "mean_token_accuracy": 0.19351499229669572,
      "num_tokens": 15570873.0,
      "step": 7490
    },
    {
      "entropy": 5.517694282531738,
      "epoch": 0.6798802612481858,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960868178247344,
      "loss": 5.4704,
      "mean_token_accuracy": 0.17666775584220887,
      "num_tokens": 15581229.0,
      "step": 7495
    },
    {
      "entropy": 5.497984743118286,
      "epoch": 0.68033381712627,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960808072239929,
      "loss": 5.3298,
      "mean_token_accuracy": 0.1761513829231262,
      "num_tokens": 15591686.0,
      "step": 7500
    },
    {
      "entropy": 5.607373332977295,
      "epoch": 0.6807873730043541,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004960747920512439,
      "loss": 5.3475,
      "mean_token_accuracy": 0.18001337945461274,
      "num_tokens": 15601869.0,
      "step": 7505
    },
    {
      "entropy": 5.4866539478302006,
      "epoch": 0.6812409288824384,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000496068772306612,
      "loss": 5.3332,
      "mean_token_accuracy": 0.1745358318090439,
      "num_tokens": 15612946.0,
      "step": 7510
    },
    {
      "entropy": 5.529070234298706,
      "epoch": 0.6816944847605225,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960627479902214,
      "loss": 5.3561,
      "mean_token_accuracy": 0.17982869893312453,
      "num_tokens": 15623761.0,
      "step": 7515
    },
    {
      "entropy": 5.523828363418579,
      "epoch": 0.6821480406386067,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000496056719102197,
      "loss": 5.3549,
      "mean_token_accuracy": 0.17677418738603592,
      "num_tokens": 15634500.0,
      "step": 7520
    },
    {
      "entropy": 5.451393985748291,
      "epoch": 0.6826015965166908,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960506856426631,
      "loss": 5.2098,
      "mean_token_accuracy": 0.1831522673368454,
      "num_tokens": 15644896.0,
      "step": 7525
    },
    {
      "entropy": 5.528468227386474,
      "epoch": 0.683055152394775,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960446476117448,
      "loss": 5.3356,
      "mean_token_accuracy": 0.1845097690820694,
      "num_tokens": 15655008.0,
      "step": 7530
    },
    {
      "entropy": 5.501651287078857,
      "epoch": 0.6835087082728593,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004960386050095668,
      "loss": 5.2435,
      "mean_token_accuracy": 0.18057304620742798,
      "num_tokens": 15665944.0,
      "step": 7535
    },
    {
      "entropy": 5.510901641845703,
      "epoch": 0.6839622641509434,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004960325578362543,
      "loss": 5.3456,
      "mean_token_accuracy": 0.18406951576471328,
      "num_tokens": 15677099.0,
      "step": 7540
    },
    {
      "entropy": 5.578004693984985,
      "epoch": 0.6844158200290276,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000496026506091932,
      "loss": 5.3118,
      "mean_token_accuracy": 0.18596543669700621,
      "num_tokens": 15687531.0,
      "step": 7545
    },
    {
      "entropy": 5.537898349761963,
      "epoch": 0.6848693759071117,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004960204497767253,
      "loss": 5.323,
      "mean_token_accuracy": 0.18122205436229705,
      "num_tokens": 15697002.0,
      "step": 7550
    },
    {
      "entropy": 5.39183406829834,
      "epoch": 0.685322931785196,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960143888907594,
      "loss": 5.1951,
      "mean_token_accuracy": 0.1852844089269638,
      "num_tokens": 15707249.0,
      "step": 7555
    },
    {
      "entropy": 5.48224287033081,
      "epoch": 0.6857764876632801,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004960083234341595,
      "loss": 5.275,
      "mean_token_accuracy": 0.18008209019899368,
      "num_tokens": 15718764.0,
      "step": 7560
    },
    {
      "entropy": 5.485287809371949,
      "epoch": 0.6862300435413643,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960022534070513,
      "loss": 5.223,
      "mean_token_accuracy": 0.18764107674360275,
      "num_tokens": 15728854.0,
      "step": 7565
    },
    {
      "entropy": 5.415549468994141,
      "epoch": 0.6866835994194485,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049599617880956,
      "loss": 5.2349,
      "mean_token_accuracy": 0.18654119819402695,
      "num_tokens": 15739682.0,
      "step": 7570
    },
    {
      "entropy": 5.4649327278137205,
      "epoch": 0.6871371552975326,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959900996418115,
      "loss": 5.2324,
      "mean_token_accuracy": 0.18843563199043273,
      "num_tokens": 15749628.0,
      "step": 7575
    },
    {
      "entropy": 5.513627624511718,
      "epoch": 0.6875907111756169,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004959840159039315,
      "loss": 5.3056,
      "mean_token_accuracy": 0.18155467957258226,
      "num_tokens": 15760137.0,
      "step": 7580
    },
    {
      "entropy": 5.538849496841431,
      "epoch": 0.688044267053701,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004959779275960456,
      "loss": 5.2556,
      "mean_token_accuracy": 0.1825544536113739,
      "num_tokens": 15769889.0,
      "step": 7585
    },
    {
      "entropy": 5.480565595626831,
      "epoch": 0.6884978229317852,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049597183471828,
      "loss": 5.3771,
      "mean_token_accuracy": 0.18086028099060059,
      "num_tokens": 15780912.0,
      "step": 7590
    },
    {
      "entropy": 5.501943922042846,
      "epoch": 0.6889513788098693,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004959657372707604,
      "loss": 5.2532,
      "mean_token_accuracy": 0.189553838968277,
      "num_tokens": 15791719.0,
      "step": 7595
    },
    {
      "entropy": 5.527008438110352,
      "epoch": 0.6894049346879536,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004959596352536132,
      "loss": 5.2763,
      "mean_token_accuracy": 0.18491054624319075,
      "num_tokens": 15802551.0,
      "step": 7600
    },
    {
      "entropy": 5.546923732757568,
      "epoch": 0.6898584905660378,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959535286669643,
      "loss": 5.3388,
      "mean_token_accuracy": 0.17994246035814285,
      "num_tokens": 15813551.0,
      "step": 7605
    },
    {
      "entropy": 5.498889207839966,
      "epoch": 0.6903120464441219,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004959474175109402,
      "loss": 5.3034,
      "mean_token_accuracy": 0.18564930260181428,
      "num_tokens": 15824262.0,
      "step": 7610
    },
    {
      "entropy": 5.550105953216553,
      "epoch": 0.6907656023222061,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004959413017856672,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1801539644598961,
      "num_tokens": 15835493.0,
      "step": 7615
    },
    {
      "entropy": 5.419754505157471,
      "epoch": 0.6912191582002902,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004959351814912716,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1831692114472389,
      "num_tokens": 15846224.0,
      "step": 7620
    },
    {
      "entropy": 5.5195647239685055,
      "epoch": 0.6916727140783745,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004959290566278803,
      "loss": 5.2845,
      "mean_token_accuracy": 0.18517156839370727,
      "num_tokens": 15855622.0,
      "step": 7625
    },
    {
      "entropy": 5.537323617935181,
      "epoch": 0.6921262699564587,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004959229271956199,
      "loss": 5.344,
      "mean_token_accuracy": 0.18175625056028366,
      "num_tokens": 15865835.0,
      "step": 7630
    },
    {
      "entropy": 5.5008909702301025,
      "epoch": 0.6925798258345428,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004959167931946169,
      "loss": 5.2823,
      "mean_token_accuracy": 0.18331976979970932,
      "num_tokens": 15875915.0,
      "step": 7635
    },
    {
      "entropy": 5.514472818374633,
      "epoch": 0.693033381712627,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004959106546249984,
      "loss": 5.2829,
      "mean_token_accuracy": 0.1838786020874977,
      "num_tokens": 15887111.0,
      "step": 7640
    },
    {
      "entropy": 5.530413293838501,
      "epoch": 0.6934869375907112,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004959045114868913,
      "loss": 5.39,
      "mean_token_accuracy": 0.1729227066040039,
      "num_tokens": 15899528.0,
      "step": 7645
    },
    {
      "entropy": 5.5232024669647215,
      "epoch": 0.6939404934687954,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004958983637804225,
      "loss": 5.3086,
      "mean_token_accuracy": 0.19027511328458785,
      "num_tokens": 15910881.0,
      "step": 7650
    },
    {
      "entropy": 5.569855737686157,
      "epoch": 0.6943940493468795,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004958922115057194,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17807182669639587,
      "num_tokens": 15921189.0,
      "step": 7655
    },
    {
      "entropy": 5.563821744918823,
      "epoch": 0.6948476052249637,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004958860546629089,
      "loss": 5.3297,
      "mean_token_accuracy": 0.18766747564077377,
      "num_tokens": 15931276.0,
      "step": 7660
    },
    {
      "entropy": 5.550361442565918,
      "epoch": 0.6953011611030478,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958798932521186,
      "loss": 5.4139,
      "mean_token_accuracy": 0.17329273372888565,
      "num_tokens": 15941506.0,
      "step": 7665
    },
    {
      "entropy": 5.4555493831634525,
      "epoch": 0.6957547169811321,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958737272734759,
      "loss": 5.245,
      "mean_token_accuracy": 0.18558478802442552,
      "num_tokens": 15951158.0,
      "step": 7670
    },
    {
      "entropy": 5.522123765945435,
      "epoch": 0.6962082728592163,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004958675567271081,
      "loss": 5.2866,
      "mean_token_accuracy": 0.18499088436365127,
      "num_tokens": 15960852.0,
      "step": 7675
    },
    {
      "entropy": 5.5311461925506595,
      "epoch": 0.6966618287373004,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000495861381613143,
      "loss": 5.2857,
      "mean_token_accuracy": 0.179577599465847,
      "num_tokens": 15971291.0,
      "step": 7680
    },
    {
      "entropy": 5.507929944992066,
      "epoch": 0.6971153846153846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004958552019317082,
      "loss": 5.3468,
      "mean_token_accuracy": 0.18295310884714128,
      "num_tokens": 15982180.0,
      "step": 7685
    },
    {
      "entropy": 5.420698976516723,
      "epoch": 0.6975689404934688,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004958490176829316,
      "loss": 5.2349,
      "mean_token_accuracy": 0.18435447812080383,
      "num_tokens": 15992363.0,
      "step": 7690
    },
    {
      "entropy": 5.496499824523926,
      "epoch": 0.698022496371553,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000495842828866941,
      "loss": 5.3217,
      "mean_token_accuracy": 0.17525662034749984,
      "num_tokens": 16003477.0,
      "step": 7695
    },
    {
      "entropy": 5.558476448059082,
      "epoch": 0.6984760522496372,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004958366354838645,
      "loss": 5.2953,
      "mean_token_accuracy": 0.18492270410060882,
      "num_tokens": 16013747.0,
      "step": 7700
    },
    {
      "entropy": 5.521188449859619,
      "epoch": 0.6989296081277213,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00049583043753383,
      "loss": 5.3281,
      "mean_token_accuracy": 0.18768803924322128,
      "num_tokens": 16025662.0,
      "step": 7705
    },
    {
      "entropy": 5.532258319854736,
      "epoch": 0.6993831640058055,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004958242350169658,
      "loss": 5.3933,
      "mean_token_accuracy": 0.18370112180709838,
      "num_tokens": 16037315.0,
      "step": 7710
    },
    {
      "entropy": 5.446314811706543,
      "epoch": 0.6998367198838897,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004958180279334001,
      "loss": 5.2128,
      "mean_token_accuracy": 0.18639328330755234,
      "num_tokens": 16047297.0,
      "step": 7715
    },
    {
      "entropy": 5.467506217956543,
      "epoch": 0.7002902757619739,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958118162832614,
      "loss": 5.1837,
      "mean_token_accuracy": 0.19042861312627793,
      "num_tokens": 16057573.0,
      "step": 7720
    },
    {
      "entropy": 5.496659231185913,
      "epoch": 0.700743831640058,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958056000666781,
      "loss": 5.3569,
      "mean_token_accuracy": 0.18165837675333024,
      "num_tokens": 16067926.0,
      "step": 7725
    },
    {
      "entropy": 5.4943725109100345,
      "epoch": 0.7011973875181422,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004957993792837786,
      "loss": 5.2781,
      "mean_token_accuracy": 0.18150821179151536,
      "num_tokens": 16077938.0,
      "step": 7730
    },
    {
      "entropy": 5.42755537033081,
      "epoch": 0.7016509433962265,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004957931539346917,
      "loss": 5.175,
      "mean_token_accuracy": 0.18916916996240615,
      "num_tokens": 16088163.0,
      "step": 7735
    },
    {
      "entropy": 5.556448745727539,
      "epoch": 0.7021044992743106,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004957869240195462,
      "loss": 5.394,
      "mean_token_accuracy": 0.17612502872943878,
      "num_tokens": 16100640.0,
      "step": 7740
    },
    {
      "entropy": 5.486884880065918,
      "epoch": 0.7025580551523948,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004957806895384707,
      "loss": 5.2049,
      "mean_token_accuracy": 0.1893024906516075,
      "num_tokens": 16111018.0,
      "step": 7745
    },
    {
      "entropy": 5.448366641998291,
      "epoch": 0.7030116110304789,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004957744504915944,
      "loss": 5.2455,
      "mean_token_accuracy": 0.1839633345603943,
      "num_tokens": 16120873.0,
      "step": 7750
    },
    {
      "entropy": 5.529433631896973,
      "epoch": 0.7034651669085631,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004957682068790463,
      "loss": 5.2736,
      "mean_token_accuracy": 0.18772852271795273,
      "num_tokens": 16131178.0,
      "step": 7755
    },
    {
      "entropy": 5.466021537780762,
      "epoch": 0.7039187227866474,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957619587009553,
      "loss": 5.278,
      "mean_token_accuracy": 0.1836032971739769,
      "num_tokens": 16141156.0,
      "step": 7760
    },
    {
      "entropy": 5.52126636505127,
      "epoch": 0.7043722786647315,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957557059574507,
      "loss": 5.3781,
      "mean_token_accuracy": 0.17983128130435944,
      "num_tokens": 16150077.0,
      "step": 7765
    },
    {
      "entropy": 5.486139154434204,
      "epoch": 0.7048258345428157,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957494486486618,
      "loss": 5.296,
      "mean_token_accuracy": 0.1811152681708336,
      "num_tokens": 16160514.0,
      "step": 7770
    },
    {
      "entropy": 5.528773260116577,
      "epoch": 0.7052793904208998,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004957431867747182,
      "loss": 5.3239,
      "mean_token_accuracy": 0.18625716120004654,
      "num_tokens": 16170811.0,
      "step": 7775
    },
    {
      "entropy": 5.452904558181762,
      "epoch": 0.7057329462989841,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957369203357492,
      "loss": 5.2389,
      "mean_token_accuracy": 0.18305610865354538,
      "num_tokens": 16181553.0,
      "step": 7780
    },
    {
      "entropy": 5.484501028060913,
      "epoch": 0.7061865021770682,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004957306493318844,
      "loss": 5.2562,
      "mean_token_accuracy": 0.19118969291448593,
      "num_tokens": 16191308.0,
      "step": 7785
    },
    {
      "entropy": 5.503818416595459,
      "epoch": 0.7066400580551524,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957243737632535,
      "loss": 5.2377,
      "mean_token_accuracy": 0.19155516624450683,
      "num_tokens": 16201311.0,
      "step": 7790
    },
    {
      "entropy": 5.545148134231567,
      "epoch": 0.7070936139332366,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004957180936299863,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1893450289964676,
      "num_tokens": 16211105.0,
      "step": 7795
    },
    {
      "entropy": 5.520404529571533,
      "epoch": 0.7075471698113207,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004957118089322127,
      "loss": 5.2874,
      "mean_token_accuracy": 0.17717190086841583,
      "num_tokens": 16220836.0,
      "step": 7800
    },
    {
      "entropy": 5.462588691711426,
      "epoch": 0.708000725689405,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004957055196700628,
      "loss": 5.1898,
      "mean_token_accuracy": 0.18572719693183898,
      "num_tokens": 16231630.0,
      "step": 7805
    },
    {
      "entropy": 5.494060945510864,
      "epoch": 0.7084542815674891,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004956992258436663,
      "loss": 5.2909,
      "mean_token_accuracy": 0.18368653506040572,
      "num_tokens": 16242272.0,
      "step": 7810
    },
    {
      "entropy": 5.399591398239136,
      "epoch": 0.7089078374455733,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004956929274531536,
      "loss": 5.1914,
      "mean_token_accuracy": 0.1873309940099716,
      "num_tokens": 16252555.0,
      "step": 7815
    },
    {
      "entropy": 5.486877584457398,
      "epoch": 0.7093613933236574,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495686624498655,
      "loss": 5.3216,
      "mean_token_accuracy": 0.18683241903781891,
      "num_tokens": 16261964.0,
      "step": 7820
    },
    {
      "entropy": 5.574159717559814,
      "epoch": 0.7098149492017417,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004956803169803007,
      "loss": 5.3752,
      "mean_token_accuracy": 0.175976400077343,
      "num_tokens": 16274543.0,
      "step": 7825
    },
    {
      "entropy": 5.467151403427124,
      "epoch": 0.7102685050798259,
      "grad_norm": 0.875,
      "learning_rate": 0.0004956740048982212,
      "loss": 5.28,
      "mean_token_accuracy": 0.17519069761037825,
      "num_tokens": 16285124.0,
      "step": 7830
    },
    {
      "entropy": 5.484114265441894,
      "epoch": 0.71072206095791,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004956676882525472,
      "loss": 5.3414,
      "mean_token_accuracy": 0.18310529738664627,
      "num_tokens": 16295727.0,
      "step": 7835
    },
    {
      "entropy": 5.574627113342285,
      "epoch": 0.7111756168359942,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000495661367043409,
      "loss": 5.2398,
      "mean_token_accuracy": 0.19142441600561141,
      "num_tokens": 16305496.0,
      "step": 7840
    },
    {
      "entropy": 5.482682752609253,
      "epoch": 0.7116291727140783,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004956550412709376,
      "loss": 5.222,
      "mean_token_accuracy": 0.19398910850286483,
      "num_tokens": 16315112.0,
      "step": 7845
    },
    {
      "entropy": 5.463079977035522,
      "epoch": 0.7120827285921626,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004956487109352637,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18460012674331666,
      "num_tokens": 16325613.0,
      "step": 7850
    },
    {
      "entropy": 5.49309515953064,
      "epoch": 0.7125362844702468,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956423760365181,
      "loss": 5.2163,
      "mean_token_accuracy": 0.18853310793638228,
      "num_tokens": 16335317.0,
      "step": 7855
    },
    {
      "entropy": 5.528402090072632,
      "epoch": 0.7129898403483309,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004956360365748321,
      "loss": 5.3547,
      "mean_token_accuracy": 0.17705050855875015,
      "num_tokens": 16345003.0,
      "step": 7860
    },
    {
      "entropy": 5.511237478256225,
      "epoch": 0.7134433962264151,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004956296925503367,
      "loss": 5.3451,
      "mean_token_accuracy": 0.18143388777971267,
      "num_tokens": 16356108.0,
      "step": 7865
    },
    {
      "entropy": 5.448657751083374,
      "epoch": 0.7138969521044993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004956233439631628,
      "loss": 5.1452,
      "mean_token_accuracy": 0.18960337191820145,
      "num_tokens": 16364733.0,
      "step": 7870
    },
    {
      "entropy": 5.506351900100708,
      "epoch": 0.7143505079825835,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004956169908134421,
      "loss": 5.2525,
      "mean_token_accuracy": 0.18633940815925598,
      "num_tokens": 16375229.0,
      "step": 7875
    },
    {
      "entropy": 5.485099172592163,
      "epoch": 0.7148040638606676,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004956106331013057,
      "loss": 5.339,
      "mean_token_accuracy": 0.18616413027048112,
      "num_tokens": 16386366.0,
      "step": 7880
    },
    {
      "entropy": 5.479721736907959,
      "epoch": 0.7152576197387518,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004956042708268853,
      "loss": 5.2205,
      "mean_token_accuracy": 0.19334444254636765,
      "num_tokens": 16397067.0,
      "step": 7885
    },
    {
      "entropy": 5.424285650253296,
      "epoch": 0.715711175616836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955979039903123,
      "loss": 5.187,
      "mean_token_accuracy": 0.18900911509990692,
      "num_tokens": 16407256.0,
      "step": 7890
    },
    {
      "entropy": 5.455164909362793,
      "epoch": 0.7161647314949202,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004955915325917183,
      "loss": 5.2539,
      "mean_token_accuracy": 0.1819542557001114,
      "num_tokens": 16418701.0,
      "step": 7895
    },
    {
      "entropy": 5.432131624221801,
      "epoch": 0.7166182873730044,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004955851566312354,
      "loss": 5.1672,
      "mean_token_accuracy": 0.19276488721370696,
      "num_tokens": 16428019.0,
      "step": 7900
    },
    {
      "entropy": 5.4328320026397705,
      "epoch": 0.7170718432510885,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004955787761089952,
      "loss": 5.2523,
      "mean_token_accuracy": 0.1807563617825508,
      "num_tokens": 16440122.0,
      "step": 7905
    },
    {
      "entropy": 5.56105408668518,
      "epoch": 0.7175253991291727,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955723910251296,
      "loss": 5.338,
      "mean_token_accuracy": 0.1794456034898758,
      "num_tokens": 16449801.0,
      "step": 7910
    },
    {
      "entropy": 5.44252028465271,
      "epoch": 0.717978955007257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004955660013797709,
      "loss": 5.1941,
      "mean_token_accuracy": 0.189080049097538,
      "num_tokens": 16460155.0,
      "step": 7915
    },
    {
      "entropy": 5.547311067581177,
      "epoch": 0.7184325108853411,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495559607173051,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17179529070854188,
      "num_tokens": 16470656.0,
      "step": 7920
    },
    {
      "entropy": 5.511463022232055,
      "epoch": 0.7188860667634253,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955532084051022,
      "loss": 5.2813,
      "mean_token_accuracy": 0.18540270030498504,
      "num_tokens": 16481449.0,
      "step": 7925
    },
    {
      "entropy": 5.474109745025634,
      "epoch": 0.7193396226415094,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004955468050760568,
      "loss": 5.2993,
      "mean_token_accuracy": 0.1859189748764038,
      "num_tokens": 16492032.0,
      "step": 7930
    },
    {
      "entropy": 5.499914312362671,
      "epoch": 0.7197931785195936,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004955403971860474,
      "loss": 5.2276,
      "mean_token_accuracy": 0.18490062803030013,
      "num_tokens": 16502221.0,
      "step": 7935
    },
    {
      "entropy": 5.50188307762146,
      "epoch": 0.7202467343976778,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955339847352063,
      "loss": 5.2127,
      "mean_token_accuracy": 0.18936268985271454,
      "num_tokens": 16512021.0,
      "step": 7940
    },
    {
      "entropy": 5.486470651626587,
      "epoch": 0.720700290275762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955275677236662,
      "loss": 5.27,
      "mean_token_accuracy": 0.1855945870280266,
      "num_tokens": 16522060.0,
      "step": 7945
    },
    {
      "entropy": 5.504107522964477,
      "epoch": 0.7211538461538461,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004955211461515599,
      "loss": 5.2715,
      "mean_token_accuracy": 0.18720357716083527,
      "num_tokens": 16532827.0,
      "step": 7950
    },
    {
      "entropy": 5.490920114517212,
      "epoch": 0.7216074020319303,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00049551472001902,
      "loss": 5.3106,
      "mean_token_accuracy": 0.18881746232509614,
      "num_tokens": 16542838.0,
      "step": 7955
    },
    {
      "entropy": 5.46091480255127,
      "epoch": 0.7220609579100146,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004955082893261796,
      "loss": 5.252,
      "mean_token_accuracy": 0.18916592001914978,
      "num_tokens": 16552615.0,
      "step": 7960
    },
    {
      "entropy": 5.521874952316284,
      "epoch": 0.7225145137880987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955018540731714,
      "loss": 5.3237,
      "mean_token_accuracy": 0.1830597385764122,
      "num_tokens": 16561777.0,
      "step": 7965
    },
    {
      "entropy": 5.519312000274658,
      "epoch": 0.7229680696661829,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004954954142601288,
      "loss": 5.2888,
      "mean_token_accuracy": 0.182749043405056,
      "num_tokens": 16572111.0,
      "step": 7970
    },
    {
      "entropy": 5.5174407958984375,
      "epoch": 0.723421625544267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004954889698871849,
      "loss": 5.3275,
      "mean_token_accuracy": 0.18790917694568635,
      "num_tokens": 16581674.0,
      "step": 7975
    },
    {
      "entropy": 5.516995191574097,
      "epoch": 0.7238751814223512,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004954825209544728,
      "loss": 5.3103,
      "mean_token_accuracy": 0.1911083474755287,
      "num_tokens": 16591802.0,
      "step": 7980
    },
    {
      "entropy": 5.4708818912506105,
      "epoch": 0.7243287373004355,
      "grad_norm": 0.9375,
      "learning_rate": 0.000495476067462126,
      "loss": 5.202,
      "mean_token_accuracy": 0.1951405331492424,
      "num_tokens": 16602772.0,
      "step": 7985
    },
    {
      "entropy": 5.5349139213562015,
      "epoch": 0.7247822931785196,
      "grad_norm": 0.96875,
      "learning_rate": 0.000495469609410278,
      "loss": 5.2517,
      "mean_token_accuracy": 0.18888139277696608,
      "num_tokens": 16613526.0,
      "step": 7990
    },
    {
      "entropy": 5.402984237670898,
      "epoch": 0.7252358490566038,
      "grad_norm": 1.125,
      "learning_rate": 0.0004954631467990622,
      "loss": 5.26,
      "mean_token_accuracy": 0.19152738451957702,
      "num_tokens": 16623583.0,
      "step": 7995
    },
    {
      "entropy": 5.496296977996826,
      "epoch": 0.7256894049346879,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954566796286124,
      "loss": 5.3182,
      "mean_token_accuracy": 0.18329164236783982,
      "num_tokens": 16635068.0,
      "step": 8000
    },
    {
      "entropy": 5.568232727050781,
      "epoch": 0.7261429608127722,
      "grad_norm": 1.0,
      "learning_rate": 0.0004954502078990623,
      "loss": 5.2167,
      "mean_token_accuracy": 0.18837719559669494,
      "num_tokens": 16645021.0,
      "step": 8005
    },
    {
      "entropy": 5.470982456207276,
      "epoch": 0.7265965166908563,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004954437316105457,
      "loss": 5.3,
      "mean_token_accuracy": 0.17855734527111053,
      "num_tokens": 16656021.0,
      "step": 8010
    },
    {
      "entropy": 5.408538675308227,
      "epoch": 0.7270500725689405,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004954372507631965,
      "loss": 5.1978,
      "mean_token_accuracy": 0.19202053397893906,
      "num_tokens": 16665236.0,
      "step": 8015
    },
    {
      "entropy": 5.490168190002441,
      "epoch": 0.7275036284470247,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004954307653571489,
      "loss": 5.396,
      "mean_token_accuracy": 0.1767436221241951,
      "num_tokens": 16676690.0,
      "step": 8020
    },
    {
      "entropy": 5.514575815200805,
      "epoch": 0.7279571843251088,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004954242753925367,
      "loss": 5.2762,
      "mean_token_accuracy": 0.17500143349170685,
      "num_tokens": 16687288.0,
      "step": 8025
    },
    {
      "entropy": 5.487605571746826,
      "epoch": 0.7284107402031931,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004954177808694946,
      "loss": 5.2849,
      "mean_token_accuracy": 0.18090034574270247,
      "num_tokens": 16697873.0,
      "step": 8030
    },
    {
      "entropy": 5.501708555221557,
      "epoch": 0.7288642960812772,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004954112817881565,
      "loss": 5.3056,
      "mean_token_accuracy": 0.18123792260885238,
      "num_tokens": 16709002.0,
      "step": 8035
    },
    {
      "entropy": 5.468709087371826,
      "epoch": 0.7293178519593614,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004954047781486568,
      "loss": 5.2637,
      "mean_token_accuracy": 0.18713146299123765,
      "num_tokens": 16718989.0,
      "step": 8040
    },
    {
      "entropy": 5.572985076904297,
      "epoch": 0.7297714078374455,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004953982699511303,
      "loss": 5.355,
      "mean_token_accuracy": 0.18397848755121232,
      "num_tokens": 16730425.0,
      "step": 8045
    },
    {
      "entropy": 5.519139575958252,
      "epoch": 0.7302249637155298,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004953917571957114,
      "loss": 5.2586,
      "mean_token_accuracy": 0.18431825190782547,
      "num_tokens": 16740216.0,
      "step": 8050
    },
    {
      "entropy": 5.4953625202178955,
      "epoch": 0.730678519593614,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953852398825346,
      "loss": 5.2423,
      "mean_token_accuracy": 0.1867049127817154,
      "num_tokens": 16752436.0,
      "step": 8055
    },
    {
      "entropy": 5.4429257869720455,
      "epoch": 0.7311320754716981,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495378718011735,
      "loss": 5.2692,
      "mean_token_accuracy": 0.1828521177172661,
      "num_tokens": 16762631.0,
      "step": 8060
    },
    {
      "entropy": 5.514633131027222,
      "epoch": 0.7315856313497823,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953721915834473,
      "loss": 5.2774,
      "mean_token_accuracy": 0.18744696229696273,
      "num_tokens": 16772613.0,
      "step": 8065
    },
    {
      "entropy": 5.547576761245727,
      "epoch": 0.7320391872278664,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004953656605978067,
      "loss": 5.2805,
      "mean_token_accuracy": 0.17784342020750046,
      "num_tokens": 16782963.0,
      "step": 8070
    },
    {
      "entropy": 5.477340126037598,
      "epoch": 0.7324927431059507,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000495359125054948,
      "loss": 5.205,
      "mean_token_accuracy": 0.1900916039943695,
      "num_tokens": 16793892.0,
      "step": 8075
    },
    {
      "entropy": 5.3859124183654785,
      "epoch": 0.7329462989840348,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004953525849550063,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18191331773996353,
      "num_tokens": 16803388.0,
      "step": 8080
    },
    {
      "entropy": 5.457746791839599,
      "epoch": 0.733399854862119,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953460402981172,
      "loss": 5.1696,
      "mean_token_accuracy": 0.19219024181365968,
      "num_tokens": 16813201.0,
      "step": 8085
    },
    {
      "entropy": 5.542439699172974,
      "epoch": 0.7338534107402032,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004953394910844157,
      "loss": 5.4383,
      "mean_token_accuracy": 0.178664830327034,
      "num_tokens": 16823265.0,
      "step": 8090
    },
    {
      "entropy": 5.487712240219116,
      "epoch": 0.7343069666182874,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004953329373140374,
      "loss": 5.2248,
      "mean_token_accuracy": 0.18390753567218782,
      "num_tokens": 16833057.0,
      "step": 8095
    },
    {
      "entropy": 5.401983737945557,
      "epoch": 0.7347605224963716,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004953263789871179,
      "loss": 5.1818,
      "mean_token_accuracy": 0.18794455230236054,
      "num_tokens": 16843488.0,
      "step": 8100
    },
    {
      "entropy": 5.413467597961426,
      "epoch": 0.7352140783744557,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004953198161037925,
      "loss": 5.1973,
      "mean_token_accuracy": 0.19814343005418777,
      "num_tokens": 16852448.0,
      "step": 8105
    },
    {
      "entropy": 5.396668195724487,
      "epoch": 0.7356676342525399,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004953132486641975,
      "loss": 5.2457,
      "mean_token_accuracy": 0.18623815029859542,
      "num_tokens": 16863944.0,
      "step": 8110
    },
    {
      "entropy": 5.5121378898620605,
      "epoch": 0.736121190130624,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000495306676668468,
      "loss": 5.1688,
      "mean_token_accuracy": 0.19186616986989974,
      "num_tokens": 16873923.0,
      "step": 8115
    },
    {
      "entropy": 5.5024964809417725,
      "epoch": 0.7365747460087083,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004953001001167406,
      "loss": 5.2731,
      "mean_token_accuracy": 0.18318339437246323,
      "num_tokens": 16884373.0,
      "step": 8120
    },
    {
      "entropy": 5.522818374633789,
      "epoch": 0.7370283018867925,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004952935190091508,
      "loss": 5.4047,
      "mean_token_accuracy": 0.17949366867542266,
      "num_tokens": 16895869.0,
      "step": 8125
    },
    {
      "entropy": 5.411563777923584,
      "epoch": 0.7374818577648766,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004952869333458349,
      "loss": 5.2367,
      "mean_token_accuracy": 0.18748896867036818,
      "num_tokens": 16906737.0,
      "step": 8130
    },
    {
      "entropy": 5.4336510181427,
      "epoch": 0.7379354136429608,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004952803431269292,
      "loss": 5.1823,
      "mean_token_accuracy": 0.1944849282503128,
      "num_tokens": 16916083.0,
      "step": 8135
    },
    {
      "entropy": 5.546496820449829,
      "epoch": 0.738388969521045,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004952737483525699,
      "loss": 5.346,
      "mean_token_accuracy": 0.1862415462732315,
      "num_tokens": 16926419.0,
      "step": 8140
    },
    {
      "entropy": 5.537788152694702,
      "epoch": 0.7388425253991292,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952671490228932,
      "loss": 5.2821,
      "mean_token_accuracy": 0.18204317539930343,
      "num_tokens": 16936067.0,
      "step": 8145
    },
    {
      "entropy": 5.551399993896484,
      "epoch": 0.7392960812772134,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004952605451380357,
      "loss": 5.2271,
      "mean_token_accuracy": 0.18934836983680725,
      "num_tokens": 16946571.0,
      "step": 8150
    },
    {
      "entropy": 5.41324372291565,
      "epoch": 0.7397496371552975,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004952539366981342,
      "loss": 5.2941,
      "mean_token_accuracy": 0.18353919684886932,
      "num_tokens": 16956764.0,
      "step": 8155
    },
    {
      "entropy": 5.457695579528808,
      "epoch": 0.7402031930333817,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495247323703325,
      "loss": 5.2137,
      "mean_token_accuracy": 0.18735239505767823,
      "num_tokens": 16966002.0,
      "step": 8160
    },
    {
      "entropy": 5.569545698165894,
      "epoch": 0.7406567489114659,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000495240706153745,
      "loss": 5.3036,
      "mean_token_accuracy": 0.1865418493747711,
      "num_tokens": 16975705.0,
      "step": 8165
    },
    {
      "entropy": 5.4822554111480715,
      "epoch": 0.7411103047895501,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004952340840495312,
      "loss": 5.2087,
      "mean_token_accuracy": 0.18633690029382705,
      "num_tokens": 16985945.0,
      "step": 8170
    },
    {
      "entropy": 5.4184746742248535,
      "epoch": 0.7415638606676342,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004952274573908203,
      "loss": 5.189,
      "mean_token_accuracy": 0.18653265982866288,
      "num_tokens": 16995475.0,
      "step": 8175
    },
    {
      "entropy": 5.464861154556274,
      "epoch": 0.7420174165457184,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004952208261777496,
      "loss": 5.2251,
      "mean_token_accuracy": 0.18869319707155227,
      "num_tokens": 17005028.0,
      "step": 8180
    },
    {
      "entropy": 5.505099534988403,
      "epoch": 0.7424709724238027,
      "grad_norm": 0.953125,
      "learning_rate": 0.000495214190410456,
      "loss": 5.2747,
      "mean_token_accuracy": 0.18964312076568604,
      "num_tokens": 17014786.0,
      "step": 8185
    },
    {
      "entropy": 5.522277116775513,
      "epoch": 0.7429245283018868,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952075500890768,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1866888329386711,
      "num_tokens": 17024431.0,
      "step": 8190
    },
    {
      "entropy": 5.457682704925537,
      "epoch": 0.743378084179971,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004952009052137493,
      "loss": 5.259,
      "mean_token_accuracy": 0.18620658963918685,
      "num_tokens": 17034368.0,
      "step": 8195
    },
    {
      "entropy": 5.472687244415283,
      "epoch": 0.7438316400580551,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951942557846112,
      "loss": 5.2908,
      "mean_token_accuracy": 0.1834196001291275,
      "num_tokens": 17043818.0,
      "step": 8200
    },
    {
      "entropy": 5.5276368141174315,
      "epoch": 0.7442851959361393,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004951876018017997,
      "loss": 5.3028,
      "mean_token_accuracy": 0.18539572805166243,
      "num_tokens": 17054843.0,
      "step": 8205
    },
    {
      "entropy": 5.494559049606323,
      "epoch": 0.7447387518142236,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004951809432654524,
      "loss": 5.3072,
      "mean_token_accuracy": 0.1883242279291153,
      "num_tokens": 17065539.0,
      "step": 8210
    },
    {
      "entropy": 5.442562866210937,
      "epoch": 0.7451923076923077,
      "grad_norm": 0.984375,
      "learning_rate": 0.000495174280175707,
      "loss": 5.2166,
      "mean_token_accuracy": 0.1864542081952095,
      "num_tokens": 17075443.0,
      "step": 8215
    },
    {
      "entropy": 5.560985994338989,
      "epoch": 0.7456458635703919,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004951676125327015,
      "loss": 5.3336,
      "mean_token_accuracy": 0.19198401570320128,
      "num_tokens": 17085490.0,
      "step": 8220
    },
    {
      "entropy": 5.551355648040771,
      "epoch": 0.746099419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004951609403365737,
      "loss": 5.3245,
      "mean_token_accuracy": 0.18634077459573745,
      "num_tokens": 17094803.0,
      "step": 8225
    },
    {
      "entropy": 5.491112995147705,
      "epoch": 0.7465529753265603,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004951542635874615,
      "loss": 5.18,
      "mean_token_accuracy": 0.18782074451446534,
      "num_tokens": 17104373.0,
      "step": 8230
    },
    {
      "entropy": 5.529582977294922,
      "epoch": 0.7470065312046444,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004951475822855029,
      "loss": 5.2702,
      "mean_token_accuracy": 0.1870327353477478,
      "num_tokens": 17115469.0,
      "step": 8235
    },
    {
      "entropy": 5.464807987213135,
      "epoch": 0.7474600870827286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004951408964308364,
      "loss": 5.2378,
      "mean_token_accuracy": 0.1856093615293503,
      "num_tokens": 17125443.0,
      "step": 8240
    },
    {
      "entropy": 5.451996564865112,
      "epoch": 0.7479136429608128,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951342060236,
      "loss": 5.2143,
      "mean_token_accuracy": 0.18845360726118088,
      "num_tokens": 17136253.0,
      "step": 8245
    },
    {
      "entropy": 5.5428102016448975,
      "epoch": 0.7483671988388969,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951275110639321,
      "loss": 5.275,
      "mean_token_accuracy": 0.18393643498420714,
      "num_tokens": 17145544.0,
      "step": 8250
    },
    {
      "entropy": 5.495501470565796,
      "epoch": 0.7488207547169812,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951208115519712,
      "loss": 5.2632,
      "mean_token_accuracy": 0.18630698323249817,
      "num_tokens": 17155043.0,
      "step": 8255
    },
    {
      "entropy": 5.464247703552246,
      "epoch": 0.7492743105950653,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951141074878558,
      "loss": 5.308,
      "mean_token_accuracy": 0.1831713780760765,
      "num_tokens": 17164885.0,
      "step": 8260
    },
    {
      "entropy": 5.5879473209381105,
      "epoch": 0.7497278664731495,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004951073988717246,
      "loss": 5.3179,
      "mean_token_accuracy": 0.18246327489614486,
      "num_tokens": 17175681.0,
      "step": 8265
    },
    {
      "entropy": 5.449235534667968,
      "epoch": 0.7501814223512336,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004951006857037163,
      "loss": 5.184,
      "mean_token_accuracy": 0.18728435188531875,
      "num_tokens": 17186002.0,
      "step": 8270
    },
    {
      "entropy": 5.469281768798828,
      "epoch": 0.7506349782293179,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004950939679839696,
      "loss": 5.2305,
      "mean_token_accuracy": 0.1872097671031952,
      "num_tokens": 17195499.0,
      "step": 8275
    },
    {
      "entropy": 5.444141530990601,
      "epoch": 0.7510885341074021,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004950872457126237,
      "loss": 5.24,
      "mean_token_accuracy": 0.18864094465970993,
      "num_tokens": 17204684.0,
      "step": 8280
    },
    {
      "entropy": 5.476453733444214,
      "epoch": 0.7515420899854862,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004950805188898175,
      "loss": 5.2717,
      "mean_token_accuracy": 0.1793087586760521,
      "num_tokens": 17213987.0,
      "step": 8285
    },
    {
      "entropy": 5.512142181396484,
      "epoch": 0.7519956458635704,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00049507378751569,
      "loss": 5.2851,
      "mean_token_accuracy": 0.18152845948934554,
      "num_tokens": 17224512.0,
      "step": 8290
    },
    {
      "entropy": 5.550980758666992,
      "epoch": 0.7524492017416545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004950670515903805,
      "loss": 5.3058,
      "mean_token_accuracy": 0.18777280449867248,
      "num_tokens": 17234362.0,
      "step": 8295
    },
    {
      "entropy": 5.5337158203125,
      "epoch": 0.7529027576197388,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004950603111140283,
      "loss": 5.299,
      "mean_token_accuracy": 0.1910579890012741,
      "num_tokens": 17244241.0,
      "step": 8300
    },
    {
      "entropy": 5.466256761550904,
      "epoch": 0.753356313497823,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004950535660867728,
      "loss": 5.2312,
      "mean_token_accuracy": 0.19583994895219803,
      "num_tokens": 17254689.0,
      "step": 8305
    },
    {
      "entropy": 5.5242081642150875,
      "epoch": 0.7538098693759071,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950468165087535,
      "loss": 5.3076,
      "mean_token_accuracy": 0.18249109089374543,
      "num_tokens": 17265506.0,
      "step": 8310
    },
    {
      "entropy": 5.509704494476319,
      "epoch": 0.7542634252539913,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004950400623801098,
      "loss": 5.3849,
      "mean_token_accuracy": 0.18589098006486893,
      "num_tokens": 17276376.0,
      "step": 8315
    },
    {
      "entropy": 5.559766817092895,
      "epoch": 0.7547169811320755,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004950333037009818,
      "loss": 5.2798,
      "mean_token_accuracy": 0.175719191133976,
      "num_tokens": 17287005.0,
      "step": 8320
    },
    {
      "entropy": 5.537489891052246,
      "epoch": 0.7551705370101597,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004950265404715089,
      "loss": 5.2791,
      "mean_token_accuracy": 0.18351641595363616,
      "num_tokens": 17297595.0,
      "step": 8325
    },
    {
      "entropy": 5.4321057319641115,
      "epoch": 0.7556240928882438,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000495019772691831,
      "loss": 5.1553,
      "mean_token_accuracy": 0.18927870392799379,
      "num_tokens": 17307319.0,
      "step": 8330
    },
    {
      "entropy": 5.5272448539733885,
      "epoch": 0.756077648766328,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004950130003620882,
      "loss": 5.3748,
      "mean_token_accuracy": 0.18703329414129258,
      "num_tokens": 17317978.0,
      "step": 8335
    },
    {
      "entropy": 5.455478620529175,
      "epoch": 0.7565312046444121,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004950062234824204,
      "loss": 5.2836,
      "mean_token_accuracy": 0.1799110844731331,
      "num_tokens": 17328339.0,
      "step": 8340
    },
    {
      "entropy": 5.499893665313721,
      "epoch": 0.7569847605224964,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004949994420529679,
      "loss": 5.317,
      "mean_token_accuracy": 0.17965888679027558,
      "num_tokens": 17338668.0,
      "step": 8345
    },
    {
      "entropy": 5.474080753326416,
      "epoch": 0.7574383164005806,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004949926560738709,
      "loss": 5.2619,
      "mean_token_accuracy": 0.17932692617177964,
      "num_tokens": 17348609.0,
      "step": 8350
    },
    {
      "entropy": 5.548439264297485,
      "epoch": 0.7578918722786647,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004949858655452697,
      "loss": 5.2713,
      "mean_token_accuracy": 0.18396163880825042,
      "num_tokens": 17358014.0,
      "step": 8355
    },
    {
      "entropy": 5.541851425170899,
      "epoch": 0.7583454281567489,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949790704673048,
      "loss": 5.2734,
      "mean_token_accuracy": 0.18673214316368103,
      "num_tokens": 17367975.0,
      "step": 8360
    },
    {
      "entropy": 5.51186900138855,
      "epoch": 0.7587989840348331,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949722708401166,
      "loss": 5.2276,
      "mean_token_accuracy": 0.18283698856830596,
      "num_tokens": 17377960.0,
      "step": 8365
    },
    {
      "entropy": 5.555224323272705,
      "epoch": 0.7592525399129173,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004949654666638457,
      "loss": 5.3064,
      "mean_token_accuracy": 0.1900601267814636,
      "num_tokens": 17387334.0,
      "step": 8370
    },
    {
      "entropy": 5.493747091293335,
      "epoch": 0.7597060957910015,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000494958657938633,
      "loss": 5.2611,
      "mean_token_accuracy": 0.18564350605010987,
      "num_tokens": 17396521.0,
      "step": 8375
    },
    {
      "entropy": 5.420101881027222,
      "epoch": 0.7601596516690856,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004949518446646192,
      "loss": 5.2373,
      "mean_token_accuracy": 0.18725302666425706,
      "num_tokens": 17407217.0,
      "step": 8380
    },
    {
      "entropy": 5.617413759231567,
      "epoch": 0.7606132075471698,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949450268419451,
      "loss": 5.3636,
      "mean_token_accuracy": 0.17854755967855454,
      "num_tokens": 17416982.0,
      "step": 8385
    },
    {
      "entropy": 5.44600477218628,
      "epoch": 0.761066763425254,
      "grad_norm": 0.9375,
      "learning_rate": 0.000494938204470752,
      "loss": 5.1947,
      "mean_token_accuracy": 0.19048412889242172,
      "num_tokens": 17427636.0,
      "step": 8390
    },
    {
      "entropy": 5.491252088546753,
      "epoch": 0.7615203193033382,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949313775511806,
      "loss": 5.3197,
      "mean_token_accuracy": 0.17598724365234375,
      "num_tokens": 17437488.0,
      "step": 8395
    },
    {
      "entropy": 5.467138290405273,
      "epoch": 0.7619738751814223,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004949245460833723,
      "loss": 5.1021,
      "mean_token_accuracy": 0.19133297652006148,
      "num_tokens": 17447838.0,
      "step": 8400
    },
    {
      "entropy": 5.547179698944092,
      "epoch": 0.7624274310595065,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004949177100674684,
      "loss": 5.3324,
      "mean_token_accuracy": 0.1857564628124237,
      "num_tokens": 17457257.0,
      "step": 8405
    },
    {
      "entropy": 5.489846420288086,
      "epoch": 0.7628809869375908,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004949108695036102,
      "loss": 5.2398,
      "mean_token_accuracy": 0.18621718734502793,
      "num_tokens": 17467729.0,
      "step": 8410
    },
    {
      "entropy": 5.480379867553711,
      "epoch": 0.7633345428156749,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949040243919392,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18046046197414398,
      "num_tokens": 17477692.0,
      "step": 8415
    },
    {
      "entropy": 5.470694160461425,
      "epoch": 0.7637880986937591,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004948971747325968,
      "loss": 5.2666,
      "mean_token_accuracy": 0.18498124331235885,
      "num_tokens": 17488730.0,
      "step": 8420
    },
    {
      "entropy": 5.538025569915772,
      "epoch": 0.7642416545718432,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494890320525725,
      "loss": 5.3334,
      "mean_token_accuracy": 0.1852163478732109,
      "num_tokens": 17498374.0,
      "step": 8425
    },
    {
      "entropy": 5.477238607406616,
      "epoch": 0.7646952104499274,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004948834617714652,
      "loss": 5.192,
      "mean_token_accuracy": 0.18794307559728624,
      "num_tokens": 17508601.0,
      "step": 8430
    },
    {
      "entropy": 5.487441873550415,
      "epoch": 0.7651487663280117,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948765984699595,
      "loss": 5.2338,
      "mean_token_accuracy": 0.18388939797878265,
      "num_tokens": 17518164.0,
      "step": 8435
    },
    {
      "entropy": 5.520015287399292,
      "epoch": 0.7656023222060958,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948697306213494,
      "loss": 5.2144,
      "mean_token_accuracy": 0.19726908057928086,
      "num_tokens": 17527244.0,
      "step": 8440
    },
    {
      "entropy": 5.4783806800842285,
      "epoch": 0.76605587808418,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948628582257775,
      "loss": 5.2786,
      "mean_token_accuracy": 0.18394037932157517,
      "num_tokens": 17538045.0,
      "step": 8445
    },
    {
      "entropy": 5.4761889457702635,
      "epoch": 0.7665094339622641,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004948559812833857,
      "loss": 5.2373,
      "mean_token_accuracy": 0.18717231154441832,
      "num_tokens": 17547584.0,
      "step": 8450
    },
    {
      "entropy": 5.484969186782837,
      "epoch": 0.7669629898403484,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004948490997943162,
      "loss": 5.2864,
      "mean_token_accuracy": 0.18479120582342148,
      "num_tokens": 17558925.0,
      "step": 8455
    },
    {
      "entropy": 5.506250715255737,
      "epoch": 0.7674165457184325,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004948422137587111,
      "loss": 5.3651,
      "mean_token_accuracy": 0.17608188539743425,
      "num_tokens": 17569355.0,
      "step": 8460
    },
    {
      "entropy": 5.5164354801177975,
      "epoch": 0.7678701015965167,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004948353231767131,
      "loss": 5.2246,
      "mean_token_accuracy": 0.18756223171949388,
      "num_tokens": 17579782.0,
      "step": 8465
    },
    {
      "entropy": 5.50532283782959,
      "epoch": 0.7683236574746009,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948284280484645,
      "loss": 5.2839,
      "mean_token_accuracy": 0.19143763184547424,
      "num_tokens": 17589772.0,
      "step": 8470
    },
    {
      "entropy": 5.406950950622559,
      "epoch": 0.768777213352685,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004948215283741081,
      "loss": 5.1837,
      "mean_token_accuracy": 0.1878790646791458,
      "num_tokens": 17599613.0,
      "step": 8475
    },
    {
      "entropy": 5.485948133468628,
      "epoch": 0.7692307692307693,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004948146241537866,
      "loss": 5.2951,
      "mean_token_accuracy": 0.1790087714791298,
      "num_tokens": 17609056.0,
      "step": 8480
    },
    {
      "entropy": 5.497835874557495,
      "epoch": 0.7696843251088534,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004948077153876425,
      "loss": 5.214,
      "mean_token_accuracy": 0.18795309066772461,
      "num_tokens": 17618620.0,
      "step": 8485
    },
    {
      "entropy": 5.446607255935669,
      "epoch": 0.7701378809869376,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948008020758188,
      "loss": 5.1809,
      "mean_token_accuracy": 0.19390195310115815,
      "num_tokens": 17627630.0,
      "step": 8490
    },
    {
      "entropy": 5.36888632774353,
      "epoch": 0.7705914368650217,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004947938842184585,
      "loss": 5.1691,
      "mean_token_accuracy": 0.1886688217520714,
      "num_tokens": 17638903.0,
      "step": 8495
    },
    {
      "entropy": 5.444310665130615,
      "epoch": 0.771044992743106,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004947869618157048,
      "loss": 5.2229,
      "mean_token_accuracy": 0.18465732783079147,
      "num_tokens": 17649257.0,
      "step": 8500
    },
    {
      "entropy": 5.456356811523437,
      "epoch": 0.7714985486211902,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004947800348677006,
      "loss": 5.1987,
      "mean_token_accuracy": 0.19576434046030045,
      "num_tokens": 17659495.0,
      "step": 8505
    },
    {
      "entropy": 5.399709033966064,
      "epoch": 0.7719521044992743,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004947731033745893,
      "loss": 5.1733,
      "mean_token_accuracy": 0.18452270925045014,
      "num_tokens": 17669671.0,
      "step": 8510
    },
    {
      "entropy": 5.529618835449218,
      "epoch": 0.7724056603773585,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004947661673365142,
      "loss": 5.3506,
      "mean_token_accuracy": 0.18083200752735137,
      "num_tokens": 17680927.0,
      "step": 8515
    },
    {
      "entropy": 5.483820056915283,
      "epoch": 0.7728592162554426,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004947592267536188,
      "loss": 5.2098,
      "mean_token_accuracy": 0.18810635805130005,
      "num_tokens": 17691378.0,
      "step": 8520
    },
    {
      "entropy": 5.477559185028076,
      "epoch": 0.7733127721335269,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004947522816260466,
      "loss": 5.2962,
      "mean_token_accuracy": 0.1867695137858391,
      "num_tokens": 17700942.0,
      "step": 8525
    },
    {
      "entropy": 5.467412757873535,
      "epoch": 0.773766328011611,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004947453319539411,
      "loss": 5.2025,
      "mean_token_accuracy": 0.18547020852565765,
      "num_tokens": 17710076.0,
      "step": 8530
    },
    {
      "entropy": 5.474762058258056,
      "epoch": 0.7742198838896952,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947383777374462,
      "loss": 5.2309,
      "mean_token_accuracy": 0.1894417494535446,
      "num_tokens": 17721009.0,
      "step": 8535
    },
    {
      "entropy": 5.449787902832031,
      "epoch": 0.7746734397677794,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004947314189767058,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18772920221090317,
      "num_tokens": 17731821.0,
      "step": 8540
    },
    {
      "entropy": 5.430331468582153,
      "epoch": 0.7751269956458636,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004947244556718635,
      "loss": 5.1503,
      "mean_token_accuracy": 0.19465826600790023,
      "num_tokens": 17741425.0,
      "step": 8545
    },
    {
      "entropy": 5.462887001037598,
      "epoch": 0.7755805515239478,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947174878230635,
      "loss": 5.1994,
      "mean_token_accuracy": 0.1927179977297783,
      "num_tokens": 17752381.0,
      "step": 8550
    },
    {
      "entropy": 5.409137153625489,
      "epoch": 0.7760341074020319,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004947105154304498,
      "loss": 5.1963,
      "mean_token_accuracy": 0.187833933532238,
      "num_tokens": 17762768.0,
      "step": 8555
    },
    {
      "entropy": 5.462284898757934,
      "epoch": 0.7764876632801161,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004947035384941667,
      "loss": 5.2253,
      "mean_token_accuracy": 0.18217650353908538,
      "num_tokens": 17774032.0,
      "step": 8560
    },
    {
      "entropy": 5.554944324493408,
      "epoch": 0.7769412191582002,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946965570143584,
      "loss": 5.3676,
      "mean_token_accuracy": 0.1783722385764122,
      "num_tokens": 17785098.0,
      "step": 8565
    },
    {
      "entropy": 5.524931526184082,
      "epoch": 0.7773947750362845,
      "grad_norm": 1.0,
      "learning_rate": 0.0004946895709911695,
      "loss": 5.2255,
      "mean_token_accuracy": 0.18829830288887023,
      "num_tokens": 17795487.0,
      "step": 8570
    },
    {
      "entropy": 5.490832090377808,
      "epoch": 0.7778483309143687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004946825804247442,
      "loss": 5.345,
      "mean_token_accuracy": 0.17945600003004075,
      "num_tokens": 17806045.0,
      "step": 8575
    },
    {
      "entropy": 5.543007564544678,
      "epoch": 0.7783018867924528,
      "grad_norm": 0.921875,
      "learning_rate": 0.000494675585315227,
      "loss": 5.2967,
      "mean_token_accuracy": 0.17957210540771484,
      "num_tokens": 17816948.0,
      "step": 8580
    },
    {
      "entropy": 5.540530586242676,
      "epoch": 0.778755442670537,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004946685856627628,
      "loss": 5.2552,
      "mean_token_accuracy": 0.18615331947803498,
      "num_tokens": 17826929.0,
      "step": 8585
    },
    {
      "entropy": 5.514763975143433,
      "epoch": 0.7792089985486212,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004946615814674963,
      "loss": 5.3319,
      "mean_token_accuracy": 0.18369301408529282,
      "num_tokens": 17837780.0,
      "step": 8590
    },
    {
      "entropy": 5.387348318099976,
      "epoch": 0.7796625544267054,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004946545727295724,
      "loss": 5.1051,
      "mean_token_accuracy": 0.19867068976163865,
      "num_tokens": 17847835.0,
      "step": 8595
    },
    {
      "entropy": 5.452921438217163,
      "epoch": 0.7801161103047896,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004946475594491359,
      "loss": 5.1948,
      "mean_token_accuracy": 0.19558655917644502,
      "num_tokens": 17858324.0,
      "step": 8600
    },
    {
      "entropy": 5.481823539733886,
      "epoch": 0.7805696661828737,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004946405416263319,
      "loss": 5.2549,
      "mean_token_accuracy": 0.18132758736610413,
      "num_tokens": 17868455.0,
      "step": 8605
    },
    {
      "entropy": 5.487615919113159,
      "epoch": 0.7810232220609579,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946335192613054,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18736030757427216,
      "num_tokens": 17879038.0,
      "step": 8610
    },
    {
      "entropy": 5.404131507873535,
      "epoch": 0.7814767779390421,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004946264923542019,
      "loss": 5.1126,
      "mean_token_accuracy": 0.1924211636185646,
      "num_tokens": 17888153.0,
      "step": 8615
    },
    {
      "entropy": 5.440420913696289,
      "epoch": 0.7819303338171263,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004946194609051664,
      "loss": 5.2392,
      "mean_token_accuracy": 0.1910945162177086,
      "num_tokens": 17898289.0,
      "step": 8620
    },
    {
      "entropy": 5.474812173843384,
      "epoch": 0.7823838896952104,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004946124249143447,
      "loss": 5.2419,
      "mean_token_accuracy": 0.1902774378657341,
      "num_tokens": 17909192.0,
      "step": 8625
    },
    {
      "entropy": 5.445071649551392,
      "epoch": 0.7828374455732946,
      "grad_norm": 0.921875,
      "learning_rate": 0.000494605384381882,
      "loss": 5.1791,
      "mean_token_accuracy": 0.1815720483660698,
      "num_tokens": 17918937.0,
      "step": 8630
    },
    {
      "entropy": 5.420489835739136,
      "epoch": 0.7832910014513788,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004945983393079239,
      "loss": 5.2201,
      "mean_token_accuracy": 0.19512285739183427,
      "num_tokens": 17929346.0,
      "step": 8635
    },
    {
      "entropy": 5.47286205291748,
      "epoch": 0.783744557329463,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945912896926163,
      "loss": 5.2335,
      "mean_token_accuracy": 0.18775832206010817,
      "num_tokens": 17939479.0,
      "step": 8640
    },
    {
      "entropy": 5.573257970809936,
      "epoch": 0.7841981132075472,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004945842355361049,
      "loss": 5.3215,
      "mean_token_accuracy": 0.1810726061463356,
      "num_tokens": 17949841.0,
      "step": 8645
    },
    {
      "entropy": 5.429520606994629,
      "epoch": 0.7846516690856313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004945771768385355,
      "loss": 5.2357,
      "mean_token_accuracy": 0.19100900441408158,
      "num_tokens": 17960334.0,
      "step": 8650
    },
    {
      "entropy": 5.42313871383667,
      "epoch": 0.7851052249637155,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000494570113600054,
      "loss": 5.1873,
      "mean_token_accuracy": 0.1927325651049614,
      "num_tokens": 17970595.0,
      "step": 8655
    },
    {
      "entropy": 5.488664007186889,
      "epoch": 0.7855587808417998,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945630458208068,
      "loss": 5.3267,
      "mean_token_accuracy": 0.17650770395994186,
      "num_tokens": 17981301.0,
      "step": 8660
    },
    {
      "entropy": 5.435780143737793,
      "epoch": 0.7860123367198839,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945559735009397,
      "loss": 5.1522,
      "mean_token_accuracy": 0.1920144185423851,
      "num_tokens": 17991262.0,
      "step": 8665
    },
    {
      "entropy": 5.522634553909302,
      "epoch": 0.7864658925979681,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945488966405993,
      "loss": 5.1962,
      "mean_token_accuracy": 0.1927980050444603,
      "num_tokens": 18001212.0,
      "step": 8670
    },
    {
      "entropy": 5.468816947937012,
      "epoch": 0.7869194484760522,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945418152399317,
      "loss": 5.2098,
      "mean_token_accuracy": 0.19336934387683868,
      "num_tokens": 18010432.0,
      "step": 8675
    },
    {
      "entropy": 5.452950716018677,
      "epoch": 0.7873730043541364,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004945347292990833,
      "loss": 5.2545,
      "mean_token_accuracy": 0.17933656126260758,
      "num_tokens": 18020902.0,
      "step": 8680
    },
    {
      "entropy": 5.4519189357757565,
      "epoch": 0.7878265602322206,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945276388182008,
      "loss": 5.1303,
      "mean_token_accuracy": 0.1907524511218071,
      "num_tokens": 18031133.0,
      "step": 8685
    },
    {
      "entropy": 5.5043336868286135,
      "epoch": 0.7882801161103048,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004945205437974309,
      "loss": 5.2221,
      "mean_token_accuracy": 0.19138137102127076,
      "num_tokens": 18042050.0,
      "step": 8690
    },
    {
      "entropy": 5.376919841766357,
      "epoch": 0.788733671988389,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945134442369201,
      "loss": 5.201,
      "mean_token_accuracy": 0.18819751739501953,
      "num_tokens": 18051642.0,
      "step": 8695
    },
    {
      "entropy": 5.506156253814697,
      "epoch": 0.7891872278664731,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004945063401368154,
      "loss": 5.2433,
      "mean_token_accuracy": 0.18314868658781053,
      "num_tokens": 18062180.0,
      "step": 8700
    },
    {
      "entropy": 5.473495769500732,
      "epoch": 0.7896407837445574,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004944992314972637,
      "loss": 5.2047,
      "mean_token_accuracy": 0.18621009290218354,
      "num_tokens": 18072302.0,
      "step": 8705
    },
    {
      "entropy": 5.4587818622589115,
      "epoch": 0.7900943396226415,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004944921183184118,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1914062112569809,
      "num_tokens": 18081824.0,
      "step": 8710
    },
    {
      "entropy": 5.457731771469116,
      "epoch": 0.7905478955007257,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000494485000600407,
      "loss": 5.2583,
      "mean_token_accuracy": 0.19050151854753494,
      "num_tokens": 18092696.0,
      "step": 8715
    },
    {
      "entropy": 5.413540267944336,
      "epoch": 0.7910014513788098,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004944778783433965,
      "loss": 5.1693,
      "mean_token_accuracy": 0.19154656678438187,
      "num_tokens": 18101790.0,
      "step": 8720
    },
    {
      "entropy": 5.525717163085938,
      "epoch": 0.791455007256894,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004944707515475276,
      "loss": 5.3025,
      "mean_token_accuracy": 0.1849859058856964,
      "num_tokens": 18111572.0,
      "step": 8725
    },
    {
      "entropy": 5.505932140350342,
      "epoch": 0.7919085631349783,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004944636202129475,
      "loss": 5.2822,
      "mean_token_accuracy": 0.1949717327952385,
      "num_tokens": 18120650.0,
      "step": 8730
    },
    {
      "entropy": 5.435335779190064,
      "epoch": 0.7923621190130624,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004944564843398039,
      "loss": 5.1569,
      "mean_token_accuracy": 0.1990776017308235,
      "num_tokens": 18131055.0,
      "step": 8735
    },
    {
      "entropy": 5.546191930770874,
      "epoch": 0.7928156748911466,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944493439282443,
      "loss": 5.2667,
      "mean_token_accuracy": 0.18769678622484207,
      "num_tokens": 18141423.0,
      "step": 8740
    },
    {
      "entropy": 5.409640216827393,
      "epoch": 0.7932692307692307,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004944421989784162,
      "loss": 5.1811,
      "mean_token_accuracy": 0.18719988465309143,
      "num_tokens": 18152015.0,
      "step": 8745
    },
    {
      "entropy": 5.4817914962768555,
      "epoch": 0.793722786647315,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944350494904676,
      "loss": 5.2799,
      "mean_token_accuracy": 0.18807127922773362,
      "num_tokens": 18161934.0,
      "step": 8750
    },
    {
      "entropy": 5.461260223388672,
      "epoch": 0.7941763425253991,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004944278954645463,
      "loss": 5.1952,
      "mean_token_accuracy": 0.18708185702562333,
      "num_tokens": 18172287.0,
      "step": 8755
    },
    {
      "entropy": 5.47312331199646,
      "epoch": 0.7946298984034833,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004944207369008002,
      "loss": 5.2517,
      "mean_token_accuracy": 0.19569893479347228,
      "num_tokens": 18181015.0,
      "step": 8760
    },
    {
      "entropy": 5.543885898590088,
      "epoch": 0.7950834542815675,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004944135737993773,
      "loss": 5.3502,
      "mean_token_accuracy": 0.18109244257211685,
      "num_tokens": 18192382.0,
      "step": 8765
    },
    {
      "entropy": 5.452446174621582,
      "epoch": 0.7955370101596516,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944064061604258,
      "loss": 5.1944,
      "mean_token_accuracy": 0.19768597930669785,
      "num_tokens": 18202597.0,
      "step": 8770
    },
    {
      "entropy": 5.4939878463745115,
      "epoch": 0.7959905660377359,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004943992339840939,
      "loss": 5.264,
      "mean_token_accuracy": 0.18397243022918702,
      "num_tokens": 18213756.0,
      "step": 8775
    },
    {
      "entropy": 5.4102050304412845,
      "epoch": 0.79644412191582,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049439205727053,
      "loss": 5.1772,
      "mean_token_accuracy": 0.18465391248464585,
      "num_tokens": 18223669.0,
      "step": 8780
    },
    {
      "entropy": 5.444017505645752,
      "epoch": 0.7968976777939042,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943848760198823,
      "loss": 5.202,
      "mean_token_accuracy": 0.19344062507152557,
      "num_tokens": 18234884.0,
      "step": 8785
    },
    {
      "entropy": 5.4118711948394775,
      "epoch": 0.7973512336719883,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004943776902322995,
      "loss": 5.1329,
      "mean_token_accuracy": 0.19479400366544725,
      "num_tokens": 18245155.0,
      "step": 8790
    },
    {
      "entropy": 5.349363708496094,
      "epoch": 0.7978047895500726,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004943704999079302,
      "loss": 5.0747,
      "mean_token_accuracy": 0.19781140685081483,
      "num_tokens": 18254686.0,
      "step": 8795
    },
    {
      "entropy": 5.433163785934449,
      "epoch": 0.7982583454281568,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004943633050469231,
      "loss": 5.1091,
      "mean_token_accuracy": 0.19448212683200836,
      "num_tokens": 18265473.0,
      "step": 8800
    },
    {
      "entropy": 5.5096200466156,
      "epoch": 0.7987119013062409,
      "grad_norm": 1.0,
      "learning_rate": 0.0004943561056494269,
      "loss": 5.306,
      "mean_token_accuracy": 0.18551455140113832,
      "num_tokens": 18276024.0,
      "step": 8805
    },
    {
      "entropy": 5.466209077835083,
      "epoch": 0.7991654571843251,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004943489017155906,
      "loss": 5.2774,
      "mean_token_accuracy": 0.17907433956861496,
      "num_tokens": 18287261.0,
      "step": 8810
    },
    {
      "entropy": 5.505891227722168,
      "epoch": 0.7996190130624092,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004943416932455631,
      "loss": 5.295,
      "mean_token_accuracy": 0.18324223309755325,
      "num_tokens": 18299374.0,
      "step": 8815
    },
    {
      "entropy": 5.438243436813354,
      "epoch": 0.8000725689404935,
      "grad_norm": 1.0,
      "learning_rate": 0.0004943344802394935,
      "loss": 5.2572,
      "mean_token_accuracy": 0.18090363442897797,
      "num_tokens": 18309344.0,
      "step": 8820
    },
    {
      "entropy": 5.50924162864685,
      "epoch": 0.8005261248185777,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004943272626975309,
      "loss": 5.2282,
      "mean_token_accuracy": 0.1878451332449913,
      "num_tokens": 18319778.0,
      "step": 8825
    },
    {
      "entropy": 5.483279848098755,
      "epoch": 0.8009796806966618,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004943200406198246,
      "loss": 5.2424,
      "mean_token_accuracy": 0.18929843008518218,
      "num_tokens": 18329787.0,
      "step": 8830
    },
    {
      "entropy": 5.42649998664856,
      "epoch": 0.801433236574746,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000494312814006524,
      "loss": 5.2021,
      "mean_token_accuracy": 0.1958215445280075,
      "num_tokens": 18340670.0,
      "step": 8835
    },
    {
      "entropy": 5.4468505859375,
      "epoch": 0.8018867924528302,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004943055828577784,
      "loss": 5.1937,
      "mean_token_accuracy": 0.1845944657921791,
      "num_tokens": 18352484.0,
      "step": 8840
    },
    {
      "entropy": 5.418262958526611,
      "epoch": 0.8023403483309144,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942983471737376,
      "loss": 5.1532,
      "mean_token_accuracy": 0.19169240295886994,
      "num_tokens": 18363103.0,
      "step": 8845
    },
    {
      "entropy": 5.37873272895813,
      "epoch": 0.8027939042089985,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942911069545511,
      "loss": 5.2267,
      "mean_token_accuracy": 0.1946313798427582,
      "num_tokens": 18373925.0,
      "step": 8850
    },
    {
      "entropy": 5.49118709564209,
      "epoch": 0.8032474600870827,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942838622003686,
      "loss": 5.2814,
      "mean_token_accuracy": 0.18206191956996917,
      "num_tokens": 18384081.0,
      "step": 8855
    },
    {
      "entropy": 5.541353750228882,
      "epoch": 0.8037010159651669,
      "grad_norm": 1.0,
      "learning_rate": 0.0004942766129113399,
      "loss": 5.1539,
      "mean_token_accuracy": 0.19194665402173997,
      "num_tokens": 18392510.0,
      "step": 8860
    },
    {
      "entropy": 5.532504987716675,
      "epoch": 0.8041545718432511,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942693590876151,
      "loss": 5.2968,
      "mean_token_accuracy": 0.18387772738933564,
      "num_tokens": 18402809.0,
      "step": 8865
    },
    {
      "entropy": 5.542223501205444,
      "epoch": 0.8046081277213353,
      "grad_norm": 1.09375,
      "learning_rate": 0.000494262100729344,
      "loss": 5.3171,
      "mean_token_accuracy": 0.18584613353013993,
      "num_tokens": 18413342.0,
      "step": 8870
    },
    {
      "entropy": 5.513447427749634,
      "epoch": 0.8050616835994194,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004942548378366768,
      "loss": 5.3013,
      "mean_token_accuracy": 0.18480869382619858,
      "num_tokens": 18424735.0,
      "step": 8875
    },
    {
      "entropy": 5.482922840118408,
      "epoch": 0.8055152394775036,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004942475704097635,
      "loss": 5.2014,
      "mean_token_accuracy": 0.19012280106544494,
      "num_tokens": 18434797.0,
      "step": 8880
    },
    {
      "entropy": 5.44505763053894,
      "epoch": 0.8059687953555879,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004942402984487548,
      "loss": 5.2439,
      "mean_token_accuracy": 0.18273085951805115,
      "num_tokens": 18445456.0,
      "step": 8885
    },
    {
      "entropy": 5.477434682846069,
      "epoch": 0.806422351233672,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942330219538008,
      "loss": 5.2316,
      "mean_token_accuracy": 0.1858235090970993,
      "num_tokens": 18456728.0,
      "step": 8890
    },
    {
      "entropy": 5.699557685852051,
      "epoch": 0.8068759071117562,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000494225740925052,
      "loss": 5.3628,
      "mean_token_accuracy": 0.18159916400909423,
      "num_tokens": 18468813.0,
      "step": 8895
    },
    {
      "entropy": 5.512146091461181,
      "epoch": 0.8073294629898403,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004942184553626591,
      "loss": 5.2507,
      "mean_token_accuracy": 0.18834080696105956,
      "num_tokens": 18477912.0,
      "step": 8900
    },
    {
      "entropy": 5.40484938621521,
      "epoch": 0.8077830188679245,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942111652667726,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1940469965338707,
      "num_tokens": 18488115.0,
      "step": 8905
    },
    {
      "entropy": 5.466488409042358,
      "epoch": 0.8082365747460087,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004942038706375433,
      "loss": 5.2326,
      "mean_token_accuracy": 0.18689305633306502,
      "num_tokens": 18498853.0,
      "step": 8910
    },
    {
      "entropy": 5.460456275939942,
      "epoch": 0.8086901306240929,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004941965714751222,
      "loss": 5.2812,
      "mean_token_accuracy": 0.18107305467128754,
      "num_tokens": 18510047.0,
      "step": 8915
    },
    {
      "entropy": 5.472559976577759,
      "epoch": 0.809143686502177,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941892677796601,
      "loss": 5.258,
      "mean_token_accuracy": 0.19089266210794448,
      "num_tokens": 18519673.0,
      "step": 8920
    },
    {
      "entropy": 5.475661754608154,
      "epoch": 0.8095972423802612,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004941819595513081,
      "loss": 5.2702,
      "mean_token_accuracy": 0.18570117205381392,
      "num_tokens": 18530090.0,
      "step": 8925
    },
    {
      "entropy": 5.567102146148682,
      "epoch": 0.8100507982583455,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004941746467902173,
      "loss": 5.3165,
      "mean_token_accuracy": 0.17997350692749023,
      "num_tokens": 18541862.0,
      "step": 8930
    },
    {
      "entropy": 5.540829753875732,
      "epoch": 0.8105043541364296,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000494167329496539,
      "loss": 5.3158,
      "mean_token_accuracy": 0.17742486000061036,
      "num_tokens": 18551624.0,
      "step": 8935
    },
    {
      "entropy": 5.449011516571045,
      "epoch": 0.8109579100145138,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004941600076704244,
      "loss": 5.2422,
      "mean_token_accuracy": 0.18720260560512542,
      "num_tokens": 18561784.0,
      "step": 8940
    },
    {
      "entropy": 5.501886558532715,
      "epoch": 0.8114114658925979,
      "grad_norm": 1.0,
      "learning_rate": 0.0004941526813120251,
      "loss": 5.3173,
      "mean_token_accuracy": 0.17972312569618226,
      "num_tokens": 18572110.0,
      "step": 8945
    },
    {
      "entropy": 5.539577198028565,
      "epoch": 0.8118650217706821,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004941453504214924,
      "loss": 5.2618,
      "mean_token_accuracy": 0.18143393248319625,
      "num_tokens": 18582012.0,
      "step": 8950
    },
    {
      "entropy": 5.399498987197876,
      "epoch": 0.8123185776487664,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004941380149989781,
      "loss": 5.0988,
      "mean_token_accuracy": 0.198584945499897,
      "num_tokens": 18592850.0,
      "step": 8955
    },
    {
      "entropy": 5.40629301071167,
      "epoch": 0.8127721335268505,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004941306750446336,
      "loss": 5.2769,
      "mean_token_accuracy": 0.18415168970823287,
      "num_tokens": 18602590.0,
      "step": 8960
    },
    {
      "entropy": 5.431480407714844,
      "epoch": 0.8132256894049347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000494123330558611,
      "loss": 5.1835,
      "mean_token_accuracy": 0.19291519969701768,
      "num_tokens": 18612607.0,
      "step": 8965
    },
    {
      "entropy": 5.474500274658203,
      "epoch": 0.8136792452830188,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004941159815410622,
      "loss": 5.166,
      "mean_token_accuracy": 0.19350909292697907,
      "num_tokens": 18622936.0,
      "step": 8970
    },
    {
      "entropy": 5.4926918029785154,
      "epoch": 0.8141328011611031,
      "grad_norm": 1.03125,
      "learning_rate": 0.000494108627992139,
      "loss": 5.3095,
      "mean_token_accuracy": 0.17892413288354875,
      "num_tokens": 18633997.0,
      "step": 8975
    },
    {
      "entropy": 5.414634323120117,
      "epoch": 0.8145863570391872,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004941012699119936,
      "loss": 5.1478,
      "mean_token_accuracy": 0.1955949380993843,
      "num_tokens": 18644347.0,
      "step": 8980
    },
    {
      "entropy": 5.4861256122589115,
      "epoch": 0.8150399129172714,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940939073007779,
      "loss": 5.2313,
      "mean_token_accuracy": 0.1909121960401535,
      "num_tokens": 18654989.0,
      "step": 8985
    },
    {
      "entropy": 5.464990901947021,
      "epoch": 0.8154934687953556,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940865401586446,
      "loss": 5.2109,
      "mean_token_accuracy": 0.19030434787273406,
      "num_tokens": 18665526.0,
      "step": 8990
    },
    {
      "entropy": 5.438522624969482,
      "epoch": 0.8159470246734397,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004940791684857458,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19021283239126205,
      "num_tokens": 18676226.0,
      "step": 8995
    },
    {
      "entropy": 5.461506366729736,
      "epoch": 0.816400580551524,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004940717922822339,
      "loss": 5.1254,
      "mean_token_accuracy": 0.1934579148888588,
      "num_tokens": 18686499.0,
      "step": 9000
    },
    {
      "epoch": 0.816400580551524,
      "eval_entropy": 5.295148694784933,
      "eval_loss": 5.239758491516113,
      "eval_mean_token_accuracy": 0.1953426175653867,
      "eval_num_tokens": 18686499.0,
      "eval_runtime": 20.5731,
      "eval_samples_per_second": 1718.747,
      "eval_steps_per_second": 214.843,
      "step": 9000
    },
    {
      "entropy": 5.512953996658325,
      "epoch": 0.8168541364296081,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004940644115482615,
      "loss": 5.2113,
      "mean_token_accuracy": 0.18784044086933135,
      "num_tokens": 18696543.0,
      "step": 9005
    },
    {
      "entropy": 5.411309862136841,
      "epoch": 0.8173076923076923,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940570262839812,
      "loss": 5.2609,
      "mean_token_accuracy": 0.18269165456295014,
      "num_tokens": 18707029.0,
      "step": 9010
    },
    {
      "entropy": 5.458065605163574,
      "epoch": 0.8177612481857764,
      "grad_norm": 1.0,
      "learning_rate": 0.0004940496364895457,
      "loss": 5.2382,
      "mean_token_accuracy": 0.1901217520236969,
      "num_tokens": 18716937.0,
      "step": 9015
    },
    {
      "entropy": 5.589706897735596,
      "epoch": 0.8182148040638607,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004940422421651082,
      "loss": 5.2344,
      "mean_token_accuracy": 0.18793250024318695,
      "num_tokens": 18727478.0,
      "step": 9020
    },
    {
      "entropy": 5.449050188064575,
      "epoch": 0.8186683599419449,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004940348433108211,
      "loss": 5.1626,
      "mean_token_accuracy": 0.18938371986150743,
      "num_tokens": 18737339.0,
      "step": 9025
    },
    {
      "entropy": 5.455372714996338,
      "epoch": 0.819121915820029,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004940274399268377,
      "loss": 5.278,
      "mean_token_accuracy": 0.1877202123403549,
      "num_tokens": 18747810.0,
      "step": 9030
    },
    {
      "entropy": 5.471349382400513,
      "epoch": 0.8195754716981132,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940200320133109,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1876297190785408,
      "num_tokens": 18757564.0,
      "step": 9035
    },
    {
      "entropy": 5.529532814025879,
      "epoch": 0.8200290275761973,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004940126195703941,
      "loss": 5.2642,
      "mean_token_accuracy": 0.18516496270895005,
      "num_tokens": 18768196.0,
      "step": 9040
    },
    {
      "entropy": 5.503971004486084,
      "epoch": 0.8204825834542816,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004940052025982405,
      "loss": 5.2378,
      "mean_token_accuracy": 0.19783169627189637,
      "num_tokens": 18778350.0,
      "step": 9045
    },
    {
      "entropy": 5.457395792007446,
      "epoch": 0.8209361393323658,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004939977810970035,
      "loss": 5.1951,
      "mean_token_accuracy": 0.18221480399370193,
      "num_tokens": 18789825.0,
      "step": 9050
    },
    {
      "entropy": 5.469322443008423,
      "epoch": 0.8213896952104499,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939903550668366,
      "loss": 5.1901,
      "mean_token_accuracy": 0.189435413479805,
      "num_tokens": 18799793.0,
      "step": 9055
    },
    {
      "entropy": 5.452561235427856,
      "epoch": 0.8218432510885341,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004939829245078934,
      "loss": 5.236,
      "mean_token_accuracy": 0.19126893430948258,
      "num_tokens": 18809771.0,
      "step": 9060
    },
    {
      "entropy": 5.465332746505737,
      "epoch": 0.8222968069666183,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004939754894203275,
      "loss": 5.2524,
      "mean_token_accuracy": 0.1888252705335617,
      "num_tokens": 18820529.0,
      "step": 9065
    },
    {
      "entropy": 5.393265581130981,
      "epoch": 0.8227503628447025,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004939680498042926,
      "loss": 5.1626,
      "mean_token_accuracy": 0.20150044560432434,
      "num_tokens": 18830797.0,
      "step": 9070
    },
    {
      "entropy": 5.421929931640625,
      "epoch": 0.8232039187227866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004939606056599428,
      "loss": 5.1794,
      "mean_token_accuracy": 0.19247416406869888,
      "num_tokens": 18840526.0,
      "step": 9075
    },
    {
      "entropy": 5.450492906570434,
      "epoch": 0.8236574746008708,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004939531569874316,
      "loss": 5.2073,
      "mean_token_accuracy": 0.18442903757095336,
      "num_tokens": 18851201.0,
      "step": 9080
    },
    {
      "entropy": 5.366197156906128,
      "epoch": 0.824111030478955,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939457037869135,
      "loss": 5.0236,
      "mean_token_accuracy": 0.21173077523708345,
      "num_tokens": 18861676.0,
      "step": 9085
    },
    {
      "entropy": 5.387629175186158,
      "epoch": 0.8245645863570392,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939382460585424,
      "loss": 5.1667,
      "mean_token_accuracy": 0.1845756560564041,
      "num_tokens": 18872669.0,
      "step": 9090
    },
    {
      "entropy": 5.379079055786133,
      "epoch": 0.8250181422351234,
      "grad_norm": 1.0,
      "learning_rate": 0.0004939307838024726,
      "loss": 5.2288,
      "mean_token_accuracy": 0.1850636750459671,
      "num_tokens": 18882130.0,
      "step": 9095
    },
    {
      "entropy": 5.423450231552124,
      "epoch": 0.8254716981132075,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939233170188585,
      "loss": 5.1872,
      "mean_token_accuracy": 0.19012628942728044,
      "num_tokens": 18892457.0,
      "step": 9100
    },
    {
      "entropy": 5.511678886413574,
      "epoch": 0.8259252539912917,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004939158457078543,
      "loss": 5.2185,
      "mean_token_accuracy": 0.18623092919588088,
      "num_tokens": 18903152.0,
      "step": 9105
    },
    {
      "entropy": 5.443718433380127,
      "epoch": 0.826378809869376,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004939083698696146,
      "loss": 5.1775,
      "mean_token_accuracy": 0.18756801933050155,
      "num_tokens": 18913740.0,
      "step": 9110
    },
    {
      "entropy": 5.431342744827271,
      "epoch": 0.8268323657474601,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493900889504294,
      "loss": 5.2354,
      "mean_token_accuracy": 0.18877937346696855,
      "num_tokens": 18923501.0,
      "step": 9115
    },
    {
      "entropy": 5.4776379585266115,
      "epoch": 0.8272859216255443,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004938934046120473,
      "loss": 5.223,
      "mean_token_accuracy": 0.1891995906829834,
      "num_tokens": 18933255.0,
      "step": 9120
    },
    {
      "entropy": 5.438175106048584,
      "epoch": 0.8277394775036284,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938859151930293,
      "loss": 5.1909,
      "mean_token_accuracy": 0.18860452771186828,
      "num_tokens": 18943649.0,
      "step": 9125
    },
    {
      "entropy": 5.4302948951721195,
      "epoch": 0.8281930333817126,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004938784212473947,
      "loss": 5.093,
      "mean_token_accuracy": 0.198411026597023,
      "num_tokens": 18953187.0,
      "step": 9130
    },
    {
      "entropy": 5.519911003112793,
      "epoch": 0.8286465892597968,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004938709227752985,
      "loss": 5.1945,
      "mean_token_accuracy": 0.19375702738761902,
      "num_tokens": 18963015.0,
      "step": 9135
    },
    {
      "entropy": 5.448123931884766,
      "epoch": 0.829100145137881,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000493863419776896,
      "loss": 5.2526,
      "mean_token_accuracy": 0.18133300095796584,
      "num_tokens": 18974184.0,
      "step": 9140
    },
    {
      "entropy": 5.43465895652771,
      "epoch": 0.8295537010159652,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004938559122523421,
      "loss": 5.2019,
      "mean_token_accuracy": 0.18903249651193618,
      "num_tokens": 18985036.0,
      "step": 9145
    },
    {
      "entropy": 5.530106878280639,
      "epoch": 0.8300072568940493,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004938484002017921,
      "loss": 5.3025,
      "mean_token_accuracy": 0.18530908972024918,
      "num_tokens": 18995237.0,
      "step": 9150
    },
    {
      "entropy": 5.515827941894531,
      "epoch": 0.8304608127721336,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004938408836254015,
      "loss": 5.3055,
      "mean_token_accuracy": 0.17978544384241105,
      "num_tokens": 19006272.0,
      "step": 9155
    },
    {
      "entropy": 5.436789894104004,
      "epoch": 0.8309143686502177,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004938333625233258,
      "loss": 5.2131,
      "mean_token_accuracy": 0.19587559551000594,
      "num_tokens": 19016728.0,
      "step": 9160
    },
    {
      "entropy": 5.479961681365967,
      "epoch": 0.8313679245283019,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004938258368957202,
      "loss": 5.2667,
      "mean_token_accuracy": 0.1806613489985466,
      "num_tokens": 19028024.0,
      "step": 9165
    },
    {
      "entropy": 5.475996828079223,
      "epoch": 0.831821480406386,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004938183067427406,
      "loss": 5.206,
      "mean_token_accuracy": 0.19370831400156022,
      "num_tokens": 19038218.0,
      "step": 9170
    },
    {
      "entropy": 5.427654123306274,
      "epoch": 0.8322750362844702,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004938107720645426,
      "loss": 5.246,
      "mean_token_accuracy": 0.18549474328756332,
      "num_tokens": 19048574.0,
      "step": 9175
    },
    {
      "entropy": 5.439662885665894,
      "epoch": 0.8327285921625545,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004938032328612821,
      "loss": 5.2156,
      "mean_token_accuracy": 0.187961608171463,
      "num_tokens": 19057721.0,
      "step": 9180
    },
    {
      "entropy": 5.447835779190063,
      "epoch": 0.8331821480406386,
      "grad_norm": 0.96875,
      "learning_rate": 0.000493795689133115,
      "loss": 5.1821,
      "mean_token_accuracy": 0.18198906183242797,
      "num_tokens": 19067582.0,
      "step": 9185
    },
    {
      "entropy": 5.447179794311523,
      "epoch": 0.8336357039187228,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004937881408801974,
      "loss": 5.1484,
      "mean_token_accuracy": 0.19615552127361296,
      "num_tokens": 19077896.0,
      "step": 9190
    },
    {
      "entropy": 5.443184280395508,
      "epoch": 0.8340892597968069,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000493780588102685,
      "loss": 5.2339,
      "mean_token_accuracy": 0.19482969790697097,
      "num_tokens": 19088082.0,
      "step": 9195
    },
    {
      "entropy": 5.4979020118713375,
      "epoch": 0.8345428156748912,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937730308007346,
      "loss": 5.2671,
      "mean_token_accuracy": 0.18952690958976745,
      "num_tokens": 19097176.0,
      "step": 9200
    },
    {
      "entropy": 5.460204553604126,
      "epoch": 0.8349963715529753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004937654689745021,
      "loss": 5.1457,
      "mean_token_accuracy": 0.19767686128616332,
      "num_tokens": 19107835.0,
      "step": 9205
    },
    {
      "entropy": 5.40765962600708,
      "epoch": 0.8354499274310595,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004937579026241439,
      "loss": 5.1612,
      "mean_token_accuracy": 0.19384574443101882,
      "num_tokens": 19118028.0,
      "step": 9210
    },
    {
      "entropy": 5.465951061248779,
      "epoch": 0.8359034833091437,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937503317498165,
      "loss": 5.2119,
      "mean_token_accuracy": 0.19141697138547897,
      "num_tokens": 19129223.0,
      "step": 9215
    },
    {
      "entropy": 5.4654120922088625,
      "epoch": 0.8363570391872278,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004937427563516765,
      "loss": 5.3005,
      "mean_token_accuracy": 0.17993318885564805,
      "num_tokens": 19140048.0,
      "step": 9220
    },
    {
      "entropy": 5.52028489112854,
      "epoch": 0.8368105950653121,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937351764298805,
      "loss": 5.279,
      "mean_token_accuracy": 0.17879910469055177,
      "num_tokens": 19150853.0,
      "step": 9225
    },
    {
      "entropy": 5.499497604370117,
      "epoch": 0.8372641509433962,
      "grad_norm": 1.0,
      "learning_rate": 0.0004937275919845855,
      "loss": 5.2588,
      "mean_token_accuracy": 0.18826455175876616,
      "num_tokens": 19161645.0,
      "step": 9230
    },
    {
      "entropy": 5.476880359649658,
      "epoch": 0.8377177068214804,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004937200030159481,
      "loss": 5.3458,
      "mean_token_accuracy": 0.17791823595762252,
      "num_tokens": 19173276.0,
      "step": 9235
    },
    {
      "entropy": 5.52899203300476,
      "epoch": 0.8381712626995645,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004937124095241252,
      "loss": 5.2262,
      "mean_token_accuracy": 0.19551825225353242,
      "num_tokens": 19182949.0,
      "step": 9240
    },
    {
      "entropy": 5.539991760253907,
      "epoch": 0.8386248185776488,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004937048115092739,
      "loss": 5.2211,
      "mean_token_accuracy": 0.18341253250837325,
      "num_tokens": 19193504.0,
      "step": 9245
    },
    {
      "entropy": 5.45371150970459,
      "epoch": 0.839078374455733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004936972089715515,
      "loss": 5.2937,
      "mean_token_accuracy": 0.18242430686950684,
      "num_tokens": 19203406.0,
      "step": 9250
    },
    {
      "entropy": 5.4398195266723635,
      "epoch": 0.8395319303338171,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493689601911115,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18913086950778962,
      "num_tokens": 19214325.0,
      "step": 9255
    },
    {
      "entropy": 5.470502233505249,
      "epoch": 0.8399854862119013,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004936819903281219,
      "loss": 5.2876,
      "mean_token_accuracy": 0.17488500624895095,
      "num_tokens": 19226466.0,
      "step": 9260
    },
    {
      "entropy": 5.565874528884888,
      "epoch": 0.8404390420899854,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004936743742227295,
      "loss": 5.3116,
      "mean_token_accuracy": 0.18711801916360854,
      "num_tokens": 19237032.0,
      "step": 9265
    },
    {
      "entropy": 5.536182641983032,
      "epoch": 0.8408925979680697,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004936667535950952,
      "loss": 5.2901,
      "mean_token_accuracy": 0.1812775194644928,
      "num_tokens": 19248009.0,
      "step": 9270
    },
    {
      "entropy": 5.523895597457885,
      "epoch": 0.8413461538461539,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004936591284453768,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1860367998480797,
      "num_tokens": 19259363.0,
      "step": 9275
    },
    {
      "entropy": 5.519369649887085,
      "epoch": 0.841799709724238,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004936514987737319,
      "loss": 5.2392,
      "mean_token_accuracy": 0.1884769007563591,
      "num_tokens": 19269191.0,
      "step": 9280
    },
    {
      "entropy": 5.515477180480957,
      "epoch": 0.8422532656023222,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004936438645803182,
      "loss": 5.2103,
      "mean_token_accuracy": 0.18393854945898055,
      "num_tokens": 19280278.0,
      "step": 9285
    },
    {
      "entropy": 5.453181600570678,
      "epoch": 0.8427068214804064,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004936362258652938,
      "loss": 5.2026,
      "mean_token_accuracy": 0.19249465465545654,
      "num_tokens": 19290130.0,
      "step": 9290
    },
    {
      "entropy": 5.559951686859131,
      "epoch": 0.8431603773584906,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004936285826288165,
      "loss": 5.3212,
      "mean_token_accuracy": 0.19275434464216232,
      "num_tokens": 19301669.0,
      "step": 9295
    },
    {
      "entropy": 5.448169231414795,
      "epoch": 0.8436139332365747,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004936209348710444,
      "loss": 5.1782,
      "mean_token_accuracy": 0.19250596314668655,
      "num_tokens": 19312490.0,
      "step": 9300
    },
    {
      "entropy": 5.4438097953796385,
      "epoch": 0.8440674891146589,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004936132825921355,
      "loss": 5.2086,
      "mean_token_accuracy": 0.18785629719495772,
      "num_tokens": 19321743.0,
      "step": 9305
    },
    {
      "entropy": 5.431347608566284,
      "epoch": 0.844521044992743,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004936056257922484,
      "loss": 5.1323,
      "mean_token_accuracy": 0.1909039482474327,
      "num_tokens": 19332149.0,
      "step": 9310
    },
    {
      "entropy": 5.466032552719116,
      "epoch": 0.8449746008708273,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004935979644715412,
      "loss": 5.2232,
      "mean_token_accuracy": 0.1855064868927002,
      "num_tokens": 19341655.0,
      "step": 9315
    },
    {
      "entropy": 5.519804620742798,
      "epoch": 0.8454281567489115,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004935902986301724,
      "loss": 5.2669,
      "mean_token_accuracy": 0.18666884005069734,
      "num_tokens": 19351404.0,
      "step": 9320
    },
    {
      "entropy": 5.51761212348938,
      "epoch": 0.8458817126269956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004935826282683005,
      "loss": 5.306,
      "mean_token_accuracy": 0.18108808994293213,
      "num_tokens": 19361788.0,
      "step": 9325
    },
    {
      "entropy": 5.4535401344299315,
      "epoch": 0.8463352685050798,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004935749533860842,
      "loss": 5.2601,
      "mean_token_accuracy": 0.1881358101963997,
      "num_tokens": 19372902.0,
      "step": 9330
    },
    {
      "entropy": 5.440549039840699,
      "epoch": 0.846788824383164,
      "grad_norm": 0.921875,
      "learning_rate": 0.000493567273983682,
      "loss": 5.1421,
      "mean_token_accuracy": 0.18869245946407318,
      "num_tokens": 19382181.0,
      "step": 9335
    },
    {
      "entropy": 5.489796781539917,
      "epoch": 0.8472423802612482,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000493559590061253,
      "loss": 5.2085,
      "mean_token_accuracy": 0.18647148460149765,
      "num_tokens": 19393692.0,
      "step": 9340
    },
    {
      "entropy": 5.487021207809448,
      "epoch": 0.8476959361393324,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004935519016189559,
      "loss": 5.2518,
      "mean_token_accuracy": 0.19533468037843704,
      "num_tokens": 19404587.0,
      "step": 9345
    },
    {
      "entropy": 5.476869916915893,
      "epoch": 0.8481494920174165,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004935442086569498,
      "loss": 5.2227,
      "mean_token_accuracy": 0.1869882181286812,
      "num_tokens": 19415296.0,
      "step": 9350
    },
    {
      "entropy": 5.532775974273681,
      "epoch": 0.8486030478955007,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935365111753939,
      "loss": 5.2712,
      "mean_token_accuracy": 0.18344380110502242,
      "num_tokens": 19426119.0,
      "step": 9355
    },
    {
      "entropy": 5.515724086761475,
      "epoch": 0.8490566037735849,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004935288091744472,
      "loss": 5.2665,
      "mean_token_accuracy": 0.18467478454113007,
      "num_tokens": 19436828.0,
      "step": 9360
    },
    {
      "entropy": 5.534423828125,
      "epoch": 0.8495101596516691,
      "grad_norm": 1.0,
      "learning_rate": 0.000493521102654269,
      "loss": 5.2324,
      "mean_token_accuracy": 0.188790924847126,
      "num_tokens": 19446583.0,
      "step": 9365
    },
    {
      "entropy": 5.423384094238282,
      "epoch": 0.8499637155297532,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004935133916150187,
      "loss": 5.1168,
      "mean_token_accuracy": 0.19754848927259444,
      "num_tokens": 19456757.0,
      "step": 9370
    },
    {
      "entropy": 5.427696084976196,
      "epoch": 0.8504172714078374,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004935056760568558,
      "loss": 5.2023,
      "mean_token_accuracy": 0.18450350314378738,
      "num_tokens": 19467624.0,
      "step": 9375
    },
    {
      "entropy": 5.441474723815918,
      "epoch": 0.8508708272859217,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004934979559799399,
      "loss": 5.157,
      "mean_token_accuracy": 0.18935077488422394,
      "num_tokens": 19477759.0,
      "step": 9380
    },
    {
      "entropy": 5.460384941101074,
      "epoch": 0.8513243831640058,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004934902313844306,
      "loss": 5.1669,
      "mean_token_accuracy": 0.1911305770277977,
      "num_tokens": 19488877.0,
      "step": 9385
    },
    {
      "entropy": 5.406986904144287,
      "epoch": 0.85177793904209,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004934825022704875,
      "loss": 5.1616,
      "mean_token_accuracy": 0.1952897772192955,
      "num_tokens": 19499170.0,
      "step": 9390
    },
    {
      "entropy": 5.454417419433594,
      "epoch": 0.8522314949201741,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004934747686382707,
      "loss": 5.2065,
      "mean_token_accuracy": 0.1808039888739586,
      "num_tokens": 19509944.0,
      "step": 9395
    },
    {
      "entropy": 5.47445969581604,
      "epoch": 0.8526850507982583,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004934670304879398,
      "loss": 5.1694,
      "mean_token_accuracy": 0.19298242628574372,
      "num_tokens": 19519014.0,
      "step": 9400
    },
    {
      "entropy": 5.4161628723144535,
      "epoch": 0.8531386066763426,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004934592878196553,
      "loss": 5.1766,
      "mean_token_accuracy": 0.1934538498520851,
      "num_tokens": 19528586.0,
      "step": 9405
    },
    {
      "entropy": 5.484393167495727,
      "epoch": 0.8535921625544267,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000493451540633577,
      "loss": 5.3113,
      "mean_token_accuracy": 0.18901212215423585,
      "num_tokens": 19539912.0,
      "step": 9410
    },
    {
      "entropy": 5.476560163497925,
      "epoch": 0.8540457184325109,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004934437889298651,
      "loss": 5.2646,
      "mean_token_accuracy": 0.17929179668426515,
      "num_tokens": 19549860.0,
      "step": 9415
    },
    {
      "entropy": 5.498154592514038,
      "epoch": 0.854499274310595,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004934360327086799,
      "loss": 5.2267,
      "mean_token_accuracy": 0.184198497235775,
      "num_tokens": 19559446.0,
      "step": 9420
    },
    {
      "entropy": 5.525316429138184,
      "epoch": 0.8549528301886793,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493428271970182,
      "loss": 5.2297,
      "mean_token_accuracy": 0.18811480402946473,
      "num_tokens": 19570282.0,
      "step": 9425
    },
    {
      "entropy": 5.475820112228393,
      "epoch": 0.8554063860667634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004934205067145318,
      "loss": 5.1505,
      "mean_token_accuracy": 0.18823657035827637,
      "num_tokens": 19581087.0,
      "step": 9430
    },
    {
      "entropy": 5.425998735427856,
      "epoch": 0.8558599419448476,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004934127369418898,
      "loss": 5.2456,
      "mean_token_accuracy": 0.18897894769906998,
      "num_tokens": 19592100.0,
      "step": 9435
    },
    {
      "entropy": 5.4478099822998045,
      "epoch": 0.8563134978229318,
      "grad_norm": 1.0,
      "learning_rate": 0.0004934049626524168,
      "loss": 5.233,
      "mean_token_accuracy": 0.18636782914400102,
      "num_tokens": 19602557.0,
      "step": 9440
    },
    {
      "entropy": 5.522075128555298,
      "epoch": 0.8567670537010159,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933971838462734,
      "loss": 5.2775,
      "mean_token_accuracy": 0.18067894577980043,
      "num_tokens": 19613643.0,
      "step": 9445
    },
    {
      "entropy": 5.467205333709717,
      "epoch": 0.8572206095791002,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004933894005236207,
      "loss": 5.1977,
      "mean_token_accuracy": 0.18942511677742005,
      "num_tokens": 19624180.0,
      "step": 9450
    },
    {
      "entropy": 5.476292467117309,
      "epoch": 0.8576741654571843,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004933816126846195,
      "loss": 5.2077,
      "mean_token_accuracy": 0.189624784886837,
      "num_tokens": 19634799.0,
      "step": 9455
    },
    {
      "entropy": 5.411914920806884,
      "epoch": 0.8581277213352685,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004933738203294309,
      "loss": 5.1713,
      "mean_token_accuracy": 0.200113882124424,
      "num_tokens": 19644618.0,
      "step": 9460
    },
    {
      "entropy": 5.411866044998169,
      "epoch": 0.8585812772133526,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000493366023458216,
      "loss": 5.1845,
      "mean_token_accuracy": 0.18326595425605774,
      "num_tokens": 19655550.0,
      "step": 9465
    },
    {
      "entropy": 5.4921434879302975,
      "epoch": 0.8590348330914369,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004933582220711362,
      "loss": 5.2532,
      "mean_token_accuracy": 0.19442976713180543,
      "num_tokens": 19665734.0,
      "step": 9470
    },
    {
      "entropy": 5.509519481658936,
      "epoch": 0.8594883889695211,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004933504161683526,
      "loss": 5.2118,
      "mean_token_accuracy": 0.18399560004472731,
      "num_tokens": 19677130.0,
      "step": 9475
    },
    {
      "entropy": 5.414972305297852,
      "epoch": 0.8599419448476052,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004933426057500269,
      "loss": 5.1227,
      "mean_token_accuracy": 0.19844300150871277,
      "num_tokens": 19687374.0,
      "step": 9480
    },
    {
      "entropy": 5.498849821090698,
      "epoch": 0.8603955007256894,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004933347908163203,
      "loss": 5.2317,
      "mean_token_accuracy": 0.18636648654937743,
      "num_tokens": 19698182.0,
      "step": 9485
    },
    {
      "entropy": 5.457022476196289,
      "epoch": 0.8608490566037735,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933269713673947,
      "loss": 5.1897,
      "mean_token_accuracy": 0.18997897803783417,
      "num_tokens": 19707664.0,
      "step": 9490
    },
    {
      "entropy": 5.358859634399414,
      "epoch": 0.8613026124818578,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933191474034116,
      "loss": 5.2329,
      "mean_token_accuracy": 0.18852289021015167,
      "num_tokens": 19717647.0,
      "step": 9495
    },
    {
      "entropy": 5.440193605422974,
      "epoch": 0.861756168359942,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493311318924533,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18900999724864959,
      "num_tokens": 19727444.0,
      "step": 9500
    },
    {
      "entropy": 5.504938125610352,
      "epoch": 0.8622097242380261,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004933034859309208,
      "loss": 5.2762,
      "mean_token_accuracy": 0.17947544306516647,
      "num_tokens": 19738719.0,
      "step": 9505
    },
    {
      "entropy": 5.543647336959839,
      "epoch": 0.8626632801161103,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004932956484227367,
      "loss": 5.2594,
      "mean_token_accuracy": 0.1917922765016556,
      "num_tokens": 19749786.0,
      "step": 9510
    },
    {
      "entropy": 5.467603588104248,
      "epoch": 0.8631168359941945,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000493287806400143,
      "loss": 5.2005,
      "mean_token_accuracy": 0.19913371056318283,
      "num_tokens": 19759623.0,
      "step": 9515
    },
    {
      "entropy": 5.470474147796631,
      "epoch": 0.8635703918722787,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004932799598633018,
      "loss": 5.2177,
      "mean_token_accuracy": 0.18518185019493102,
      "num_tokens": 19770117.0,
      "step": 9520
    },
    {
      "entropy": 5.495015096664429,
      "epoch": 0.8640239477503628,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004932721088123753,
      "loss": 5.3056,
      "mean_token_accuracy": 0.1860126957297325,
      "num_tokens": 19780748.0,
      "step": 9525
    },
    {
      "entropy": 5.509607982635498,
      "epoch": 0.864477503628447,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004932642532475262,
      "loss": 5.2589,
      "mean_token_accuracy": 0.19054854065179824,
      "num_tokens": 19791816.0,
      "step": 9530
    },
    {
      "entropy": 5.537548017501831,
      "epoch": 0.8649310595065312,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004932563931689168,
      "loss": 5.3026,
      "mean_token_accuracy": 0.18602988123893738,
      "num_tokens": 19801096.0,
      "step": 9535
    },
    {
      "entropy": 5.464199876785278,
      "epoch": 0.8653846153846154,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004932485285767093,
      "loss": 5.1812,
      "mean_token_accuracy": 0.19209689795970916,
      "num_tokens": 19811258.0,
      "step": 9540
    },
    {
      "entropy": 5.499051570892334,
      "epoch": 0.8658381712626996,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004932406594710667,
      "loss": 5.2517,
      "mean_token_accuracy": 0.1938839554786682,
      "num_tokens": 19821765.0,
      "step": 9545
    },
    {
      "entropy": 5.470533132553101,
      "epoch": 0.8662917271407837,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004932327858521517,
      "loss": 5.114,
      "mean_token_accuracy": 0.19665188640356063,
      "num_tokens": 19831537.0,
      "step": 9550
    },
    {
      "entropy": 5.406108570098877,
      "epoch": 0.8667452830188679,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493224907720127,
      "loss": 5.1387,
      "mean_token_accuracy": 0.1899405762553215,
      "num_tokens": 19841830.0,
      "step": 9555
    },
    {
      "entropy": 5.479475736618042,
      "epoch": 0.8671988388969522,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004932170250751556,
      "loss": 5.2427,
      "mean_token_accuracy": 0.18139030337333678,
      "num_tokens": 19851896.0,
      "step": 9560
    },
    {
      "entropy": 5.500799608230591,
      "epoch": 0.8676523947750363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004932091379174006,
      "loss": 5.2659,
      "mean_token_accuracy": 0.1821811392903328,
      "num_tokens": 19861735.0,
      "step": 9565
    },
    {
      "entropy": 5.479330682754517,
      "epoch": 0.8681059506531205,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493201246247025,
      "loss": 5.1533,
      "mean_token_accuracy": 0.19605274200439454,
      "num_tokens": 19872333.0,
      "step": 9570
    },
    {
      "entropy": 5.469655227661133,
      "epoch": 0.8685595065312046,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000493193350064192,
      "loss": 5.189,
      "mean_token_accuracy": 0.18709672391414642,
      "num_tokens": 19882896.0,
      "step": 9575
    },
    {
      "entropy": 5.439773464202881,
      "epoch": 0.8690130624092888,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004931854493690649,
      "loss": 5.1423,
      "mean_token_accuracy": 0.19140439331531525,
      "num_tokens": 19893767.0,
      "step": 9580
    },
    {
      "entropy": 5.377742004394531,
      "epoch": 0.869466618287373,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004931775441618071,
      "loss": 5.1093,
      "mean_token_accuracy": 0.19865830689668657,
      "num_tokens": 19904242.0,
      "step": 9585
    },
    {
      "entropy": 5.438934183120727,
      "epoch": 0.8699201741654572,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004931696344425821,
      "loss": 5.25,
      "mean_token_accuracy": 0.19252071380615235,
      "num_tokens": 19913898.0,
      "step": 9590
    },
    {
      "entropy": 5.5521914005279545,
      "epoch": 0.8703737300435413,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004931617202115535,
      "loss": 5.2507,
      "mean_token_accuracy": 0.1872472956776619,
      "num_tokens": 19923756.0,
      "step": 9595
    },
    {
      "entropy": 5.4616528987884525,
      "epoch": 0.8708272859216255,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004931538014688849,
      "loss": 5.1822,
      "mean_token_accuracy": 0.1898686945438385,
      "num_tokens": 19932720.0,
      "step": 9600
    },
    {
      "entropy": 5.530621290206909,
      "epoch": 0.8712808417997098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004931458782147401,
      "loss": 5.3842,
      "mean_token_accuracy": 0.17831998318433762,
      "num_tokens": 19942962.0,
      "step": 9605
    },
    {
      "entropy": 5.552232170104981,
      "epoch": 0.8717343976777939,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000493137950449283,
      "loss": 5.2713,
      "mean_token_accuracy": 0.1898573726415634,
      "num_tokens": 19953735.0,
      "step": 9610
    },
    {
      "entropy": 5.449032592773437,
      "epoch": 0.8721879535558781,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004931300181726775,
      "loss": 5.2334,
      "mean_token_accuracy": 0.18857302963733674,
      "num_tokens": 19964416.0,
      "step": 9615
    },
    {
      "entropy": 5.468363428115845,
      "epoch": 0.8726415094339622,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004931220813850875,
      "loss": 5.2418,
      "mean_token_accuracy": 0.18374694138765335,
      "num_tokens": 19974531.0,
      "step": 9620
    },
    {
      "entropy": 5.505137968063354,
      "epoch": 0.8730950653120464,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004931141400866773,
      "loss": 5.2263,
      "mean_token_accuracy": 0.19138203859329223,
      "num_tokens": 19984964.0,
      "step": 9625
    },
    {
      "entropy": 5.444048309326172,
      "epoch": 0.8735486211901307,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004931061942776112,
      "loss": 5.1759,
      "mean_token_accuracy": 0.19382957071065904,
      "num_tokens": 19995568.0,
      "step": 9630
    },
    {
      "entropy": 5.571946382522583,
      "epoch": 0.8740021770682148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930982439580534,
      "loss": 5.3621,
      "mean_token_accuracy": 0.17532415837049484,
      "num_tokens": 20005856.0,
      "step": 9635
    },
    {
      "entropy": 5.51176905632019,
      "epoch": 0.874455732946299,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004930902891281683,
      "loss": 5.2643,
      "mean_token_accuracy": 0.18537067621946335,
      "num_tokens": 20015683.0,
      "step": 9640
    },
    {
      "entropy": 5.3699976921081545,
      "epoch": 0.8749092888243831,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004930823297881205,
      "loss": 5.0767,
      "mean_token_accuracy": 0.1967955693602562,
      "num_tokens": 20025998.0,
      "step": 9645
    },
    {
      "entropy": 5.497171115875244,
      "epoch": 0.8753628447024674,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930743659380745,
      "loss": 5.2379,
      "mean_token_accuracy": 0.18628917187452315,
      "num_tokens": 20036486.0,
      "step": 9650
    },
    {
      "entropy": 5.4253710269927975,
      "epoch": 0.8758164005805515,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000493066397578195,
      "loss": 5.1297,
      "mean_token_accuracy": 0.19823774695396423,
      "num_tokens": 20046784.0,
      "step": 9655
    },
    {
      "entropy": 5.450145435333252,
      "epoch": 0.8762699564586357,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004930584247086468,
      "loss": 5.2119,
      "mean_token_accuracy": 0.1873118743300438,
      "num_tokens": 20056999.0,
      "step": 9660
    },
    {
      "entropy": 5.436940431594849,
      "epoch": 0.8767235123367199,
      "grad_norm": 0.96875,
      "learning_rate": 0.000493050447329595,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19763690680265428,
      "num_tokens": 20067612.0,
      "step": 9665
    },
    {
      "entropy": 5.494093751907348,
      "epoch": 0.877177068214804,
      "grad_norm": 1.0,
      "learning_rate": 0.0004930424654412043,
      "loss": 5.2578,
      "mean_token_accuracy": 0.19012722074985505,
      "num_tokens": 20077330.0,
      "step": 9670
    },
    {
      "entropy": 5.48297963142395,
      "epoch": 0.8776306240928883,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004930344790436398,
      "loss": 5.256,
      "mean_token_accuracy": 0.18199552744626998,
      "num_tokens": 20086801.0,
      "step": 9675
    },
    {
      "entropy": 5.499620962142944,
      "epoch": 0.8780841799709724,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930264881370666,
      "loss": 5.1915,
      "mean_token_accuracy": 0.18674761801958084,
      "num_tokens": 20097047.0,
      "step": 9680
    },
    {
      "entropy": 5.465418910980224,
      "epoch": 0.8785377358490566,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004930184927216504,
      "loss": 5.2343,
      "mean_token_accuracy": 0.191551773250103,
      "num_tokens": 20107185.0,
      "step": 9685
    },
    {
      "entropy": 5.489413642883301,
      "epoch": 0.8789912917271407,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004930104927975559,
      "loss": 5.1704,
      "mean_token_accuracy": 0.19116564840078354,
      "num_tokens": 20118493.0,
      "step": 9690
    },
    {
      "entropy": 5.467175674438477,
      "epoch": 0.879444847605225,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004930024883649489,
      "loss": 5.2275,
      "mean_token_accuracy": 0.18803865611553192,
      "num_tokens": 20129294.0,
      "step": 9695
    },
    {
      "entropy": 5.431062269210815,
      "epoch": 0.8798984034833092,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000492994479423995,
      "loss": 5.1918,
      "mean_token_accuracy": 0.19232005923986434,
      "num_tokens": 20139194.0,
      "step": 9700
    },
    {
      "entropy": 5.45842924118042,
      "epoch": 0.8803519593613933,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004929864659748597,
      "loss": 5.2081,
      "mean_token_accuracy": 0.18481118232011795,
      "num_tokens": 20149681.0,
      "step": 9705
    },
    {
      "entropy": 5.508456516265869,
      "epoch": 0.8808055152394775,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004929784480177087,
      "loss": 5.2977,
      "mean_token_accuracy": 0.1847716510295868,
      "num_tokens": 20160557.0,
      "step": 9710
    },
    {
      "entropy": 5.457950210571289,
      "epoch": 0.8812590711175616,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004929704255527078,
      "loss": 5.169,
      "mean_token_accuracy": 0.1884682521224022,
      "num_tokens": 20171047.0,
      "step": 9715
    },
    {
      "entropy": 5.476606464385986,
      "epoch": 0.8817126269956459,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492962398580023,
      "loss": 5.3521,
      "mean_token_accuracy": 0.1803833708167076,
      "num_tokens": 20182148.0,
      "step": 9720
    },
    {
      "entropy": 5.409409523010254,
      "epoch": 0.88216618287373,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004929543670998203,
      "loss": 5.1017,
      "mean_token_accuracy": 0.19479717761278154,
      "num_tokens": 20192131.0,
      "step": 9725
    },
    {
      "entropy": 5.4837233543396,
      "epoch": 0.8826197387518142,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004929463311122657,
      "loss": 5.219,
      "mean_token_accuracy": 0.1879323437809944,
      "num_tokens": 20202805.0,
      "step": 9730
    },
    {
      "entropy": 5.467478561401367,
      "epoch": 0.8830732946298984,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004929382906175253,
      "loss": 5.1584,
      "mean_token_accuracy": 0.193536177277565,
      "num_tokens": 20212702.0,
      "step": 9735
    },
    {
      "entropy": 5.4351178169250485,
      "epoch": 0.8835268505079826,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929302456157657,
      "loss": 5.1812,
      "mean_token_accuracy": 0.1951988160610199,
      "num_tokens": 20222661.0,
      "step": 9740
    },
    {
      "entropy": 5.474312114715576,
      "epoch": 0.8839804063860668,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004929221961071531,
      "loss": 5.2279,
      "mean_token_accuracy": 0.18725454956293106,
      "num_tokens": 20233012.0,
      "step": 9745
    },
    {
      "entropy": 5.498449611663818,
      "epoch": 0.8844339622641509,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929141420918538,
      "loss": 5.2431,
      "mean_token_accuracy": 0.19418645799160003,
      "num_tokens": 20242918.0,
      "step": 9750
    },
    {
      "entropy": 5.457444906234741,
      "epoch": 0.8848875181422351,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004929060835700347,
      "loss": 5.1631,
      "mean_token_accuracy": 0.19698089808225633,
      "num_tokens": 20253254.0,
      "step": 9755
    },
    {
      "entropy": 5.45939040184021,
      "epoch": 0.8853410740203193,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000492898020541862,
      "loss": 5.2168,
      "mean_token_accuracy": 0.18872032910585404,
      "num_tokens": 20263355.0,
      "step": 9760
    },
    {
      "entropy": 5.419777965545654,
      "epoch": 0.8857946298984035,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004928899530075029,
      "loss": 5.1452,
      "mean_token_accuracy": 0.19159601032733917,
      "num_tokens": 20274716.0,
      "step": 9765
    },
    {
      "entropy": 5.386063766479492,
      "epoch": 0.8862481857764877,
      "grad_norm": 1.0,
      "learning_rate": 0.0004928818809671241,
      "loss": 5.1268,
      "mean_token_accuracy": 0.19315028339624404,
      "num_tokens": 20285631.0,
      "step": 9770
    },
    {
      "entropy": 5.49028263092041,
      "epoch": 0.8867017416545718,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004928738044208923,
      "loss": 5.2228,
      "mean_token_accuracy": 0.19549182206392288,
      "num_tokens": 20295717.0,
      "step": 9775
    },
    {
      "entropy": 5.394902086257934,
      "epoch": 0.887155297532656,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004928657233689748,
      "loss": 5.1927,
      "mean_token_accuracy": 0.19078177809715272,
      "num_tokens": 20306637.0,
      "step": 9780
    },
    {
      "entropy": 5.46283483505249,
      "epoch": 0.8876088534107403,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928576378115387,
      "loss": 5.1645,
      "mean_token_accuracy": 0.18774652034044265,
      "num_tokens": 20316629.0,
      "step": 9785
    },
    {
      "entropy": 5.560376405715942,
      "epoch": 0.8880624092888244,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004928495477487509,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1772662043571472,
      "num_tokens": 20327935.0,
      "step": 9790
    },
    {
      "entropy": 5.461290884017944,
      "epoch": 0.8885159651669086,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004928414531807792,
      "loss": 5.1561,
      "mean_token_accuracy": 0.18659840822219848,
      "num_tokens": 20339142.0,
      "step": 9795
    },
    {
      "entropy": 5.415150117874146,
      "epoch": 0.8889695210449927,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004928333541077906,
      "loss": 5.1661,
      "mean_token_accuracy": 0.20060190111398696,
      "num_tokens": 20348474.0,
      "step": 9800
    },
    {
      "entropy": 5.486493635177612,
      "epoch": 0.8894230769230769,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928252505299527,
      "loss": 5.2328,
      "mean_token_accuracy": 0.18882507979869842,
      "num_tokens": 20359440.0,
      "step": 9805
    },
    {
      "entropy": 5.55666823387146,
      "epoch": 0.8898766328011611,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004928171424474332,
      "loss": 5.2732,
      "mean_token_accuracy": 0.18582544177770616,
      "num_tokens": 20370355.0,
      "step": 9810
    },
    {
      "entropy": 5.410488796234131,
      "epoch": 0.8903301886792453,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004928090298603996,
      "loss": 5.1534,
      "mean_token_accuracy": 0.19132491052150727,
      "num_tokens": 20380214.0,
      "step": 9815
    },
    {
      "entropy": 5.48265962600708,
      "epoch": 0.8907837445573294,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004928009127690199,
      "loss": 5.1669,
      "mean_token_accuracy": 0.18830980807542802,
      "num_tokens": 20389190.0,
      "step": 9820
    },
    {
      "entropy": 5.463851690292358,
      "epoch": 0.8912373004354136,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004927927911734617,
      "loss": 5.2319,
      "mean_token_accuracy": 0.18881546407938005,
      "num_tokens": 20399531.0,
      "step": 9825
    },
    {
      "entropy": 5.42667145729065,
      "epoch": 0.8916908563134979,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004927846650738931,
      "loss": 5.2032,
      "mean_token_accuracy": 0.1890368089079857,
      "num_tokens": 20410201.0,
      "step": 9830
    },
    {
      "entropy": 5.436289834976196,
      "epoch": 0.892144412191582,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004927765344704821,
      "loss": 5.1757,
      "mean_token_accuracy": 0.1938535362482071,
      "num_tokens": 20421579.0,
      "step": 9835
    },
    {
      "entropy": 5.458463096618653,
      "epoch": 0.8925979680696662,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004927683993633969,
      "loss": 5.2256,
      "mean_token_accuracy": 0.1929639533162117,
      "num_tokens": 20432048.0,
      "step": 9840
    },
    {
      "entropy": 5.534075403213501,
      "epoch": 0.8930515239477503,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004927602597528057,
      "loss": 5.2518,
      "mean_token_accuracy": 0.18371923565864562,
      "num_tokens": 20441752.0,
      "step": 9845
    },
    {
      "entropy": 5.4015154361724855,
      "epoch": 0.8935050798258345,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004927521156388769,
      "loss": 5.1202,
      "mean_token_accuracy": 0.19143102169036866,
      "num_tokens": 20451693.0,
      "step": 9850
    },
    {
      "entropy": 5.4664863586425785,
      "epoch": 0.8939586357039188,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004927439670217788,
      "loss": 5.2549,
      "mean_token_accuracy": 0.18985022753477096,
      "num_tokens": 20462312.0,
      "step": 9855
    },
    {
      "entropy": 5.4695860862731935,
      "epoch": 0.8944121915820029,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049273581390168,
      "loss": 5.1755,
      "mean_token_accuracy": 0.18485859036445618,
      "num_tokens": 20472546.0,
      "step": 9860
    },
    {
      "entropy": 5.461061286926269,
      "epoch": 0.8948657474600871,
      "grad_norm": 0.875,
      "learning_rate": 0.000492727656278749,
      "loss": 5.1854,
      "mean_token_accuracy": 0.19223169088363648,
      "num_tokens": 20483266.0,
      "step": 9865
    },
    {
      "entropy": 5.432494115829468,
      "epoch": 0.8953193033381712,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004927194941531547,
      "loss": 5.1281,
      "mean_token_accuracy": 0.18978212475776673,
      "num_tokens": 20492490.0,
      "step": 9870
    },
    {
      "entropy": 5.45404806137085,
      "epoch": 0.8957728592162555,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004927113275250657,
      "loss": 5.1775,
      "mean_token_accuracy": 0.19612860679626465,
      "num_tokens": 20503077.0,
      "step": 9875
    },
    {
      "entropy": 5.425263690948486,
      "epoch": 0.8962264150943396,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004927031563946511,
      "loss": 5.2018,
      "mean_token_accuracy": 0.18478720486164094,
      "num_tokens": 20513771.0,
      "step": 9880
    },
    {
      "entropy": 5.487907648086548,
      "epoch": 0.8966799709724238,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004926949807620796,
      "loss": 5.2973,
      "mean_token_accuracy": 0.18408113718032837,
      "num_tokens": 20523616.0,
      "step": 9885
    },
    {
      "entropy": 5.4724946975708,
      "epoch": 0.897133526850508,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004926868006275204,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19698286801576614,
      "num_tokens": 20533613.0,
      "step": 9890
    },
    {
      "entropy": 5.463048982620239,
      "epoch": 0.8975870827285921,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004926786159911427,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1881907179951668,
      "num_tokens": 20543812.0,
      "step": 9895
    },
    {
      "entropy": 5.4218991756439205,
      "epoch": 0.8980406386066764,
      "grad_norm": 1.046875,
      "learning_rate": 0.000492670426853116,
      "loss": 5.1287,
      "mean_token_accuracy": 0.1955796554684639,
      "num_tokens": 20553785.0,
      "step": 9900
    },
    {
      "entropy": 5.5498435497283936,
      "epoch": 0.8984941944847605,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004926622332136092,
      "loss": 5.2431,
      "mean_token_accuracy": 0.190711410343647,
      "num_tokens": 20563326.0,
      "step": 9905
    },
    {
      "entropy": 5.443296480178833,
      "epoch": 0.8989477503628447,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004926540350727918,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19946378320455552,
      "num_tokens": 20573643.0,
      "step": 9910
    },
    {
      "entropy": 5.396259069442749,
      "epoch": 0.8994013062409288,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926458324308337,
      "loss": 5.1156,
      "mean_token_accuracy": 0.1931847482919693,
      "num_tokens": 20583916.0,
      "step": 9915
    },
    {
      "entropy": 5.465553617477417,
      "epoch": 0.8998548621190131,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004926376252879043,
      "loss": 5.2767,
      "mean_token_accuracy": 0.1940779507160187,
      "num_tokens": 20593984.0,
      "step": 9920
    },
    {
      "entropy": 5.466677331924439,
      "epoch": 0.9003084179970973,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004926294136441733,
      "loss": 5.1092,
      "mean_token_accuracy": 0.19847935438156128,
      "num_tokens": 20603927.0,
      "step": 9925
    },
    {
      "entropy": 5.486162567138672,
      "epoch": 0.9007619738751814,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926211974998106,
      "loss": 5.2377,
      "mean_token_accuracy": 0.19301336407661437,
      "num_tokens": 20614147.0,
      "step": 9930
    },
    {
      "entropy": 5.497865104675293,
      "epoch": 0.9012155297532656,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000492612976854986,
      "loss": 5.2085,
      "mean_token_accuracy": 0.19398757964372634,
      "num_tokens": 20623631.0,
      "step": 9935
    },
    {
      "entropy": 5.412996292114258,
      "epoch": 0.9016690856313497,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004926047517098697,
      "loss": 5.0911,
      "mean_token_accuracy": 0.19451865702867507,
      "num_tokens": 20632566.0,
      "step": 9940
    },
    {
      "entropy": 5.490465974807739,
      "epoch": 0.902122641509434,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004925965220646315,
      "loss": 5.2074,
      "mean_token_accuracy": 0.1901567503809929,
      "num_tokens": 20643246.0,
      "step": 9945
    },
    {
      "entropy": 5.451488637924195,
      "epoch": 0.9025761973875182,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004925882879194419,
      "loss": 5.2783,
      "mean_token_accuracy": 0.18662670105695725,
      "num_tokens": 20653372.0,
      "step": 9950
    },
    {
      "entropy": 5.446879196166992,
      "epoch": 0.9030297532656023,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004925800492744712,
      "loss": 5.2239,
      "mean_token_accuracy": 0.18523058593273162,
      "num_tokens": 20662687.0,
      "step": 9955
    },
    {
      "entropy": 5.436415195465088,
      "epoch": 0.9034833091436865,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004925718061298895,
      "loss": 5.1343,
      "mean_token_accuracy": 0.19940126687288284,
      "num_tokens": 20673168.0,
      "step": 9960
    },
    {
      "entropy": 5.574391031265259,
      "epoch": 0.9039368650217707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004925635584858675,
      "loss": 5.2616,
      "mean_token_accuracy": 0.1847979247570038,
      "num_tokens": 20683100.0,
      "step": 9965
    },
    {
      "entropy": 5.563992643356324,
      "epoch": 0.9043904208998549,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004925553063425755,
      "loss": 5.2711,
      "mean_token_accuracy": 0.18730091899633408,
      "num_tokens": 20694097.0,
      "step": 9970
    },
    {
      "entropy": 5.4666440963745115,
      "epoch": 0.904843976777939,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004925470497001844,
      "loss": 5.2329,
      "mean_token_accuracy": 0.19150329530239105,
      "num_tokens": 20704640.0,
      "step": 9975
    },
    {
      "entropy": 5.427973079681396,
      "epoch": 0.9052975326560232,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004925387885588648,
      "loss": 5.179,
      "mean_token_accuracy": 0.1923139587044716,
      "num_tokens": 20714419.0,
      "step": 9980
    },
    {
      "entropy": 5.5230906963348385,
      "epoch": 0.9057510885341074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004925305229187877,
      "loss": 5.2626,
      "mean_token_accuracy": 0.1851169213652611,
      "num_tokens": 20724705.0,
      "step": 9985
    },
    {
      "entropy": 5.553748750686646,
      "epoch": 0.9062046444121916,
      "grad_norm": 1.09375,
      "learning_rate": 0.000492522252780124,
      "loss": 5.3253,
      "mean_token_accuracy": 0.18897947520017624,
      "num_tokens": 20734753.0,
      "step": 9990
    },
    {
      "entropy": 5.435711288452149,
      "epoch": 0.9066582002902758,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004925139781430446,
      "loss": 5.1086,
      "mean_token_accuracy": 0.19855519235134125,
      "num_tokens": 20745389.0,
      "step": 9995
    },
    {
      "entropy": 5.46137261390686,
      "epoch": 0.9071117561683599,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004925056990077207,
      "loss": 5.1865,
      "mean_token_accuracy": 0.1954839289188385,
      "num_tokens": 20755702.0,
      "step": 10000
    },
    {
      "entropy": 5.471224355697632,
      "epoch": 0.9075653120464441,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004924974153743235,
      "loss": 5.0828,
      "mean_token_accuracy": 0.2000062495470047,
      "num_tokens": 20766046.0,
      "step": 10005
    },
    {
      "entropy": 5.457051086425781,
      "epoch": 0.9080188679245284,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004924891272430244,
      "loss": 5.1963,
      "mean_token_accuracy": 0.19161363542079926,
      "num_tokens": 20776654.0,
      "step": 10010
    },
    {
      "entropy": 5.489421749114991,
      "epoch": 0.9084724238026125,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924808346139948,
      "loss": 5.2516,
      "mean_token_accuracy": 0.19583165645599365,
      "num_tokens": 20787066.0,
      "step": 10015
    },
    {
      "entropy": 5.47434401512146,
      "epoch": 0.9089259796806967,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004924725374874059,
      "loss": 5.1592,
      "mean_token_accuracy": 0.1912713587284088,
      "num_tokens": 20796822.0,
      "step": 10020
    },
    {
      "entropy": 5.468144750595092,
      "epoch": 0.9093795355587808,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004924642358634297,
      "loss": 5.2155,
      "mean_token_accuracy": 0.1913888320326805,
      "num_tokens": 20807602.0,
      "step": 10025
    },
    {
      "entropy": 5.5108407497406,
      "epoch": 0.909833091436865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004924559297422376,
      "loss": 5.1719,
      "mean_token_accuracy": 0.19738072901964188,
      "num_tokens": 20819046.0,
      "step": 10030
    },
    {
      "entropy": 5.51841721534729,
      "epoch": 0.9102866473149492,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924476191240014,
      "loss": 5.1916,
      "mean_token_accuracy": 0.18738922476768494,
      "num_tokens": 20828989.0,
      "step": 10035
    },
    {
      "entropy": 5.4257941246032715,
      "epoch": 0.9107402031930334,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000492439304008893,
      "loss": 5.2032,
      "mean_token_accuracy": 0.18497975170612335,
      "num_tokens": 20839366.0,
      "step": 10040
    },
    {
      "entropy": 5.4617719650268555,
      "epoch": 0.9111937590711175,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004924309843970847,
      "loss": 5.1952,
      "mean_token_accuracy": 0.19048805832862853,
      "num_tokens": 20849513.0,
      "step": 10045
    },
    {
      "entropy": 5.54318790435791,
      "epoch": 0.9116473149492017,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004924226602887479,
      "loss": 5.2508,
      "mean_token_accuracy": 0.18627595603466035,
      "num_tokens": 20861406.0,
      "step": 10050
    },
    {
      "entropy": 5.462168550491333,
      "epoch": 0.912100870827286,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004924143316840552,
      "loss": 5.1538,
      "mean_token_accuracy": 0.1994623914361,
      "num_tokens": 20871190.0,
      "step": 10055
    },
    {
      "entropy": 5.488011312484741,
      "epoch": 0.9125544267053701,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004924059985831787,
      "loss": 5.3135,
      "mean_token_accuracy": 0.18306103050708772,
      "num_tokens": 20881589.0,
      "step": 10060
    },
    {
      "entropy": 5.480007028579712,
      "epoch": 0.9130079825834543,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004923976609862908,
      "loss": 5.2285,
      "mean_token_accuracy": 0.18074655085802077,
      "num_tokens": 20892072.0,
      "step": 10065
    },
    {
      "entropy": 5.519109153747559,
      "epoch": 0.9134615384615384,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923893188935638,
      "loss": 5.2174,
      "mean_token_accuracy": 0.18683876395225524,
      "num_tokens": 20902616.0,
      "step": 10070
    },
    {
      "entropy": 5.5475568771362305,
      "epoch": 0.9139150943396226,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004923809723051702,
      "loss": 5.218,
      "mean_token_accuracy": 0.1869152545928955,
      "num_tokens": 20913832.0,
      "step": 10075
    },
    {
      "entropy": 5.40401291847229,
      "epoch": 0.9143686502177069,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004923726212212829,
      "loss": 5.1051,
      "mean_token_accuracy": 0.1958935961127281,
      "num_tokens": 20923322.0,
      "step": 10080
    },
    {
      "entropy": 5.413718605041504,
      "epoch": 0.914822206095791,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004923642656420742,
      "loss": 5.1963,
      "mean_token_accuracy": 0.1901403397321701,
      "num_tokens": 20933334.0,
      "step": 10085
    },
    {
      "entropy": 5.449005460739135,
      "epoch": 0.9152757619738752,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004923559055677173,
      "loss": 5.1714,
      "mean_token_accuracy": 0.18672747910022736,
      "num_tokens": 20944171.0,
      "step": 10090
    },
    {
      "entropy": 5.463436222076416,
      "epoch": 0.9157293178519593,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004923475409983848,
      "loss": 5.2149,
      "mean_token_accuracy": 0.19056689441204072,
      "num_tokens": 20954200.0,
      "step": 10095
    },
    {
      "entropy": 5.472059440612793,
      "epoch": 0.9161828737300436,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004923391719342497,
      "loss": 5.2217,
      "mean_token_accuracy": 0.186924010515213,
      "num_tokens": 20964935.0,
      "step": 10100
    },
    {
      "entropy": 5.444699954986572,
      "epoch": 0.9166364296081277,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004923307983754852,
      "loss": 5.2157,
      "mean_token_accuracy": 0.19322005957365035,
      "num_tokens": 20975499.0,
      "step": 10105
    },
    {
      "entropy": 5.40584716796875,
      "epoch": 0.9170899854862119,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004923224203222645,
      "loss": 5.1483,
      "mean_token_accuracy": 0.19008391797542573,
      "num_tokens": 20986199.0,
      "step": 10110
    },
    {
      "entropy": 5.468102693557739,
      "epoch": 0.9175435413642961,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004923140377747606,
      "loss": 5.2175,
      "mean_token_accuracy": 0.19170612692832947,
      "num_tokens": 20996313.0,
      "step": 10115
    },
    {
      "entropy": 5.4784221172332765,
      "epoch": 0.9179970972423802,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004923056507331471,
      "loss": 5.198,
      "mean_token_accuracy": 0.18825269341468812,
      "num_tokens": 21006133.0,
      "step": 10120
    },
    {
      "entropy": 5.489863491058349,
      "epoch": 0.9184506531204645,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922972591975974,
      "loss": 5.1674,
      "mean_token_accuracy": 0.19215502738952636,
      "num_tokens": 21016448.0,
      "step": 10125
    },
    {
      "entropy": 5.407008695602417,
      "epoch": 0.9189042089985486,
      "grad_norm": 1.0,
      "learning_rate": 0.0004922888631682849,
      "loss": 5.1062,
      "mean_token_accuracy": 0.19257669001817704,
      "num_tokens": 21026458.0,
      "step": 10130
    },
    {
      "entropy": 5.399579048156738,
      "epoch": 0.9193577648766328,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922804626453834,
      "loss": 5.1513,
      "mean_token_accuracy": 0.19592053592205047,
      "num_tokens": 21037111.0,
      "step": 10135
    },
    {
      "entropy": 5.428416681289673,
      "epoch": 0.9198113207547169,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004922720576290665,
      "loss": 5.202,
      "mean_token_accuracy": 0.19176458269357682,
      "num_tokens": 21048041.0,
      "step": 10140
    },
    {
      "entropy": 5.43221607208252,
      "epoch": 0.9202648766328012,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922636481195082,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2073404759168625,
      "num_tokens": 21058099.0,
      "step": 10145
    },
    {
      "entropy": 5.4595221996307375,
      "epoch": 0.9207184325108854,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004922552341168823,
      "loss": 5.2142,
      "mean_token_accuracy": 0.18525523990392684,
      "num_tokens": 21068137.0,
      "step": 10150
    },
    {
      "entropy": 5.39358959197998,
      "epoch": 0.9211719883889695,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004922468156213627,
      "loss": 5.0924,
      "mean_token_accuracy": 0.19592654556035996,
      "num_tokens": 21077556.0,
      "step": 10155
    },
    {
      "entropy": 5.444814157485962,
      "epoch": 0.9216255442670537,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004922383926331236,
      "loss": 5.1259,
      "mean_token_accuracy": 0.1955080047249794,
      "num_tokens": 21087524.0,
      "step": 10160
    },
    {
      "entropy": 5.4274458408355715,
      "epoch": 0.9220791001451378,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922299651523392,
      "loss": 5.2001,
      "mean_token_accuracy": 0.18721571266651155,
      "num_tokens": 21098446.0,
      "step": 10165
    },
    {
      "entropy": 5.437089157104492,
      "epoch": 0.9225326560232221,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004922215331791837,
      "loss": 5.2132,
      "mean_token_accuracy": 0.1849062278866768,
      "num_tokens": 21109291.0,
      "step": 10170
    },
    {
      "entropy": 5.444094038009643,
      "epoch": 0.9229862119013063,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004922130967138316,
      "loss": 5.089,
      "mean_token_accuracy": 0.19493570625782014,
      "num_tokens": 21120043.0,
      "step": 10175
    },
    {
      "entropy": 5.47716703414917,
      "epoch": 0.9234397677793904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004922046557564574,
      "loss": 5.2076,
      "mean_token_accuracy": 0.1816290646791458,
      "num_tokens": 21130339.0,
      "step": 10180
    },
    {
      "entropy": 5.493131637573242,
      "epoch": 0.9238933236574746,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921962103072355,
      "loss": 5.2752,
      "mean_token_accuracy": 0.18547841012477875,
      "num_tokens": 21142307.0,
      "step": 10185
    },
    {
      "entropy": 5.484469032287597,
      "epoch": 0.9243468795355588,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004921877603663406,
      "loss": 5.1808,
      "mean_token_accuracy": 0.19219895750284194,
      "num_tokens": 21151843.0,
      "step": 10190
    },
    {
      "entropy": 5.4950824737548825,
      "epoch": 0.924800435413643,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004921793059339474,
      "loss": 5.1816,
      "mean_token_accuracy": 0.19396258443593978,
      "num_tokens": 21162574.0,
      "step": 10195
    },
    {
      "entropy": 5.452712297439575,
      "epoch": 0.9252539912917271,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004921708470102309,
      "loss": 5.1382,
      "mean_token_accuracy": 0.19424404352903366,
      "num_tokens": 21172637.0,
      "step": 10200
    },
    {
      "entropy": 5.42418475151062,
      "epoch": 0.9257075471698113,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492162383595366,
      "loss": 5.1475,
      "mean_token_accuracy": 0.1976383075118065,
      "num_tokens": 21182272.0,
      "step": 10205
    },
    {
      "entropy": 5.523120403289795,
      "epoch": 0.9261611030478955,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004921539156895275,
      "loss": 5.2767,
      "mean_token_accuracy": 0.18169392496347428,
      "num_tokens": 21193712.0,
      "step": 10210
    },
    {
      "entropy": 5.541693687438965,
      "epoch": 0.9266146589259797,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004921454432928908,
      "loss": 5.3006,
      "mean_token_accuracy": 0.1801042675971985,
      "num_tokens": 21205104.0,
      "step": 10215
    },
    {
      "entropy": 5.4941614151000975,
      "epoch": 0.9270682148040639,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492136966405631,
      "loss": 5.1576,
      "mean_token_accuracy": 0.19228855818510054,
      "num_tokens": 21215495.0,
      "step": 10220
    },
    {
      "entropy": 5.437613391876221,
      "epoch": 0.927521770682148,
      "grad_norm": 1.125,
      "learning_rate": 0.0004921284850279233,
      "loss": 5.1911,
      "mean_token_accuracy": 0.18829042315483094,
      "num_tokens": 21226047.0,
      "step": 10225
    },
    {
      "entropy": 5.4525219917297365,
      "epoch": 0.9279753265602322,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004921199991599433,
      "loss": 5.2463,
      "mean_token_accuracy": 0.18756283074617386,
      "num_tokens": 21235784.0,
      "step": 10230
    },
    {
      "entropy": 5.526726865768433,
      "epoch": 0.9284288824383164,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004921115088018665,
      "loss": 5.1999,
      "mean_token_accuracy": 0.18950969576835633,
      "num_tokens": 21245493.0,
      "step": 10235
    },
    {
      "entropy": 5.584573888778687,
      "epoch": 0.9288824383164006,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492103013953868,
      "loss": 5.2826,
      "mean_token_accuracy": 0.1842719316482544,
      "num_tokens": 21255541.0,
      "step": 10240
    },
    {
      "entropy": 5.423892831802368,
      "epoch": 0.9293359941944848,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004920945146161242,
      "loss": 5.1304,
      "mean_token_accuracy": 0.196283920109272,
      "num_tokens": 21265427.0,
      "step": 10245
    },
    {
      "entropy": 5.45904951095581,
      "epoch": 0.9297895500725689,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004920860107888105,
      "loss": 5.1504,
      "mean_token_accuracy": 0.19452783316373826,
      "num_tokens": 21276597.0,
      "step": 10250
    },
    {
      "entropy": 5.411054039001465,
      "epoch": 0.9302431059506531,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004920775024721026,
      "loss": 5.1693,
      "mean_token_accuracy": 0.19338110834360123,
      "num_tokens": 21286416.0,
      "step": 10255
    },
    {
      "entropy": 5.446629858016967,
      "epoch": 0.9306966618287373,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920689896661767,
      "loss": 5.1831,
      "mean_token_accuracy": 0.1925641641020775,
      "num_tokens": 21296064.0,
      "step": 10260
    },
    {
      "entropy": 5.426272392272949,
      "epoch": 0.9311502177068215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004920604723712088,
      "loss": 5.1953,
      "mean_token_accuracy": 0.1939927965402603,
      "num_tokens": 21306836.0,
      "step": 10265
    },
    {
      "entropy": 5.4682684421539305,
      "epoch": 0.9316037735849056,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004920519505873751,
      "loss": 5.1732,
      "mean_token_accuracy": 0.19016034603118898,
      "num_tokens": 21316936.0,
      "step": 10270
    },
    {
      "entropy": 5.444418096542359,
      "epoch": 0.9320573294629898,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004920434243148515,
      "loss": 5.1067,
      "mean_token_accuracy": 0.1991023138165474,
      "num_tokens": 21326892.0,
      "step": 10275
    },
    {
      "entropy": 5.4291871070861815,
      "epoch": 0.9325108853410741,
      "grad_norm": 1.0,
      "learning_rate": 0.0004920348935538147,
      "loss": 5.1313,
      "mean_token_accuracy": 0.18786720633506776,
      "num_tokens": 21337270.0,
      "step": 10280
    },
    {
      "entropy": 5.484059762954712,
      "epoch": 0.9329644412191582,
      "grad_norm": 0.984375,
      "learning_rate": 0.000492026358304441,
      "loss": 5.2205,
      "mean_token_accuracy": 0.18756954073905946,
      "num_tokens": 21347968.0,
      "step": 10285
    },
    {
      "entropy": 5.4217315196990965,
      "epoch": 0.9334179970972424,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000492017818566907,
      "loss": 5.1521,
      "mean_token_accuracy": 0.19510474503040315,
      "num_tokens": 21358312.0,
      "step": 10290
    },
    {
      "entropy": 5.430413627624512,
      "epoch": 0.9338715529753265,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920092743413892,
      "loss": 5.1203,
      "mean_token_accuracy": 0.19392232745885848,
      "num_tokens": 21368794.0,
      "step": 10295
    },
    {
      "entropy": 5.517674350738526,
      "epoch": 0.9343251088534107,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004920007256280642,
      "loss": 5.1994,
      "mean_token_accuracy": 0.19540511071681976,
      "num_tokens": 21380080.0,
      "step": 10300
    },
    {
      "entropy": 5.465408277511597,
      "epoch": 0.934778664731495,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000491992172427109,
      "loss": 5.1687,
      "mean_token_accuracy": 0.1919153705239296,
      "num_tokens": 21389961.0,
      "step": 10305
    },
    {
      "entropy": 5.423232889175415,
      "epoch": 0.9352322206095791,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004919836147387002,
      "loss": 5.2048,
      "mean_token_accuracy": 0.19409374743700028,
      "num_tokens": 21399869.0,
      "step": 10310
    },
    {
      "entropy": 5.502434301376343,
      "epoch": 0.9356857764876633,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004919750525630151,
      "loss": 5.2018,
      "mean_token_accuracy": 0.19113861620426179,
      "num_tokens": 21410343.0,
      "step": 10315
    },
    {
      "entropy": 5.52030725479126,
      "epoch": 0.9361393323657474,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004919664859002306,
      "loss": 5.1763,
      "mean_token_accuracy": 0.19153531044721603,
      "num_tokens": 21420915.0,
      "step": 10320
    },
    {
      "entropy": 5.468138742446899,
      "epoch": 0.9365928882438317,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004919579147505239,
      "loss": 5.1393,
      "mean_token_accuracy": 0.19598333686590194,
      "num_tokens": 21430879.0,
      "step": 10325
    },
    {
      "entropy": 5.482166004180908,
      "epoch": 0.9370464441219158,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004919493391140721,
      "loss": 5.2311,
      "mean_token_accuracy": 0.19225272685289382,
      "num_tokens": 21440843.0,
      "step": 10330
    },
    {
      "entropy": 5.5462892055511475,
      "epoch": 0.9375,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004919407589910529,
      "loss": 5.3852,
      "mean_token_accuracy": 0.18182432949542998,
      "num_tokens": 21452132.0,
      "step": 10335
    },
    {
      "entropy": 5.4647728443145756,
      "epoch": 0.9379535558780842,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004919321743816434,
      "loss": 5.1714,
      "mean_token_accuracy": 0.1902553766965866,
      "num_tokens": 21461861.0,
      "step": 10340
    },
    {
      "entropy": 5.461749696731568,
      "epoch": 0.9384071117561683,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004919235852860213,
      "loss": 5.1619,
      "mean_token_accuracy": 0.18584073036909105,
      "num_tokens": 21472244.0,
      "step": 10345
    },
    {
      "entropy": 5.505767917633056,
      "epoch": 0.9388606676342526,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004919149917043641,
      "loss": 5.2325,
      "mean_token_accuracy": 0.18217187970876694,
      "num_tokens": 21482394.0,
      "step": 10350
    },
    {
      "entropy": 5.476301479339599,
      "epoch": 0.9393142235123367,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004919063936368496,
      "loss": 5.2623,
      "mean_token_accuracy": 0.1858245924115181,
      "num_tokens": 21493334.0,
      "step": 10355
    },
    {
      "entropy": 5.526564693450927,
      "epoch": 0.9397677793904209,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004918977910836557,
      "loss": 5.2172,
      "mean_token_accuracy": 0.1870238721370697,
      "num_tokens": 21504443.0,
      "step": 10360
    },
    {
      "entropy": 5.524574422836304,
      "epoch": 0.940221335268505,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004918891840449601,
      "loss": 5.2329,
      "mean_token_accuracy": 0.1889067441225052,
      "num_tokens": 21513908.0,
      "step": 10365
    },
    {
      "entropy": 5.509088134765625,
      "epoch": 0.9406748911465893,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000491880572520941,
      "loss": 5.2706,
      "mean_token_accuracy": 0.18197775781154632,
      "num_tokens": 21524349.0,
      "step": 10370
    },
    {
      "entropy": 5.601503610610962,
      "epoch": 0.9411284470246735,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004918719565117763,
      "loss": 5.3169,
      "mean_token_accuracy": 0.19112387597560881,
      "num_tokens": 21535003.0,
      "step": 10375
    },
    {
      "entropy": 5.616733551025391,
      "epoch": 0.9415820029027576,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004918633360176444,
      "loss": 5.3396,
      "mean_token_accuracy": 0.17721167653799058,
      "num_tokens": 21545045.0,
      "step": 10380
    },
    {
      "entropy": 5.420244455337524,
      "epoch": 0.9420355587808418,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004918547110387234,
      "loss": 5.1642,
      "mean_token_accuracy": 0.19451433718204497,
      "num_tokens": 21554830.0,
      "step": 10385
    },
    {
      "entropy": 5.470901536941528,
      "epoch": 0.9424891146589259,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004918460815751916,
      "loss": 5.147,
      "mean_token_accuracy": 0.1933375909924507,
      "num_tokens": 21564693.0,
      "step": 10390
    },
    {
      "entropy": 5.455016374588013,
      "epoch": 0.9429426705370102,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004918374476272277,
      "loss": 5.151,
      "mean_token_accuracy": 0.18868277966976166,
      "num_tokens": 21575475.0,
      "step": 10395
    },
    {
      "entropy": 5.444744110107422,
      "epoch": 0.9433962264150944,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004918288091950101,
      "loss": 5.1625,
      "mean_token_accuracy": 0.19193150252103805,
      "num_tokens": 21586944.0,
      "step": 10400
    },
    {
      "entropy": 5.508098602294922,
      "epoch": 0.9438497822931785,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918201662787174,
      "loss": 5.1872,
      "mean_token_accuracy": 0.188102026283741,
      "num_tokens": 21598170.0,
      "step": 10405
    },
    {
      "entropy": 5.457108020782471,
      "epoch": 0.9443033381712627,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004918115188785285,
      "loss": 5.2038,
      "mean_token_accuracy": 0.1924883246421814,
      "num_tokens": 21609170.0,
      "step": 10410
    },
    {
      "entropy": 5.5134035587310795,
      "epoch": 0.9447568940493469,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004918028669946221,
      "loss": 5.2712,
      "mean_token_accuracy": 0.1856304958462715,
      "num_tokens": 21620171.0,
      "step": 10415
    },
    {
      "entropy": 5.444842720031739,
      "epoch": 0.9452104499274311,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004917942106271771,
      "loss": 5.204,
      "mean_token_accuracy": 0.18975074589252472,
      "num_tokens": 21631072.0,
      "step": 10420
    },
    {
      "entropy": 5.446480655670166,
      "epoch": 0.9456640058055152,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004917855497763727,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19692519158124924,
      "num_tokens": 21641543.0,
      "step": 10425
    },
    {
      "entropy": 5.450199317932129,
      "epoch": 0.9461175616835994,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004917768844423878,
      "loss": 5.2013,
      "mean_token_accuracy": 0.19204503446817398,
      "num_tokens": 21651342.0,
      "step": 10430
    },
    {
      "entropy": 5.479935264587402,
      "epoch": 0.9465711175616836,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004917682146254017,
      "loss": 5.158,
      "mean_token_accuracy": 0.19622128307819367,
      "num_tokens": 21661043.0,
      "step": 10435
    },
    {
      "entropy": 5.386088800430298,
      "epoch": 0.9470246734397678,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917595403255938,
      "loss": 5.04,
      "mean_token_accuracy": 0.20307991057634353,
      "num_tokens": 21671197.0,
      "step": 10440
    },
    {
      "entropy": 5.388993835449218,
      "epoch": 0.947478229317852,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004917508615431431,
      "loss": 5.1074,
      "mean_token_accuracy": 0.19641754031181335,
      "num_tokens": 21682066.0,
      "step": 10445
    },
    {
      "entropy": 5.509283256530762,
      "epoch": 0.9479317851959361,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004917421782782296,
      "loss": 5.1637,
      "mean_token_accuracy": 0.1928715631365776,
      "num_tokens": 21692117.0,
      "step": 10450
    },
    {
      "entropy": 5.497783470153808,
      "epoch": 0.9483853410740203,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004917334905310325,
      "loss": 5.2899,
      "mean_token_accuracy": 0.18302855789661407,
      "num_tokens": 21703544.0,
      "step": 10455
    },
    {
      "entropy": 5.399119567871094,
      "epoch": 0.9488388969521045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004917247983017317,
      "loss": 5.1063,
      "mean_token_accuracy": 0.19429357051849366,
      "num_tokens": 21712850.0,
      "step": 10460
    },
    {
      "entropy": 5.465300464630127,
      "epoch": 0.9492924528301887,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004917161015905067,
      "loss": 5.2568,
      "mean_token_accuracy": 0.19032784849405288,
      "num_tokens": 21723015.0,
      "step": 10465
    },
    {
      "entropy": 5.50637149810791,
      "epoch": 0.9497460087082729,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004917074003975376,
      "loss": 5.2109,
      "mean_token_accuracy": 0.1906416416168213,
      "num_tokens": 21732453.0,
      "step": 10470
    },
    {
      "entropy": 5.502040147781372,
      "epoch": 0.950199564586357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004916986947230042,
      "loss": 5.2851,
      "mean_token_accuracy": 0.1872340992093086,
      "num_tokens": 21743251.0,
      "step": 10475
    },
    {
      "entropy": 5.475094842910766,
      "epoch": 0.9506531204644412,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004916899845670866,
      "loss": 5.1852,
      "mean_token_accuracy": 0.19052052050828933,
      "num_tokens": 21752891.0,
      "step": 10480
    },
    {
      "entropy": 5.50803575515747,
      "epoch": 0.9511066763425254,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004916812699299649,
      "loss": 5.1881,
      "mean_token_accuracy": 0.19168722480535508,
      "num_tokens": 21763230.0,
      "step": 10485
    },
    {
      "entropy": 5.468595266342163,
      "epoch": 0.9515602322206096,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004916725508118193,
      "loss": 5.1464,
      "mean_token_accuracy": 0.19173318296670913,
      "num_tokens": 21773663.0,
      "step": 10490
    },
    {
      "entropy": 5.454308080673218,
      "epoch": 0.9520137880986937,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004916638272128301,
      "loss": 5.2005,
      "mean_token_accuracy": 0.19195641875267028,
      "num_tokens": 21784265.0,
      "step": 10495
    },
    {
      "entropy": 5.421849203109741,
      "epoch": 0.9524673439767779,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004916550991331777,
      "loss": 5.1688,
      "mean_token_accuracy": 0.19283314794301987,
      "num_tokens": 21795056.0,
      "step": 10500
    },
    {
      "entropy": 5.451410961151123,
      "epoch": 0.9529208998548622,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004916463665730427,
      "loss": 5.1743,
      "mean_token_accuracy": 0.19450462907552718,
      "num_tokens": 21806204.0,
      "step": 10505
    },
    {
      "entropy": 5.452659845352173,
      "epoch": 0.9533744557329463,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004916376295326057,
      "loss": 5.1935,
      "mean_token_accuracy": 0.19136037677526474,
      "num_tokens": 21816993.0,
      "step": 10510
    },
    {
      "entropy": 5.424757862091065,
      "epoch": 0.9538280116110305,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004916288880120472,
      "loss": 5.0334,
      "mean_token_accuracy": 0.206775464117527,
      "num_tokens": 21827457.0,
      "step": 10515
    },
    {
      "entropy": 5.468567562103272,
      "epoch": 0.9542815674891146,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004916201420115482,
      "loss": 5.1734,
      "mean_token_accuracy": 0.18725205212831497,
      "num_tokens": 21837555.0,
      "step": 10520
    },
    {
      "entropy": 5.469878101348877,
      "epoch": 0.9547351233671988,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004916113915312895,
      "loss": 5.2008,
      "mean_token_accuracy": 0.19283771514892578,
      "num_tokens": 21847074.0,
      "step": 10525
    },
    {
      "entropy": 5.503349781036377,
      "epoch": 0.9551886792452831,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916026365714519,
      "loss": 5.1922,
      "mean_token_accuracy": 0.18674940168857573,
      "num_tokens": 21858333.0,
      "step": 10530
    },
    {
      "entropy": 5.4334405899047855,
      "epoch": 0.9556422351233672,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004915938771322165,
      "loss": 5.1958,
      "mean_token_accuracy": 0.19297861605882644,
      "num_tokens": 21869132.0,
      "step": 10535
    },
    {
      "entropy": 5.437691497802734,
      "epoch": 0.9560957910014514,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004915851132137646,
      "loss": 5.2101,
      "mean_token_accuracy": 0.18776627331972123,
      "num_tokens": 21879153.0,
      "step": 10540
    },
    {
      "entropy": 5.471747541427613,
      "epoch": 0.9565493468795355,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004915763448162774,
      "loss": 5.1079,
      "mean_token_accuracy": 0.1977652981877327,
      "num_tokens": 21889063.0,
      "step": 10545
    },
    {
      "entropy": 5.468826198577881,
      "epoch": 0.9570029027576198,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004915675719399362,
      "loss": 5.2302,
      "mean_token_accuracy": 0.18763369768857957,
      "num_tokens": 21899005.0,
      "step": 10550
    },
    {
      "entropy": 5.450291156768799,
      "epoch": 0.9574564586357039,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004915587945849225,
      "loss": 5.0986,
      "mean_token_accuracy": 0.20017891079187394,
      "num_tokens": 21908391.0,
      "step": 10555
    },
    {
      "entropy": 5.456761217117309,
      "epoch": 0.9579100145137881,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004915500127514176,
      "loss": 5.1713,
      "mean_token_accuracy": 0.19450551569461821,
      "num_tokens": 21917655.0,
      "step": 10560
    },
    {
      "entropy": 5.545481443405151,
      "epoch": 0.9583635703918723,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004915412264396034,
      "loss": 5.232,
      "mean_token_accuracy": 0.19058077186346054,
      "num_tokens": 21928038.0,
      "step": 10565
    },
    {
      "entropy": 5.486158227920532,
      "epoch": 0.9588171262699564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004915324356496613,
      "loss": 5.1218,
      "mean_token_accuracy": 0.19642450511455536,
      "num_tokens": 21938106.0,
      "step": 10570
    },
    {
      "entropy": 5.421289777755737,
      "epoch": 0.9592706821480407,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004915236403817733,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19307896196842195,
      "num_tokens": 21947484.0,
      "step": 10575
    },
    {
      "entropy": 5.461407136917114,
      "epoch": 0.9597242380261248,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004915148406361213,
      "loss": 5.21,
      "mean_token_accuracy": 0.19016327410936357,
      "num_tokens": 21957139.0,
      "step": 10580
    },
    {
      "entropy": 5.465814828872681,
      "epoch": 0.960177793904209,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004915060364128872,
      "loss": 5.2068,
      "mean_token_accuracy": 0.18789542019367217,
      "num_tokens": 21967030.0,
      "step": 10585
    },
    {
      "entropy": 5.424587106704712,
      "epoch": 0.9606313497822931,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004914972277122532,
      "loss": 5.1502,
      "mean_token_accuracy": 0.1932281047105789,
      "num_tokens": 21976418.0,
      "step": 10590
    },
    {
      "entropy": 5.436279344558716,
      "epoch": 0.9610849056603774,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004914884145344012,
      "loss": 5.15,
      "mean_token_accuracy": 0.19504899382591248,
      "num_tokens": 21987284.0,
      "step": 10595
    },
    {
      "entropy": 5.402978086471558,
      "epoch": 0.9615384615384616,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004914795968795137,
      "loss": 5.0709,
      "mean_token_accuracy": 0.20060221701860428,
      "num_tokens": 21997106.0,
      "step": 10600
    },
    {
      "entropy": 5.435127687454224,
      "epoch": 0.9619920174165457,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004914707747477728,
      "loss": 5.0591,
      "mean_token_accuracy": 0.19728294610977173,
      "num_tokens": 22007589.0,
      "step": 10605
    },
    {
      "entropy": 5.447371244430542,
      "epoch": 0.9624455732946299,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914619481393613,
      "loss": 5.2022,
      "mean_token_accuracy": 0.1917535647749901,
      "num_tokens": 22017669.0,
      "step": 10610
    },
    {
      "entropy": 5.585604906082153,
      "epoch": 0.962899129172714,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004914531170544615,
      "loss": 5.3528,
      "mean_token_accuracy": 0.182457534968853,
      "num_tokens": 22029330.0,
      "step": 10615
    },
    {
      "entropy": 5.460028839111328,
      "epoch": 0.9633526850507983,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004914442814932561,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19587075859308242,
      "num_tokens": 22039877.0,
      "step": 10620
    },
    {
      "entropy": 5.451121139526367,
      "epoch": 0.9638062409288825,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004914354414559278,
      "loss": 5.1776,
      "mean_token_accuracy": 0.19791433960199356,
      "num_tokens": 22049579.0,
      "step": 10625
    },
    {
      "entropy": 5.337468099594116,
      "epoch": 0.9642597968069666,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004914265969426594,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20568925887346268,
      "num_tokens": 22059608.0,
      "step": 10630
    },
    {
      "entropy": 5.454187726974487,
      "epoch": 0.9647133526850508,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004914177479536338,
      "loss": 5.2144,
      "mean_token_accuracy": 0.19067026823759078,
      "num_tokens": 22070036.0,
      "step": 10635
    },
    {
      "entropy": 5.490912675857544,
      "epoch": 0.965166908563135,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000491408894489034,
      "loss": 5.2014,
      "mean_token_accuracy": 0.18741769790649415,
      "num_tokens": 22080679.0,
      "step": 10640
    },
    {
      "entropy": 5.437967729568482,
      "epoch": 0.9656204644412192,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004914000365490432,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20011549443006516,
      "num_tokens": 22092192.0,
      "step": 10645
    },
    {
      "entropy": 5.446996021270752,
      "epoch": 0.9660740203193033,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004913911741338445,
      "loss": 5.2316,
      "mean_token_accuracy": 0.18467727899551392,
      "num_tokens": 22102722.0,
      "step": 10650
    },
    {
      "entropy": 5.398031520843506,
      "epoch": 0.9665275761973875,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004913823072436211,
      "loss": 5.1069,
      "mean_token_accuracy": 0.20055763721466063,
      "num_tokens": 22113521.0,
      "step": 10655
    },
    {
      "entropy": 5.489092683792114,
      "epoch": 0.9669811320754716,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004913734358785565,
      "loss": 5.1951,
      "mean_token_accuracy": 0.191675665974617,
      "num_tokens": 22124891.0,
      "step": 10660
    },
    {
      "entropy": 5.45971531867981,
      "epoch": 0.9674346879535559,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004913645600388341,
      "loss": 5.1955,
      "mean_token_accuracy": 0.1867385983467102,
      "num_tokens": 22135972.0,
      "step": 10665
    },
    {
      "entropy": 5.504754066467285,
      "epoch": 0.9678882438316401,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004913556797246374,
      "loss": 5.2719,
      "mean_token_accuracy": 0.17639271914958954,
      "num_tokens": 22146623.0,
      "step": 10670
    },
    {
      "entropy": 5.51257095336914,
      "epoch": 0.9683417997097242,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004913467949361503,
      "loss": 5.1694,
      "mean_token_accuracy": 0.1888551414012909,
      "num_tokens": 22157619.0,
      "step": 10675
    },
    {
      "entropy": 5.4886709690094,
      "epoch": 0.9687953555878084,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004913379056735561,
      "loss": 5.1681,
      "mean_token_accuracy": 0.1849115401506424,
      "num_tokens": 22168655.0,
      "step": 10680
    },
    {
      "entropy": 5.448235177993775,
      "epoch": 0.9692489114658926,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000491329011937039,
      "loss": 5.2101,
      "mean_token_accuracy": 0.1861863613128662,
      "num_tokens": 22178500.0,
      "step": 10685
    },
    {
      "entropy": 5.498290205001831,
      "epoch": 0.9697024673439768,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004913201137267828,
      "loss": 5.209,
      "mean_token_accuracy": 0.18552537113428116,
      "num_tokens": 22188873.0,
      "step": 10690
    },
    {
      "entropy": 5.423058652877808,
      "epoch": 0.970156023222061,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004913112110429715,
      "loss": 5.1115,
      "mean_token_accuracy": 0.1923189640045166,
      "num_tokens": 22199743.0,
      "step": 10695
    },
    {
      "entropy": 5.42342176437378,
      "epoch": 0.9706095791001451,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004913023038857892,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2012026712298393,
      "num_tokens": 22210301.0,
      "step": 10700
    },
    {
      "entropy": 5.42089877128601,
      "epoch": 0.9710631349782293,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004912933922554203,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19924921542406082,
      "num_tokens": 22220771.0,
      "step": 10705
    },
    {
      "entropy": 5.436170768737793,
      "epoch": 0.9715166908563135,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004912844761520487,
      "loss": 5.189,
      "mean_token_accuracy": 0.1894242450594902,
      "num_tokens": 22231100.0,
      "step": 10710
    },
    {
      "entropy": 5.427358102798462,
      "epoch": 0.9719702467343977,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004912755555758592,
      "loss": 5.1704,
      "mean_token_accuracy": 0.1919427692890167,
      "num_tokens": 22240997.0,
      "step": 10715
    },
    {
      "entropy": 5.469578981399536,
      "epoch": 0.9724238026124818,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491266630527036,
      "loss": 5.2051,
      "mean_token_accuracy": 0.18859542161226273,
      "num_tokens": 22251795.0,
      "step": 10720
    },
    {
      "entropy": 5.541397523880005,
      "epoch": 0.972877358490566,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004912577010057638,
      "loss": 5.2499,
      "mean_token_accuracy": 0.18489183038473128,
      "num_tokens": 22262465.0,
      "step": 10725
    },
    {
      "entropy": 5.585089445114136,
      "epoch": 0.9733309143686503,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004912487670122273,
      "loss": 5.3521,
      "mean_token_accuracy": 0.17896877229213715,
      "num_tokens": 22272399.0,
      "step": 10730
    },
    {
      "entropy": 5.510864639282227,
      "epoch": 0.9737844702467344,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004912398285466111,
      "loss": 5.2631,
      "mean_token_accuracy": 0.18806591629981995,
      "num_tokens": 22282545.0,
      "step": 10735
    },
    {
      "entropy": 5.4886369705200195,
      "epoch": 0.9742380261248186,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004912308856091002,
      "loss": 5.211,
      "mean_token_accuracy": 0.19261860400438308,
      "num_tokens": 22293154.0,
      "step": 10740
    },
    {
      "entropy": 5.491091871261597,
      "epoch": 0.9746915820029027,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004912219381998795,
      "loss": 5.1295,
      "mean_token_accuracy": 0.2008547693490982,
      "num_tokens": 22302309.0,
      "step": 10745
    },
    {
      "entropy": 5.527703189849854,
      "epoch": 0.9751451378809869,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004912129863191339,
      "loss": 5.241,
      "mean_token_accuracy": 0.17985348105430604,
      "num_tokens": 22312316.0,
      "step": 10750
    },
    {
      "entropy": 5.4772176265716555,
      "epoch": 0.9755986937590712,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004912040299670487,
      "loss": 5.2116,
      "mean_token_accuracy": 0.19279262274503708,
      "num_tokens": 22322387.0,
      "step": 10755
    },
    {
      "entropy": 5.465480613708496,
      "epoch": 0.9760522496371553,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004911950691438091,
      "loss": 5.2052,
      "mean_token_accuracy": 0.1968354642391205,
      "num_tokens": 22332759.0,
      "step": 10760
    },
    {
      "entropy": 5.4949132919311525,
      "epoch": 0.9765058055152395,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004911861038496004,
      "loss": 5.245,
      "mean_token_accuracy": 0.19074769914150239,
      "num_tokens": 22344056.0,
      "step": 10765
    },
    {
      "entropy": 5.44970850944519,
      "epoch": 0.9769593613933236,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004911771340846079,
      "loss": 5.194,
      "mean_token_accuracy": 0.18383531719446183,
      "num_tokens": 22353877.0,
      "step": 10770
    },
    {
      "entropy": 5.5462226390838625,
      "epoch": 0.9774129172714079,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911681598490172,
      "loss": 5.2222,
      "mean_token_accuracy": 0.1833869755268097,
      "num_tokens": 22363549.0,
      "step": 10775
    },
    {
      "entropy": 5.534874534606933,
      "epoch": 0.977866473149492,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911591811430139,
      "loss": 5.2567,
      "mean_token_accuracy": 0.18992500454187394,
      "num_tokens": 22374040.0,
      "step": 10780
    },
    {
      "entropy": 5.518569135665894,
      "epoch": 0.9783200290275762,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004911501979667836,
      "loss": 5.3026,
      "mean_token_accuracy": 0.18968116790056228,
      "num_tokens": 22384819.0,
      "step": 10785
    },
    {
      "entropy": 5.458084917068481,
      "epoch": 0.9787735849056604,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004911412103205122,
      "loss": 5.1802,
      "mean_token_accuracy": 0.1892141282558441,
      "num_tokens": 22395130.0,
      "step": 10790
    },
    {
      "entropy": 5.475216674804687,
      "epoch": 0.9792271407837445,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004911322182043855,
      "loss": 5.1936,
      "mean_token_accuracy": 0.18980710357427596,
      "num_tokens": 22405093.0,
      "step": 10795
    },
    {
      "entropy": 5.490931606292724,
      "epoch": 0.9796806966618288,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004911232216185893,
      "loss": 5.2105,
      "mean_token_accuracy": 0.18691559731960297,
      "num_tokens": 22415008.0,
      "step": 10800
    },
    {
      "entropy": 5.51223521232605,
      "epoch": 0.9801342525399129,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004911142205633099,
      "loss": 5.2572,
      "mean_token_accuracy": 0.18417985886335372,
      "num_tokens": 22426486.0,
      "step": 10805
    },
    {
      "entropy": 5.445620203018189,
      "epoch": 0.9805878084179971,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004911052150387334,
      "loss": 5.1643,
      "mean_token_accuracy": 0.1907837927341461,
      "num_tokens": 22436844.0,
      "step": 10810
    },
    {
      "entropy": 5.5398115634918215,
      "epoch": 0.9810413642960812,
      "grad_norm": 1.125,
      "learning_rate": 0.0004910962050450458,
      "loss": 5.2869,
      "mean_token_accuracy": 0.18634007424116134,
      "num_tokens": 22446630.0,
      "step": 10815
    },
    {
      "entropy": 5.5155932903289795,
      "epoch": 0.9814949201741655,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910871905824338,
      "loss": 5.1246,
      "mean_token_accuracy": 0.1972308024764061,
      "num_tokens": 22456406.0,
      "step": 10820
    },
    {
      "entropy": 5.421662712097168,
      "epoch": 0.9819484760522497,
      "grad_norm": 1.0,
      "learning_rate": 0.0004910781716510837,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19659640938043593,
      "num_tokens": 22466210.0,
      "step": 10825
    },
    {
      "entropy": 5.473699998855591,
      "epoch": 0.9824020319303338,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004910691482511817,
      "loss": 5.2212,
      "mean_token_accuracy": 0.19028413742780687,
      "num_tokens": 22476978.0,
      "step": 10830
    },
    {
      "entropy": 5.563931512832641,
      "epoch": 0.982855587808418,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004910601203829148,
      "loss": 5.1888,
      "mean_token_accuracy": 0.18743186444044113,
      "num_tokens": 22488613.0,
      "step": 10835
    },
    {
      "entropy": 5.440819263458252,
      "epoch": 0.9833091436865021,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004910510880464696,
      "loss": 5.0876,
      "mean_token_accuracy": 0.2051035374403,
      "num_tokens": 22498312.0,
      "step": 10840
    },
    {
      "entropy": 5.437257242202759,
      "epoch": 0.9837626995645864,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004910420512420327,
      "loss": 5.2154,
      "mean_token_accuracy": 0.19098950028419495,
      "num_tokens": 22508131.0,
      "step": 10845
    },
    {
      "entropy": 5.461184930801392,
      "epoch": 0.9842162554426706,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004910330099697913,
      "loss": 5.1836,
      "mean_token_accuracy": 0.19517870992422104,
      "num_tokens": 22517842.0,
      "step": 10850
    },
    {
      "entropy": 5.508009815216065,
      "epoch": 0.9846698113207547,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004910239642299321,
      "loss": 5.1597,
      "mean_token_accuracy": 0.19345295876264573,
      "num_tokens": 22528239.0,
      "step": 10855
    },
    {
      "entropy": 5.465402603149414,
      "epoch": 0.9851233671988389,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004910149140226422,
      "loss": 5.1465,
      "mean_token_accuracy": 0.19278796762228012,
      "num_tokens": 22539734.0,
      "step": 10860
    },
    {
      "entropy": 5.471485900878906,
      "epoch": 0.9855769230769231,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004910058593481089,
      "loss": 5.14,
      "mean_token_accuracy": 0.19574241936206818,
      "num_tokens": 22549796.0,
      "step": 10865
    },
    {
      "entropy": 5.4115073680877686,
      "epoch": 0.9860304789550073,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909968002065195,
      "loss": 5.1119,
      "mean_token_accuracy": 0.19815587401390075,
      "num_tokens": 22559934.0,
      "step": 10870
    },
    {
      "entropy": 5.553992795944214,
      "epoch": 0.9864840348330914,
      "grad_norm": 1.078125,
      "learning_rate": 0.000490987736598061,
      "loss": 5.2505,
      "mean_token_accuracy": 0.17646464705467224,
      "num_tokens": 22569864.0,
      "step": 10875
    },
    {
      "entropy": 5.48905930519104,
      "epoch": 0.9869375907111756,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004909786685229211,
      "loss": 5.1793,
      "mean_token_accuracy": 0.1971484124660492,
      "num_tokens": 22580687.0,
      "step": 10880
    },
    {
      "entropy": 5.515979480743408,
      "epoch": 0.9873911465892597,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004909695959812875,
      "loss": 5.2313,
      "mean_token_accuracy": 0.19113008826971054,
      "num_tokens": 22590507.0,
      "step": 10885
    },
    {
      "entropy": 5.498454666137695,
      "epoch": 0.987844702467344,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004909605189733474,
      "loss": 5.2535,
      "mean_token_accuracy": 0.19267081320285798,
      "num_tokens": 22601901.0,
      "step": 10890
    },
    {
      "entropy": 5.446087169647217,
      "epoch": 0.9882982583454282,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004909514374992889,
      "loss": 5.1967,
      "mean_token_accuracy": 0.19195072799921037,
      "num_tokens": 22613003.0,
      "step": 10895
    },
    {
      "entropy": 5.4279900074005125,
      "epoch": 0.9887518142235123,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004909423515592996,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19864177107810974,
      "num_tokens": 22623401.0,
      "step": 10900
    },
    {
      "entropy": 5.49190239906311,
      "epoch": 0.9892053701015965,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004909332611535675,
      "loss": 5.1861,
      "mean_token_accuracy": 0.18905673325061798,
      "num_tokens": 22633147.0,
      "step": 10905
    },
    {
      "entropy": 5.422115516662598,
      "epoch": 0.9896589259796807,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004909241662822804,
      "loss": 5.043,
      "mean_token_accuracy": 0.20871215909719468,
      "num_tokens": 22642990.0,
      "step": 10910
    },
    {
      "entropy": 5.443415689468384,
      "epoch": 0.9901124818577649,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909150669456265,
      "loss": 5.2646,
      "mean_token_accuracy": 0.18424686342477797,
      "num_tokens": 22652734.0,
      "step": 10915
    },
    {
      "entropy": 5.483167934417724,
      "epoch": 0.9905660377358491,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000490905963143794,
      "loss": 5.2804,
      "mean_token_accuracy": 0.18312346786260605,
      "num_tokens": 22662968.0,
      "step": 10920
    },
    {
      "entropy": 5.407788610458374,
      "epoch": 0.9910195936139332,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004908968548769713,
      "loss": 5.102,
      "mean_token_accuracy": 0.2041473627090454,
      "num_tokens": 22673157.0,
      "step": 10925
    },
    {
      "entropy": 5.471360683441162,
      "epoch": 0.9914731494920174,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908877421453466,
      "loss": 5.2468,
      "mean_token_accuracy": 0.18830858021974564,
      "num_tokens": 22683105.0,
      "step": 10930
    },
    {
      "entropy": 5.453107118606567,
      "epoch": 0.9919267053701016,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004908786249491082,
      "loss": 5.1757,
      "mean_token_accuracy": 0.20476199686527252,
      "num_tokens": 22693793.0,
      "step": 10935
    },
    {
      "entropy": 5.566500282287597,
      "epoch": 0.9923802612481858,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000490869503288445,
      "loss": 5.2202,
      "mean_token_accuracy": 0.1911487653851509,
      "num_tokens": 22703884.0,
      "step": 10940
    },
    {
      "entropy": 5.522388029098511,
      "epoch": 0.99283381712627,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908603771635454,
      "loss": 5.2657,
      "mean_token_accuracy": 0.18617092519998552,
      "num_tokens": 22713699.0,
      "step": 10945
    },
    {
      "entropy": 5.451653718948364,
      "epoch": 0.9932873730043541,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004908512465745982,
      "loss": 5.1013,
      "mean_token_accuracy": 0.1970734789967537,
      "num_tokens": 22722871.0,
      "step": 10950
    },
    {
      "entropy": 5.477705287933349,
      "epoch": 0.9937409288824384,
      "grad_norm": 1.0,
      "learning_rate": 0.0004908421115217922,
      "loss": 5.1089,
      "mean_token_accuracy": 0.19224777370691298,
      "num_tokens": 22733314.0,
      "step": 10955
    },
    {
      "entropy": 5.404214572906494,
      "epoch": 0.9941944847605225,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004908329720053164,
      "loss": 5.1123,
      "mean_token_accuracy": 0.2003941997885704,
      "num_tokens": 22743255.0,
      "step": 10960
    },
    {
      "entropy": 5.388226270675659,
      "epoch": 0.9946480406386067,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004908238280253597,
      "loss": 5.0903,
      "mean_token_accuracy": 0.20025001019239425,
      "num_tokens": 22753857.0,
      "step": 10965
    },
    {
      "entropy": 5.4696001529693605,
      "epoch": 0.9951015965166908,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004908146795821112,
      "loss": 5.0843,
      "mean_token_accuracy": 0.1949297696352005,
      "num_tokens": 22764326.0,
      "step": 10970
    },
    {
      "entropy": 5.480047273635864,
      "epoch": 0.995555152394775,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004908055266757602,
      "loss": 5.1715,
      "mean_token_accuracy": 0.19572446793317794,
      "num_tokens": 22774958.0,
      "step": 10975
    },
    {
      "entropy": 5.403993844985962,
      "epoch": 0.9960087082728593,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907963693064958,
      "loss": 5.0864,
      "mean_token_accuracy": 0.19230775833129882,
      "num_tokens": 22785616.0,
      "step": 10980
    },
    {
      "entropy": 5.529326391220093,
      "epoch": 0.9964622641509434,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004907872074745075,
      "loss": 5.2676,
      "mean_token_accuracy": 0.18333887457847595,
      "num_tokens": 22795967.0,
      "step": 10985
    },
    {
      "entropy": 5.450800800323487,
      "epoch": 0.9969158200290276,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004907780411799849,
      "loss": 5.267,
      "mean_token_accuracy": 0.18713125139474868,
      "num_tokens": 22806335.0,
      "step": 10990
    },
    {
      "entropy": 5.5267292022705075,
      "epoch": 0.9973693759071117,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004907688704231173,
      "loss": 5.1945,
      "mean_token_accuracy": 0.19679880142211914,
      "num_tokens": 22816390.0,
      "step": 10995
    },
    {
      "entropy": 5.578009223937988,
      "epoch": 0.997822931785196,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004907596952040945,
      "loss": 5.2575,
      "mean_token_accuracy": 0.19109320491552353,
      "num_tokens": 22826395.0,
      "step": 11000
    },
    {
      "entropy": 5.526041650772095,
      "epoch": 0.9982764876632801,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004907505155231063,
      "loss": 5.2196,
      "mean_token_accuracy": 0.19320557415485382,
      "num_tokens": 22836790.0,
      "step": 11005
    },
    {
      "entropy": 5.450096607208252,
      "epoch": 0.9987300435413643,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004907413313803424,
      "loss": 5.1121,
      "mean_token_accuracy": 0.19249580502510072,
      "num_tokens": 22846687.0,
      "step": 11010
    },
    {
      "entropy": 5.496992063522339,
      "epoch": 0.9991835994194485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004907321427759928,
      "loss": 5.1698,
      "mean_token_accuracy": 0.19380525201559068,
      "num_tokens": 22857208.0,
      "step": 11015
    },
    {
      "entropy": 5.507448673248291,
      "epoch": 0.9996371552975326,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004907229497102475,
      "loss": 5.2677,
      "mean_token_accuracy": 0.18637495636940002,
      "num_tokens": 22868556.0,
      "step": 11020
    },
    {
      "entropy": 5.520956563949585,
      "epoch": 1.0000907111756168,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004907137521832966,
      "loss": 5.2354,
      "mean_token_accuracy": 0.19049685299396515,
      "num_tokens": 22880081.0,
      "step": 11025
    },
    {
      "entropy": 5.454957485198975,
      "epoch": 1.000544267053701,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004907045501953304,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19731134325265884,
      "num_tokens": 22892141.0,
      "step": 11030
    },
    {
      "entropy": 5.452963924407959,
      "epoch": 1.0009978229317853,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000490695343746539,
      "loss": 5.0744,
      "mean_token_accuracy": 0.19878874123096466,
      "num_tokens": 22902688.0,
      "step": 11035
    },
    {
      "entropy": 5.38298077583313,
      "epoch": 1.0014513788098693,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004906861328371131,
      "loss": 5.0207,
      "mean_token_accuracy": 0.19931923151016234,
      "num_tokens": 22912453.0,
      "step": 11040
    },
    {
      "entropy": 5.4921776294708256,
      "epoch": 1.0019049346879536,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004906769174672429,
      "loss": 5.182,
      "mean_token_accuracy": 0.18716212958097458,
      "num_tokens": 22923167.0,
      "step": 11045
    },
    {
      "entropy": 5.5100603103637695,
      "epoch": 1.0023584905660377,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000490667697637119,
      "loss": 5.1439,
      "mean_token_accuracy": 0.19183895885944366,
      "num_tokens": 22934753.0,
      "step": 11050
    },
    {
      "entropy": 5.430588102340698,
      "epoch": 1.002812046444122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004906584733469324,
      "loss": 5.0642,
      "mean_token_accuracy": 0.20371449440717698,
      "num_tokens": 22945277.0,
      "step": 11055
    },
    {
      "entropy": 5.500294542312622,
      "epoch": 1.0032656023222062,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004906492445968735,
      "loss": 5.1419,
      "mean_token_accuracy": 0.18651341646909714,
      "num_tokens": 22955797.0,
      "step": 11060
    },
    {
      "entropy": 5.483213472366333,
      "epoch": 1.0037191582002902,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004906400113871332,
      "loss": 5.109,
      "mean_token_accuracy": 0.19797017574310302,
      "num_tokens": 22965969.0,
      "step": 11065
    },
    {
      "entropy": 5.449386882781982,
      "epoch": 1.0041727140783745,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004906307737179026,
      "loss": 5.094,
      "mean_token_accuracy": 0.20706432759761811,
      "num_tokens": 22975323.0,
      "step": 11070
    },
    {
      "entropy": 5.4158118724822994,
      "epoch": 1.0046262699564585,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004906215315893726,
      "loss": 5.0854,
      "mean_token_accuracy": 0.20272112637758255,
      "num_tokens": 22985063.0,
      "step": 11075
    },
    {
      "entropy": 5.557787370681763,
      "epoch": 1.0050798258345428,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004906122850017344,
      "loss": 5.2222,
      "mean_token_accuracy": 0.18601493388414383,
      "num_tokens": 22996001.0,
      "step": 11080
    },
    {
      "entropy": 5.504201650619507,
      "epoch": 1.005533381712627,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004906030339551792,
      "loss": 5.132,
      "mean_token_accuracy": 0.19287194013595582,
      "num_tokens": 23006896.0,
      "step": 11085
    },
    {
      "entropy": 5.462795066833496,
      "epoch": 1.0059869375907111,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905937784498983,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19932139068841934,
      "num_tokens": 23017080.0,
      "step": 11090
    },
    {
      "entropy": 5.472736549377442,
      "epoch": 1.0064404934687954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004905845184860831,
      "loss": 5.0633,
      "mean_token_accuracy": 0.1989378586411476,
      "num_tokens": 23027168.0,
      "step": 11095
    },
    {
      "entropy": 5.474420785903931,
      "epoch": 1.0068940493468794,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905752540639252,
      "loss": 5.1048,
      "mean_token_accuracy": 0.19014271944761277,
      "num_tokens": 23037364.0,
      "step": 11100
    },
    {
      "entropy": 5.379934692382813,
      "epoch": 1.0073476052249637,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004905659851836161,
      "loss": 4.9947,
      "mean_token_accuracy": 0.20249858796596526,
      "num_tokens": 23047609.0,
      "step": 11105
    },
    {
      "entropy": 5.465517282485962,
      "epoch": 1.007801161103048,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905567118453475,
      "loss": 5.1455,
      "mean_token_accuracy": 0.19380562752485275,
      "num_tokens": 23057942.0,
      "step": 11110
    },
    {
      "entropy": 5.455103206634521,
      "epoch": 1.008254716981132,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004905474340493111,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19476013481616974,
      "num_tokens": 23069119.0,
      "step": 11115
    },
    {
      "entropy": 5.412275934219361,
      "epoch": 1.0087082728592163,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004905381517956988,
      "loss": 5.0439,
      "mean_token_accuracy": 0.19710911959409713,
      "num_tokens": 23080060.0,
      "step": 11120
    },
    {
      "entropy": 5.417080545425415,
      "epoch": 1.0091618287373005,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004905288650847027,
      "loss": 5.0792,
      "mean_token_accuracy": 0.1906353548169136,
      "num_tokens": 23090645.0,
      "step": 11125
    },
    {
      "entropy": 5.480456209182739,
      "epoch": 1.0096153846153846,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004905195739165146,
      "loss": 5.0965,
      "mean_token_accuracy": 0.19660176038742067,
      "num_tokens": 23100449.0,
      "step": 11130
    },
    {
      "entropy": 5.476227140426635,
      "epoch": 1.0100689404934688,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490510278291327,
      "loss": 5.0782,
      "mean_token_accuracy": 0.19399780929088592,
      "num_tokens": 23110415.0,
      "step": 11135
    },
    {
      "entropy": 5.424685478210449,
      "epoch": 1.010522496371553,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004905009782093318,
      "loss": 5.07,
      "mean_token_accuracy": 0.19324882477521896,
      "num_tokens": 23120336.0,
      "step": 11140
    },
    {
      "entropy": 5.4333597183227536,
      "epoch": 1.0109760522496372,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004904916736707214,
      "loss": 5.1458,
      "mean_token_accuracy": 0.18603336960077285,
      "num_tokens": 23130568.0,
      "step": 11145
    },
    {
      "entropy": 5.512174463272094,
      "epoch": 1.0114296081277214,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004904823646756885,
      "loss": 5.1057,
      "mean_token_accuracy": 0.19878295361995696,
      "num_tokens": 23140080.0,
      "step": 11150
    },
    {
      "entropy": 5.445178365707397,
      "epoch": 1.0118831640058055,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004904730512244252,
      "loss": 5.031,
      "mean_token_accuracy": 0.19515729397535325,
      "num_tokens": 23150360.0,
      "step": 11155
    },
    {
      "entropy": 5.403223466873169,
      "epoch": 1.0123367198838897,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904637333171243,
      "loss": 5.0357,
      "mean_token_accuracy": 0.19459904134273528,
      "num_tokens": 23160620.0,
      "step": 11160
    },
    {
      "entropy": 5.518907165527343,
      "epoch": 1.0127902757619738,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004904544109539785,
      "loss": 5.1258,
      "mean_token_accuracy": 0.19950608909130096,
      "num_tokens": 23171920.0,
      "step": 11165
    },
    {
      "entropy": 5.475254821777344,
      "epoch": 1.013243831640058,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004904450841351805,
      "loss": 5.0886,
      "mean_token_accuracy": 0.20122201591730118,
      "num_tokens": 23182929.0,
      "step": 11170
    },
    {
      "entropy": 5.488019275665283,
      "epoch": 1.0136973875181423,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004904357528609236,
      "loss": 5.0553,
      "mean_token_accuracy": 0.18875425010919572,
      "num_tokens": 23193020.0,
      "step": 11175
    },
    {
      "entropy": 5.40572509765625,
      "epoch": 1.0141509433962264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004904264171314,
      "loss": 5.0859,
      "mean_token_accuracy": 0.1987457439303398,
      "num_tokens": 23204073.0,
      "step": 11180
    },
    {
      "entropy": 5.39017915725708,
      "epoch": 1.0146044992743106,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004904170769468035,
      "loss": 4.9797,
      "mean_token_accuracy": 0.19448689222335816,
      "num_tokens": 23214376.0,
      "step": 11185
    },
    {
      "entropy": 5.484990119934082,
      "epoch": 1.0150580551523947,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004904077323073268,
      "loss": 5.1377,
      "mean_token_accuracy": 0.19592861980199813,
      "num_tokens": 23224304.0,
      "step": 11190
    },
    {
      "entropy": 5.456968784332275,
      "epoch": 1.015511611030479,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004903983832131633,
      "loss": 5.1484,
      "mean_token_accuracy": 0.1907399609684944,
      "num_tokens": 23235324.0,
      "step": 11195
    },
    {
      "entropy": 5.451511812210083,
      "epoch": 1.0159651669085632,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004903890296645063,
      "loss": 5.1226,
      "mean_token_accuracy": 0.19383998960256577,
      "num_tokens": 23245774.0,
      "step": 11200
    },
    {
      "entropy": 5.491146993637085,
      "epoch": 1.0164187227866472,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004903796716615495,
      "loss": 5.0392,
      "mean_token_accuracy": 0.19635346233844758,
      "num_tokens": 23256274.0,
      "step": 11205
    },
    {
      "entropy": 5.450030946731568,
      "epoch": 1.0168722786647315,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004903703092044861,
      "loss": 5.081,
      "mean_token_accuracy": 0.19349735528230666,
      "num_tokens": 23266304.0,
      "step": 11210
    },
    {
      "entropy": 5.372950410842895,
      "epoch": 1.0173258345428158,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903609422935097,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20075019150972367,
      "num_tokens": 23276552.0,
      "step": 11215
    },
    {
      "entropy": 5.465046977996826,
      "epoch": 1.0177793904208998,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004903515709288143,
      "loss": 5.1085,
      "mean_token_accuracy": 0.18825433999300004,
      "num_tokens": 23287799.0,
      "step": 11220
    },
    {
      "entropy": 5.4047778129577635,
      "epoch": 1.018232946298984,
      "grad_norm": 1.0,
      "learning_rate": 0.0004903421951105933,
      "loss": 4.9574,
      "mean_token_accuracy": 0.20188473016023636,
      "num_tokens": 23297958.0,
      "step": 11225
    },
    {
      "entropy": 5.429022884368896,
      "epoch": 1.0186865021770681,
      "grad_norm": 1.125,
      "learning_rate": 0.0004903328148390411,
      "loss": 5.158,
      "mean_token_accuracy": 0.18890633732080458,
      "num_tokens": 23307523.0,
      "step": 11230
    },
    {
      "entropy": 5.496990394592285,
      "epoch": 1.0191400580551524,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004903234301143513,
      "loss": 5.1286,
      "mean_token_accuracy": 0.19909049868583678,
      "num_tokens": 23317970.0,
      "step": 11235
    },
    {
      "entropy": 5.454283809661865,
      "epoch": 1.0195936139332367,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004903140409367181,
      "loss": 5.1051,
      "mean_token_accuracy": 0.19640181660652162,
      "num_tokens": 23328612.0,
      "step": 11240
    },
    {
      "entropy": 5.433233976364136,
      "epoch": 1.0200471698113207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004903046473063357,
      "loss": 5.0575,
      "mean_token_accuracy": 0.19017990976572036,
      "num_tokens": 23338104.0,
      "step": 11245
    },
    {
      "entropy": 5.4058473110198975,
      "epoch": 1.020500725689405,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004902952492233983,
      "loss": 4.987,
      "mean_token_accuracy": 0.20831747949123383,
      "num_tokens": 23348105.0,
      "step": 11250
    },
    {
      "entropy": 5.413752603530884,
      "epoch": 1.020954281567489,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004902858466881002,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20240961015224457,
      "num_tokens": 23357150.0,
      "step": 11255
    },
    {
      "entropy": 5.453649282455444,
      "epoch": 1.0214078374455733,
      "grad_norm": 1.015625,
      "learning_rate": 0.000490276439700636,
      "loss": 5.1546,
      "mean_token_accuracy": 0.19585868269205092,
      "num_tokens": 23367223.0,
      "step": 11260
    },
    {
      "entropy": 5.470883369445801,
      "epoch": 1.0218613933236576,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004902670282612002,
      "loss": 5.0934,
      "mean_token_accuracy": 0.1947690561413765,
      "num_tokens": 23377664.0,
      "step": 11265
    },
    {
      "entropy": 5.406785202026367,
      "epoch": 1.0223149492017416,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004902576123699873,
      "loss": 4.9868,
      "mean_token_accuracy": 0.19861319065093994,
      "num_tokens": 23388057.0,
      "step": 11270
    },
    {
      "entropy": 5.334575366973877,
      "epoch": 1.0227685050798259,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004902481920271922,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2003134861588478,
      "num_tokens": 23397808.0,
      "step": 11275
    },
    {
      "entropy": 5.366316890716552,
      "epoch": 1.02322206095791,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004902387672330097,
      "loss": 5.0174,
      "mean_token_accuracy": 0.19368760138750077,
      "num_tokens": 23406979.0,
      "step": 11280
    },
    {
      "entropy": 5.426238918304444,
      "epoch": 1.0236756168359942,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004902293379876345,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19691926538944243,
      "num_tokens": 23416545.0,
      "step": 11285
    },
    {
      "entropy": 5.450316762924194,
      "epoch": 1.0241291727140784,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004902199042912618,
      "loss": 5.0831,
      "mean_token_accuracy": 0.19223443567752838,
      "num_tokens": 23427464.0,
      "step": 11290
    },
    {
      "entropy": 5.466953468322754,
      "epoch": 1.0245827285921625,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004902104661440868,
      "loss": 5.0991,
      "mean_token_accuracy": 0.1991693541407585,
      "num_tokens": 23436690.0,
      "step": 11295
    },
    {
      "entropy": 5.4285167217254635,
      "epoch": 1.0250362844702468,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902010235463042,
      "loss": 5.0272,
      "mean_token_accuracy": 0.19876836538314818,
      "num_tokens": 23446701.0,
      "step": 11300
    },
    {
      "entropy": 5.519870901107788,
      "epoch": 1.025489840348331,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004901915764981098,
      "loss": 5.131,
      "mean_token_accuracy": 0.189468115568161,
      "num_tokens": 23456966.0,
      "step": 11305
    },
    {
      "entropy": 5.4783522605896,
      "epoch": 1.025943396226415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004901821249996988,
      "loss": 5.0943,
      "mean_token_accuracy": 0.2010599821805954,
      "num_tokens": 23467300.0,
      "step": 11310
    },
    {
      "entropy": 5.496423721313477,
      "epoch": 1.0263969521044993,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004901726690512665,
      "loss": 5.1383,
      "mean_token_accuracy": 0.19087745994329453,
      "num_tokens": 23478457.0,
      "step": 11315
    },
    {
      "entropy": 5.416741752624512,
      "epoch": 1.0268505079825834,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004901632086530087,
      "loss": 5.0217,
      "mean_token_accuracy": 0.19998096525669098,
      "num_tokens": 23488066.0,
      "step": 11320
    },
    {
      "entropy": 5.435812044143677,
      "epoch": 1.0273040638606676,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004901537438051207,
      "loss": 5.1075,
      "mean_token_accuracy": 0.19100387990474701,
      "num_tokens": 23498856.0,
      "step": 11325
    },
    {
      "entropy": 5.46122555732727,
      "epoch": 1.027757619738752,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901442745077986,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19592872262001038,
      "num_tokens": 23509845.0,
      "step": 11330
    },
    {
      "entropy": 5.523153638839721,
      "epoch": 1.028211175616836,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004901348007612381,
      "loss": 5.1921,
      "mean_token_accuracy": 0.19549354612827302,
      "num_tokens": 23520322.0,
      "step": 11335
    },
    {
      "entropy": 5.4768232822418215,
      "epoch": 1.0286647314949202,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901253225656351,
      "loss": 5.1097,
      "mean_token_accuracy": 0.19431154578924179,
      "num_tokens": 23530030.0,
      "step": 11340
    },
    {
      "entropy": 5.5160911083221436,
      "epoch": 1.0291182873730043,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004901158399211855,
      "loss": 5.1606,
      "mean_token_accuracy": 0.19508347511291504,
      "num_tokens": 23541063.0,
      "step": 11345
    },
    {
      "entropy": 5.4110572814941404,
      "epoch": 1.0295718432510885,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004901063528280856,
      "loss": 5.0839,
      "mean_token_accuracy": 0.1935800477862358,
      "num_tokens": 23551693.0,
      "step": 11350
    },
    {
      "entropy": 5.448520803451538,
      "epoch": 1.0300253991291728,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004900968612865314,
      "loss": 5.11,
      "mean_token_accuracy": 0.20178464502096177,
      "num_tokens": 23561758.0,
      "step": 11355
    },
    {
      "entropy": 5.408980846405029,
      "epoch": 1.0304789550072568,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900873652967195,
      "loss": 5.0612,
      "mean_token_accuracy": 0.19118053913116456,
      "num_tokens": 23571912.0,
      "step": 11360
    },
    {
      "entropy": 5.424785518646241,
      "epoch": 1.030932510885341,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004900778648588459,
      "loss": 5.0172,
      "mean_token_accuracy": 0.20222554206848145,
      "num_tokens": 23582600.0,
      "step": 11365
    },
    {
      "entropy": 5.40379490852356,
      "epoch": 1.0313860667634251,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004900683599731073,
      "loss": 4.9951,
      "mean_token_accuracy": 0.2006585642695427,
      "num_tokens": 23592761.0,
      "step": 11370
    },
    {
      "entropy": 5.512581634521484,
      "epoch": 1.0318396226415094,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004900588506397002,
      "loss": 5.1455,
      "mean_token_accuracy": 0.19584969729185103,
      "num_tokens": 23603119.0,
      "step": 11375
    },
    {
      "entropy": 5.387804555892944,
      "epoch": 1.0322931785195937,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004900493368588213,
      "loss": 5.048,
      "mean_token_accuracy": 0.20134272575378417,
      "num_tokens": 23613916.0,
      "step": 11380
    },
    {
      "entropy": 5.507238960266113,
      "epoch": 1.0327467343976777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004900398186306672,
      "loss": 5.1862,
      "mean_token_accuracy": 0.19214550256729127,
      "num_tokens": 23623956.0,
      "step": 11385
    },
    {
      "entropy": 5.428924179077148,
      "epoch": 1.033200290275762,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000490030295955435,
      "loss": 5.0863,
      "mean_token_accuracy": 0.1945862203836441,
      "num_tokens": 23634666.0,
      "step": 11390
    },
    {
      "entropy": 5.396199321746826,
      "epoch": 1.0336538461538463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004900207688333215,
      "loss": 5.0342,
      "mean_token_accuracy": 0.19014710038900376,
      "num_tokens": 23645601.0,
      "step": 11395
    },
    {
      "entropy": 5.490629100799561,
      "epoch": 1.0341074020319303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004900112372645236,
      "loss": 5.1063,
      "mean_token_accuracy": 0.19617590457201003,
      "num_tokens": 23656339.0,
      "step": 11400
    },
    {
      "entropy": 5.4447340965271,
      "epoch": 1.0345609579100146,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004900017012492386,
      "loss": 5.0659,
      "mean_token_accuracy": 0.1966501459479332,
      "num_tokens": 23667433.0,
      "step": 11405
    },
    {
      "entropy": 5.4997495174407955,
      "epoch": 1.0350145137880986,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004899921607876634,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19640792310237884,
      "num_tokens": 23678186.0,
      "step": 11410
    },
    {
      "entropy": 5.467590475082398,
      "epoch": 1.0354680696661829,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899826158799958,
      "loss": 5.0607,
      "mean_token_accuracy": 0.1908168226480484,
      "num_tokens": 23688317.0,
      "step": 11415
    },
    {
      "entropy": 5.428335046768188,
      "epoch": 1.0359216255442671,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004899730665264328,
      "loss": 5.0081,
      "mean_token_accuracy": 0.19815969318151475,
      "num_tokens": 23697752.0,
      "step": 11420
    },
    {
      "entropy": 5.3815539360046385,
      "epoch": 1.0363751814223512,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004899635127271721,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20386362820863724,
      "num_tokens": 23707803.0,
      "step": 11425
    },
    {
      "entropy": 5.378336381912232,
      "epoch": 1.0368287373004355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004899539544824111,
      "loss": 5.0019,
      "mean_token_accuracy": 0.20478526055812835,
      "num_tokens": 23717630.0,
      "step": 11430
    },
    {
      "entropy": 5.455833387374878,
      "epoch": 1.0372822931785195,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004899443917923476,
      "loss": 5.1178,
      "mean_token_accuracy": 0.19574906826019287,
      "num_tokens": 23728836.0,
      "step": 11435
    },
    {
      "entropy": 5.418558263778687,
      "epoch": 1.0377358490566038,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004899348246571792,
      "loss": 5.1006,
      "mean_token_accuracy": 0.19457397907972335,
      "num_tokens": 23740145.0,
      "step": 11440
    },
    {
      "entropy": 5.45527195930481,
      "epoch": 1.038189404934688,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000489925253077104,
      "loss": 5.0693,
      "mean_token_accuracy": 0.19532546550035476,
      "num_tokens": 23751162.0,
      "step": 11445
    },
    {
      "entropy": 5.454708051681519,
      "epoch": 1.038642960812772,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899156770523196,
      "loss": 5.0814,
      "mean_token_accuracy": 0.1966884195804596,
      "num_tokens": 23761634.0,
      "step": 11450
    },
    {
      "entropy": 5.4408070087432865,
      "epoch": 1.0390965166908563,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004899060965830244,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20172560662031175,
      "num_tokens": 23772483.0,
      "step": 11455
    },
    {
      "entropy": 5.417911100387573,
      "epoch": 1.0395500725689404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004898965116694162,
      "loss": 5.0371,
      "mean_token_accuracy": 0.1978759214282036,
      "num_tokens": 23782298.0,
      "step": 11460
    },
    {
      "entropy": 5.4759174346923825,
      "epoch": 1.0400036284470247,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898869223116935,
      "loss": 5.1069,
      "mean_token_accuracy": 0.19060955047607422,
      "num_tokens": 23792492.0,
      "step": 11465
    },
    {
      "entropy": 5.512869071960449,
      "epoch": 1.040457184325109,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004898773285100546,
      "loss": 5.2549,
      "mean_token_accuracy": 0.1875505417585373,
      "num_tokens": 23802999.0,
      "step": 11470
    },
    {
      "entropy": 5.4411492347717285,
      "epoch": 1.040910740203193,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004898677302646975,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19900441765785218,
      "num_tokens": 23812541.0,
      "step": 11475
    },
    {
      "entropy": 5.45083065032959,
      "epoch": 1.0413642960812772,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004898581275758211,
      "loss": 5.1469,
      "mean_token_accuracy": 0.18596832305192948,
      "num_tokens": 23823244.0,
      "step": 11480
    },
    {
      "entropy": 5.434124565124511,
      "epoch": 1.0418178519593615,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898485204436239,
      "loss": 4.9841,
      "mean_token_accuracy": 0.202264666557312,
      "num_tokens": 23832346.0,
      "step": 11485
    },
    {
      "entropy": 5.484871912002563,
      "epoch": 1.0422714078374455,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004898389088683047,
      "loss": 5.117,
      "mean_token_accuracy": 0.19769508838653566,
      "num_tokens": 23842554.0,
      "step": 11490
    },
    {
      "entropy": 5.402813959121704,
      "epoch": 1.0427249637155298,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004898292928500619,
      "loss": 4.9592,
      "mean_token_accuracy": 0.20696790814399718,
      "num_tokens": 23852938.0,
      "step": 11495
    },
    {
      "entropy": 5.412786197662354,
      "epoch": 1.0431785195936139,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004898196723890947,
      "loss": 4.9841,
      "mean_token_accuracy": 0.20367277711629866,
      "num_tokens": 23862367.0,
      "step": 11500
    },
    {
      "entropy": 5.41459174156189,
      "epoch": 1.0436320754716981,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004898100474856019,
      "loss": 5.1084,
      "mean_token_accuracy": 0.19728990495204926,
      "num_tokens": 23873249.0,
      "step": 11505
    },
    {
      "entropy": 5.454684019088745,
      "epoch": 1.0440856313497824,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004898004181397824,
      "loss": 5.1245,
      "mean_token_accuracy": 0.18645377606153488,
      "num_tokens": 23883830.0,
      "step": 11510
    },
    {
      "entropy": 5.490788984298706,
      "epoch": 1.0445391872278664,
      "grad_norm": 1.0,
      "learning_rate": 0.0004897907843518358,
      "loss": 5.033,
      "mean_token_accuracy": 0.1962950274348259,
      "num_tokens": 23894359.0,
      "step": 11515
    },
    {
      "entropy": 5.436550092697144,
      "epoch": 1.0449927431059507,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004897811461219609,
      "loss": 5.042,
      "mean_token_accuracy": 0.19545938819646835,
      "num_tokens": 23903779.0,
      "step": 11520
    },
    {
      "entropy": 5.438144254684448,
      "epoch": 1.0454462989840347,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004897715034503571,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19297188222408296,
      "num_tokens": 23914523.0,
      "step": 11525
    },
    {
      "entropy": 5.4399497509002686,
      "epoch": 1.045899854862119,
      "grad_norm": 0.96875,
      "learning_rate": 0.000489761856337224,
      "loss": 5.08,
      "mean_token_accuracy": 0.19704519659280778,
      "num_tokens": 23924996.0,
      "step": 11530
    },
    {
      "entropy": 5.451943254470825,
      "epoch": 1.0463534107402033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004897522047827609,
      "loss": 5.0481,
      "mean_token_accuracy": 0.2018464282155037,
      "num_tokens": 23935843.0,
      "step": 11535
    },
    {
      "entropy": 5.422876787185669,
      "epoch": 1.0468069666182873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004897425487871676,
      "loss": 5.009,
      "mean_token_accuracy": 0.1973980635404587,
      "num_tokens": 23945044.0,
      "step": 11540
    },
    {
      "entropy": 5.415144586563111,
      "epoch": 1.0472605224963716,
      "grad_norm": 1.0,
      "learning_rate": 0.0004897328883506436,
      "loss": 5.0648,
      "mean_token_accuracy": 0.19418827295303345,
      "num_tokens": 23955933.0,
      "step": 11545
    },
    {
      "entropy": 5.455891418457031,
      "epoch": 1.0477140783744556,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004897232234733887,
      "loss": 5.1506,
      "mean_token_accuracy": 0.18856869488954545,
      "num_tokens": 23966320.0,
      "step": 11550
    },
    {
      "entropy": 5.4573126316070555,
      "epoch": 1.04816763425254,
      "grad_norm": 1.0,
      "learning_rate": 0.0004897135541556028,
      "loss": 5.0344,
      "mean_token_accuracy": 0.19937969446182252,
      "num_tokens": 23976233.0,
      "step": 11555
    },
    {
      "entropy": 5.468496084213257,
      "epoch": 1.0486211901306242,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000489703880397486,
      "loss": 5.0975,
      "mean_token_accuracy": 0.19293475151062012,
      "num_tokens": 23987854.0,
      "step": 11560
    },
    {
      "entropy": 5.446594524383545,
      "epoch": 1.0490747460087082,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004896942021992382,
      "loss": 5.1068,
      "mean_token_accuracy": 0.1926787257194519,
      "num_tokens": 23997973.0,
      "step": 11565
    },
    {
      "entropy": 5.448502826690674,
      "epoch": 1.0495283018867925,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004896845195610595,
      "loss": 5.0942,
      "mean_token_accuracy": 0.19503124207258224,
      "num_tokens": 24007313.0,
      "step": 11570
    },
    {
      "entropy": 5.430074739456177,
      "epoch": 1.0499818577648767,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004896748324831504,
      "loss": 5.1129,
      "mean_token_accuracy": 0.1907646909356117,
      "num_tokens": 24018254.0,
      "step": 11575
    },
    {
      "entropy": 5.468389320373535,
      "epoch": 1.0504354136429608,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004896651409657109,
      "loss": 5.1175,
      "mean_token_accuracy": 0.1924986645579338,
      "num_tokens": 24029799.0,
      "step": 11580
    },
    {
      "entropy": 5.381548690795898,
      "epoch": 1.050888969521045,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004896554450089417,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20224028080701828,
      "num_tokens": 24040972.0,
      "step": 11585
    },
    {
      "entropy": 5.401288747787476,
      "epoch": 1.051342525399129,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004896457446130431,
      "loss": 4.9488,
      "mean_token_accuracy": 0.2015111193060875,
      "num_tokens": 24050781.0,
      "step": 11590
    },
    {
      "entropy": 5.395908451080322,
      "epoch": 1.0517960812772134,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004896360397782159,
      "loss": 5.0786,
      "mean_token_accuracy": 0.18981412649154664,
      "num_tokens": 24060184.0,
      "step": 11595
    },
    {
      "entropy": 5.503989028930664,
      "epoch": 1.0522496371552976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004896263305046607,
      "loss": 5.1437,
      "mean_token_accuracy": 0.19646002352237701,
      "num_tokens": 24070777.0,
      "step": 11600
    },
    {
      "entropy": 5.475661325454712,
      "epoch": 1.0527031930333817,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004896166167925782,
      "loss": 5.169,
      "mean_token_accuracy": 0.19187893271446227,
      "num_tokens": 24080878.0,
      "step": 11605
    },
    {
      "entropy": 5.473572683334351,
      "epoch": 1.053156748911466,
      "grad_norm": 1.0,
      "learning_rate": 0.0004896068986421693,
      "loss": 5.1247,
      "mean_token_accuracy": 0.2013508141040802,
      "num_tokens": 24090038.0,
      "step": 11610
    },
    {
      "entropy": 5.542226314544678,
      "epoch": 1.05361030478955,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895971760536351,
      "loss": 5.148,
      "mean_token_accuracy": 0.1930503010749817,
      "num_tokens": 24100425.0,
      "step": 11615
    },
    {
      "entropy": 5.510891008377075,
      "epoch": 1.0540638606676342,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004895874490271768,
      "loss": 5.2155,
      "mean_token_accuracy": 0.18790028393268585,
      "num_tokens": 24112182.0,
      "step": 11620
    },
    {
      "entropy": 5.470790004730224,
      "epoch": 1.0545174165457185,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004895777175629952,
      "loss": 5.0586,
      "mean_token_accuracy": 0.19576068520545958,
      "num_tokens": 24122182.0,
      "step": 11625
    },
    {
      "entropy": 5.5000087261199955,
      "epoch": 1.0549709724238026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004895679816612917,
      "loss": 5.0537,
      "mean_token_accuracy": 0.2022800475358963,
      "num_tokens": 24131930.0,
      "step": 11630
    },
    {
      "entropy": 5.490646553039551,
      "epoch": 1.0554245283018868,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004895582413222678,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20672089010477065,
      "num_tokens": 24141575.0,
      "step": 11635
    },
    {
      "entropy": 5.49963755607605,
      "epoch": 1.0558780841799709,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004895484965461247,
      "loss": 5.1974,
      "mean_token_accuracy": 0.1874408856034279,
      "num_tokens": 24151720.0,
      "step": 11640
    },
    {
      "entropy": 5.440736627578735,
      "epoch": 1.0563316400580551,
      "grad_norm": 1.0625,
      "learning_rate": 0.000489538747333064,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19491846710443497,
      "num_tokens": 24161881.0,
      "step": 11645
    },
    {
      "entropy": 5.440998458862305,
      "epoch": 1.0567851959361394,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004895289936832874,
      "loss": 5.0397,
      "mean_token_accuracy": 0.2015311449766159,
      "num_tokens": 24171474.0,
      "step": 11650
    },
    {
      "entropy": 5.508123254776001,
      "epoch": 1.0572387518142234,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004895192355969966,
      "loss": 5.1487,
      "mean_token_accuracy": 0.19003770053386687,
      "num_tokens": 24182186.0,
      "step": 11655
    },
    {
      "entropy": 5.491469430923462,
      "epoch": 1.0576923076923077,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004895094730743932,
      "loss": 5.0807,
      "mean_token_accuracy": 0.1989141210913658,
      "num_tokens": 24192318.0,
      "step": 11660
    },
    {
      "entropy": 5.486510848999023,
      "epoch": 1.058145863570392,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894997061156793,
      "loss": 5.1427,
      "mean_token_accuracy": 0.19250682592391968,
      "num_tokens": 24202918.0,
      "step": 11665
    },
    {
      "entropy": 5.416793775558472,
      "epoch": 1.058599419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004894899347210567,
      "loss": 5.0251,
      "mean_token_accuracy": 0.19752568751573563,
      "num_tokens": 24213218.0,
      "step": 11670
    },
    {
      "entropy": 5.421510839462281,
      "epoch": 1.0590529753265603,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004894801588907278,
      "loss": 5.0285,
      "mean_token_accuracy": 0.19644063711166382,
      "num_tokens": 24223229.0,
      "step": 11675
    },
    {
      "entropy": 5.5234157085418705,
      "epoch": 1.0595065312046443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004894703786248944,
      "loss": 5.0986,
      "mean_token_accuracy": 0.1940212830901146,
      "num_tokens": 24233637.0,
      "step": 11680
    },
    {
      "entropy": 5.477429342269898,
      "epoch": 1.0599600870827286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000489460593923759,
      "loss": 5.1354,
      "mean_token_accuracy": 0.19680089503526688,
      "num_tokens": 24244251.0,
      "step": 11685
    },
    {
      "entropy": 5.442391777038575,
      "epoch": 1.0604136429608129,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004894508047875237,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19454250782728194,
      "num_tokens": 24254380.0,
      "step": 11690
    },
    {
      "entropy": 5.318802404403686,
      "epoch": 1.060867198838897,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004894410112163913,
      "loss": 4.9794,
      "mean_token_accuracy": 0.19969397485256196,
      "num_tokens": 24264841.0,
      "step": 11695
    },
    {
      "entropy": 5.373764562606811,
      "epoch": 1.0613207547169812,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004894312132105641,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20056951344013213,
      "num_tokens": 24274383.0,
      "step": 11700
    },
    {
      "entropy": 5.514595413208008,
      "epoch": 1.0617743105950652,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004894214107702447,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19603312909603118,
      "num_tokens": 24284140.0,
      "step": 11705
    },
    {
      "entropy": 5.563332653045654,
      "epoch": 1.0622278664731495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000489411603895636,
      "loss": 5.1639,
      "mean_token_accuracy": 0.19430688321590422,
      "num_tokens": 24295044.0,
      "step": 11710
    },
    {
      "entropy": 5.4479235172271725,
      "epoch": 1.0626814223512338,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004894017925869406,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19257737845182418,
      "num_tokens": 24305080.0,
      "step": 11715
    },
    {
      "entropy": 5.455956172943115,
      "epoch": 1.0631349782293178,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893919768443615,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19278850853443147,
      "num_tokens": 24316306.0,
      "step": 11720
    },
    {
      "entropy": 5.403377342224121,
      "epoch": 1.063588534107402,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004893821566681017,
      "loss": 5.0886,
      "mean_token_accuracy": 0.19761644750833512,
      "num_tokens": 24326740.0,
      "step": 11725
    },
    {
      "entropy": 5.464867639541626,
      "epoch": 1.0640420899854863,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004893723320583644,
      "loss": 5.1241,
      "mean_token_accuracy": 0.18957178741693498,
      "num_tokens": 24337621.0,
      "step": 11730
    },
    {
      "entropy": 5.38532977104187,
      "epoch": 1.0644956458635704,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004893625030153525,
      "loss": 4.9783,
      "mean_token_accuracy": 0.20226975977420808,
      "num_tokens": 24347732.0,
      "step": 11735
    },
    {
      "entropy": 5.4007179737091064,
      "epoch": 1.0649492017416546,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004893526695392694,
      "loss": 4.9978,
      "mean_token_accuracy": 0.2070948839187622,
      "num_tokens": 24359173.0,
      "step": 11740
    },
    {
      "entropy": 5.41707181930542,
      "epoch": 1.0654027576197387,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004893428316303185,
      "loss": 5.0104,
      "mean_token_accuracy": 0.19994874447584152,
      "num_tokens": 24369465.0,
      "step": 11745
    },
    {
      "entropy": 5.376338243484497,
      "epoch": 1.065856313497823,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004893329892887031,
      "loss": 5.0463,
      "mean_token_accuracy": 0.19964333474636078,
      "num_tokens": 24378899.0,
      "step": 11750
    },
    {
      "entropy": 5.45647497177124,
      "epoch": 1.0663098693759072,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004893231425146271,
      "loss": 5.1578,
      "mean_token_accuracy": 0.19202471226453782,
      "num_tokens": 24388919.0,
      "step": 11755
    },
    {
      "entropy": 5.399653387069702,
      "epoch": 1.0667634252539913,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004893132913082937,
      "loss": 5.0326,
      "mean_token_accuracy": 0.19771426767110825,
      "num_tokens": 24399632.0,
      "step": 11760
    },
    {
      "entropy": 5.473707294464111,
      "epoch": 1.0672169811320755,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004893034356699069,
      "loss": 5.071,
      "mean_token_accuracy": 0.19982456862926484,
      "num_tokens": 24410817.0,
      "step": 11765
    },
    {
      "entropy": 5.410086917877197,
      "epoch": 1.0676705370101596,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004892935755996705,
      "loss": 5.0781,
      "mean_token_accuracy": 0.1969650834798813,
      "num_tokens": 24421058.0,
      "step": 11770
    },
    {
      "entropy": 5.388855934143066,
      "epoch": 1.0681240928882438,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004892837110977883,
      "loss": 4.9924,
      "mean_token_accuracy": 0.19559596329927445,
      "num_tokens": 24431522.0,
      "step": 11775
    },
    {
      "entropy": 5.419882726669312,
      "epoch": 1.068577648766328,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004892738421644644,
      "loss": 5.0341,
      "mean_token_accuracy": 0.19629716575145723,
      "num_tokens": 24441746.0,
      "step": 11780
    },
    {
      "entropy": 5.366162490844727,
      "epoch": 1.0690312046444121,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004892639687999027,
      "loss": 4.95,
      "mean_token_accuracy": 0.20535225570201873,
      "num_tokens": 24451916.0,
      "step": 11785
    },
    {
      "entropy": 5.41597089767456,
      "epoch": 1.0694847605224964,
      "grad_norm": 1.0,
      "learning_rate": 0.0004892540910043076,
      "loss": 5.0739,
      "mean_token_accuracy": 0.20218056589365005,
      "num_tokens": 24461935.0,
      "step": 11790
    },
    {
      "entropy": 5.413613605499267,
      "epoch": 1.0699383164005805,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004892442087778833,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20131217688322067,
      "num_tokens": 24472437.0,
      "step": 11795
    },
    {
      "entropy": 5.4258464813232425,
      "epoch": 1.0703918722786647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004892343221208342,
      "loss": 5.0914,
      "mean_token_accuracy": 0.1990344926714897,
      "num_tokens": 24481726.0,
      "step": 11800
    },
    {
      "entropy": 5.487134218215942,
      "epoch": 1.070845428156749,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004892244310333646,
      "loss": 5.1581,
      "mean_token_accuracy": 0.19333946853876113,
      "num_tokens": 24492340.0,
      "step": 11805
    },
    {
      "entropy": 5.4427143096923825,
      "epoch": 1.071298984034833,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004892145355156793,
      "loss": 5.0169,
      "mean_token_accuracy": 0.20001528561115264,
      "num_tokens": 24502138.0,
      "step": 11810
    },
    {
      "entropy": 5.435838603973389,
      "epoch": 1.0717525399129173,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004892046355679829,
      "loss": 5.0742,
      "mean_token_accuracy": 0.19548026472330093,
      "num_tokens": 24512723.0,
      "step": 11815
    },
    {
      "entropy": 5.462838745117187,
      "epoch": 1.0722060957910013,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004891947311904799,
      "loss": 5.1137,
      "mean_token_accuracy": 0.19110726565122604,
      "num_tokens": 24523766.0,
      "step": 11820
    },
    {
      "entropy": 5.464926099777221,
      "epoch": 1.0726596516690856,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004891848223833753,
      "loss": 5.0878,
      "mean_token_accuracy": 0.1985626518726349,
      "num_tokens": 24534571.0,
      "step": 11825
    },
    {
      "entropy": 5.450968503952026,
      "epoch": 1.0731132075471699,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489174909146874,
      "loss": 5.0981,
      "mean_token_accuracy": 0.20022153854370117,
      "num_tokens": 24545265.0,
      "step": 11830
    },
    {
      "entropy": 5.505740213394165,
      "epoch": 1.073566763425254,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004891649914811811,
      "loss": 5.1771,
      "mean_token_accuracy": 0.1910492166876793,
      "num_tokens": 24555524.0,
      "step": 11835
    },
    {
      "entropy": 5.40087103843689,
      "epoch": 1.0740203193033382,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004891550693865014,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2117801010608673,
      "num_tokens": 24565420.0,
      "step": 11840
    },
    {
      "entropy": 5.4563820362091064,
      "epoch": 1.0744738751814222,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004891451428630405,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19624711722135543,
      "num_tokens": 24576794.0,
      "step": 11845
    },
    {
      "entropy": 5.445266962051392,
      "epoch": 1.0749274310595065,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004891352119110033,
      "loss": 5.0809,
      "mean_token_accuracy": 0.1940919429063797,
      "num_tokens": 24586320.0,
      "step": 11850
    },
    {
      "entropy": 5.3934401512146,
      "epoch": 1.0753809869375908,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891252765305954,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20423324257135392,
      "num_tokens": 24596524.0,
      "step": 11855
    },
    {
      "entropy": 5.46302227973938,
      "epoch": 1.0758345428156748,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004891153367220222,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20729490667581557,
      "num_tokens": 24605877.0,
      "step": 11860
    },
    {
      "entropy": 5.5180091381073,
      "epoch": 1.076288098693759,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004891053924854894,
      "loss": 5.2121,
      "mean_token_accuracy": 0.1921609804034233,
      "num_tokens": 24616719.0,
      "step": 11865
    },
    {
      "entropy": 5.402059555053711,
      "epoch": 1.0767416545718433,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004890954438212024,
      "loss": 4.96,
      "mean_token_accuracy": 0.20680669397115709,
      "num_tokens": 24626882.0,
      "step": 11870
    },
    {
      "entropy": 5.457244205474853,
      "epoch": 1.0771952104499274,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890854907293671,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19212204962968826,
      "num_tokens": 24637884.0,
      "step": 11875
    },
    {
      "entropy": 5.399580287933349,
      "epoch": 1.0776487663280117,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004890755332101893,
      "loss": 5.0726,
      "mean_token_accuracy": 0.19744679927825928,
      "num_tokens": 24647389.0,
      "step": 11880
    },
    {
      "entropy": 5.430721712112427,
      "epoch": 1.0781023222060957,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004890655712638749,
      "loss": 5.0856,
      "mean_token_accuracy": 0.1923668220639229,
      "num_tokens": 24657595.0,
      "step": 11885
    },
    {
      "entropy": 5.4359862327575685,
      "epoch": 1.07855587808418,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048905560489063,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20309585630893706,
      "num_tokens": 24667683.0,
      "step": 11890
    },
    {
      "entropy": 5.470280981063842,
      "epoch": 1.0790094339622642,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004890456340906605,
      "loss": 5.1483,
      "mean_token_accuracy": 0.1932810515165329,
      "num_tokens": 24678041.0,
      "step": 11895
    },
    {
      "entropy": 5.421089029312133,
      "epoch": 1.0794629898403483,
      "grad_norm": 1.0,
      "learning_rate": 0.0004890356588641729,
      "loss": 5.1296,
      "mean_token_accuracy": 0.20076433718204498,
      "num_tokens": 24688569.0,
      "step": 11900
    },
    {
      "entropy": 5.457909011840821,
      "epoch": 1.0799165457184325,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004890256792113733,
      "loss": 5.0963,
      "mean_token_accuracy": 0.19976598918437957,
      "num_tokens": 24699913.0,
      "step": 11905
    },
    {
      "entropy": 5.5236265659332275,
      "epoch": 1.0803701015965168,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004890156951324681,
      "loss": 5.1152,
      "mean_token_accuracy": 0.1960355222225189,
      "num_tokens": 24710682.0,
      "step": 11910
    },
    {
      "entropy": 5.541093349456787,
      "epoch": 1.0808236574746009,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004890057066276636,
      "loss": 5.1933,
      "mean_token_accuracy": 0.18897853046655655,
      "num_tokens": 24721087.0,
      "step": 11915
    },
    {
      "entropy": 5.404382562637329,
      "epoch": 1.0812772133526851,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004889957136971668,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2035218745470047,
      "num_tokens": 24731323.0,
      "step": 11920
    },
    {
      "entropy": 5.405501651763916,
      "epoch": 1.0817307692307692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000488985716341184,
      "loss": 5.0808,
      "mean_token_accuracy": 0.19516118317842485,
      "num_tokens": 24742559.0,
      "step": 11925
    },
    {
      "entropy": 5.49205756187439,
      "epoch": 1.0821843251088534,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004889757145599222,
      "loss": 5.0874,
      "mean_token_accuracy": 0.19950402975082399,
      "num_tokens": 24752307.0,
      "step": 11930
    },
    {
      "entropy": 5.3561821460723875,
      "epoch": 1.0826378809869377,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000488965708353588,
      "loss": 5.0244,
      "mean_token_accuracy": 0.19089826941490173,
      "num_tokens": 24762792.0,
      "step": 11935
    },
    {
      "entropy": 5.476434326171875,
      "epoch": 1.0830914368650217,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004889556977223883,
      "loss": 5.1384,
      "mean_token_accuracy": 0.19093414396047592,
      "num_tokens": 24774377.0,
      "step": 11940
    },
    {
      "entropy": 5.4761061668396,
      "epoch": 1.083544992743106,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004889456826665304,
      "loss": 5.0632,
      "mean_token_accuracy": 0.19885472059249878,
      "num_tokens": 24785919.0,
      "step": 11945
    },
    {
      "entropy": 5.484697580337524,
      "epoch": 1.08399854862119,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004889356631862211,
      "loss": 5.1019,
      "mean_token_accuracy": 0.1904277101159096,
      "num_tokens": 24796594.0,
      "step": 11950
    },
    {
      "entropy": 5.427489614486694,
      "epoch": 1.0844521044992743,
      "grad_norm": 0.953125,
      "learning_rate": 0.000488925639281668,
      "loss": 5.0979,
      "mean_token_accuracy": 0.1992164745926857,
      "num_tokens": 24807322.0,
      "step": 11955
    },
    {
      "entropy": 5.385541248321533,
      "epoch": 1.0849056603773586,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004889156109530781,
      "loss": 4.985,
      "mean_token_accuracy": 0.19826211035251617,
      "num_tokens": 24817593.0,
      "step": 11960
    },
    {
      "entropy": 5.427872514724731,
      "epoch": 1.0853592162554426,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004889055782006588,
      "loss": 4.9904,
      "mean_token_accuracy": 0.205525766313076,
      "num_tokens": 24827577.0,
      "step": 11965
    },
    {
      "entropy": 5.401228189468384,
      "epoch": 1.085812772133527,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004888955410246176,
      "loss": 5.0438,
      "mean_token_accuracy": 0.19906434863805772,
      "num_tokens": 24838148.0,
      "step": 11970
    },
    {
      "entropy": 5.435687494277954,
      "epoch": 1.086266328011611,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004888854994251621,
      "loss": 4.995,
      "mean_token_accuracy": 0.2041756436228752,
      "num_tokens": 24848451.0,
      "step": 11975
    },
    {
      "entropy": 5.477515411376953,
      "epoch": 1.0867198838896952,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004888754534025,
      "loss": 5.1223,
      "mean_token_accuracy": 0.18711033910512925,
      "num_tokens": 24859619.0,
      "step": 11980
    },
    {
      "entropy": 5.460151338577271,
      "epoch": 1.0871734397677795,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004888654029568391,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1858536958694458,
      "num_tokens": 24869863.0,
      "step": 11985
    },
    {
      "entropy": 5.4626185417175295,
      "epoch": 1.0876269956458635,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004888553480883871,
      "loss": 5.1,
      "mean_token_accuracy": 0.18833720684051514,
      "num_tokens": 24880524.0,
      "step": 11990
    },
    {
      "entropy": 5.428664207458496,
      "epoch": 1.0880805515239478,
      "grad_norm": 0.890625,
      "learning_rate": 0.000488845288797352,
      "loss": 5.0514,
      "mean_token_accuracy": 0.19847525656223297,
      "num_tokens": 24891514.0,
      "step": 11995
    },
    {
      "entropy": 5.419693374633789,
      "epoch": 1.0885341074020318,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004888352250839418,
      "loss": 5.012,
      "mean_token_accuracy": 0.200258307158947,
      "num_tokens": 24900842.0,
      "step": 12000
    },
    {
      "epoch": 1.0885341074020318,
      "eval_entropy": 5.213240118803482,
      "eval_loss": 5.159305095672607,
      "eval_mean_token_accuracy": 0.20191680841923298,
      "eval_num_tokens": 24900842.0,
      "eval_runtime": 20.5739,
      "eval_samples_per_second": 1718.686,
      "eval_steps_per_second": 214.836,
      "step": 12000
    },
    {
      "entropy": 5.403762531280518,
      "epoch": 1.088987663280116,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004888251569483647,
      "loss": 5.0113,
      "mean_token_accuracy": 0.19987739324569703,
      "num_tokens": 24911845.0,
      "step": 12005
    },
    {
      "entropy": 5.361207818984985,
      "epoch": 1.0894412191582004,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004888150843908288,
      "loss": 5.0266,
      "mean_token_accuracy": 0.2027574062347412,
      "num_tokens": 24922211.0,
      "step": 12010
    },
    {
      "entropy": 5.489847183227539,
      "epoch": 1.0898947750362844,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004888050074115426,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19196625500917436,
      "num_tokens": 24932129.0,
      "step": 12015
    },
    {
      "entropy": 5.494090127944946,
      "epoch": 1.0903483309143687,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887949260107143,
      "loss": 5.1334,
      "mean_token_accuracy": 0.1945919945836067,
      "num_tokens": 24942869.0,
      "step": 12020
    },
    {
      "entropy": 5.415091466903687,
      "epoch": 1.0908018867924527,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004887848401885524,
      "loss": 5.0506,
      "mean_token_accuracy": 0.2002049967646599,
      "num_tokens": 24952621.0,
      "step": 12025
    },
    {
      "entropy": 5.534883880615235,
      "epoch": 1.091255442670537,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004887747499452656,
      "loss": 5.1812,
      "mean_token_accuracy": 0.1875045657157898,
      "num_tokens": 24963587.0,
      "step": 12030
    },
    {
      "entropy": 5.480429410934448,
      "epoch": 1.0917089985486212,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004887646552810624,
      "loss": 5.0306,
      "mean_token_accuracy": 0.19918842613697052,
      "num_tokens": 24974122.0,
      "step": 12035
    },
    {
      "entropy": 5.440608882904053,
      "epoch": 1.0921625544267053,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004887545561961517,
      "loss": 5.1152,
      "mean_token_accuracy": 0.19135283827781677,
      "num_tokens": 24985145.0,
      "step": 12040
    },
    {
      "entropy": 5.395934009552002,
      "epoch": 1.0926161103047896,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004887444526907424,
      "loss": 5.0364,
      "mean_token_accuracy": 0.19766356348991393,
      "num_tokens": 24995511.0,
      "step": 12045
    },
    {
      "entropy": 5.450332164764404,
      "epoch": 1.0930696661828738,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004887343447650432,
      "loss": 5.061,
      "mean_token_accuracy": 0.1917600452899933,
      "num_tokens": 25006193.0,
      "step": 12050
    },
    {
      "entropy": 5.46754002571106,
      "epoch": 1.0935232220609579,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004887242324192634,
      "loss": 5.0732,
      "mean_token_accuracy": 0.1974289357662201,
      "num_tokens": 25016444.0,
      "step": 12055
    },
    {
      "entropy": 5.467090892791748,
      "epoch": 1.0939767779390421,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004887141156536119,
      "loss": 5.0745,
      "mean_token_accuracy": 0.1957561507821083,
      "num_tokens": 25027898.0,
      "step": 12060
    },
    {
      "entropy": 5.428858900070191,
      "epoch": 1.0944303338171262,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004887039944682981,
      "loss": 4.9934,
      "mean_token_accuracy": 0.19940518587827682,
      "num_tokens": 25038201.0,
      "step": 12065
    },
    {
      "entropy": 5.53903865814209,
      "epoch": 1.0948838896952104,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886938688635313,
      "loss": 5.1722,
      "mean_token_accuracy": 0.18696350902318953,
      "num_tokens": 25048466.0,
      "step": 12070
    },
    {
      "entropy": 5.558661270141601,
      "epoch": 1.0953374455732947,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004886837388395207,
      "loss": 5.1272,
      "mean_token_accuracy": 0.197776260972023,
      "num_tokens": 25059064.0,
      "step": 12075
    },
    {
      "entropy": 5.527079391479492,
      "epoch": 1.0957910014513788,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000488673604396476,
      "loss": 5.2062,
      "mean_token_accuracy": 0.18161393254995345,
      "num_tokens": 25068834.0,
      "step": 12080
    },
    {
      "entropy": 5.487286567687988,
      "epoch": 1.096244557329463,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886634655346067,
      "loss": 5.1407,
      "mean_token_accuracy": 0.19002335518598557,
      "num_tokens": 25080058.0,
      "step": 12085
    },
    {
      "entropy": 5.416148138046265,
      "epoch": 1.0966981132075473,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886533222541225,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20018682926893233,
      "num_tokens": 25089751.0,
      "step": 12090
    },
    {
      "entropy": 5.448303937911987,
      "epoch": 1.0971516690856313,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004886431745552331,
      "loss": 5.0962,
      "mean_token_accuracy": 0.19731398224830626,
      "num_tokens": 25099055.0,
      "step": 12095
    },
    {
      "entropy": 5.538101100921631,
      "epoch": 1.0976052249637156,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004886330224381485,
      "loss": 5.1922,
      "mean_token_accuracy": 0.18713112622499467,
      "num_tokens": 25109793.0,
      "step": 12100
    },
    {
      "entropy": 5.500201988220215,
      "epoch": 1.0980587808417996,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004886228659030785,
      "loss": 5.092,
      "mean_token_accuracy": 0.19966764748096466,
      "num_tokens": 25119814.0,
      "step": 12105
    },
    {
      "entropy": 5.498597908020019,
      "epoch": 1.098512336719884,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004886127049502331,
      "loss": 5.1462,
      "mean_token_accuracy": 0.19277023226022721,
      "num_tokens": 25130747.0,
      "step": 12110
    },
    {
      "entropy": 5.426317882537842,
      "epoch": 1.0989658925979682,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004886025395798228,
      "loss": 5.0515,
      "mean_token_accuracy": 0.1940399870276451,
      "num_tokens": 25142235.0,
      "step": 12115
    },
    {
      "entropy": 5.343563508987427,
      "epoch": 1.0994194484760522,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885923697920573,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2048935681581497,
      "num_tokens": 25152047.0,
      "step": 12120
    },
    {
      "entropy": 5.41880784034729,
      "epoch": 1.0998730043541365,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004885821955871472,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20211029052734375,
      "num_tokens": 25162678.0,
      "step": 12125
    },
    {
      "entropy": 5.490183734893799,
      "epoch": 1.1003265602322205,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004885720169653028,
      "loss": 5.0378,
      "mean_token_accuracy": 0.19948583245277404,
      "num_tokens": 25172973.0,
      "step": 12130
    },
    {
      "entropy": 5.490416479110718,
      "epoch": 1.1007801161103048,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004885618339267348,
      "loss": 5.1721,
      "mean_token_accuracy": 0.1911870390176773,
      "num_tokens": 25183458.0,
      "step": 12135
    },
    {
      "entropy": 5.395343351364136,
      "epoch": 1.101233671988389,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004885516464716537,
      "loss": 5.0467,
      "mean_token_accuracy": 0.19833064079284668,
      "num_tokens": 25193827.0,
      "step": 12140
    },
    {
      "entropy": 5.504849767684936,
      "epoch": 1.101687227866473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048854145460027,
      "loss": 5.0821,
      "mean_token_accuracy": 0.18926733285188674,
      "num_tokens": 25204448.0,
      "step": 12145
    },
    {
      "entropy": 5.47561469078064,
      "epoch": 1.1021407837445574,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885312583127946,
      "loss": 5.0308,
      "mean_token_accuracy": 0.2008280038833618,
      "num_tokens": 25215041.0,
      "step": 12150
    },
    {
      "entropy": 5.443638896942138,
      "epoch": 1.1025943396226414,
      "grad_norm": 1.0,
      "learning_rate": 0.0004885210576094385,
      "loss": 5.0681,
      "mean_token_accuracy": 0.19479611963033677,
      "num_tokens": 25225937.0,
      "step": 12155
    },
    {
      "entropy": 5.407575702667236,
      "epoch": 1.1030478955007257,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004885108524904124,
      "loss": 4.986,
      "mean_token_accuracy": 0.20156417191028594,
      "num_tokens": 25235270.0,
      "step": 12160
    },
    {
      "entropy": 5.43190450668335,
      "epoch": 1.10350145137881,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004885006429559275,
      "loss": 5.0864,
      "mean_token_accuracy": 0.18922053277492523,
      "num_tokens": 25247080.0,
      "step": 12165
    },
    {
      "entropy": 5.489084720611572,
      "epoch": 1.103955007256894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004884904290061949,
      "loss": 5.0778,
      "mean_token_accuracy": 0.18807390183210373,
      "num_tokens": 25256718.0,
      "step": 12170
    },
    {
      "entropy": 5.487191200256348,
      "epoch": 1.1044085631349783,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004884802106414258,
      "loss": 5.1941,
      "mean_token_accuracy": 0.18723392188549043,
      "num_tokens": 25266632.0,
      "step": 12175
    },
    {
      "entropy": 5.464863109588623,
      "epoch": 1.1048621190130623,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004884699878618316,
      "loss": 5.0549,
      "mean_token_accuracy": 0.1978112742304802,
      "num_tokens": 25275598.0,
      "step": 12180
    },
    {
      "entropy": 5.517183065414429,
      "epoch": 1.1053156748911466,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004884597606676236,
      "loss": 5.1461,
      "mean_token_accuracy": 0.1934637799859047,
      "num_tokens": 25286598.0,
      "step": 12185
    },
    {
      "entropy": 5.4533368110656735,
      "epoch": 1.1057692307692308,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004884495290590134,
      "loss": 5.0586,
      "mean_token_accuracy": 0.1986728146672249,
      "num_tokens": 25296730.0,
      "step": 12190
    },
    {
      "entropy": 5.412643527984619,
      "epoch": 1.1062227866473149,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004884392930362124,
      "loss": 4.9533,
      "mean_token_accuracy": 0.20431596785783768,
      "num_tokens": 25306326.0,
      "step": 12195
    },
    {
      "entropy": 5.470515298843384,
      "epoch": 1.1066763425253991,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004884290525994326,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19730253219604493,
      "num_tokens": 25316342.0,
      "step": 12200
    },
    {
      "entropy": 5.545531749725342,
      "epoch": 1.1071298984034832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004884188077488855,
      "loss": 5.1291,
      "mean_token_accuracy": 0.18943889439105988,
      "num_tokens": 25326129.0,
      "step": 12205
    },
    {
      "entropy": 5.431648206710816,
      "epoch": 1.1075834542815675,
      "grad_norm": 1.0,
      "learning_rate": 0.0004884085584847831,
      "loss": 5.087,
      "mean_token_accuracy": 0.19083992540836334,
      "num_tokens": 25337156.0,
      "step": 12210
    },
    {
      "entropy": 5.507016372680664,
      "epoch": 1.1080370101596517,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004883983048073375,
      "loss": 5.2059,
      "mean_token_accuracy": 0.18340847343206407,
      "num_tokens": 25347632.0,
      "step": 12215
    },
    {
      "entropy": 5.433973884582519,
      "epoch": 1.1084905660377358,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004883880467167603,
      "loss": 4.9614,
      "mean_token_accuracy": 0.20440714359283446,
      "num_tokens": 25359186.0,
      "step": 12220
    },
    {
      "entropy": 5.452834415435791,
      "epoch": 1.10894412191582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000488377784213264,
      "loss": 5.0833,
      "mean_token_accuracy": 0.1990295484662056,
      "num_tokens": 25368993.0,
      "step": 12225
    },
    {
      "entropy": 5.404776620864868,
      "epoch": 1.1093976777939043,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004883675172970608,
      "loss": 5.0585,
      "mean_token_accuracy": 0.19580052495002748,
      "num_tokens": 25379880.0,
      "step": 12230
    },
    {
      "entropy": 5.507901239395141,
      "epoch": 1.1098512336719883,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000488357245968363,
      "loss": 5.2194,
      "mean_token_accuracy": 0.1886405259370804,
      "num_tokens": 25389942.0,
      "step": 12235
    },
    {
      "entropy": 5.460845708847046,
      "epoch": 1.1103047895500726,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004883469702273829,
      "loss": 5.0248,
      "mean_token_accuracy": 0.20065219849348068,
      "num_tokens": 25400836.0,
      "step": 12240
    },
    {
      "entropy": 5.439810752868652,
      "epoch": 1.1107583454281567,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004883366900743331,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20459828972816468,
      "num_tokens": 25410502.0,
      "step": 12245
    },
    {
      "entropy": 5.412273931503296,
      "epoch": 1.111211901306241,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004883264055094262,
      "loss": 4.9971,
      "mean_token_accuracy": 0.2025846540927887,
      "num_tokens": 25420301.0,
      "step": 12250
    },
    {
      "entropy": 5.423372793197632,
      "epoch": 1.1116654571843252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004883161165328749,
      "loss": 5.1022,
      "mean_token_accuracy": 0.19204796105623245,
      "num_tokens": 25431107.0,
      "step": 12255
    },
    {
      "entropy": 5.422839641571045,
      "epoch": 1.1121190130624092,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000488305823144892,
      "loss": 5.0848,
      "mean_token_accuracy": 0.19618319123983383,
      "num_tokens": 25441256.0,
      "step": 12260
    },
    {
      "entropy": 5.490327835083008,
      "epoch": 1.1125725689404935,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004882955253456902,
      "loss": 5.0207,
      "mean_token_accuracy": 0.20296041369438172,
      "num_tokens": 25451627.0,
      "step": 12265
    },
    {
      "entropy": 5.496054744720459,
      "epoch": 1.1130261248185778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004882852231354826,
      "loss": 5.148,
      "mean_token_accuracy": 0.18812839686870575,
      "num_tokens": 25461273.0,
      "step": 12270
    },
    {
      "entropy": 5.3958149433135985,
      "epoch": 1.1134796806966618,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004882749165144822,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20185253024101257,
      "num_tokens": 25471949.0,
      "step": 12275
    },
    {
      "entropy": 5.490109395980835,
      "epoch": 1.113933236574746,
      "grad_norm": 1.0,
      "learning_rate": 0.0004882646054829023,
      "loss": 5.056,
      "mean_token_accuracy": 0.19314002394676208,
      "num_tokens": 25482175.0,
      "step": 12280
    },
    {
      "entropy": 5.491643333435059,
      "epoch": 1.1143867924528301,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048825429004095597,
      "loss": 5.1249,
      "mean_token_accuracy": 0.19887231439352035,
      "num_tokens": 25491786.0,
      "step": 12285
    },
    {
      "entropy": 5.425946044921875,
      "epoch": 1.1148403483309144,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004882439701888566,
      "loss": 5.0449,
      "mean_token_accuracy": 0.20476123839616775,
      "num_tokens": 25502103.0,
      "step": 12290
    },
    {
      "entropy": 5.43491415977478,
      "epoch": 1.1152939042089987,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048823364592681754,
      "loss": 5.0132,
      "mean_token_accuracy": 0.1989473059773445,
      "num_tokens": 25512530.0,
      "step": 12295
    },
    {
      "entropy": 5.418636131286621,
      "epoch": 1.1157474600870827,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048822331725505234,
      "loss": 5.0166,
      "mean_token_accuracy": 0.1979011669754982,
      "num_tokens": 25522905.0,
      "step": 12300
    },
    {
      "entropy": 5.410347890853882,
      "epoch": 1.116201015965167,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048821298417377456,
      "loss": 5.0001,
      "mean_token_accuracy": 0.20536549240350724,
      "num_tokens": 25532963.0,
      "step": 12305
    },
    {
      "entropy": 5.36260495185852,
      "epoch": 1.116654571843251,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000488202646683198,
      "loss": 5.0772,
      "mean_token_accuracy": 0.19717641025781632,
      "num_tokens": 25542583.0,
      "step": 12310
    },
    {
      "entropy": 5.430088186264038,
      "epoch": 1.1171081277213353,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004881923047835364,
      "loss": 5.0414,
      "mean_token_accuracy": 0.19662002921104432,
      "num_tokens": 25553367.0,
      "step": 12315
    },
    {
      "entropy": 5.577167987823486,
      "epoch": 1.1175616835994195,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004881819584750036,
      "loss": 5.1457,
      "mean_token_accuracy": 0.1906937912106514,
      "num_tokens": 25564823.0,
      "step": 12320
    },
    {
      "entropy": 5.47358980178833,
      "epoch": 1.1180152394775036,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048817160775781356,
      "loss": 5.1128,
      "mean_token_accuracy": 0.19616676568984986,
      "num_tokens": 25574946.0,
      "step": 12325
    },
    {
      "entropy": 5.454680919647217,
      "epoch": 1.1184687953555879,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048816125263218023,
      "loss": 5.1098,
      "mean_token_accuracy": 0.19523930549621582,
      "num_tokens": 25584779.0,
      "step": 12330
    },
    {
      "entropy": 5.5475927829742435,
      "epoch": 1.118922351233672,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048815089309831803,
      "loss": 5.1617,
      "mean_token_accuracy": 0.1924641713500023,
      "num_tokens": 25594725.0,
      "step": 12335
    },
    {
      "entropy": 5.4552511215209964,
      "epoch": 1.1193759071117562,
      "grad_norm": 1.015625,
      "learning_rate": 0.000488140529156441,
      "loss": 5.1385,
      "mean_token_accuracy": 0.19338975101709366,
      "num_tokens": 25605499.0,
      "step": 12340
    },
    {
      "entropy": 5.427136659622192,
      "epoch": 1.1198294629898404,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004881301608067635,
      "loss": 5.0939,
      "mean_token_accuracy": 0.1955274298787117,
      "num_tokens": 25616443.0,
      "step": 12345
    },
    {
      "entropy": 5.462292909622192,
      "epoch": 1.1202830188679245,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048811978804949997,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19406415075063704,
      "num_tokens": 25627762.0,
      "step": 12350
    },
    {
      "entropy": 5.545721483230591,
      "epoch": 1.1207365747460087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000488109410884865,
      "loss": 5.1339,
      "mean_token_accuracy": 0.19659575074911118,
      "num_tokens": 25637904.0,
      "step": 12355
    },
    {
      "entropy": 5.44553017616272,
      "epoch": 1.1211901306240928,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048809902931307294,
      "loss": 5.0098,
      "mean_token_accuracy": 0.19970433562994003,
      "num_tokens": 25649058.0,
      "step": 12360
    },
    {
      "entropy": 5.398366451263428,
      "epoch": 1.121643686502177,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004880886433343388,
      "loss": 5.0741,
      "mean_token_accuracy": 0.20112952291965486,
      "num_tokens": 25659161.0,
      "step": 12365
    },
    {
      "entropy": 5.4166340827941895,
      "epoch": 1.1220972423802613,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004880782529488771,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20340972095727922,
      "num_tokens": 25669336.0,
      "step": 12370
    },
    {
      "entropy": 5.492840766906738,
      "epoch": 1.1225507982583454,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004880678581569029,
      "loss": 5.1135,
      "mean_token_accuracy": 0.1954827070236206,
      "num_tokens": 25680443.0,
      "step": 12375
    },
    {
      "entropy": 5.4376270294189455,
      "epoch": 1.1230043541364296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048805745895863105,
      "loss": 5.0297,
      "mean_token_accuracy": 0.19975681155920028,
      "num_tokens": 25690202.0,
      "step": 12380
    },
    {
      "entropy": 5.4068975925445555,
      "epoch": 1.1234579100145137,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004880470553542767,
      "loss": 5.0979,
      "mean_token_accuracy": 0.19868273437023162,
      "num_tokens": 25699255.0,
      "step": 12385
    },
    {
      "entropy": 5.454531192779541,
      "epoch": 1.123911465892598,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048803664734405495,
      "loss": 5.0702,
      "mean_token_accuracy": 0.19199754446744918,
      "num_tokens": 25710714.0,
      "step": 12390
    },
    {
      "entropy": 5.463699913024902,
      "epoch": 1.1243650217706822,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000488026234928181,
      "loss": 5.0572,
      "mean_token_accuracy": 0.1986502930521965,
      "num_tokens": 25721722.0,
      "step": 12395
    },
    {
      "entropy": 5.453849935531617,
      "epoch": 1.1248185776487662,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004880158181068703,
      "loss": 5.0366,
      "mean_token_accuracy": 0.2029099076986313,
      "num_tokens": 25732096.0,
      "step": 12400
    },
    {
      "entropy": 5.465212678909301,
      "epoch": 1.1252721335268505,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048800539688033807,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19555932879447938,
      "num_tokens": 25741499.0,
      "step": 12405
    },
    {
      "entropy": 5.475707721710205,
      "epoch": 1.1257256894049348,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004879949712488001,
      "loss": 5.1854,
      "mean_token_accuracy": 0.19263612776994704,
      "num_tokens": 25751913.0,
      "step": 12410
    },
    {
      "entropy": 5.4851563453674315,
      "epoch": 1.1261792452830188,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048798454121247177,
      "loss": 5.1184,
      "mean_token_accuracy": 0.1930095836520195,
      "num_tokens": 25762602.0,
      "step": 12415
    },
    {
      "entropy": 5.414087772369385,
      "epoch": 1.126632801161103,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004879741067715688,
      "loss": 5.0738,
      "mean_token_accuracy": 0.1986461728811264,
      "num_tokens": 25773872.0,
      "step": 12420
    },
    {
      "entropy": 5.476527690887451,
      "epoch": 1.1270863570391871,
      "grad_norm": 1.0,
      "learning_rate": 0.00048796366792630715,
      "loss": 5.072,
      "mean_token_accuracy": 0.19697064459323882,
      "num_tokens": 25784723.0,
      "step": 12425
    },
    {
      "entropy": 5.449933576583862,
      "epoch": 1.1275399129172714,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004879532246769025,
      "loss": 5.0951,
      "mean_token_accuracy": 0.19810961186885834,
      "num_tokens": 25794897.0,
      "step": 12430
    },
    {
      "entropy": 5.416313743591308,
      "epoch": 1.1279934687953557,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004879427770235709,
      "loss": 4.9869,
      "mean_token_accuracy": 0.20813877135515213,
      "num_tokens": 25804445.0,
      "step": 12435
    },
    {
      "entropy": 5.491560363769532,
      "epoch": 1.1284470246734397,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048793232496652846,
      "loss": 5.1432,
      "mean_token_accuracy": 0.1873038589954376,
      "num_tokens": 25815495.0,
      "step": 12440
    },
    {
      "entropy": 5.480292654037475,
      "epoch": 1.128900580551524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048792186850599124,
      "loss": 5.0759,
      "mean_token_accuracy": 0.2019079625606537,
      "num_tokens": 25826120.0,
      "step": 12445
    },
    {
      "entropy": 5.44529709815979,
      "epoch": 1.1293541364296082,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004879114076421755,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20093466639518737,
      "num_tokens": 25835926.0,
      "step": 12450
    },
    {
      "entropy": 5.4184167861938475,
      "epoch": 1.1298076923076923,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048790094237529763,
      "loss": 5.1001,
      "mean_token_accuracy": 0.1918276235461235,
      "num_tokens": 25846095.0,
      "step": 12455
    },
    {
      "entropy": 5.508404397964478,
      "epoch": 1.1302612481857766,
      "grad_norm": 1.046875,
      "learning_rate": 0.000487890472705574,
      "loss": 5.1076,
      "mean_token_accuracy": 0.19151043891906738,
      "num_tokens": 25858541.0,
      "step": 12460
    },
    {
      "entropy": 5.484144592285157,
      "epoch": 1.1307148040638606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048787999863322113,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2003431960940361,
      "num_tokens": 25867664.0,
      "step": 12465
    },
    {
      "entropy": 5.443308925628662,
      "epoch": 1.1311683599419449,
      "grad_norm": 1.0,
      "learning_rate": 0.0004878695201584557,
      "loss": 5.1018,
      "mean_token_accuracy": 0.1985948473215103,
      "num_tokens": 25877300.0,
      "step": 12470
    },
    {
      "entropy": 5.424301815032959,
      "epoch": 1.1316219158200291,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004878590372814943,
      "loss": 5.0575,
      "mean_token_accuracy": 0.20260118544101716,
      "num_tokens": 25888009.0,
      "step": 12475
    },
    {
      "entropy": 5.50917501449585,
      "epoch": 1.1320754716981132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048784855000255374,
      "loss": 5.1167,
      "mean_token_accuracy": 0.19358260482549666,
      "num_tokens": 25897285.0,
      "step": 12480
    },
    {
      "entropy": 5.4255023956298825,
      "epoch": 1.1325290275761974,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048783805832185097,
      "loss": 5.009,
      "mean_token_accuracy": 0.20263294130563736,
      "num_tokens": 25907355.0,
      "step": 12485
    },
    {
      "entropy": 5.472705078125,
      "epoch": 1.1329825834542815,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004878275622396029,
      "loss": 5.0844,
      "mean_token_accuracy": 0.20127519220113754,
      "num_tokens": 25918566.0,
      "step": 12490
    },
    {
      "entropy": 5.463294839859008,
      "epoch": 1.1334361393323658,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048781706175602665,
      "loss": 5.1393,
      "mean_token_accuracy": 0.19933076500892638,
      "num_tokens": 25929145.0,
      "step": 12495
    },
    {
      "entropy": 5.4120849609375,
      "epoch": 1.13388969521045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048780655687133927,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20007957518100739,
      "num_tokens": 25938163.0,
      "step": 12500
    },
    {
      "entropy": 5.403029680252075,
      "epoch": 1.134343251088534,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048779604758575806,
      "loss": 5.0394,
      "mean_token_accuracy": 0.1949640303850174,
      "num_tokens": 25948962.0,
      "step": 12505
    },
    {
      "entropy": 5.389069318771362,
      "epoch": 1.1347968069666183,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004877855338995004,
      "loss": 5.0095,
      "mean_token_accuracy": 0.19757837504148484,
      "num_tokens": 25959653.0,
      "step": 12510
    },
    {
      "entropy": 5.512422895431518,
      "epoch": 1.1352503628447024,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048777501581278367,
      "loss": 5.1007,
      "mean_token_accuracy": 0.1961492881178856,
      "num_tokens": 25970268.0,
      "step": 12515
    },
    {
      "entropy": 5.425408792495728,
      "epoch": 1.1357039187227866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004877644933258254,
      "loss": 5.0506,
      "mean_token_accuracy": 0.19497006237506867,
      "num_tokens": 25980321.0,
      "step": 12520
    },
    {
      "entropy": 5.386345720291137,
      "epoch": 1.136157474600871,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004877539664388432,
      "loss": 5.0559,
      "mean_token_accuracy": 0.19168413579463958,
      "num_tokens": 25992466.0,
      "step": 12525
    },
    {
      "entropy": 5.461630630493164,
      "epoch": 1.136611030478955,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048774343515205464,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20293886661529542,
      "num_tokens": 26003553.0,
      "step": 12530
    },
    {
      "entropy": 5.46403136253357,
      "epoch": 1.1370645863570392,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004877328994656778,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20149170607328415,
      "num_tokens": 26013299.0,
      "step": 12535
    },
    {
      "entropy": 5.416072607040405,
      "epoch": 1.1375181422351233,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004877223593799303,
      "loss": 5.0718,
      "mean_token_accuracy": 0.19764264523983002,
      "num_tokens": 26022913.0,
      "step": 12540
    },
    {
      "entropy": 5.3480541706085205,
      "epoch": 1.1379716981132075,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004877118148950302,
      "loss": 5.0585,
      "mean_token_accuracy": 0.20063163489103317,
      "num_tokens": 26033191.0,
      "step": 12545
    },
    {
      "entropy": 5.589880514144897,
      "epoch": 1.1384252539912918,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048770126601119555,
      "loss": 5.1971,
      "mean_token_accuracy": 0.1850689321756363,
      "num_tokens": 26043517.0,
      "step": 12550
    },
    {
      "entropy": 5.570900964736938,
      "epoch": 1.1388788098693758,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048769071272864454,
      "loss": 5.1594,
      "mean_token_accuracy": 0.19274065196514129,
      "num_tokens": 26054173.0,
      "step": 12555
    },
    {
      "entropy": 5.5047448635101315,
      "epoch": 1.13933236574746,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048768015504759536,
      "loss": 5.1242,
      "mean_token_accuracy": 0.1928042858839035,
      "num_tokens": 26064929.0,
      "step": 12560
    },
    {
      "entropy": 5.408150243759155,
      "epoch": 1.1397859216255442,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004876695929682665,
      "loss": 5.022,
      "mean_token_accuracy": 0.20337846577167512,
      "num_tokens": 26074872.0,
      "step": 12565
    },
    {
      "entropy": 5.414556503295898,
      "epoch": 1.1402394775036284,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048765902649087605,
      "loss": 4.9541,
      "mean_token_accuracy": 0.20306820124387742,
      "num_tokens": 26086328.0,
      "step": 12570
    },
    {
      "entropy": 5.485406827926636,
      "epoch": 1.1406930333817127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048764845561564285,
      "loss": 5.0623,
      "mean_token_accuracy": 0.19497212320566176,
      "num_tokens": 26096477.0,
      "step": 12575
    },
    {
      "entropy": 5.446309804916382,
      "epoch": 1.1411465892597967,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048763788034278545,
      "loss": 4.9958,
      "mean_token_accuracy": 0.20379463136196135,
      "num_tokens": 26106183.0,
      "step": 12580
    },
    {
      "entropy": 5.419105434417725,
      "epoch": 1.141600145137881,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004876273006725225,
      "loss": 5.0759,
      "mean_token_accuracy": 0.19942841231822966,
      "num_tokens": 26117932.0,
      "step": 12585
    },
    {
      "entropy": 5.397345113754272,
      "epoch": 1.1420537010159653,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048761671660507266,
      "loss": 5.0232,
      "mean_token_accuracy": 0.1995558112859726,
      "num_tokens": 26128616.0,
      "step": 12590
    },
    {
      "entropy": 5.464322662353515,
      "epoch": 1.1425072568940493,
      "grad_norm": 0.9375,
      "learning_rate": 0.000487606128140655,
      "loss": 5.0477,
      "mean_token_accuracy": 0.19695389568805693,
      "num_tokens": 26138758.0,
      "step": 12595
    },
    {
      "entropy": 5.420771551132202,
      "epoch": 1.1429608127721336,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004875955352794885,
      "loss": 5.0502,
      "mean_token_accuracy": 0.19705618023872376,
      "num_tokens": 26149199.0,
      "step": 12600
    },
    {
      "entropy": 5.4241548538208,
      "epoch": 1.1434143686502176,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004875849380217921,
      "loss": 5.1327,
      "mean_token_accuracy": 0.20088202506303787,
      "num_tokens": 26160276.0,
      "step": 12605
    },
    {
      "entropy": 5.4528889656066895,
      "epoch": 1.1438679245283019,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000487574336367785,
      "loss": 5.0169,
      "mean_token_accuracy": 0.2016328141093254,
      "num_tokens": 26169564.0,
      "step": 12610
    },
    {
      "entropy": 5.366958856582642,
      "epoch": 1.1443214804063861,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004875637303176865,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20262449383735656,
      "num_tokens": 26179135.0,
      "step": 12615
    },
    {
      "entropy": 5.459455728530884,
      "epoch": 1.1447750362844702,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048755311987171585,
      "loss": 5.1171,
      "mean_token_accuracy": 0.19743296205997468,
      "num_tokens": 26190116.0,
      "step": 12620
    },
    {
      "entropy": 5.479267597198486,
      "epoch": 1.1452285921625545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048754250503009244,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1911403179168701,
      "num_tokens": 26200847.0,
      "step": 12625
    },
    {
      "entropy": 5.545922040939331,
      "epoch": 1.1456821480406387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048753188579303596,
      "loss": 5.1417,
      "mean_token_accuracy": 0.1953682154417038,
      "num_tokens": 26211151.0,
      "step": 12630
    },
    {
      "entropy": 5.4858533382415775,
      "epoch": 1.1461357039187228,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004875212621607659,
      "loss": 5.0533,
      "mean_token_accuracy": 0.20282673686742783,
      "num_tokens": 26221273.0,
      "step": 12635
    },
    {
      "entropy": 5.466365957260132,
      "epoch": 1.146589259796807,
      "grad_norm": 1.0,
      "learning_rate": 0.00048751063413350203,
      "loss": 5.163,
      "mean_token_accuracy": 0.194965261220932,
      "num_tokens": 26231770.0,
      "step": 12640
    },
    {
      "entropy": 5.495484018325806,
      "epoch": 1.147042815674891,
      "grad_norm": 1.09375,
      "learning_rate": 0.000487500001711464,
      "loss": 5.1197,
      "mean_token_accuracy": 0.18718278259038926,
      "num_tokens": 26241386.0,
      "step": 12645
    },
    {
      "entropy": 5.574727201461792,
      "epoch": 1.1474963715529753,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004874893648948719,
      "loss": 5.1869,
      "mean_token_accuracy": 0.19222330152988434,
      "num_tokens": 26251991.0,
      "step": 12650
    },
    {
      "entropy": 5.574859189987182,
      "epoch": 1.1479499274310596,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048747872368394546,
      "loss": 5.2441,
      "mean_token_accuracy": 0.19536986351013183,
      "num_tokens": 26262331.0,
      "step": 12655
    },
    {
      "entropy": 5.503897953033447,
      "epoch": 1.1484034833091437,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048746807807890496,
      "loss": 5.1156,
      "mean_token_accuracy": 0.19601597040891647,
      "num_tokens": 26272939.0,
      "step": 12660
    },
    {
      "entropy": 5.452318811416626,
      "epoch": 1.148857039187228,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048745742807997046,
      "loss": 5.0617,
      "mean_token_accuracy": 0.20496115684509278,
      "num_tokens": 26282849.0,
      "step": 12665
    },
    {
      "entropy": 5.424884271621704,
      "epoch": 1.149310595065312,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048744677368736216,
      "loss": 5.0783,
      "mean_token_accuracy": 0.20587577670812607,
      "num_tokens": 26292449.0,
      "step": 12670
    },
    {
      "entropy": 5.362032032012939,
      "epoch": 1.1497641509433962,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048743611490130044,
      "loss": 5.0219,
      "mean_token_accuracy": 0.19552327394485475,
      "num_tokens": 26302506.0,
      "step": 12675
    },
    {
      "entropy": 5.492298984527588,
      "epoch": 1.1502177068214805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048742545172200576,
      "loss": 5.0863,
      "mean_token_accuracy": 0.20193991661071778,
      "num_tokens": 26312346.0,
      "step": 12680
    },
    {
      "entropy": 5.507142162322998,
      "epoch": 1.1506712626995645,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004874147841496986,
      "loss": 5.0933,
      "mean_token_accuracy": 0.2021457850933075,
      "num_tokens": 26323173.0,
      "step": 12685
    },
    {
      "entropy": 5.518513441085815,
      "epoch": 1.1511248185776488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004874041121845996,
      "loss": 5.1545,
      "mean_token_accuracy": 0.1896779328584671,
      "num_tokens": 26334371.0,
      "step": 12690
    },
    {
      "entropy": 5.42510199546814,
      "epoch": 1.1515783744557329,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873934358269294,
      "loss": 5.0614,
      "mean_token_accuracy": 0.2023568034172058,
      "num_tokens": 26344506.0,
      "step": 12695
    },
    {
      "entropy": 5.5277252197265625,
      "epoch": 1.1520319303338171,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004873827550769088,
      "loss": 5.1029,
      "mean_token_accuracy": 0.18706048280000687,
      "num_tokens": 26354534.0,
      "step": 12700
    },
    {
      "entropy": 5.45455904006958,
      "epoch": 1.1524854862119014,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004873720699347587,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19750375896692277,
      "num_tokens": 26365682.0,
      "step": 12705
    },
    {
      "entropy": 5.53352952003479,
      "epoch": 1.1529390420899854,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004873613804007001,
      "loss": 5.1866,
      "mean_token_accuracy": 0.1976620763540268,
      "num_tokens": 26376702.0,
      "step": 12710
    },
    {
      "entropy": 5.440504360198974,
      "epoch": 1.1533925979680697,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004873506864749541,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20106828659772874,
      "num_tokens": 26387154.0,
      "step": 12715
    },
    {
      "entropy": 5.5413073062896725,
      "epoch": 1.1538461538461537,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004873399881577417,
      "loss": 5.1618,
      "mean_token_accuracy": 0.19441997408866882,
      "num_tokens": 26397883.0,
      "step": 12720
    },
    {
      "entropy": 5.380645418167115,
      "epoch": 1.154299709724238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004873292854492842,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20258790552616118,
      "num_tokens": 26407613.0,
      "step": 12725
    },
    {
      "entropy": 5.344910144805908,
      "epoch": 1.1547532656023223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004873185783498031,
      "loss": 4.994,
      "mean_token_accuracy": 0.19510757476091384,
      "num_tokens": 26417221.0,
      "step": 12730
    },
    {
      "entropy": 5.446285247802734,
      "epoch": 1.1552068214804063,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004873078668595197,
      "loss": 5.0874,
      "mean_token_accuracy": 0.1940292626619339,
      "num_tokens": 26427976.0,
      "step": 12735
    },
    {
      "entropy": 5.473560428619384,
      "epoch": 1.1556603773584906,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048729715097865534,
      "loss": 5.1114,
      "mean_token_accuracy": 0.20027161091566087,
      "num_tokens": 26439213.0,
      "step": 12740
    },
    {
      "entropy": 5.435029172897339,
      "epoch": 1.1561139332365746,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004872864307074319,
      "loss": 5.0492,
      "mean_token_accuracy": 0.20389191210269927,
      "num_tokens": 26449763.0,
      "step": 12745
    },
    {
      "entropy": 5.520690250396728,
      "epoch": 1.156567489114659,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048727570604607096,
      "loss": 5.1146,
      "mean_token_accuracy": 0.19712771773338317,
      "num_tokens": 26460133.0,
      "step": 12750
    },
    {
      "entropy": 5.5018998146057125,
      "epoch": 1.1570210449927432,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048726497699479437,
      "loss": 5.0798,
      "mean_token_accuracy": 0.19398481249809266,
      "num_tokens": 26470164.0,
      "step": 12755
    },
    {
      "entropy": 5.442802476882934,
      "epoch": 1.1574746008708272,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000487254243553824,
      "loss": 5.0914,
      "mean_token_accuracy": 0.19554944038391114,
      "num_tokens": 26480314.0,
      "step": 12760
    },
    {
      "entropy": 5.3819260597229,
      "epoch": 1.1579281567489115,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004872435057233817,
      "loss": 5.038,
      "mean_token_accuracy": 0.20284716337919234,
      "num_tokens": 26491184.0,
      "step": 12765
    },
    {
      "entropy": 5.406965637207032,
      "epoch": 1.1583817126269957,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048723276350368957,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20368421971797943,
      "num_tokens": 26501432.0,
      "step": 12770
    },
    {
      "entropy": 5.4483438491821286,
      "epoch": 1.1588352685050798,
      "grad_norm": 1.0,
      "learning_rate": 0.00048722201689496984,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19702692180871964,
      "num_tokens": 26510379.0,
      "step": 12775
    },
    {
      "entropy": 5.429786205291748,
      "epoch": 1.159288824383164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004872112658974447,
      "loss": 5.0561,
      "mean_token_accuracy": 0.20342138409614563,
      "num_tokens": 26519454.0,
      "step": 12780
    },
    {
      "entropy": 5.387827634811401,
      "epoch": 1.159742380261248,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048720051051133657,
      "loss": 4.9709,
      "mean_token_accuracy": 0.20228770673274993,
      "num_tokens": 26531325.0,
      "step": 12785
    },
    {
      "entropy": 5.440437602996826,
      "epoch": 1.1601959361393324,
      "grad_norm": 1.0,
      "learning_rate": 0.0004871897507368678,
      "loss": 5.0725,
      "mean_token_accuracy": 0.20184554904699326,
      "num_tokens": 26541510.0,
      "step": 12790
    },
    {
      "entropy": 5.493928003311157,
      "epoch": 1.1606494920174166,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048717898657426093,
      "loss": 5.2491,
      "mean_token_accuracy": 0.18356596529483796,
      "num_tokens": 26553400.0,
      "step": 12795
    },
    {
      "entropy": 5.481597852706909,
      "epoch": 1.1611030478955007,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048716821802373856,
      "loss": 5.0769,
      "mean_token_accuracy": 0.19700282961130142,
      "num_tokens": 26564340.0,
      "step": 12800
    },
    {
      "entropy": 5.448103237152099,
      "epoch": 1.161556603773585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048715744508552334,
      "loss": 5.0507,
      "mean_token_accuracy": 0.19755611568689346,
      "num_tokens": 26573615.0,
      "step": 12805
    },
    {
      "entropy": 5.4913605690002445,
      "epoch": 1.1620101596516692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004871466677598381,
      "loss": 5.1896,
      "mean_token_accuracy": 0.18805204033851625,
      "num_tokens": 26584149.0,
      "step": 12810
    },
    {
      "entropy": 5.4462597370147705,
      "epoch": 1.1624637155297532,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004871358860469057,
      "loss": 5.0908,
      "mean_token_accuracy": 0.1880527302622795,
      "num_tokens": 26595064.0,
      "step": 12815
    },
    {
      "entropy": 5.4592596054077145,
      "epoch": 1.1629172714078375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048712509994694903,
      "loss": 5.0957,
      "mean_token_accuracy": 0.19753992557525635,
      "num_tokens": 26605636.0,
      "step": 12820
    },
    {
      "entropy": 5.525445079803466,
      "epoch": 1.1633708272859216,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004871143094601914,
      "loss": 5.1863,
      "mean_token_accuracy": 0.19185950756072997,
      "num_tokens": 26616065.0,
      "step": 12825
    },
    {
      "entropy": 5.518282413482666,
      "epoch": 1.1638243831640058,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004871035145868557,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19677670150995255,
      "num_tokens": 26627246.0,
      "step": 12830
    },
    {
      "entropy": 5.459026336669922,
      "epoch": 1.16427793904209,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048709271532716524,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19623770713806152,
      "num_tokens": 26637630.0,
      "step": 12835
    },
    {
      "entropy": 5.516048812866211,
      "epoch": 1.1647314949201741,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004870819116813434,
      "loss": 5.1509,
      "mean_token_accuracy": 0.19567939937114714,
      "num_tokens": 26648683.0,
      "step": 12840
    },
    {
      "entropy": 5.5269402980804445,
      "epoch": 1.1651850507982584,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048707110364961363,
      "loss": 5.1139,
      "mean_token_accuracy": 0.1932143375277519,
      "num_tokens": 26659573.0,
      "step": 12845
    },
    {
      "entropy": 5.423376846313476,
      "epoch": 1.1656386066763424,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048706029123219937,
      "loss": 5.0102,
      "mean_token_accuracy": 0.2039925828576088,
      "num_tokens": 26669384.0,
      "step": 12850
    },
    {
      "entropy": 5.482315349578857,
      "epoch": 1.1660921625544267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048704947442932426,
      "loss": 5.1805,
      "mean_token_accuracy": 0.18662424385547638,
      "num_tokens": 26679721.0,
      "step": 12855
    },
    {
      "entropy": 5.531332302093506,
      "epoch": 1.166545718432511,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048703865324121196,
      "loss": 5.027,
      "mean_token_accuracy": 0.20461533665657045,
      "num_tokens": 26689923.0,
      "step": 12860
    },
    {
      "entropy": 5.401912021636963,
      "epoch": 1.166999274310595,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048702782766808637,
      "loss": 5.0387,
      "mean_token_accuracy": 0.19950828552246094,
      "num_tokens": 26699953.0,
      "step": 12865
    },
    {
      "entropy": 5.479466533660888,
      "epoch": 1.1674528301886793,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004870169977101711,
      "loss": 5.2542,
      "mean_token_accuracy": 0.185440394282341,
      "num_tokens": 26712099.0,
      "step": 12870
    },
    {
      "entropy": 5.518303060531617,
      "epoch": 1.1679063860667633,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048700616336769037,
      "loss": 5.087,
      "mean_token_accuracy": 0.1994853988289833,
      "num_tokens": 26722334.0,
      "step": 12875
    },
    {
      "entropy": 5.479982852935791,
      "epoch": 1.1683599419448476,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048699532464086817,
      "loss": 5.0563,
      "mean_token_accuracy": 0.199135123193264,
      "num_tokens": 26732470.0,
      "step": 12880
    },
    {
      "entropy": 5.441677236557007,
      "epoch": 1.1688134978229319,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048698448152992865,
      "loss": 5.0901,
      "mean_token_accuracy": 0.19643069058656693,
      "num_tokens": 26742015.0,
      "step": 12885
    },
    {
      "entropy": 5.417946290969849,
      "epoch": 1.169267053701016,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048697363403509605,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19372081607580185,
      "num_tokens": 26753231.0,
      "step": 12890
    },
    {
      "entropy": 5.489230966567993,
      "epoch": 1.1697206095791002,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004869627821565946,
      "loss": 5.061,
      "mean_token_accuracy": 0.19762594401836395,
      "num_tokens": 26763895.0,
      "step": 12895
    },
    {
      "entropy": 5.52664737701416,
      "epoch": 1.1701741654571842,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004869519258946489,
      "loss": 5.1666,
      "mean_token_accuracy": 0.19777561277151107,
      "num_tokens": 26774208.0,
      "step": 12900
    },
    {
      "entropy": 5.434159612655639,
      "epoch": 1.1706277213352685,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048694106524948324,
      "loss": 5.0975,
      "mean_token_accuracy": 0.19534947276115416,
      "num_tokens": 26785554.0,
      "step": 12905
    },
    {
      "entropy": 5.466801881790161,
      "epoch": 1.1710812772133528,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048693020022132243,
      "loss": 5.0105,
      "mean_token_accuracy": 0.19920032918453218,
      "num_tokens": 26795488.0,
      "step": 12910
    },
    {
      "entropy": 5.522245454788208,
      "epoch": 1.1715348330914368,
      "grad_norm": 0.984375,
      "learning_rate": 0.000486919330810391,
      "loss": 5.1736,
      "mean_token_accuracy": 0.18677914440631865,
      "num_tokens": 26806801.0,
      "step": 12915
    },
    {
      "entropy": 5.49633207321167,
      "epoch": 1.171988388969521,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004869084570169139,
      "loss": 5.0865,
      "mean_token_accuracy": 0.20189729630947112,
      "num_tokens": 26816869.0,
      "step": 12920
    },
    {
      "entropy": 5.430732583999633,
      "epoch": 1.172441944847605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048689757884111585,
      "loss": 5.032,
      "mean_token_accuracy": 0.19632718116044998,
      "num_tokens": 26827003.0,
      "step": 12925
    },
    {
      "entropy": 5.411961412429809,
      "epoch": 1.1728955007256894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048688669628322185,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20101807564496993,
      "num_tokens": 26837065.0,
      "step": 12930
    },
    {
      "entropy": 5.389821910858155,
      "epoch": 1.1733490566037736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000486875809343457,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20275212675333024,
      "num_tokens": 26847487.0,
      "step": 12935
    },
    {
      "entropy": 5.447001123428345,
      "epoch": 1.1738026124818577,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048686491802204645,
      "loss": 5.041,
      "mean_token_accuracy": 0.2034755453467369,
      "num_tokens": 26858407.0,
      "step": 12940
    },
    {
      "entropy": 5.481881952285766,
      "epoch": 1.174256168359942,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004868540223192154,
      "loss": 5.1027,
      "mean_token_accuracy": 0.19423448741436006,
      "num_tokens": 26868469.0,
      "step": 12945
    },
    {
      "entropy": 5.493884038925171,
      "epoch": 1.1747097242380262,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004868431222351892,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19713809788227082,
      "num_tokens": 26879580.0,
      "step": 12950
    },
    {
      "entropy": 5.488229417800904,
      "epoch": 1.1751632801161103,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048683221777019313,
      "loss": 5.0896,
      "mean_token_accuracy": 0.20102750808000563,
      "num_tokens": 26889513.0,
      "step": 12955
    },
    {
      "entropy": 5.466822147369385,
      "epoch": 1.1756168359941945,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048682130892445293,
      "loss": 5.1661,
      "mean_token_accuracy": 0.1931549623608589,
      "num_tokens": 26900792.0,
      "step": 12960
    },
    {
      "entropy": 5.437531661987305,
      "epoch": 1.1760703918722786,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000486810395698194,
      "loss": 5.0536,
      "mean_token_accuracy": 0.203791543841362,
      "num_tokens": 26911298.0,
      "step": 12965
    },
    {
      "entropy": 5.488604164123535,
      "epoch": 1.1765239477503628,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048679947809164223,
      "loss": 5.1055,
      "mean_token_accuracy": 0.19251493662595748,
      "num_tokens": 26921545.0,
      "step": 12970
    },
    {
      "entropy": 5.484994983673095,
      "epoch": 1.176977503628447,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004867885561050232,
      "loss": 5.0559,
      "mean_token_accuracy": 0.20793627351522445,
      "num_tokens": 26932140.0,
      "step": 12975
    },
    {
      "entropy": 5.481871938705444,
      "epoch": 1.1774310595065312,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004867776297385629,
      "loss": 5.1175,
      "mean_token_accuracy": 0.19122447669506074,
      "num_tokens": 26942634.0,
      "step": 12980
    },
    {
      "entropy": 5.495228290557861,
      "epoch": 1.1778846153846154,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048676669899248723,
      "loss": 5.0718,
      "mean_token_accuracy": 0.2001894935965538,
      "num_tokens": 26952758.0,
      "step": 12985
    },
    {
      "entropy": 5.39152045249939,
      "epoch": 1.1783381712626997,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004867557638670222,
      "loss": 5.0146,
      "mean_token_accuracy": 0.20257681906223296,
      "num_tokens": 26963439.0,
      "step": 12990
    },
    {
      "entropy": 5.446252346038818,
      "epoch": 1.1787917271407837,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004867448243623941,
      "loss": 5.0844,
      "mean_token_accuracy": 0.1936539277434349,
      "num_tokens": 26974331.0,
      "step": 12995
    },
    {
      "entropy": 5.441408300399781,
      "epoch": 1.179245283018868,
      "grad_norm": 1.078125,
      "learning_rate": 0.000486733880478829,
      "loss": 4.9579,
      "mean_token_accuracy": 0.20778458416461945,
      "num_tokens": 26983416.0,
      "step": 13000
    },
    {
      "entropy": 5.458328771591186,
      "epoch": 1.179698838896952,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048672293221655333,
      "loss": 5.0835,
      "mean_token_accuracy": 0.20128298848867415,
      "num_tokens": 26994176.0,
      "step": 13005
    },
    {
      "entropy": 5.438782453536987,
      "epoch": 1.1801523947750363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004867119795757935,
      "loss": 5.076,
      "mean_token_accuracy": 0.1955456778407097,
      "num_tokens": 27003295.0,
      "step": 13010
    },
    {
      "entropy": 5.462780618667603,
      "epoch": 1.1806059506531206,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000486701022556776,
      "loss": 5.0368,
      "mean_token_accuracy": 0.19585613012313843,
      "num_tokens": 27014018.0,
      "step": 13015
    },
    {
      "entropy": 5.444080924987793,
      "epoch": 1.1810595065312046,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004866900611597273,
      "loss": 5.0587,
      "mean_token_accuracy": 0.20128905773162842,
      "num_tokens": 27024706.0,
      "step": 13020
    },
    {
      "entropy": 5.487481689453125,
      "epoch": 1.1815130624092889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048667909538487427,
      "loss": 5.0813,
      "mean_token_accuracy": 0.20304662138223648,
      "num_tokens": 27035801.0,
      "step": 13025
    },
    {
      "entropy": 5.39069128036499,
      "epoch": 1.181966618287373,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004866681252324436,
      "loss": 4.952,
      "mean_token_accuracy": 0.20496505051851271,
      "num_tokens": 27046596.0,
      "step": 13030
    },
    {
      "entropy": 5.428235149383545,
      "epoch": 1.1824201741654572,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048665715070266214,
      "loss": 5.0642,
      "mean_token_accuracy": 0.19454382210969925,
      "num_tokens": 27057057.0,
      "step": 13035
    },
    {
      "entropy": 5.338984727859497,
      "epoch": 1.1828737300435415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048664617179575685,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20416733175516127,
      "num_tokens": 27067947.0,
      "step": 13040
    },
    {
      "entropy": 5.481903743743897,
      "epoch": 1.1833272859216255,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004866351885119548,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19917620718479156,
      "num_tokens": 27077503.0,
      "step": 13045
    },
    {
      "entropy": 5.51445779800415,
      "epoch": 1.1837808417997098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004866242008514832,
      "loss": 5.1751,
      "mean_token_accuracy": 0.19417932480573655,
      "num_tokens": 27088298.0,
      "step": 13050
    },
    {
      "entropy": 5.456486558914184,
      "epoch": 1.1842343976777938,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048661320881456904,
      "loss": 5.0368,
      "mean_token_accuracy": 0.19583414793014525,
      "num_tokens": 27097841.0,
      "step": 13055
    },
    {
      "entropy": 5.3624402523040775,
      "epoch": 1.184687953555878,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004866022124014399,
      "loss": 4.9876,
      "mean_token_accuracy": 0.20166653245687485,
      "num_tokens": 27108263.0,
      "step": 13060
    },
    {
      "entropy": 5.421962404251099,
      "epoch": 1.1851415094339623,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048659121161232313,
      "loss": 5.0483,
      "mean_token_accuracy": 0.20191064178943635,
      "num_tokens": 27119339.0,
      "step": 13065
    },
    {
      "entropy": 5.501571559906006,
      "epoch": 1.1855950653120464,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004865802064474461,
      "loss": 5.0946,
      "mean_token_accuracy": 0.19581292271614076,
      "num_tokens": 27130393.0,
      "step": 13070
    },
    {
      "entropy": 5.4905561923980715,
      "epoch": 1.1860486211901307,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048656919690703653,
      "loss": 5.0875,
      "mean_token_accuracy": 0.19397873133420945,
      "num_tokens": 27141156.0,
      "step": 13075
    },
    {
      "entropy": 5.503406047821045,
      "epoch": 1.1865021770682147,
      "grad_norm": 1.03125,
      "learning_rate": 0.000486558182991322,
      "loss": 5.1199,
      "mean_token_accuracy": 0.19888761788606643,
      "num_tokens": 27151630.0,
      "step": 13080
    },
    {
      "entropy": 5.421214866638183,
      "epoch": 1.186955732946299,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004865471647005304,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20177824348211287,
      "num_tokens": 27160734.0,
      "step": 13085
    },
    {
      "entropy": 5.435471391677856,
      "epoch": 1.1874092888243832,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048653614203488947,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19972168803215026,
      "num_tokens": 27171584.0,
      "step": 13090
    },
    {
      "entropy": 5.436698722839355,
      "epoch": 1.1878628447024673,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004865251149946273,
      "loss": 4.9972,
      "mean_token_accuracy": 0.2030612885951996,
      "num_tokens": 27180817.0,
      "step": 13095
    },
    {
      "entropy": 5.495074653625489,
      "epoch": 1.1883164005805515,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048651408357997173,
      "loss": 5.0673,
      "mean_token_accuracy": 0.19619399309158325,
      "num_tokens": 27190363.0,
      "step": 13100
    },
    {
      "entropy": 5.454643440246582,
      "epoch": 1.1887699564586356,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048650304779115114,
      "loss": 5.0202,
      "mean_token_accuracy": 0.20756469517946244,
      "num_tokens": 27202181.0,
      "step": 13105
    },
    {
      "entropy": 5.412134408950806,
      "epoch": 1.1892235123367199,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004864920076283935,
      "loss": 5.0849,
      "mean_token_accuracy": 0.19545914530754088,
      "num_tokens": 27212514.0,
      "step": 13110
    },
    {
      "entropy": 5.490122127532959,
      "epoch": 1.1896770682148041,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864809630919273,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20648339837789537,
      "num_tokens": 27223352.0,
      "step": 13115
    },
    {
      "entropy": 5.463384580612183,
      "epoch": 1.1901306240928882,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004864699141819809,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20400721579790115,
      "num_tokens": 27232643.0,
      "step": 13120
    },
    {
      "entropy": 5.384722566604614,
      "epoch": 1.1905841799709724,
      "grad_norm": 1.0,
      "learning_rate": 0.00048645886089878276,
      "loss": 5.0515,
      "mean_token_accuracy": 0.20001365542411803,
      "num_tokens": 27243793.0,
      "step": 13125
    },
    {
      "entropy": 5.4541233539581295,
      "epoch": 1.1910377358490567,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048644780324256153,
      "loss": 5.0645,
      "mean_token_accuracy": 0.1945493534207344,
      "num_tokens": 27254554.0,
      "step": 13130
    },
    {
      "entropy": 5.475072622299194,
      "epoch": 1.1914912917271407,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004864367412135458,
      "loss": 5.1231,
      "mean_token_accuracy": 0.19704449623823167,
      "num_tokens": 27265842.0,
      "step": 13135
    },
    {
      "entropy": 5.480845308303833,
      "epoch": 1.191944847605225,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048642567481196446,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1948697656393051,
      "num_tokens": 27277617.0,
      "step": 13140
    },
    {
      "entropy": 5.415107583999633,
      "epoch": 1.192398403483309,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004864146040380462,
      "loss": 5.0305,
      "mean_token_accuracy": 0.1978158101439476,
      "num_tokens": 27287450.0,
      "step": 13145
    },
    {
      "entropy": 5.43112621307373,
      "epoch": 1.1928519593613933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004864035288920202,
      "loss": 5.0508,
      "mean_token_accuracy": 0.19773907363414764,
      "num_tokens": 27297010.0,
      "step": 13150
    },
    {
      "entropy": 5.426964807510376,
      "epoch": 1.1933055152394776,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048639244937411516,
      "loss": 5.0932,
      "mean_token_accuracy": 0.19573701322078704,
      "num_tokens": 27306874.0,
      "step": 13155
    },
    {
      "entropy": 5.491222953796386,
      "epoch": 1.1937590711175616,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048638136548456045,
      "loss": 5.0884,
      "mean_token_accuracy": 0.1987365409731865,
      "num_tokens": 27318762.0,
      "step": 13160
    },
    {
      "entropy": 5.4339982032775875,
      "epoch": 1.194212626995646,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048637027722358536,
      "loss": 4.968,
      "mean_token_accuracy": 0.214048133790493,
      "num_tokens": 27327231.0,
      "step": 13165
    },
    {
      "entropy": 5.370849227905273,
      "epoch": 1.1946661828737302,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048635918459141893,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20344783663749694,
      "num_tokens": 27338517.0,
      "step": 13170
    },
    {
      "entropy": 5.4767357349395756,
      "epoch": 1.1951197387518142,
      "grad_norm": 1.0,
      "learning_rate": 0.00048634808758829073,
      "loss": 5.0785,
      "mean_token_accuracy": 0.19354832619428636,
      "num_tokens": 27348861.0,
      "step": 13175
    },
    {
      "entropy": 5.46518726348877,
      "epoch": 1.1955732946298985,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048633698621443024,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20526906549930574,
      "num_tokens": 27358610.0,
      "step": 13180
    },
    {
      "entropy": 5.4275483131408695,
      "epoch": 1.1960268505079825,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000486325880470067,
      "loss": 5.1218,
      "mean_token_accuracy": 0.19882804602384568,
      "num_tokens": 27369707.0,
      "step": 13185
    },
    {
      "entropy": 5.4541737079620365,
      "epoch": 1.1964804063860668,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004863147703554307,
      "loss": 5.0667,
      "mean_token_accuracy": 0.19907967001199722,
      "num_tokens": 27381030.0,
      "step": 13190
    },
    {
      "entropy": 5.501768779754639,
      "epoch": 1.196933962264151,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048630365587075115,
      "loss": 5.0918,
      "mean_token_accuracy": 0.1956538215279579,
      "num_tokens": 27390866.0,
      "step": 13195
    },
    {
      "entropy": 5.478093004226684,
      "epoch": 1.197387518142235,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048629253701625816,
      "loss": 5.0222,
      "mean_token_accuracy": 0.205246639251709,
      "num_tokens": 27400724.0,
      "step": 13200
    },
    {
      "entropy": 5.450027704238892,
      "epoch": 1.1978410740203194,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004862814137921816,
      "loss": 5.0402,
      "mean_token_accuracy": 0.19982354640960692,
      "num_tokens": 27411212.0,
      "step": 13205
    },
    {
      "entropy": 5.4368507862091064,
      "epoch": 1.1982946298984034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048627028619875157,
      "loss": 5.074,
      "mean_token_accuracy": 0.19660384058952332,
      "num_tokens": 27421279.0,
      "step": 13210
    },
    {
      "entropy": 5.4540740013122555,
      "epoch": 1.1987481857764877,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004862591542361982,
      "loss": 5.0459,
      "mean_token_accuracy": 0.20009150803089143,
      "num_tokens": 27432520.0,
      "step": 13215
    },
    {
      "entropy": 5.392856073379517,
      "epoch": 1.199201741654572,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004862480179047516,
      "loss": 4.9837,
      "mean_token_accuracy": 0.20842310637235642,
      "num_tokens": 27442246.0,
      "step": 13220
    },
    {
      "entropy": 5.449037837982178,
      "epoch": 1.199655297532656,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004862368772046423,
      "loss": 5.045,
      "mean_token_accuracy": 0.19735703915357589,
      "num_tokens": 27452134.0,
      "step": 13225
    },
    {
      "entropy": 5.461572074890137,
      "epoch": 1.2001088534107403,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048622573213610057,
      "loss": 5.1387,
      "mean_token_accuracy": 0.1919248580932617,
      "num_tokens": 27462253.0,
      "step": 13230
    },
    {
      "entropy": 5.455710411071777,
      "epoch": 1.2005624092888243,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048621458269935676,
      "loss": 5.0398,
      "mean_token_accuracy": 0.2015654996037483,
      "num_tokens": 27473131.0,
      "step": 13235
    },
    {
      "entropy": 5.46015772819519,
      "epoch": 1.2010159651669086,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048620342889464153,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20629900246858596,
      "num_tokens": 27483683.0,
      "step": 13240
    },
    {
      "entropy": 5.359331655502319,
      "epoch": 1.2014695210449928,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048619227072218565,
      "loss": 5.043,
      "mean_token_accuracy": 0.19450653046369554,
      "num_tokens": 27493679.0,
      "step": 13245
    },
    {
      "entropy": 5.400630521774292,
      "epoch": 1.2019230769230769,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004861811081822198,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20626620203256607,
      "num_tokens": 27503897.0,
      "step": 13250
    },
    {
      "entropy": 5.507361555099488,
      "epoch": 1.2023766328011611,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004861699412749748,
      "loss": 5.1281,
      "mean_token_accuracy": 0.192038594186306,
      "num_tokens": 27514560.0,
      "step": 13255
    },
    {
      "entropy": 5.431878185272216,
      "epoch": 1.2028301886792452,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004861587700006816,
      "loss": 5.0268,
      "mean_token_accuracy": 0.19795208871364595,
      "num_tokens": 27525594.0,
      "step": 13260
    },
    {
      "entropy": 5.369587707519531,
      "epoch": 1.2032837445573294,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048614759435957114,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20556827783584594,
      "num_tokens": 27535676.0,
      "step": 13265
    },
    {
      "entropy": 5.386460971832276,
      "epoch": 1.2037373004354137,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004861364143518747,
      "loss": 5.0072,
      "mean_token_accuracy": 0.19516339600086213,
      "num_tokens": 27545520.0,
      "step": 13270
    },
    {
      "entropy": 5.525118112564087,
      "epoch": 1.2041908563134978,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004861252299778234,
      "loss": 5.0856,
      "mean_token_accuracy": 0.19029136896133422,
      "num_tokens": 27556866.0,
      "step": 13275
    },
    {
      "entropy": 5.426550960540771,
      "epoch": 1.204644412191582,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048611404123764855,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20249955207109452,
      "num_tokens": 27566951.0,
      "step": 13280
    },
    {
      "entropy": 5.428686857223511,
      "epoch": 1.205097968069666,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048610284813158154,
      "loss": 5.1326,
      "mean_token_accuracy": 0.19744616746902466,
      "num_tokens": 27576796.0,
      "step": 13285
    },
    {
      "entropy": 5.387802028656006,
      "epoch": 1.2055515239477503,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004860916506598538,
      "loss": 4.9886,
      "mean_token_accuracy": 0.20070512294769288,
      "num_tokens": 27586902.0,
      "step": 13290
    },
    {
      "entropy": 5.557977724075317,
      "epoch": 1.2060050798258346,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048608044882269695,
      "loss": 5.1751,
      "mean_token_accuracy": 0.1885147735476494,
      "num_tokens": 27599232.0,
      "step": 13295
    },
    {
      "entropy": 5.423964977264404,
      "epoch": 1.2064586357039186,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004860692426203427,
      "loss": 4.9695,
      "mean_token_accuracy": 0.19955171942710875,
      "num_tokens": 27609022.0,
      "step": 13300
    },
    {
      "entropy": 5.461498355865478,
      "epoch": 1.206912191582003,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004860580320530226,
      "loss": 5.1252,
      "mean_token_accuracy": 0.2006087601184845,
      "num_tokens": 27619829.0,
      "step": 13305
    },
    {
      "entropy": 5.458715867996216,
      "epoch": 1.2073657474600872,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004860468171209687,
      "loss": 5.1272,
      "mean_token_accuracy": 0.205771704018116,
      "num_tokens": 27629337.0,
      "step": 13310
    },
    {
      "entropy": 5.517880821228028,
      "epoch": 1.2078193033381712,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004860355978244128,
      "loss": 5.1399,
      "mean_token_accuracy": 0.19886357337236404,
      "num_tokens": 27639377.0,
      "step": 13315
    },
    {
      "entropy": 5.468346691131591,
      "epoch": 1.2082728592162555,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000486024374163587,
      "loss": 5.0213,
      "mean_token_accuracy": 0.19874342381954194,
      "num_tokens": 27649989.0,
      "step": 13320
    },
    {
      "entropy": 5.4235265254974365,
      "epoch": 1.2087264150943395,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048601314613872334,
      "loss": 5.1085,
      "mean_token_accuracy": 0.18839994966983795,
      "num_tokens": 27659967.0,
      "step": 13325
    },
    {
      "entropy": 5.462990427017212,
      "epoch": 1.2091799709724238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000486001913750054,
      "loss": 5.0514,
      "mean_token_accuracy": 0.2000204548239708,
      "num_tokens": 27670099.0,
      "step": 13330
    },
    {
      "entropy": 5.4266736030578615,
      "epoch": 1.209633526850508,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859906769978113,
      "loss": 5.1068,
      "mean_token_accuracy": 0.190760999917984,
      "num_tokens": 27681497.0,
      "step": 13335
    },
    {
      "entropy": 5.539730072021484,
      "epoch": 1.2100870827285921,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048597943588222776,
      "loss": 5.1977,
      "mean_token_accuracy": 0.1931617110967636,
      "num_tokens": 27691828.0,
      "step": 13340
    },
    {
      "entropy": 5.5186830997467045,
      "epoch": 1.2105406386066764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048596819040353564,
      "loss": 5.0064,
      "mean_token_accuracy": 0.2051544338464737,
      "num_tokens": 27701426.0,
      "step": 13345
    },
    {
      "entropy": 5.463600921630859,
      "epoch": 1.2109941944847606,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859569405619676,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20318714678287506,
      "num_tokens": 27711112.0,
      "step": 13350
    },
    {
      "entropy": 5.464934396743774,
      "epoch": 1.2114477503628447,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004859456863577562,
      "loss": 5.126,
      "mean_token_accuracy": 0.19313183426856995,
      "num_tokens": 27723160.0,
      "step": 13355
    },
    {
      "entropy": 5.474137687683106,
      "epoch": 1.211901306240929,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048593442779113437,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19417185932397843,
      "num_tokens": 27733143.0,
      "step": 13360
    },
    {
      "entropy": 5.44663667678833,
      "epoch": 1.212354862119013,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004859231648623348,
      "loss": 5.0624,
      "mean_token_accuracy": 0.19898829460144044,
      "num_tokens": 27743956.0,
      "step": 13365
    },
    {
      "entropy": 5.47234354019165,
      "epoch": 1.2128084179970973,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048591189757159035,
      "loss": 5.0764,
      "mean_token_accuracy": 0.2030792161822319,
      "num_tokens": 27754192.0,
      "step": 13370
    },
    {
      "entropy": 5.442453050613404,
      "epoch": 1.2132619738751815,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004859006259191342,
      "loss": 5.1129,
      "mean_token_accuracy": 0.18974043130874635,
      "num_tokens": 27763939.0,
      "step": 13375
    },
    {
      "entropy": 5.45769248008728,
      "epoch": 1.2137155297532656,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004858893499051993,
      "loss": 5.0787,
      "mean_token_accuracy": 0.20038015097379686,
      "num_tokens": 27774743.0,
      "step": 13380
    },
    {
      "entropy": 5.4658379554748535,
      "epoch": 1.2141690856313498,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004858780695300189,
      "loss": 5.0665,
      "mean_token_accuracy": 0.19677289873361586,
      "num_tokens": 27785344.0,
      "step": 13385
    },
    {
      "entropy": 5.48243556022644,
      "epoch": 1.2146226415094339,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004858667847938263,
      "loss": 5.1281,
      "mean_token_accuracy": 0.1950848549604416,
      "num_tokens": 27797101.0,
      "step": 13390
    },
    {
      "entropy": 5.41499433517456,
      "epoch": 1.2150761973875182,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048585549569685486,
      "loss": 5.0171,
      "mean_token_accuracy": 0.2033864140510559,
      "num_tokens": 27807581.0,
      "step": 13395
    },
    {
      "entropy": 5.413809013366699,
      "epoch": 1.2155297532656024,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048584420223933805,
      "loss": 4.9873,
      "mean_token_accuracy": 0.20077995657920839,
      "num_tokens": 27818170.0,
      "step": 13400
    },
    {
      "entropy": 5.400589656829834,
      "epoch": 1.2159833091436865,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004858329044215094,
      "loss": 5.0175,
      "mean_token_accuracy": 0.19965792447328568,
      "num_tokens": 27828153.0,
      "step": 13405
    },
    {
      "entropy": 5.457147598266602,
      "epoch": 1.2164368650217707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048582160224360253,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20848435014486313,
      "num_tokens": 27837681.0,
      "step": 13410
    },
    {
      "entropy": 5.440137815475464,
      "epoch": 1.2168904208998548,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004858102957058513,
      "loss": 5.0955,
      "mean_token_accuracy": 0.2026829332113266,
      "num_tokens": 27847447.0,
      "step": 13415
    },
    {
      "entropy": 5.371809101104736,
      "epoch": 1.217343976777939,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048579898480848933,
      "loss": 4.9514,
      "mean_token_accuracy": 0.2035052314400673,
      "num_tokens": 27858586.0,
      "step": 13420
    },
    {
      "entropy": 5.458023548126221,
      "epoch": 1.2177975326560233,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004857876695517507,
      "loss": 5.0823,
      "mean_token_accuracy": 0.19574685841798783,
      "num_tokens": 27868832.0,
      "step": 13425
    },
    {
      "entropy": 5.450125980377197,
      "epoch": 1.2182510885341074,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048577634993586923,
      "loss": 5.0781,
      "mean_token_accuracy": 0.1965117871761322,
      "num_tokens": 27879339.0,
      "step": 13430
    },
    {
      "entropy": 5.49401478767395,
      "epoch": 1.2187046444121916,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048576502596107924,
      "loss": 5.0731,
      "mean_token_accuracy": 0.19752632081508636,
      "num_tokens": 27889964.0,
      "step": 13435
    },
    {
      "entropy": 5.494561100006104,
      "epoch": 1.2191582002902757,
      "grad_norm": 1.0,
      "learning_rate": 0.0004857536976276148,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19315129667520523,
      "num_tokens": 27901004.0,
      "step": 13440
    },
    {
      "entropy": 5.422154235839844,
      "epoch": 1.21961175616836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004857423649357102,
      "loss": 5.0427,
      "mean_token_accuracy": 0.19650053530931472,
      "num_tokens": 27910788.0,
      "step": 13445
    },
    {
      "entropy": 5.49261589050293,
      "epoch": 1.2200653120464442,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048573102788559974,
      "loss": 5.082,
      "mean_token_accuracy": 0.20031410306692124,
      "num_tokens": 27921360.0,
      "step": 13450
    },
    {
      "entropy": 5.472883510589599,
      "epoch": 1.2205188679245282,
      "grad_norm": 1.0,
      "learning_rate": 0.0004857196864775179,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20247834622859956,
      "num_tokens": 27931889.0,
      "step": 13455
    },
    {
      "entropy": 5.419800710678101,
      "epoch": 1.2209724238026125,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048570834071169926,
      "loss": 5.0273,
      "mean_token_accuracy": 0.20334334075450897,
      "num_tokens": 27942049.0,
      "step": 13460
    },
    {
      "entropy": 5.491580820083618,
      "epoch": 1.2214259796806965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048569699058837856,
      "loss": 5.1455,
      "mean_token_accuracy": 0.20272215604782104,
      "num_tokens": 27952416.0,
      "step": 13465
    },
    {
      "entropy": 5.4474114894866945,
      "epoch": 1.2218795355587808,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004856856361077903,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20347577780485154,
      "num_tokens": 27962009.0,
      "step": 13470
    },
    {
      "entropy": 5.501205205917358,
      "epoch": 1.222333091436865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004856742772701694,
      "loss": 5.0844,
      "mean_token_accuracy": 0.19673027098178864,
      "num_tokens": 27972175.0,
      "step": 13475
    },
    {
      "entropy": 5.474965524673462,
      "epoch": 1.2227866473149491,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048566291407575073,
      "loss": 5.1027,
      "mean_token_accuracy": 0.2038879156112671,
      "num_tokens": 27982899.0,
      "step": 13480
    },
    {
      "entropy": 5.474923753738404,
      "epoch": 1.2232402031930334,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048565154652476937,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20354434847831726,
      "num_tokens": 27993231.0,
      "step": 13485
    },
    {
      "entropy": 5.419674634933472,
      "epoch": 1.2236937590711174,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004856401746174603,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20534394532442093,
      "num_tokens": 28003623.0,
      "step": 13490
    },
    {
      "entropy": 5.452385377883911,
      "epoch": 1.2241473149492017,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048562879835405874,
      "loss": 5.0147,
      "mean_token_accuracy": 0.20308402925729752,
      "num_tokens": 28013878.0,
      "step": 13495
    },
    {
      "entropy": 5.40763578414917,
      "epoch": 1.224600870827286,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004856174177348,
      "loss": 5.0931,
      "mean_token_accuracy": 0.20048608928918837,
      "num_tokens": 28024224.0,
      "step": 13500
    },
    {
      "entropy": 5.432199096679687,
      "epoch": 1.22505442670537,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004856060327599194,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2000358670949936,
      "num_tokens": 28034819.0,
      "step": 13505
    },
    {
      "entropy": 5.412619400024414,
      "epoch": 1.2255079825834543,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048559464342965233,
      "loss": 5.0721,
      "mean_token_accuracy": 0.19996053576469422,
      "num_tokens": 28044240.0,
      "step": 13510
    },
    {
      "entropy": 5.546529722213745,
      "epoch": 1.2259615384615385,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004855832497442344,
      "loss": 5.2026,
      "mean_token_accuracy": 0.1869696170091629,
      "num_tokens": 28055046.0,
      "step": 13515
    },
    {
      "entropy": 5.427810525894165,
      "epoch": 1.2264150943396226,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004855718517039012,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20430445671081543,
      "num_tokens": 28065272.0,
      "step": 13520
    },
    {
      "entropy": 5.476324367523193,
      "epoch": 1.2268686502177069,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048556044930888844,
      "loss": 4.9943,
      "mean_token_accuracy": 0.20883602052927017,
      "num_tokens": 28076255.0,
      "step": 13525
    },
    {
      "entropy": 5.46715235710144,
      "epoch": 1.2273222060957911,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004855490425594319,
      "loss": 5.1216,
      "mean_token_accuracy": 0.1979125589132309,
      "num_tokens": 28086258.0,
      "step": 13530
    },
    {
      "entropy": 5.460022735595703,
      "epoch": 1.2277757619738752,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004855376314557675,
      "loss": 5.0307,
      "mean_token_accuracy": 0.20612121671438216,
      "num_tokens": 28096497.0,
      "step": 13535
    },
    {
      "entropy": 5.485916566848755,
      "epoch": 1.2282293178519594,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004855262159981313,
      "loss": 5.0505,
      "mean_token_accuracy": 0.20545612424612045,
      "num_tokens": 28106282.0,
      "step": 13540
    },
    {
      "entropy": 5.418528652191162,
      "epoch": 1.2286828737300435,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048551479618675925,
      "loss": 5.0958,
      "mean_token_accuracy": 0.19765927642583847,
      "num_tokens": 28115771.0,
      "step": 13545
    },
    {
      "entropy": 5.4977963924407955,
      "epoch": 1.2291364296081277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048550337202188756,
      "loss": 5.02,
      "mean_token_accuracy": 0.20294783860445023,
      "num_tokens": 28125477.0,
      "step": 13550
    },
    {
      "entropy": 5.503011703491211,
      "epoch": 1.229589985486212,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004854919435037525,
      "loss": 5.064,
      "mean_token_accuracy": 0.19926833361387253,
      "num_tokens": 28135800.0,
      "step": 13555
    },
    {
      "entropy": 5.374794340133667,
      "epoch": 1.230043541364296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004854805106325904,
      "loss": 4.9404,
      "mean_token_accuracy": 0.20187063962221147,
      "num_tokens": 28146779.0,
      "step": 13560
    },
    {
      "entropy": 5.445360422134399,
      "epoch": 1.2304970972423803,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004854690734086378,
      "loss": 5.0992,
      "mean_token_accuracy": 0.1955277666449547,
      "num_tokens": 28156819.0,
      "step": 13565
    },
    {
      "entropy": 5.462018251419067,
      "epoch": 1.2309506531204644,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004854576318321311,
      "loss": 5.0786,
      "mean_token_accuracy": 0.19740512669086457,
      "num_tokens": 28168553.0,
      "step": 13570
    },
    {
      "entropy": 5.506776571273804,
      "epoch": 1.2314042089985486,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048544618590330685,
      "loss": 5.041,
      "mean_token_accuracy": 0.19851386845111846,
      "num_tokens": 28178064.0,
      "step": 13575
    },
    {
      "entropy": 5.536520624160767,
      "epoch": 1.231857764876633,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048543473562240194,
      "loss": 5.1733,
      "mean_token_accuracy": 0.19981323778629304,
      "num_tokens": 28188580.0,
      "step": 13580
    },
    {
      "entropy": 5.484105062484741,
      "epoch": 1.232311320754717,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000485423280989653,
      "loss": 5.143,
      "mean_token_accuracy": 0.19147918820381166,
      "num_tokens": 28199955.0,
      "step": 13585
    },
    {
      "entropy": 5.497591495513916,
      "epoch": 1.2327648766328012,
      "grad_norm": 0.984375,
      "learning_rate": 0.000485411822005297,
      "loss": 5.1,
      "mean_token_accuracy": 0.19627357721328736,
      "num_tokens": 28210506.0,
      "step": 13590
    },
    {
      "entropy": 5.454776954650879,
      "epoch": 1.2332184325108853,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048540035866957097,
      "loss": 5.0279,
      "mean_token_accuracy": 0.20769426971673965,
      "num_tokens": 28220486.0,
      "step": 13595
    },
    {
      "entropy": 5.52288236618042,
      "epoch": 1.2336719883889695,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004853888909827118,
      "loss": 5.1499,
      "mean_token_accuracy": 0.19164299666881562,
      "num_tokens": 28232668.0,
      "step": 13600
    },
    {
      "entropy": 5.452500867843628,
      "epoch": 1.2341255442670538,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004853774189449568,
      "loss": 5.0993,
      "mean_token_accuracy": 0.1957123190164566,
      "num_tokens": 28243733.0,
      "step": 13605
    },
    {
      "entropy": 5.523473739624023,
      "epoch": 1.2345791001451378,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048536594255654315,
      "loss": 5.1259,
      "mean_token_accuracy": 0.19753974080085754,
      "num_tokens": 28254953.0,
      "step": 13610
    },
    {
      "entropy": 5.504914999008179,
      "epoch": 1.235032656023222,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004853544618177083,
      "loss": 5.0639,
      "mean_token_accuracy": 0.2029399037361145,
      "num_tokens": 28264959.0,
      "step": 13615
    },
    {
      "entropy": 5.5076741695404055,
      "epoch": 1.2354862119013061,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048534297672868945,
      "loss": 5.0522,
      "mean_token_accuracy": 0.20373029857873917,
      "num_tokens": 28275509.0,
      "step": 13620
    },
    {
      "entropy": 5.466273307800293,
      "epoch": 1.2359397677793904,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048533148728972434,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20362340658903122,
      "num_tokens": 28285983.0,
      "step": 13625
    },
    {
      "entropy": 5.508877849578857,
      "epoch": 1.2363933236574747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048531999350105037,
      "loss": 5.1072,
      "mean_token_accuracy": 0.19323109984397888,
      "num_tokens": 28296178.0,
      "step": 13630
    },
    {
      "entropy": 5.4554966449737545,
      "epoch": 1.2368468795355587,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048530849536290537,
      "loss": 5.0641,
      "mean_token_accuracy": 0.19579949975013733,
      "num_tokens": 28305741.0,
      "step": 13635
    },
    {
      "entropy": 5.405771970748901,
      "epoch": 1.237300435413643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004852969928755271,
      "loss": 5.0228,
      "mean_token_accuracy": 0.2004196122288704,
      "num_tokens": 28315607.0,
      "step": 13640
    },
    {
      "entropy": 5.41153917312622,
      "epoch": 1.237753991291727,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048528548603915344,
      "loss": 5.0409,
      "mean_token_accuracy": 0.20086603313684465,
      "num_tokens": 28327349.0,
      "step": 13645
    },
    {
      "entropy": 5.455974292755127,
      "epoch": 1.2382075471698113,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048527397485402236,
      "loss": 5.0228,
      "mean_token_accuracy": 0.19990852177143098,
      "num_tokens": 28339949.0,
      "step": 13650
    },
    {
      "entropy": 5.482010841369629,
      "epoch": 1.2386611030478956,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048526245932037186,
      "loss": 5.068,
      "mean_token_accuracy": 0.202985480427742,
      "num_tokens": 28351340.0,
      "step": 13655
    },
    {
      "entropy": 5.355767774581909,
      "epoch": 1.2391146589259796,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048525093943844013,
      "loss": 4.919,
      "mean_token_accuracy": 0.209889517724514,
      "num_tokens": 28360929.0,
      "step": 13660
    },
    {
      "entropy": 5.409904336929321,
      "epoch": 1.2395682148040639,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004852394152084654,
      "loss": 5.0201,
      "mean_token_accuracy": 0.20112548023462296,
      "num_tokens": 28370477.0,
      "step": 13665
    },
    {
      "entropy": 5.431086206436158,
      "epoch": 1.240021770682148,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000485227886630686,
      "loss": 4.9522,
      "mean_token_accuracy": 0.21357503831386565,
      "num_tokens": 28380876.0,
      "step": 13670
    },
    {
      "entropy": 5.358803033828735,
      "epoch": 1.2404753265602322,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004852163537053403,
      "loss": 4.9369,
      "mean_token_accuracy": 0.19906609058380126,
      "num_tokens": 28390685.0,
      "step": 13675
    },
    {
      "entropy": 5.3954808712005615,
      "epoch": 1.2409288824383164,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004852048164326669,
      "loss": 5.0412,
      "mean_token_accuracy": 0.1940809279680252,
      "num_tokens": 28400236.0,
      "step": 13680
    },
    {
      "entropy": 5.483770275115967,
      "epoch": 1.2413824383164005,
      "grad_norm": 1.125,
      "learning_rate": 0.0004851932748129043,
      "loss": 5.0809,
      "mean_token_accuracy": 0.20474082380533218,
      "num_tokens": 28410353.0,
      "step": 13685
    },
    {
      "entropy": 5.49216742515564,
      "epoch": 1.2418359941944848,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004851817288462912,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20008640140295028,
      "num_tokens": 28420241.0,
      "step": 13690
    },
    {
      "entropy": 5.438311290740967,
      "epoch": 1.242289550072569,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004851701785330663,
      "loss": 5.0811,
      "mean_token_accuracy": 0.19697158932685851,
      "num_tokens": 28430610.0,
      "step": 13695
    },
    {
      "entropy": 5.428465366363525,
      "epoch": 1.242743105950653,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004851586238734687,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20403676927089692,
      "num_tokens": 28440026.0,
      "step": 13700
    },
    {
      "entropy": 5.454707479476928,
      "epoch": 1.2431966618287373,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004851470648677371,
      "loss": 5.0842,
      "mean_token_accuracy": 0.18985789269208908,
      "num_tokens": 28450707.0,
      "step": 13705
    },
    {
      "entropy": 5.456290054321289,
      "epoch": 1.2436502177068216,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004851355015161107,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20259017944335939,
      "num_tokens": 28462378.0,
      "step": 13710
    },
    {
      "entropy": 5.535874938964843,
      "epoch": 1.2441037735849056,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004851239338188286,
      "loss": 5.1117,
      "mean_token_accuracy": 0.1960500493645668,
      "num_tokens": 28473084.0,
      "step": 13715
    },
    {
      "entropy": 5.461009693145752,
      "epoch": 1.24455732946299,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048511236177612994,
      "loss": 5.0429,
      "mean_token_accuracy": 0.19701926559209823,
      "num_tokens": 28482767.0,
      "step": 13720
    },
    {
      "entropy": 5.410448694229126,
      "epoch": 1.245010885341074,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048510078538825413,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20349868386983871,
      "num_tokens": 28493274.0,
      "step": 13725
    },
    {
      "entropy": 5.493898391723633,
      "epoch": 1.2454644412191582,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048508920465544056,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20370924472808838,
      "num_tokens": 28503823.0,
      "step": 13730
    },
    {
      "entropy": 5.419942474365234,
      "epoch": 1.2459179970972425,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004850776195779286,
      "loss": 5.1138,
      "mean_token_accuracy": 0.1970915362238884,
      "num_tokens": 28514209.0,
      "step": 13735
    },
    {
      "entropy": 5.500412511825561,
      "epoch": 1.2463715529753265,
      "grad_norm": 1.0,
      "learning_rate": 0.000485066030155958,
      "loss": 5.0868,
      "mean_token_accuracy": 0.19385533630847931,
      "num_tokens": 28524665.0,
      "step": 13740
    },
    {
      "entropy": 5.434648895263672,
      "epoch": 1.2468251088534108,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004850544363897684,
      "loss": 4.9809,
      "mean_token_accuracy": 0.2031462699174881,
      "num_tokens": 28533639.0,
      "step": 13745
    },
    {
      "entropy": 5.353348159790039,
      "epoch": 1.2472786647314948,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048504283827959947,
      "loss": 5.0081,
      "mean_token_accuracy": 0.20723264366388322,
      "num_tokens": 28544469.0,
      "step": 13750
    },
    {
      "entropy": 5.4903758525848385,
      "epoch": 1.2477322206095791,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004850312358256912,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20680339336395265,
      "num_tokens": 28554637.0,
      "step": 13755
    },
    {
      "entropy": 5.423876667022705,
      "epoch": 1.2481857764876634,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004850196290282834,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20637150704860688,
      "num_tokens": 28564752.0,
      "step": 13760
    },
    {
      "entropy": 5.469447374343872,
      "epoch": 1.2486393323657474,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048500801788761613,
      "loss": 5.1582,
      "mean_token_accuracy": 0.19746088087558747,
      "num_tokens": 28575007.0,
      "step": 13765
    },
    {
      "entropy": 5.523878669738769,
      "epoch": 1.2490928882438317,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004849964024039296,
      "loss": 5.0473,
      "mean_token_accuracy": 0.19712385386228562,
      "num_tokens": 28584907.0,
      "step": 13770
    },
    {
      "entropy": 5.538279438018799,
      "epoch": 1.2495464441219157,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048498478257746387,
      "loss": 5.1184,
      "mean_token_accuracy": 0.2048226147890091,
      "num_tokens": 28594244.0,
      "step": 13775
    },
    {
      "entropy": 5.463101148605347,
      "epoch": 1.25,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048497315840845934,
      "loss": 5.0777,
      "mean_token_accuracy": 0.20071172267198562,
      "num_tokens": 28603816.0,
      "step": 13780
    },
    {
      "entropy": 5.521854114532471,
      "epoch": 1.2504535558780843,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004849615298971565,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19445410519838333,
      "num_tokens": 28613067.0,
      "step": 13785
    },
    {
      "entropy": 5.486951446533203,
      "epoch": 1.2509071117561683,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048494989704379556,
      "loss": 5.0502,
      "mean_token_accuracy": 0.20105471462011337,
      "num_tokens": 28623752.0,
      "step": 13790
    },
    {
      "entropy": 5.413280344009399,
      "epoch": 1.2513606676342526,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004849382598486173,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20841633677482604,
      "num_tokens": 28633190.0,
      "step": 13795
    },
    {
      "entropy": 5.55495924949646,
      "epoch": 1.2518142235123366,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048492661831186243,
      "loss": 5.1081,
      "mean_token_accuracy": 0.19206711351871492,
      "num_tokens": 28644567.0,
      "step": 13800
    },
    {
      "entropy": 5.522542428970337,
      "epoch": 1.2522677793904209,
      "grad_norm": 1.0,
      "learning_rate": 0.0004849149724337715,
      "loss": 5.1058,
      "mean_token_accuracy": 0.1908365935087204,
      "num_tokens": 28654785.0,
      "step": 13805
    },
    {
      "entropy": 5.393986463546753,
      "epoch": 1.2527213352685052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048490332221458543,
      "loss": 4.9982,
      "mean_token_accuracy": 0.19900225549936296,
      "num_tokens": 28665286.0,
      "step": 13810
    },
    {
      "entropy": 5.48044376373291,
      "epoch": 1.2531748911465892,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004848916676545452,
      "loss": 5.1052,
      "mean_token_accuracy": 0.20103529691696168,
      "num_tokens": 28675830.0,
      "step": 13815
    },
    {
      "entropy": 5.53523325920105,
      "epoch": 1.2536284470246735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048488000875389174,
      "loss": 5.1604,
      "mean_token_accuracy": 0.1919257715344429,
      "num_tokens": 28686640.0,
      "step": 13820
    },
    {
      "entropy": 5.473912048339844,
      "epoch": 1.2540820029027575,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004848683455128663,
      "loss": 5.0548,
      "mean_token_accuracy": 0.19861929714679719,
      "num_tokens": 28697677.0,
      "step": 13825
    },
    {
      "entropy": 5.460227012634277,
      "epoch": 1.2545355587808418,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048485667793170994,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19863663166761397,
      "num_tokens": 28708831.0,
      "step": 13830
    },
    {
      "entropy": 5.488374185562134,
      "epoch": 1.254989114658926,
      "grad_norm": 1.0,
      "learning_rate": 0.000484845006010664,
      "loss": 5.1024,
      "mean_token_accuracy": 0.19582780748605727,
      "num_tokens": 28718959.0,
      "step": 13835
    },
    {
      "entropy": 5.497612953186035,
      "epoch": 1.25544267053701,
      "grad_norm": 1.0,
      "learning_rate": 0.0004848333297499699,
      "loss": 5.0388,
      "mean_token_accuracy": 0.2033858522772789,
      "num_tokens": 28728719.0,
      "step": 13840
    },
    {
      "entropy": 5.57854151725769,
      "epoch": 1.2558962264150944,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048482164914986903,
      "loss": 5.1998,
      "mean_token_accuracy": 0.19002826511859894,
      "num_tokens": 28740032.0,
      "step": 13845
    },
    {
      "entropy": 5.421471977233887,
      "epoch": 1.2563497822931784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048480996421060307,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20632674247026445,
      "num_tokens": 28749374.0,
      "step": 13850
    },
    {
      "entropy": 5.39776201248169,
      "epoch": 1.2568033381712627,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048479827493241347,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19689469933509826,
      "num_tokens": 28760038.0,
      "step": 13855
    },
    {
      "entropy": 5.452365922927856,
      "epoch": 1.257256894049347,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048478658131554216,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19886355251073837,
      "num_tokens": 28770247.0,
      "step": 13860
    },
    {
      "entropy": 5.440692329406739,
      "epoch": 1.2577104499274312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004847748833602308,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20380997955799102,
      "num_tokens": 28780251.0,
      "step": 13865
    },
    {
      "entropy": 5.389808464050293,
      "epoch": 1.2581640058055152,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004847631810667215,
      "loss": 5.0149,
      "mean_token_accuracy": 0.20636868774890899,
      "num_tokens": 28790829.0,
      "step": 13870
    },
    {
      "entropy": 5.435606670379639,
      "epoch": 1.2586175616835993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004847514744352561,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20586081147193908,
      "num_tokens": 28801036.0,
      "step": 13875
    },
    {
      "entropy": 5.370066547393799,
      "epoch": 1.2590711175616836,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004847397634660768,
      "loss": 4.9946,
      "mean_token_accuracy": 0.2021748438477516,
      "num_tokens": 28811013.0,
      "step": 13880
    },
    {
      "entropy": 5.3982508182525635,
      "epoch": 1.2595246734397678,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048472804815942575,
      "loss": 5.002,
      "mean_token_accuracy": 0.19941286295652388,
      "num_tokens": 28821367.0,
      "step": 13885
    },
    {
      "entropy": 5.469687032699585,
      "epoch": 1.259978229317852,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048471632851554517,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20566950142383575,
      "num_tokens": 28831077.0,
      "step": 13890
    },
    {
      "entropy": 5.4493091106414795,
      "epoch": 1.2604317851959361,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004847046045346775,
      "loss": 5.0759,
      "mean_token_accuracy": 0.19188063442707062,
      "num_tokens": 28841490.0,
      "step": 13895
    },
    {
      "entropy": 5.484811115264892,
      "epoch": 1.2608853410740204,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048469287621706516,
      "loss": 5.0706,
      "mean_token_accuracy": 0.1937114790081978,
      "num_tokens": 28852510.0,
      "step": 13900
    },
    {
      "entropy": 5.460968494415283,
      "epoch": 1.2613388969521044,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004846811435629507,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20472006350755692,
      "num_tokens": 28863388.0,
      "step": 13905
    },
    {
      "entropy": 5.3632509231567385,
      "epoch": 1.2617924528301887,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004846694065725768,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2048925742506981,
      "num_tokens": 28873798.0,
      "step": 13910
    },
    {
      "entropy": 5.428742408752441,
      "epoch": 1.262246008708273,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004846576652461861,
      "loss": 5.017,
      "mean_token_accuracy": 0.19878183901309968,
      "num_tokens": 28883482.0,
      "step": 13915
    },
    {
      "entropy": 5.449710321426392,
      "epoch": 1.262699564586357,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048464591958402147,
      "loss": 4.9776,
      "mean_token_accuracy": 0.20800818800926207,
      "num_tokens": 28893573.0,
      "step": 13920
    },
    {
      "entropy": 5.516650629043579,
      "epoch": 1.2631531204644413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048463416958632573,
      "loss": 5.1539,
      "mean_token_accuracy": 0.19524807929992677,
      "num_tokens": 28904064.0,
      "step": 13925
    },
    {
      "entropy": 5.480606603622436,
      "epoch": 1.2636066763425253,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048462241525334206,
      "loss": 5.0882,
      "mean_token_accuracy": 0.19799073934555053,
      "num_tokens": 28915329.0,
      "step": 13930
    },
    {
      "entropy": 5.4654511451721195,
      "epoch": 1.2640602322206096,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004846106565853133,
      "loss": 5.1555,
      "mean_token_accuracy": 0.19211496114730836,
      "num_tokens": 28925228.0,
      "step": 13935
    },
    {
      "entropy": 5.443673038482666,
      "epoch": 1.2645137880986939,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048459889358248284,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20181661546230317,
      "num_tokens": 28935891.0,
      "step": 13940
    },
    {
      "entropy": 5.448412036895752,
      "epoch": 1.264967343976778,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048458712624509383,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2014430433511734,
      "num_tokens": 28946372.0,
      "step": 13945
    },
    {
      "entropy": 5.474843120574951,
      "epoch": 1.2654208998548622,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048457535457338964,
      "loss": 5.0553,
      "mean_token_accuracy": 0.19741925448179246,
      "num_tokens": 28957874.0,
      "step": 13950
    },
    {
      "entropy": 5.495023775100708,
      "epoch": 1.2658744557329462,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004845635785676137,
      "loss": 5.0469,
      "mean_token_accuracy": 0.1934681475162506,
      "num_tokens": 28968023.0,
      "step": 13955
    },
    {
      "entropy": 5.414947462081909,
      "epoch": 1.2663280116110305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048455179822800957,
      "loss": 4.9437,
      "mean_token_accuracy": 0.20423877686262132,
      "num_tokens": 28977744.0,
      "step": 13960
    },
    {
      "entropy": 5.521876525878906,
      "epoch": 1.2667815674891147,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004845400135548208,
      "loss": 5.1851,
      "mean_token_accuracy": 0.19199081808328627,
      "num_tokens": 28988875.0,
      "step": 13965
    },
    {
      "entropy": 5.399349594116211,
      "epoch": 1.2672351233671988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004845282245482912,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20364806205034255,
      "num_tokens": 28998562.0,
      "step": 13970
    },
    {
      "entropy": 5.461163949966431,
      "epoch": 1.267688679245283,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004845164312086646,
      "loss": 5.0192,
      "mean_token_accuracy": 0.2067327618598938,
      "num_tokens": 29008783.0,
      "step": 13975
    },
    {
      "entropy": 5.3779669284820555,
      "epoch": 1.268142235123367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004845046335361847,
      "loss": 4.9629,
      "mean_token_accuracy": 0.19951594322919847,
      "num_tokens": 29019303.0,
      "step": 13980
    },
    {
      "entropy": 5.491613435745239,
      "epoch": 1.2685957910014514,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048449283153109566,
      "loss": 5.1747,
      "mean_token_accuracy": 0.1931134954094887,
      "num_tokens": 29029843.0,
      "step": 13985
    },
    {
      "entropy": 5.516561651229859,
      "epoch": 1.2690493468795356,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004844810251936413,
      "loss": 5.0933,
      "mean_token_accuracy": 0.19759776145219804,
      "num_tokens": 29040603.0,
      "step": 13990
    },
    {
      "entropy": 5.44688777923584,
      "epoch": 1.2695029027576197,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004844692145240661,
      "loss": 5.0275,
      "mean_token_accuracy": 0.2032545730471611,
      "num_tokens": 29050599.0,
      "step": 13995
    },
    {
      "entropy": 5.414919519424439,
      "epoch": 1.269956458635704,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004844573995226142,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19551022350788116,
      "num_tokens": 29061292.0,
      "step": 14000
    },
    {
      "entropy": 5.406631278991699,
      "epoch": 1.270410014513788,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048444558018952984,
      "loss": 4.9989,
      "mean_token_accuracy": 0.2096165671944618,
      "num_tokens": 29071779.0,
      "step": 14005
    },
    {
      "entropy": 5.416690397262573,
      "epoch": 1.2708635703918723,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048443375652505754,
      "loss": 4.9562,
      "mean_token_accuracy": 0.20473774522542953,
      "num_tokens": 29082073.0,
      "step": 14010
    },
    {
      "entropy": 5.43922872543335,
      "epoch": 1.2713171262699565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004844219285294418,
      "loss": 4.9564,
      "mean_token_accuracy": 0.20453754663467408,
      "num_tokens": 29092032.0,
      "step": 14015
    },
    {
      "entropy": 5.404605340957642,
      "epoch": 1.2717706821480406,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048441009620292714,
      "loss": 5.0012,
      "mean_token_accuracy": 0.2056380331516266,
      "num_tokens": 29102936.0,
      "step": 14020
    },
    {
      "entropy": 5.442543697357178,
      "epoch": 1.2722242380261248,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004843982595457584,
      "loss": 5.0258,
      "mean_token_accuracy": 0.203692264854908,
      "num_tokens": 29113385.0,
      "step": 14025
    },
    {
      "entropy": 5.4288288116455075,
      "epoch": 1.2726777939042089,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048438641855818026,
      "loss": 5.076,
      "mean_token_accuracy": 0.1968202218413353,
      "num_tokens": 29125075.0,
      "step": 14030
    },
    {
      "entropy": 5.4668896198272705,
      "epoch": 1.2731313497822931,
      "grad_norm": 1.0,
      "learning_rate": 0.0004843745732404376,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20056262612342834,
      "num_tokens": 29135098.0,
      "step": 14035
    },
    {
      "entropy": 5.551703929901123,
      "epoch": 1.2735849056603774,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004843627235927755,
      "loss": 5.1132,
      "mean_token_accuracy": 0.19678700864315032,
      "num_tokens": 29145577.0,
      "step": 14040
    },
    {
      "entropy": 5.447263574600219,
      "epoch": 1.2740384615384617,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004843508696154388,
      "loss": 5.0379,
      "mean_token_accuracy": 0.2010553538799286,
      "num_tokens": 29155905.0,
      "step": 14045
    },
    {
      "entropy": 5.362573432922363,
      "epoch": 1.2744920174165457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048433901130867283,
      "loss": 4.9685,
      "mean_token_accuracy": 0.20433792024850844,
      "num_tokens": 29164926.0,
      "step": 14050
    },
    {
      "entropy": 5.383613729476929,
      "epoch": 1.2749455732946298,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004843271486727228,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20115193128585815,
      "num_tokens": 29175050.0,
      "step": 14055
    },
    {
      "entropy": 5.432337713241577,
      "epoch": 1.275399129172714,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048431528170783395,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19764190763235093,
      "num_tokens": 29186255.0,
      "step": 14060
    },
    {
      "entropy": 5.472193193435669,
      "epoch": 1.2758526850507983,
      "grad_norm": 1.0,
      "learning_rate": 0.0004843034104142517,
      "loss": 5.0465,
      "mean_token_accuracy": 0.20842560082674028,
      "num_tokens": 29195639.0,
      "step": 14065
    },
    {
      "entropy": 5.45418119430542,
      "epoch": 1.2763062409288826,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004842915347922216,
      "loss": 5.0631,
      "mean_token_accuracy": 0.19702211916446685,
      "num_tokens": 29206482.0,
      "step": 14070
    },
    {
      "entropy": 5.485725736618042,
      "epoch": 1.2767597968069666,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004842796548419893,
      "loss": 5.0404,
      "mean_token_accuracy": 0.20077505707740784,
      "num_tokens": 29217433.0,
      "step": 14075
    },
    {
      "entropy": 5.506093978881836,
      "epoch": 1.2772133526850509,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048426777056380035,
      "loss": 5.1112,
      "mean_token_accuracy": 0.1898355156183243,
      "num_tokens": 29226193.0,
      "step": 14080
    },
    {
      "entropy": 5.413127422332764,
      "epoch": 1.277666908563135,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004842558819579006,
      "loss": 5.0447,
      "mean_token_accuracy": 0.2013655796647072,
      "num_tokens": 29237061.0,
      "step": 14085
    },
    {
      "entropy": 5.393286514282226,
      "epoch": 1.2781204644412192,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004842439890245358,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20250939279794694,
      "num_tokens": 29247689.0,
      "step": 14090
    },
    {
      "entropy": 5.413403701782227,
      "epoch": 1.2785740203193035,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048423209176395204,
      "loss": 5.0048,
      "mean_token_accuracy": 0.20171325504779816,
      "num_tokens": 29258235.0,
      "step": 14095
    },
    {
      "entropy": 5.405281209945679,
      "epoch": 1.2790275761973875,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004842201901763954,
      "loss": 4.9823,
      "mean_token_accuracy": 0.2064704939723015,
      "num_tokens": 29268884.0,
      "step": 14100
    },
    {
      "entropy": 5.420878267288208,
      "epoch": 1.2794811320754718,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004842082842621118,
      "loss": 5.1251,
      "mean_token_accuracy": 0.19449977129697799,
      "num_tokens": 29280484.0,
      "step": 14105
    },
    {
      "entropy": 5.477148818969726,
      "epoch": 1.2799346879535558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004841963740213476,
      "loss": 5.0879,
      "mean_token_accuracy": 0.20096696317195892,
      "num_tokens": 29291199.0,
      "step": 14110
    },
    {
      "entropy": 5.512815713882446,
      "epoch": 1.28038824383164,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004841844594543491,
      "loss": 5.0608,
      "mean_token_accuracy": 0.1944354072213173,
      "num_tokens": 29302007.0,
      "step": 14115
    },
    {
      "entropy": 5.461197090148926,
      "epoch": 1.2808417997097243,
      "grad_norm": 1.0,
      "learning_rate": 0.0004841725405613626,
      "loss": 5.0319,
      "mean_token_accuracy": 0.2100828170776367,
      "num_tokens": 29312127.0,
      "step": 14120
    },
    {
      "entropy": 5.467263507843017,
      "epoch": 1.2812953555878084,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004841606173426347,
      "loss": 5.1469,
      "mean_token_accuracy": 0.19667672961950303,
      "num_tokens": 29323691.0,
      "step": 14125
    },
    {
      "entropy": 5.434922504425049,
      "epoch": 1.2817489114658926,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048414868979841197,
      "loss": 5.0647,
      "mean_token_accuracy": 0.19760261178016664,
      "num_tokens": 29335203.0,
      "step": 14130
    },
    {
      "entropy": 5.494853067398071,
      "epoch": 1.2822024673439767,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048413675792894105,
      "loss": 5.0908,
      "mean_token_accuracy": 0.2013104259967804,
      "num_tokens": 29346422.0,
      "step": 14135
    },
    {
      "entropy": 5.44482274055481,
      "epoch": 1.282656023222061,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004841248217344686,
      "loss": 5.0054,
      "mean_token_accuracy": 0.1991575077176094,
      "num_tokens": 29356084.0,
      "step": 14140
    },
    {
      "entropy": 5.40657606124878,
      "epoch": 1.2831095791001452,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004841128812152415,
      "loss": 5.0157,
      "mean_token_accuracy": 0.20386681109666824,
      "num_tokens": 29366093.0,
      "step": 14145
    },
    {
      "entropy": 5.477137041091919,
      "epoch": 1.2835631349782293,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004841009363715068,
      "loss": 5.0718,
      "mean_token_accuracy": 0.2001335069537163,
      "num_tokens": 29376621.0,
      "step": 14150
    },
    {
      "entropy": 5.5024326801300045,
      "epoch": 1.2840166908563135,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004840889872035115,
      "loss": 5.1196,
      "mean_token_accuracy": 0.19854445606470109,
      "num_tokens": 29387119.0,
      "step": 14155
    },
    {
      "entropy": 5.455497598648071,
      "epoch": 1.2844702467343976,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004840770337115026,
      "loss": 5.0426,
      "mean_token_accuracy": 0.20350901484489442,
      "num_tokens": 29396981.0,
      "step": 14160
    },
    {
      "entropy": 5.488083553314209,
      "epoch": 1.2849238026124818,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048406507589572747,
      "loss": 5.0299,
      "mean_token_accuracy": 0.20519718825817107,
      "num_tokens": 29406737.0,
      "step": 14165
    },
    {
      "entropy": 5.475179767608642,
      "epoch": 1.2853773584905661,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004840531137564331,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20185788571834565,
      "num_tokens": 29417300.0,
      "step": 14170
    },
    {
      "entropy": 5.436309385299682,
      "epoch": 1.2858309143686502,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004840411472938672,
      "loss": 4.966,
      "mean_token_accuracy": 0.20452154278755189,
      "num_tokens": 29426749.0,
      "step": 14175
    },
    {
      "entropy": 5.393154668807983,
      "epoch": 1.2862844702467344,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004840291765082771,
      "loss": 4.9809,
      "mean_token_accuracy": 0.19870773255825042,
      "num_tokens": 29437264.0,
      "step": 14180
    },
    {
      "entropy": 5.422954511642456,
      "epoch": 1.2867380261248185,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048401720139991036,
      "loss": 4.9818,
      "mean_token_accuracy": 0.1949951395392418,
      "num_tokens": 29448192.0,
      "step": 14185
    },
    {
      "entropy": 5.45281662940979,
      "epoch": 1.2871915820029027,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004840052219690146,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2128344178199768,
      "num_tokens": 29457923.0,
      "step": 14190
    },
    {
      "entropy": 5.366837024688721,
      "epoch": 1.287645137880987,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004839932382158376,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21100290715694428,
      "num_tokens": 29466914.0,
      "step": 14195
    },
    {
      "entropy": 5.425601863861084,
      "epoch": 1.288098693759071,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004839812501406272,
      "loss": 5.0713,
      "mean_token_accuracy": 0.19851188957691193,
      "num_tokens": 29477648.0,
      "step": 14200
    },
    {
      "entropy": 5.4876306533813475,
      "epoch": 1.2885522496371553,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004839692577436313,
      "loss": 5.0509,
      "mean_token_accuracy": 0.20532836019992828,
      "num_tokens": 29487947.0,
      "step": 14205
    },
    {
      "entropy": 5.553531694412231,
      "epoch": 1.2890058055152394,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004839572610250978,
      "loss": 5.1424,
      "mean_token_accuracy": 0.1927253320813179,
      "num_tokens": 29497667.0,
      "step": 14210
    },
    {
      "entropy": 5.460983657836914,
      "epoch": 1.2894593613933236,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048394525998527496,
      "loss": 5.0264,
      "mean_token_accuracy": 0.19897037148475646,
      "num_tokens": 29507065.0,
      "step": 14215
    },
    {
      "entropy": 5.4825396060943605,
      "epoch": 1.2899129172714079,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004839332546244109,
      "loss": 5.1198,
      "mean_token_accuracy": 0.20061064064502715,
      "num_tokens": 29517752.0,
      "step": 14220
    },
    {
      "entropy": 5.454512977600098,
      "epoch": 1.2903664731494922,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048392124494275385,
      "loss": 5.028,
      "mean_token_accuracy": 0.1958656743168831,
      "num_tokens": 29527894.0,
      "step": 14225
    },
    {
      "entropy": 5.423327112197876,
      "epoch": 1.2908200290275762,
      "grad_norm": 1.0,
      "learning_rate": 0.0004839092309405523,
      "loss": 5.0654,
      "mean_token_accuracy": 0.19844336658716202,
      "num_tokens": 29538136.0,
      "step": 14230
    },
    {
      "entropy": 5.452183866500855,
      "epoch": 1.2912735849056602,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048389721261805457,
      "loss": 5.047,
      "mean_token_accuracy": 0.19840533882379532,
      "num_tokens": 29549193.0,
      "step": 14235
    },
    {
      "entropy": 5.387442398071289,
      "epoch": 1.2917271407837445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048388518997550926,
      "loss": 4.9525,
      "mean_token_accuracy": 0.20968773365020751,
      "num_tokens": 29558889.0,
      "step": 14240
    },
    {
      "entropy": 5.454190015792847,
      "epoch": 1.2921806966618288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048387316301316493,
      "loss": 5.0601,
      "mean_token_accuracy": 0.19536134749650955,
      "num_tokens": 29569691.0,
      "step": 14245
    },
    {
      "entropy": 5.480500888824463,
      "epoch": 1.292634252539913,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004838611317312705,
      "loss": 5.1387,
      "mean_token_accuracy": 0.19596640020608902,
      "num_tokens": 29580459.0,
      "step": 14250
    },
    {
      "entropy": 5.452585601806641,
      "epoch": 1.293087808417997,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004838490961300745,
      "loss": 5.0541,
      "mean_token_accuracy": 0.19633100181818008,
      "num_tokens": 29590652.0,
      "step": 14255
    },
    {
      "entropy": 5.474935817718506,
      "epoch": 1.2935413642960814,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048383705620982606,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20783319175243378,
      "num_tokens": 29600353.0,
      "step": 14260
    },
    {
      "entropy": 5.4646180152893065,
      "epoch": 1.2939949201741654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048382501197077405,
      "loss": 5.0185,
      "mean_token_accuracy": 0.20905023217201232,
      "num_tokens": 29610561.0,
      "step": 14265
    },
    {
      "entropy": 5.398465728759765,
      "epoch": 1.2944484760522497,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004838129634131675,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20370285362005233,
      "num_tokens": 29621741.0,
      "step": 14270
    },
    {
      "entropy": 5.408251619338989,
      "epoch": 1.294902031930334,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004838009105372558,
      "loss": 4.9551,
      "mean_token_accuracy": 0.20420809090137482,
      "num_tokens": 29633097.0,
      "step": 14275
    },
    {
      "entropy": 5.448822832107544,
      "epoch": 1.295355587808418,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048378885334328793,
      "loss": 5.0664,
      "mean_token_accuracy": 0.19318899959325792,
      "num_tokens": 29644290.0,
      "step": 14280
    },
    {
      "entropy": 5.518048477172852,
      "epoch": 1.2958091436865022,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004837767918315135,
      "loss": 5.1117,
      "mean_token_accuracy": 0.19329611957073212,
      "num_tokens": 29654977.0,
      "step": 14285
    },
    {
      "entropy": 5.52895302772522,
      "epoch": 1.2962626995645863,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048376472600218174,
      "loss": 5.1034,
      "mean_token_accuracy": 0.19857558757066726,
      "num_tokens": 29665020.0,
      "step": 14290
    },
    {
      "entropy": 5.4473686695098875,
      "epoch": 1.2967162554426706,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004837526558555422,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20379104614257812,
      "num_tokens": 29675098.0,
      "step": 14295
    },
    {
      "entropy": 5.395389080047607,
      "epoch": 1.2971698113207548,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004837405813918446,
      "loss": 4.9732,
      "mean_token_accuracy": 0.2069588765501976,
      "num_tokens": 29685450.0,
      "step": 14300
    },
    {
      "entropy": 5.399443626403809,
      "epoch": 1.2976233671988389,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048372850261133856,
      "loss": 4.9946,
      "mean_token_accuracy": 0.20608473420143128,
      "num_tokens": 29694818.0,
      "step": 14305
    },
    {
      "entropy": 5.476320552825928,
      "epoch": 1.2980769230769231,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048371641951427393,
      "loss": 5.0921,
      "mean_token_accuracy": 0.19565918147563935,
      "num_tokens": 29705573.0,
      "step": 14310
    },
    {
      "entropy": 5.466921043395996,
      "epoch": 1.2985304789550072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048370433210090055,
      "loss": 4.9994,
      "mean_token_accuracy": 0.20001791417598724,
      "num_tokens": 29716366.0,
      "step": 14315
    },
    {
      "entropy": 5.4390482902526855,
      "epoch": 1.2989840348330914,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004836922403714683,
      "loss": 5.0215,
      "mean_token_accuracy": 0.20087531805038453,
      "num_tokens": 29727181.0,
      "step": 14320
    },
    {
      "entropy": 5.393351268768311,
      "epoch": 1.2994375907111757,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048368014432622746,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20365506559610366,
      "num_tokens": 29736960.0,
      "step": 14325
    },
    {
      "entropy": 5.40035457611084,
      "epoch": 1.2998911465892597,
      "grad_norm": 1.03125,
      "learning_rate": 0.000483668043965428,
      "loss": 4.93,
      "mean_token_accuracy": 0.21387912034988404,
      "num_tokens": 29746354.0,
      "step": 14330
    },
    {
      "entropy": 5.445358324050903,
      "epoch": 1.300344702467344,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004836559392893203,
      "loss": 5.0449,
      "mean_token_accuracy": 0.19463613033294677,
      "num_tokens": 29756627.0,
      "step": 14335
    },
    {
      "entropy": 5.4277238845825195,
      "epoch": 1.300798258345428,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004836438302981545,
      "loss": 5.023,
      "mean_token_accuracy": 0.20784646421670913,
      "num_tokens": 29766684.0,
      "step": 14340
    },
    {
      "entropy": 5.427753973007202,
      "epoch": 1.3012518142235123,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004836317169921811,
      "loss": 5.055,
      "mean_token_accuracy": 0.20116524547338485,
      "num_tokens": 29776556.0,
      "step": 14345
    },
    {
      "entropy": 5.52408995628357,
      "epoch": 1.3017053701015966,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048361959937165075,
      "loss": 5.1917,
      "mean_token_accuracy": 0.19422274082899094,
      "num_tokens": 29786896.0,
      "step": 14350
    },
    {
      "entropy": 5.439397239685059,
      "epoch": 1.3021589259796806,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004836074774368138,
      "loss": 4.9857,
      "mean_token_accuracy": 0.20644939094781875,
      "num_tokens": 29797332.0,
      "step": 14355
    },
    {
      "entropy": 5.44417028427124,
      "epoch": 1.302612481857765,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048359535118792116,
      "loss": 5.0807,
      "mean_token_accuracy": 0.2013534665107727,
      "num_tokens": 29806428.0,
      "step": 14360
    },
    {
      "entropy": 5.437863159179687,
      "epoch": 1.303066037735849,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004835832206252235,
      "loss": 5.055,
      "mean_token_accuracy": 0.2009737566113472,
      "num_tokens": 29816794.0,
      "step": 14365
    },
    {
      "entropy": 5.52238097190857,
      "epoch": 1.3035195936139332,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004835710857489716,
      "loss": 5.0781,
      "mean_token_accuracy": 0.20529444962739946,
      "num_tokens": 29827077.0,
      "step": 14370
    },
    {
      "entropy": 5.4323179721832275,
      "epoch": 1.3039731494920175,
      "grad_norm": 1.0,
      "learning_rate": 0.0004835589465594165,
      "loss": 5.0536,
      "mean_token_accuracy": 0.19651621133089064,
      "num_tokens": 29837868.0,
      "step": 14375
    },
    {
      "entropy": 5.445604848861694,
      "epoch": 1.3044267053701015,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048354680305680926,
      "loss": 5.0627,
      "mean_token_accuracy": 0.2027847409248352,
      "num_tokens": 29846929.0,
      "step": 14380
    },
    {
      "entropy": 5.41439905166626,
      "epoch": 1.3048802612481858,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048353465524140105,
      "loss": 5.0041,
      "mean_token_accuracy": 0.2011212959885597,
      "num_tokens": 29857905.0,
      "step": 14385
    },
    {
      "entropy": 5.437925243377686,
      "epoch": 1.3053338171262698,
      "grad_norm": 1.09375,
      "learning_rate": 0.000483522503113443,
      "loss": 5.0607,
      "mean_token_accuracy": 0.19779178202152253,
      "num_tokens": 29868230.0,
      "step": 14390
    },
    {
      "entropy": 5.440568447113037,
      "epoch": 1.305787373004354,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048351034667318634,
      "loss": 5.0212,
      "mean_token_accuracy": 0.2029603287577629,
      "num_tokens": 29879224.0,
      "step": 14395
    },
    {
      "entropy": 5.5207991123199465,
      "epoch": 1.3062409288824384,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004834981859208827,
      "loss": 5.1658,
      "mean_token_accuracy": 0.19206029772758484,
      "num_tokens": 29890416.0,
      "step": 14400
    },
    {
      "entropy": 5.47591118812561,
      "epoch": 1.3066944847605226,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048348602085678346,
      "loss": 5.1053,
      "mean_token_accuracy": 0.19681367427110671,
      "num_tokens": 29901064.0,
      "step": 14405
    },
    {
      "entropy": 5.4896162986755375,
      "epoch": 1.3071480406386067,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048347385148114004,
      "loss": 5.135,
      "mean_token_accuracy": 0.19611346125602722,
      "num_tokens": 29911084.0,
      "step": 14410
    },
    {
      "entropy": 5.397078657150269,
      "epoch": 1.3076015965166907,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048346167779420436,
      "loss": 4.986,
      "mean_token_accuracy": 0.21088134199380876,
      "num_tokens": 29921104.0,
      "step": 14415
    },
    {
      "entropy": 5.441667795181274,
      "epoch": 1.308055152394775,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048344949979622806,
      "loss": 5.0105,
      "mean_token_accuracy": 0.20658447295427323,
      "num_tokens": 29932014.0,
      "step": 14420
    },
    {
      "entropy": 5.416403722763062,
      "epoch": 1.3085087082728593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048343731748746294,
      "loss": 5.0246,
      "mean_token_accuracy": 0.20720082372426987,
      "num_tokens": 29941707.0,
      "step": 14425
    },
    {
      "entropy": 5.453537797927856,
      "epoch": 1.3089622641509435,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000483425130868161,
      "loss": 5.0319,
      "mean_token_accuracy": 0.19454951584339142,
      "num_tokens": 29952082.0,
      "step": 14430
    },
    {
      "entropy": 5.404423666000366,
      "epoch": 1.3094158200290276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004834129399385742,
      "loss": 4.8531,
      "mean_token_accuracy": 0.21807902902364731,
      "num_tokens": 29961744.0,
      "step": 14435
    },
    {
      "entropy": 5.438563823699951,
      "epoch": 1.3098693759071118,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004834007446989548,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20898614972829818,
      "num_tokens": 29972611.0,
      "step": 14440
    },
    {
      "entropy": 5.407641267776489,
      "epoch": 1.3103229317851959,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048338854514955474,
      "loss": 4.9419,
      "mean_token_accuracy": 0.2093118906021118,
      "num_tokens": 29982660.0,
      "step": 14445
    },
    {
      "entropy": 5.432706880569458,
      "epoch": 1.3107764876632801,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048337634129062655,
      "loss": 5.0649,
      "mean_token_accuracy": 0.19014720171689986,
      "num_tokens": 29993349.0,
      "step": 14450
    },
    {
      "entropy": 5.524210405349732,
      "epoch": 1.3112300435413644,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048336413312242255,
      "loss": 5.145,
      "mean_token_accuracy": 0.19443282335996628,
      "num_tokens": 30004474.0,
      "step": 14455
    },
    {
      "entropy": 5.499291038513183,
      "epoch": 1.3116835994194485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048335192064519514,
      "loss": 5.0975,
      "mean_token_accuracy": 0.19702026546001433,
      "num_tokens": 30015189.0,
      "step": 14460
    },
    {
      "entropy": 5.472742652893066,
      "epoch": 1.3121371552975327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048333970385919695,
      "loss": 4.9927,
      "mean_token_accuracy": 0.19688308984041214,
      "num_tokens": 30025491.0,
      "step": 14465
    },
    {
      "entropy": 5.4534389019012455,
      "epoch": 1.3125907111756168,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048332748276468053,
      "loss": 5.0529,
      "mean_token_accuracy": 0.2055317595601082,
      "num_tokens": 30034683.0,
      "step": 14470
    },
    {
      "entropy": 5.412027359008789,
      "epoch": 1.313044267053701,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004833152573618987,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20131270736455917,
      "num_tokens": 30044135.0,
      "step": 14475
    },
    {
      "entropy": 5.484115648269653,
      "epoch": 1.3134978229317853,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048330302765110436,
      "loss": 5.0915,
      "mean_token_accuracy": 0.20330243408679963,
      "num_tokens": 30054520.0,
      "step": 14480
    },
    {
      "entropy": 5.487312746047974,
      "epoch": 1.3139513788098693,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048329079363255024,
      "loss": 5.0749,
      "mean_token_accuracy": 0.19585316628217697,
      "num_tokens": 30065439.0,
      "step": 14485
    },
    {
      "entropy": 5.493306255340576,
      "epoch": 1.3144049346879536,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048327855530648934,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20039551556110383,
      "num_tokens": 30075362.0,
      "step": 14490
    },
    {
      "entropy": 5.432645845413208,
      "epoch": 1.3148584905660377,
      "grad_norm": 1.0,
      "learning_rate": 0.000483266312673175,
      "loss": 5.0265,
      "mean_token_accuracy": 0.19832777976989746,
      "num_tokens": 30084913.0,
      "step": 14495
    },
    {
      "entropy": 5.44658465385437,
      "epoch": 1.315312046444122,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004832540657328602,
      "loss": 5.0468,
      "mean_token_accuracy": 0.2005489856004715,
      "num_tokens": 30095704.0,
      "step": 14500
    },
    {
      "entropy": 5.435765361785888,
      "epoch": 1.3157656023222062,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004832418144857982,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20489873737096786,
      "num_tokens": 30105981.0,
      "step": 14505
    },
    {
      "entropy": 5.4482102394104,
      "epoch": 1.3162191582002902,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048322955893224245,
      "loss": 5.058,
      "mean_token_accuracy": 0.19121648818254472,
      "num_tokens": 30115772.0,
      "step": 14510
    },
    {
      "entropy": 5.466764068603515,
      "epoch": 1.3166727140783745,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048321729907244626,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20533003211021422,
      "num_tokens": 30127050.0,
      "step": 14515
    },
    {
      "entropy": 5.46243839263916,
      "epoch": 1.3171262699564585,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004832050349066634,
      "loss": 5.0496,
      "mean_token_accuracy": 0.1961934819817543,
      "num_tokens": 30138221.0,
      "step": 14520
    },
    {
      "entropy": 5.4993661403656,
      "epoch": 1.3175798258345428,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004831927664351473,
      "loss": 5.1244,
      "mean_token_accuracy": 0.1948306679725647,
      "num_tokens": 30150181.0,
      "step": 14525
    },
    {
      "entropy": 5.444983863830567,
      "epoch": 1.318033381712627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004831804936581518,
      "loss": 5.0907,
      "mean_token_accuracy": 0.20480622947216034,
      "num_tokens": 30160521.0,
      "step": 14530
    },
    {
      "entropy": 5.518583297729492,
      "epoch": 1.3184869375907111,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004831682165759305,
      "loss": 5.0528,
      "mean_token_accuracy": 0.20147155076265336,
      "num_tokens": 30170611.0,
      "step": 14535
    },
    {
      "entropy": 5.483779048919677,
      "epoch": 1.3189404934687954,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048315593518873757,
      "loss": 5.0191,
      "mean_token_accuracy": 0.2089202657341957,
      "num_tokens": 30179914.0,
      "step": 14540
    },
    {
      "entropy": 5.418671417236328,
      "epoch": 1.3193940493468794,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004831436494968268,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20389744341373445,
      "num_tokens": 30191426.0,
      "step": 14545
    },
    {
      "entropy": 5.445282936096191,
      "epoch": 1.3198476052249637,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048313135950045234,
      "loss": 4.9902,
      "mean_token_accuracy": 0.20250587314367294,
      "num_tokens": 30202813.0,
      "step": 14550
    },
    {
      "entropy": 5.443851375579834,
      "epoch": 1.320301161103048,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048311906519986824,
      "loss": 4.9323,
      "mean_token_accuracy": 0.20622701942920685,
      "num_tokens": 30213044.0,
      "step": 14555
    },
    {
      "entropy": 5.447887945175171,
      "epoch": 1.320754716981132,
      "grad_norm": 1.140625,
      "learning_rate": 0.000483106766595329,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20020903646945953,
      "num_tokens": 30223419.0,
      "step": 14560
    },
    {
      "entropy": 5.505366373062134,
      "epoch": 1.3212082728592163,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004830944636870887,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19486852139234542,
      "num_tokens": 30233203.0,
      "step": 14565
    },
    {
      "entropy": 5.464370012283325,
      "epoch": 1.3216618287373003,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048308215647540183,
      "loss": 5.0559,
      "mean_token_accuracy": 0.2010621652007103,
      "num_tokens": 30244604.0,
      "step": 14570
    },
    {
      "entropy": 5.494146108627319,
      "epoch": 1.3221153846153846,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048306984496052297,
      "loss": 5.1131,
      "mean_token_accuracy": 0.19592046290636062,
      "num_tokens": 30255355.0,
      "step": 14575
    },
    {
      "entropy": 5.54925045967102,
      "epoch": 1.3225689404934688,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048305752914270665,
      "loss": 5.1373,
      "mean_token_accuracy": 0.19726582765579223,
      "num_tokens": 30265665.0,
      "step": 14580
    },
    {
      "entropy": 5.539267683029175,
      "epoch": 1.3230224963715531,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048304520902220767,
      "loss": 5.1185,
      "mean_token_accuracy": 0.20208002030849456,
      "num_tokens": 30274347.0,
      "step": 14585
    },
    {
      "entropy": 5.495355844497681,
      "epoch": 1.3234760522496372,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004830328845992808,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20875369161367416,
      "num_tokens": 30284243.0,
      "step": 14590
    },
    {
      "entropy": 5.390436172485352,
      "epoch": 1.3239296081277212,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048302055587418074,
      "loss": 4.964,
      "mean_token_accuracy": 0.20331924855709077,
      "num_tokens": 30294327.0,
      "step": 14595
    },
    {
      "entropy": 5.410218811035156,
      "epoch": 1.3243831640058055,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004830082228471626,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19879168272018433,
      "num_tokens": 30304188.0,
      "step": 14600
    },
    {
      "entropy": 5.45996503829956,
      "epoch": 1.3248367198838897,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048299588551848146,
      "loss": 4.9852,
      "mean_token_accuracy": 0.2053225114941597,
      "num_tokens": 30314917.0,
      "step": 14605
    },
    {
      "entropy": 5.437911081314087,
      "epoch": 1.325290275761974,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048298354388839237,
      "loss": 4.9962,
      "mean_token_accuracy": 0.20336007922887803,
      "num_tokens": 30325221.0,
      "step": 14610
    },
    {
      "entropy": 5.50159010887146,
      "epoch": 1.325743831640058,
      "grad_norm": 1.0,
      "learning_rate": 0.0004829711979571506,
      "loss": 5.0996,
      "mean_token_accuracy": 0.20005955100059508,
      "num_tokens": 30336196.0,
      "step": 14615
    },
    {
      "entropy": 5.423232793807983,
      "epoch": 1.3261973875181423,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004829588477250115,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2151394471526146,
      "num_tokens": 30345546.0,
      "step": 14620
    },
    {
      "entropy": 5.445691871643066,
      "epoch": 1.3266509433962264,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004829464931922304,
      "loss": 5.0395,
      "mean_token_accuracy": 0.20275977700948716,
      "num_tokens": 30355672.0,
      "step": 14625
    },
    {
      "entropy": 5.442801475524902,
      "epoch": 1.3271044992743106,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048293413435906285,
      "loss": 5.0437,
      "mean_token_accuracy": 0.19868946373462676,
      "num_tokens": 30366233.0,
      "step": 14630
    },
    {
      "entropy": 5.471208190917968,
      "epoch": 1.327558055152395,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004829217712257644,
      "loss": 5.0438,
      "mean_token_accuracy": 0.20054723769426347,
      "num_tokens": 30377366.0,
      "step": 14635
    },
    {
      "entropy": 5.450821590423584,
      "epoch": 1.328011611030479,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048290940379259073,
      "loss": 5.0941,
      "mean_token_accuracy": 0.20034591108560562,
      "num_tokens": 30388022.0,
      "step": 14640
    },
    {
      "entropy": 5.457063865661621,
      "epoch": 1.3284651669085632,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048289703205979756,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20205169171094894,
      "num_tokens": 30399235.0,
      "step": 14645
    },
    {
      "entropy": 5.57108793258667,
      "epoch": 1.3289187227866472,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048288465602764084,
      "loss": 5.1142,
      "mean_token_accuracy": 0.20028597861528397,
      "num_tokens": 30409112.0,
      "step": 14650
    },
    {
      "entropy": 5.409694528579712,
      "epoch": 1.3293722786647315,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048287227569637645,
      "loss": 5.0199,
      "mean_token_accuracy": 0.20294926911592484,
      "num_tokens": 30420202.0,
      "step": 14655
    },
    {
      "entropy": 5.459199047088623,
      "epoch": 1.3298258345428158,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048285989106626036,
      "loss": 5.0698,
      "mean_token_accuracy": 0.20262259989976883,
      "num_tokens": 30431210.0,
      "step": 14660
    },
    {
      "entropy": 5.454660701751709,
      "epoch": 1.3302793904208998,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828475021375489,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20774459838867188,
      "num_tokens": 30441291.0,
      "step": 14665
    },
    {
      "entropy": 5.424936532974243,
      "epoch": 1.330732946298984,
      "grad_norm": 1.125,
      "learning_rate": 0.0004828351089104981,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21167429983615876,
      "num_tokens": 30450285.0,
      "step": 14670
    },
    {
      "entropy": 5.4106463432312015,
      "epoch": 1.3311865021770681,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048282271138536424,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20098748207092285,
      "num_tokens": 30460311.0,
      "step": 14675
    },
    {
      "entropy": 5.528666687011719,
      "epoch": 1.3316400580551524,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004828103095624037,
      "loss": 5.0549,
      "mean_token_accuracy": 0.2035180002450943,
      "num_tokens": 30470188.0,
      "step": 14680
    },
    {
      "entropy": 5.441724872589111,
      "epoch": 1.3320936139332367,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000482797903441873,
      "loss": 5.002,
      "mean_token_accuracy": 0.1982833057641983,
      "num_tokens": 30481406.0,
      "step": 14685
    },
    {
      "entropy": 5.374399185180664,
      "epoch": 1.3325471698113207,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004827854930240288,
      "loss": 5.004,
      "mean_token_accuracy": 0.2057785615324974,
      "num_tokens": 30492314.0,
      "step": 14690
    },
    {
      "entropy": 5.532073640823365,
      "epoch": 1.333000725689405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004827730783091276,
      "loss": 5.1381,
      "mean_token_accuracy": 0.19633910357952117,
      "num_tokens": 30503273.0,
      "step": 14695
    },
    {
      "entropy": 5.512577486038208,
      "epoch": 1.333454281567489,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048276065929742616,
      "loss": 5.0296,
      "mean_token_accuracy": 0.19977344423532487,
      "num_tokens": 30513582.0,
      "step": 14700
    },
    {
      "entropy": 5.413820219039917,
      "epoch": 1.3339078374455733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004827482359891814,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2092178627848625,
      "num_tokens": 30523261.0,
      "step": 14705
    },
    {
      "entropy": 5.468668031692505,
      "epoch": 1.3343613933236576,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048273580838465016,
      "loss": 5.076,
      "mean_token_accuracy": 0.2011701285839081,
      "num_tokens": 30533774.0,
      "step": 14710
    },
    {
      "entropy": 5.464613723754883,
      "epoch": 1.3348149492017416,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004827233764840894,
      "loss": 5.1199,
      "mean_token_accuracy": 0.19367510378360747,
      "num_tokens": 30544580.0,
      "step": 14715
    },
    {
      "entropy": 5.484494543075561,
      "epoch": 1.3352685050798259,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004827109402877564,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20472703725099564,
      "num_tokens": 30555105.0,
      "step": 14720
    },
    {
      "entropy": 5.46086688041687,
      "epoch": 1.33572206095791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004826984997959082,
      "loss": 5.1107,
      "mean_token_accuracy": 0.18970336467027665,
      "num_tokens": 30564912.0,
      "step": 14725
    },
    {
      "entropy": 5.424795866012573,
      "epoch": 1.3361756168359942,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000482686055008802,
      "loss": 5.029,
      "mean_token_accuracy": 0.19458736777305602,
      "num_tokens": 30575862.0,
      "step": 14730
    },
    {
      "entropy": 5.477669715881348,
      "epoch": 1.3366291727140784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048267360592669535,
      "loss": 5.15,
      "mean_token_accuracy": 0.19470789283514023,
      "num_tokens": 30586951.0,
      "step": 14735
    },
    {
      "entropy": 5.497680521011352,
      "epoch": 1.3370827285921625,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004826611525498456,
      "loss": 5.0709,
      "mean_token_accuracy": 0.19571490734815597,
      "num_tokens": 30597604.0,
      "step": 14740
    },
    {
      "entropy": 5.504551362991333,
      "epoch": 1.3375362844702468,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004826486948785103,
      "loss": 5.118,
      "mean_token_accuracy": 0.19376556128263472,
      "num_tokens": 30608157.0,
      "step": 14745
    },
    {
      "entropy": 5.462109375,
      "epoch": 1.3379898403483308,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048263623291294706,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20001840591430664,
      "num_tokens": 30618527.0,
      "step": 14750
    },
    {
      "entropy": 5.5271852016448975,
      "epoch": 1.338443396226415,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048262376665341355,
      "loss": 5.1486,
      "mean_token_accuracy": 0.19271586388349532,
      "num_tokens": 30629253.0,
      "step": 14755
    },
    {
      "entropy": 5.500084733963012,
      "epoch": 1.3388969521044993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004826112961001677,
      "loss": 5.107,
      "mean_token_accuracy": 0.18575113266706467,
      "num_tokens": 30639015.0,
      "step": 14760
    },
    {
      "entropy": 5.545000314712524,
      "epoch": 1.3393505079825834,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048259882125346735,
      "loss": 5.1807,
      "mean_token_accuracy": 0.18538891077041625,
      "num_tokens": 30650035.0,
      "step": 14765
    },
    {
      "entropy": 5.489197015762329,
      "epoch": 1.3398040638606676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048258634211357045,
      "loss": 5.086,
      "mean_token_accuracy": 0.2026127129793167,
      "num_tokens": 30659813.0,
      "step": 14770
    },
    {
      "entropy": 5.461983060836792,
      "epoch": 1.3402576197387517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004825738586807351,
      "loss": 5.0291,
      "mean_token_accuracy": 0.2014084503054619,
      "num_tokens": 30670243.0,
      "step": 14775
    },
    {
      "entropy": 5.427335739135742,
      "epoch": 1.340711175616836,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048256137095521945,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20176693052053452,
      "num_tokens": 30680056.0,
      "step": 14780
    },
    {
      "entropy": 5.461502981185913,
      "epoch": 1.3411647314949202,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048254887893728175,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20859941095113754,
      "num_tokens": 30690536.0,
      "step": 14785
    },
    {
      "entropy": 5.427439641952515,
      "epoch": 1.3416182873730045,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004825363826271804,
      "loss": 5.0497,
      "mean_token_accuracy": 0.19469753801822662,
      "num_tokens": 30700200.0,
      "step": 14790
    },
    {
      "entropy": 5.4225091457366945,
      "epoch": 1.3420718432510885,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004825238820251737,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20724240243434905,
      "num_tokens": 30710576.0,
      "step": 14795
    },
    {
      "entropy": 5.439230823516846,
      "epoch": 1.3425253991291728,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048251137713152027,
      "loss": 5.0538,
      "mean_token_accuracy": 0.19843484163284303,
      "num_tokens": 30721495.0,
      "step": 14800
    },
    {
      "entropy": 5.488902759552002,
      "epoch": 1.3429789550072568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004824988679464787,
      "loss": 5.0783,
      "mean_token_accuracy": 0.2005630537867546,
      "num_tokens": 30731700.0,
      "step": 14805
    },
    {
      "entropy": 5.503637075424194,
      "epoch": 1.343432510885341,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048248635447030767,
      "loss": 5.0608,
      "mean_token_accuracy": 0.19916742742061616,
      "num_tokens": 30743421.0,
      "step": 14810
    },
    {
      "entropy": 5.453477621078491,
      "epoch": 1.3438860667634254,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004824738367032659,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20747420191764832,
      "num_tokens": 30754238.0,
      "step": 14815
    },
    {
      "entropy": 5.439966249465942,
      "epoch": 1.3443396226415094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004824613146456123,
      "loss": 5.0652,
      "mean_token_accuracy": 0.19901268929243088,
      "num_tokens": 30763938.0,
      "step": 14820
    },
    {
      "entropy": 5.524941539764404,
      "epoch": 1.3447931785195937,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004824487882976059,
      "loss": 5.121,
      "mean_token_accuracy": 0.19707119315862656,
      "num_tokens": 30773654.0,
      "step": 14825
    },
    {
      "entropy": 5.533320617675781,
      "epoch": 1.3452467343976777,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048243625765950564,
      "loss": 5.103,
      "mean_token_accuracy": 0.19529523849487304,
      "num_tokens": 30784659.0,
      "step": 14830
    },
    {
      "entropy": 5.427193021774292,
      "epoch": 1.345700290275762,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004824237227315708,
      "loss": 4.9791,
      "mean_token_accuracy": 0.1971741423010826,
      "num_tokens": 30795194.0,
      "step": 14835
    },
    {
      "entropy": 5.3818628787994385,
      "epoch": 1.3461538461538463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004824111835140604,
      "loss": 4.9139,
      "mean_token_accuracy": 0.2113494724035263,
      "num_tokens": 30805623.0,
      "step": 14840
    },
    {
      "entropy": 5.438431215286255,
      "epoch": 1.3466074020319303,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000482398640007234,
      "loss": 4.9711,
      "mean_token_accuracy": 0.20331282466650008,
      "num_tokens": 30814899.0,
      "step": 14845
    },
    {
      "entropy": 5.443796539306641,
      "epoch": 1.3470609579100146,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00048238609221135076,
      "loss": 5.0826,
      "mean_token_accuracy": 0.1992559924721718,
      "num_tokens": 30825546.0,
      "step": 14850
    },
    {
      "entropy": 5.39530725479126,
      "epoch": 1.3475145137880986,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004823735401266704,
      "loss": 4.9726,
      "mean_token_accuracy": 0.20329416990280152,
      "num_tokens": 30836406.0,
      "step": 14855
    },
    {
      "entropy": 5.427872180938721,
      "epoch": 1.3479680696661829,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048236098375345227,
      "loss": 5.073,
      "mean_token_accuracy": 0.20036971867084502,
      "num_tokens": 30847966.0,
      "step": 14860
    },
    {
      "entropy": 5.550434494018555,
      "epoch": 1.3484216255442671,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004823484230919563,
      "loss": 5.0939,
      "mean_token_accuracy": 0.19483736902475357,
      "num_tokens": 30856831.0,
      "step": 14865
    },
    {
      "entropy": 5.4240264892578125,
      "epoch": 1.3488751814223512,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048233585814244195,
      "loss": 4.8837,
      "mean_token_accuracy": 0.21467597633600236,
      "num_tokens": 30866076.0,
      "step": 14870
    },
    {
      "entropy": 5.4501073360443115,
      "epoch": 1.3493287373004355,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004823232889051693,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20600891709327698,
      "num_tokens": 30876845.0,
      "step": 14875
    },
    {
      "entropy": 5.420187616348267,
      "epoch": 1.3497822931785195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048231071538039816,
      "loss": 5.041,
      "mean_token_accuracy": 0.19574736207723617,
      "num_tokens": 30887448.0,
      "step": 14880
    },
    {
      "entropy": 5.497925186157227,
      "epoch": 1.3502358490566038,
      "grad_norm": 1.0,
      "learning_rate": 0.00048229813756838863,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20030652433633805,
      "num_tokens": 30898414.0,
      "step": 14885
    },
    {
      "entropy": 5.445967340469361,
      "epoch": 1.350689404934688,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004822855554694007,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19963205456733704,
      "num_tokens": 30908559.0,
      "step": 14890
    },
    {
      "entropy": 5.457993984222412,
      "epoch": 1.351142960812772,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048227296908369475,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20594154745340348,
      "num_tokens": 30918655.0,
      "step": 14895
    },
    {
      "entropy": 5.44824628829956,
      "epoch": 1.3515965166908563,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048226037841153104,
      "loss": 5.0499,
      "mean_token_accuracy": 0.20258834958076477,
      "num_tokens": 30929188.0,
      "step": 14900
    },
    {
      "entropy": 5.501089763641358,
      "epoch": 1.3520500725689404,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004822477834531698,
      "loss": 5.1109,
      "mean_token_accuracy": 0.19966886788606644,
      "num_tokens": 30939902.0,
      "step": 14905
    },
    {
      "entropy": 5.4724814891815186,
      "epoch": 1.3525036284470247,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004822351842088716,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20224396884441376,
      "num_tokens": 30951289.0,
      "step": 14910
    },
    {
      "entropy": 5.541313600540161,
      "epoch": 1.352957184325109,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004822225806788969,
      "loss": 5.1729,
      "mean_token_accuracy": 0.18792562633752824,
      "num_tokens": 30961233.0,
      "step": 14915
    },
    {
      "entropy": 5.476714181900024,
      "epoch": 1.353410740203193,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048220997286350654,
      "loss": 4.9991,
      "mean_token_accuracy": 0.20616774260997772,
      "num_tokens": 30971922.0,
      "step": 14920
    },
    {
      "entropy": 5.43091139793396,
      "epoch": 1.3538642960812772,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048219736076296114,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20542879700660704,
      "num_tokens": 30981358.0,
      "step": 14925
    },
    {
      "entropy": 5.482666444778443,
      "epoch": 1.3543178519593613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004821847443775214,
      "loss": 5.1073,
      "mean_token_accuracy": 0.19539569616317748,
      "num_tokens": 30991984.0,
      "step": 14930
    },
    {
      "entropy": 5.469381523132324,
      "epoch": 1.3547714078374455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048217212370744853,
      "loss": 4.9967,
      "mean_token_accuracy": 0.20533296167850495,
      "num_tokens": 31002790.0,
      "step": 14935
    },
    {
      "entropy": 5.367084264755249,
      "epoch": 1.3552249637155298,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048215949875300327,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2160383567214012,
      "num_tokens": 31012827.0,
      "step": 14940
    },
    {
      "entropy": 5.369181156158447,
      "epoch": 1.3556785195936139,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004821468695144467,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20431342273950576,
      "num_tokens": 31023075.0,
      "step": 14945
    },
    {
      "entropy": 5.48018445968628,
      "epoch": 1.3561320754716981,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048213423599204015,
      "loss": 5.055,
      "mean_token_accuracy": 0.20195306688547135,
      "num_tokens": 31032454.0,
      "step": 14950
    },
    {
      "entropy": 5.569077444076538,
      "epoch": 1.3565856313497822,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048212159818604477,
      "loss": 5.1833,
      "mean_token_accuracy": 0.18612317144870758,
      "num_tokens": 31042498.0,
      "step": 14955
    },
    {
      "entropy": 5.434992074966431,
      "epoch": 1.3570391872278664,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000482108956096722,
      "loss": 5.0274,
      "mean_token_accuracy": 0.20031053572893143,
      "num_tokens": 31053188.0,
      "step": 14960
    },
    {
      "entropy": 5.452586126327515,
      "epoch": 1.3574927431059507,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048209630972433326,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20466638952493668,
      "num_tokens": 31062840.0,
      "step": 14965
    },
    {
      "entropy": 5.5303045272827145,
      "epoch": 1.357946298984035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048208365906914,
      "loss": 5.1531,
      "mean_token_accuracy": 0.2004319131374359,
      "num_tokens": 31072646.0,
      "step": 14970
    },
    {
      "entropy": 5.42915096282959,
      "epoch": 1.358399854862119,
      "grad_norm": 1.125,
      "learning_rate": 0.00048207100413140397,
      "loss": 4.9793,
      "mean_token_accuracy": 0.19790226519107817,
      "num_tokens": 31082700.0,
      "step": 14975
    },
    {
      "entropy": 5.508495330810547,
      "epoch": 1.3588534107402033,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004820583449113868,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19902369678020476,
      "num_tokens": 31093743.0,
      "step": 14980
    },
    {
      "entropy": 5.455974435806274,
      "epoch": 1.3593069666182873,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048204568140935023,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2065230205655098,
      "num_tokens": 31104293.0,
      "step": 14985
    },
    {
      "entropy": 5.384525966644287,
      "epoch": 1.3597605224963716,
      "grad_norm": 1.0,
      "learning_rate": 0.0004820330136255563,
      "loss": 5.0208,
      "mean_token_accuracy": 0.20574056953191758,
      "num_tokens": 31114244.0,
      "step": 14990
    },
    {
      "entropy": 5.420985412597656,
      "epoch": 1.3602140783744558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048202034156026683,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20137730985879898,
      "num_tokens": 31123596.0,
      "step": 14995
    },
    {
      "entropy": 5.50952730178833,
      "epoch": 1.36066763425254,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048200766521374396,
      "loss": 5.0688,
      "mean_token_accuracy": 0.19272379130125045,
      "num_tokens": 31134258.0,
      "step": 15000
    },
    {
      "epoch": 1.36066763425254,
      "eval_entropy": 5.280845000733078,
      "eval_loss": 5.10365104675293,
      "eval_mean_token_accuracy": 0.20662164339011882,
      "eval_num_tokens": 31134258.0,
      "eval_runtime": 20.4823,
      "eval_samples_per_second": 1726.364,
      "eval_steps_per_second": 215.796,
      "step": 15000
    },
    {
      "entropy": 5.4724356174469,
      "epoch": 1.3611211901306242,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004819949845862498,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20538970232009887,
      "num_tokens": 31144806.0,
      "step": 15005
    },
    {
      "entropy": 5.517314434051514,
      "epoch": 1.3615747460087082,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004819822996780466,
      "loss": 5.1393,
      "mean_token_accuracy": 0.20090858042240142,
      "num_tokens": 31155642.0,
      "step": 15010
    },
    {
      "entropy": 5.508015060424805,
      "epoch": 1.3620283018867925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004819696104893967,
      "loss": 4.9951,
      "mean_token_accuracy": 0.20463416576385499,
      "num_tokens": 31165034.0,
      "step": 15015
    },
    {
      "entropy": 5.478755807876587,
      "epoch": 1.3624818577648767,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004819569170205625,
      "loss": 5.0517,
      "mean_token_accuracy": 0.19870482832193376,
      "num_tokens": 31176491.0,
      "step": 15020
    },
    {
      "entropy": 5.47834587097168,
      "epoch": 1.3629354136429608,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048194421927180654,
      "loss": 5.0701,
      "mean_token_accuracy": 0.19554078727960586,
      "num_tokens": 31186198.0,
      "step": 15025
    },
    {
      "entropy": 5.491092252731323,
      "epoch": 1.363388969521045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004819315172433914,
      "loss": 5.0431,
      "mean_token_accuracy": 0.2019846871495247,
      "num_tokens": 31197094.0,
      "step": 15030
    },
    {
      "entropy": 5.501713037490845,
      "epoch": 1.363842525399129,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048191881093557977,
      "loss": 5.0588,
      "mean_token_accuracy": 0.19390265941619872,
      "num_tokens": 31207505.0,
      "step": 15035
    },
    {
      "entropy": 5.543484306335449,
      "epoch": 1.3642960812772134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048190610034863434,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1883800745010376,
      "num_tokens": 31218809.0,
      "step": 15040
    },
    {
      "entropy": 5.492035722732544,
      "epoch": 1.3647496371552976,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004818933854828181,
      "loss": 5.0977,
      "mean_token_accuracy": 0.19075336456298828,
      "num_tokens": 31230281.0,
      "step": 15045
    },
    {
      "entropy": 5.417478942871094,
      "epoch": 1.3652031930333817,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004818806663383939,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20595351308584214,
      "num_tokens": 31239573.0,
      "step": 15050
    },
    {
      "entropy": 5.439194059371948,
      "epoch": 1.365656748911466,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004818679429156248,
      "loss": 5.0144,
      "mean_token_accuracy": 0.1995464190840721,
      "num_tokens": 31249671.0,
      "step": 15055
    },
    {
      "entropy": 5.4043097496032715,
      "epoch": 1.36611030478955,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048185521521477385,
      "loss": 5.0422,
      "mean_token_accuracy": 0.19515294283628465,
      "num_tokens": 31259601.0,
      "step": 15060
    },
    {
      "entropy": 5.423627138137817,
      "epoch": 1.3665638606676342,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818424832361044,
      "loss": 5.0211,
      "mean_token_accuracy": 0.20258173048496247,
      "num_tokens": 31269085.0,
      "step": 15065
    },
    {
      "entropy": 5.443383264541626,
      "epoch": 1.3670174165457185,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004818297469798797,
      "loss": 5.033,
      "mean_token_accuracy": 0.19837417751550673,
      "num_tokens": 31280177.0,
      "step": 15070
    },
    {
      "entropy": 5.475092983245849,
      "epoch": 1.3674709724238026,
      "grad_norm": 1.0,
      "learning_rate": 0.00048181700644636314,
      "loss": 4.959,
      "mean_token_accuracy": 0.20413506925106048,
      "num_tokens": 31290545.0,
      "step": 15075
    },
    {
      "entropy": 5.376138925552368,
      "epoch": 1.3679245283018868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048180426163581816,
      "loss": 4.8884,
      "mean_token_accuracy": 0.20839907675981523,
      "num_tokens": 31301143.0,
      "step": 15080
    },
    {
      "entropy": 5.407545137405395,
      "epoch": 1.3683780841799709,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004817915125485083,
      "loss": 4.9994,
      "mean_token_accuracy": 0.19968184381723403,
      "num_tokens": 31311071.0,
      "step": 15085
    },
    {
      "entropy": 5.510735940933228,
      "epoch": 1.3688316400580551,
      "grad_norm": 1.0,
      "learning_rate": 0.00048177875918469735,
      "loss": 5.0838,
      "mean_token_accuracy": 0.2015668496489525,
      "num_tokens": 31322263.0,
      "step": 15090
    },
    {
      "entropy": 5.480931663513184,
      "epoch": 1.3692851959361394,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048176600154464897,
      "loss": 4.9925,
      "mean_token_accuracy": 0.20168393552303315,
      "num_tokens": 31332975.0,
      "step": 15095
    },
    {
      "entropy": 5.530743408203125,
      "epoch": 1.3697387518142234,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004817532396286269,
      "loss": 5.1169,
      "mean_token_accuracy": 0.1979136660695076,
      "num_tokens": 31345252.0,
      "step": 15100
    },
    {
      "entropy": 5.473556995391846,
      "epoch": 1.3701923076923077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004817404734368952,
      "loss": 4.9884,
      "mean_token_accuracy": 0.20655723959207534,
      "num_tokens": 31355479.0,
      "step": 15105
    },
    {
      "entropy": 5.521392488479615,
      "epoch": 1.3706458635703918,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048172770296971785,
      "loss": 5.122,
      "mean_token_accuracy": 0.19433184564113617,
      "num_tokens": 31366020.0,
      "step": 15110
    },
    {
      "entropy": 5.481007862091064,
      "epoch": 1.371099419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004817149282273589,
      "loss": 5.045,
      "mean_token_accuracy": 0.19936117976903917,
      "num_tokens": 31376055.0,
      "step": 15115
    },
    {
      "entropy": 5.4966240406036375,
      "epoch": 1.3715529753265603,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004817021492100825,
      "loss": 5.1654,
      "mean_token_accuracy": 0.19234600365161897,
      "num_tokens": 31386179.0,
      "step": 15120
    },
    {
      "entropy": 5.415863752365112,
      "epoch": 1.3720065312046443,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000481689365918153,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20340771079063416,
      "num_tokens": 31395850.0,
      "step": 15125
    },
    {
      "entropy": 5.457096195220947,
      "epoch": 1.3724600870827286,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004816765783518348,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21304902583360671,
      "num_tokens": 31405053.0,
      "step": 15130
    },
    {
      "entropy": 5.412125158309936,
      "epoch": 1.3729136429608126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004816637865113923,
      "loss": 4.9607,
      "mean_token_accuracy": 0.21003830283880234,
      "num_tokens": 31415153.0,
      "step": 15135
    },
    {
      "entropy": 5.4197296619415285,
      "epoch": 1.373367198838897,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048165099039708994,
      "loss": 5.0389,
      "mean_token_accuracy": 0.199916335940361,
      "num_tokens": 31425451.0,
      "step": 15140
    },
    {
      "entropy": 5.41675443649292,
      "epoch": 1.3738207547169812,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004816381900091925,
      "loss": 5.0165,
      "mean_token_accuracy": 0.19917270839214324,
      "num_tokens": 31436298.0,
      "step": 15145
    },
    {
      "entropy": 5.454756546020508,
      "epoch": 1.3742743105950654,
      "grad_norm": 1.0,
      "learning_rate": 0.00048162538534796465,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20529780089855193,
      "num_tokens": 31446932.0,
      "step": 15150
    },
    {
      "entropy": 5.500065851211548,
      "epoch": 1.3747278664731495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004816125764136711,
      "loss": 5.0792,
      "mean_token_accuracy": 0.19022306948900222,
      "num_tokens": 31457582.0,
      "step": 15155
    },
    {
      "entropy": 5.544494867324829,
      "epoch": 1.3751814223512338,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004815997632065769,
      "loss": 5.0484,
      "mean_token_accuracy": 0.1995017260313034,
      "num_tokens": 31468168.0,
      "step": 15160
    },
    {
      "entropy": 5.470380640029907,
      "epoch": 1.3756349782293178,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048158694572694694,
      "loss": 5.0266,
      "mean_token_accuracy": 0.2055066630244255,
      "num_tokens": 31478524.0,
      "step": 15165
    },
    {
      "entropy": 5.424317693710327,
      "epoch": 1.376088534107402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048157412397504623,
      "loss": 4.9969,
      "mean_token_accuracy": 0.2064652994275093,
      "num_tokens": 31488608.0,
      "step": 15170
    },
    {
      "entropy": 5.419355297088623,
      "epoch": 1.3765420899854863,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004815612979511401,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20743164122104646,
      "num_tokens": 31498984.0,
      "step": 15175
    },
    {
      "entropy": 5.518174743652343,
      "epoch": 1.3769956458635704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048154846765549353,
      "loss": 5.0349,
      "mean_token_accuracy": 0.20261340588331223,
      "num_tokens": 31509110.0,
      "step": 15180
    },
    {
      "entropy": 5.468982219696045,
      "epoch": 1.3774492017416546,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048153563308837215,
      "loss": 5.0076,
      "mean_token_accuracy": 0.20486426800489427,
      "num_tokens": 31519266.0,
      "step": 15185
    },
    {
      "entropy": 5.471644926071167,
      "epoch": 1.3779027576197387,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048152279425004124,
      "loss": 5.0366,
      "mean_token_accuracy": 0.20559466779232025,
      "num_tokens": 31529043.0,
      "step": 15190
    },
    {
      "entropy": 5.476575899124145,
      "epoch": 1.378356313497823,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004815099511407663,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20737977772951127,
      "num_tokens": 31538650.0,
      "step": 15195
    },
    {
      "entropy": 5.406746435165405,
      "epoch": 1.3788098693759072,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000481497103760813,
      "loss": 4.9612,
      "mean_token_accuracy": 0.20198217034339905,
      "num_tokens": 31549392.0,
      "step": 15200
    },
    {
      "entropy": 5.485914802551269,
      "epoch": 1.3792634252539913,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048148425211044684,
      "loss": 5.1425,
      "mean_token_accuracy": 0.19214161485433578,
      "num_tokens": 31559999.0,
      "step": 15205
    },
    {
      "entropy": 5.496637058258057,
      "epoch": 1.3797169811320755,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004814713961899338,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19900211691856384,
      "num_tokens": 31570108.0,
      "step": 15210
    },
    {
      "entropy": 5.464858150482177,
      "epoch": 1.3801705370101596,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048145853599953966,
      "loss": 5.0068,
      "mean_token_accuracy": 0.19859155416488647,
      "num_tokens": 31579698.0,
      "step": 15215
    },
    {
      "entropy": 5.431590795516968,
      "epoch": 1.3806240928882438,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048144567153953036,
      "loss": 5.0683,
      "mean_token_accuracy": 0.1957996219396591,
      "num_tokens": 31589292.0,
      "step": 15220
    },
    {
      "entropy": 5.504480648040771,
      "epoch": 1.381077648766328,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048143280281017197,
      "loss": 5.1295,
      "mean_token_accuracy": 0.19753700643777847,
      "num_tokens": 31599848.0,
      "step": 15225
    },
    {
      "entropy": 5.509224843978882,
      "epoch": 1.3815312046444121,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004814199298117307,
      "loss": 5.0672,
      "mean_token_accuracy": 0.20320505499839783,
      "num_tokens": 31611109.0,
      "step": 15230
    },
    {
      "entropy": 5.423985385894776,
      "epoch": 1.3819847605224964,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048140705254447256,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20411137342453003,
      "num_tokens": 31621140.0,
      "step": 15235
    },
    {
      "entropy": 5.524167060852051,
      "epoch": 1.3824383164005805,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048139417100866405,
      "loss": 5.0873,
      "mean_token_accuracy": 0.19728801548480987,
      "num_tokens": 31631268.0,
      "step": 15240
    },
    {
      "entropy": 5.503660774230957,
      "epoch": 1.3828918722786647,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048138128520457146,
      "loss": 5.0234,
      "mean_token_accuracy": 0.2008575454354286,
      "num_tokens": 31640758.0,
      "step": 15245
    },
    {
      "entropy": 5.445882511138916,
      "epoch": 1.383345428156749,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004813683951324613,
      "loss": 5.0485,
      "mean_token_accuracy": 0.20046461671590804,
      "num_tokens": 31651060.0,
      "step": 15250
    },
    {
      "entropy": 5.504006481170654,
      "epoch": 1.383798984034833,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048135550079260014,
      "loss": 5.1242,
      "mean_token_accuracy": 0.19967885166406632,
      "num_tokens": 31662282.0,
      "step": 15255
    },
    {
      "entropy": 5.463409852981568,
      "epoch": 1.3842525399129173,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004813426021852547,
      "loss": 5.0028,
      "mean_token_accuracy": 0.19269192218780518,
      "num_tokens": 31672364.0,
      "step": 15260
    },
    {
      "entropy": 5.475115585327148,
      "epoch": 1.3847060957910013,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048132969931069156,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19727637022733688,
      "num_tokens": 31683094.0,
      "step": 15265
    },
    {
      "entropy": 5.46751856803894,
      "epoch": 1.3851596516690856,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004813167921691777,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20178418308496476,
      "num_tokens": 31693087.0,
      "step": 15270
    },
    {
      "entropy": 5.473338556289673,
      "epoch": 1.3856132075471699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048130388076098003,
      "loss": 5.02,
      "mean_token_accuracy": 0.19593427181243897,
      "num_tokens": 31703708.0,
      "step": 15275
    },
    {
      "entropy": 5.428715753555298,
      "epoch": 1.386066763425254,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048129096508636547,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20823626965284348,
      "num_tokens": 31713640.0,
      "step": 15280
    },
    {
      "entropy": 5.467691898345947,
      "epoch": 1.3865203193033382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048127804514560126,
      "loss": 5.0648,
      "mean_token_accuracy": 0.2034163922071457,
      "num_tokens": 31723526.0,
      "step": 15285
    },
    {
      "entropy": 5.414355897903443,
      "epoch": 1.3869738751814222,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048126512093895447,
      "loss": 4.9901,
      "mean_token_accuracy": 0.20380796790122985,
      "num_tokens": 31733265.0,
      "step": 15290
    },
    {
      "entropy": 5.496120643615723,
      "epoch": 1.3874274310595065,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004812521924666925,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2064056247472763,
      "num_tokens": 31743832.0,
      "step": 15295
    },
    {
      "entropy": 5.495451641082764,
      "epoch": 1.3878809869375908,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004812392597290825,
      "loss": 5.0604,
      "mean_token_accuracy": 0.209172622859478,
      "num_tokens": 31754443.0,
      "step": 15300
    },
    {
      "entropy": 5.372473287582397,
      "epoch": 1.3883345428156748,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004812263227263921,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20174793899059296,
      "num_tokens": 31765304.0,
      "step": 15305
    },
    {
      "entropy": 5.432365036010742,
      "epoch": 1.388788098693759,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004812133814588888,
      "loss": 5.0556,
      "mean_token_accuracy": 0.1972840517759323,
      "num_tokens": 31775809.0,
      "step": 15310
    },
    {
      "entropy": 5.494178009033203,
      "epoch": 1.3892416545718431,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004812004359268402,
      "loss": 5.0189,
      "mean_token_accuracy": 0.20662343353033066,
      "num_tokens": 31786271.0,
      "step": 15315
    },
    {
      "entropy": 5.504286527633667,
      "epoch": 1.3896952104499274,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004811874861305141,
      "loss": 5.0737,
      "mean_token_accuracy": 0.20229720920324326,
      "num_tokens": 31797358.0,
      "step": 15320
    },
    {
      "entropy": 5.4540262699127195,
      "epoch": 1.3901487663280117,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004811745320701782,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21028562784194946,
      "num_tokens": 31808303.0,
      "step": 15325
    },
    {
      "entropy": 5.393069648742676,
      "epoch": 1.390602322206096,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004811615737461004,
      "loss": 5.0089,
      "mean_token_accuracy": 0.20239959061145782,
      "num_tokens": 31818837.0,
      "step": 15330
    },
    {
      "entropy": 5.495754814147949,
      "epoch": 1.39105587808418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048114861115854877,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20156844109296798,
      "num_tokens": 31829729.0,
      "step": 15335
    },
    {
      "entropy": 5.44893627166748,
      "epoch": 1.3915094339622642,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048113564430779136,
      "loss": 4.9727,
      "mean_token_accuracy": 0.20025082528591157,
      "num_tokens": 31839996.0,
      "step": 15340
    },
    {
      "entropy": 5.3745111465454105,
      "epoch": 1.3919629898403483,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004811226731940962,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2046112149953842,
      "num_tokens": 31850298.0,
      "step": 15345
    },
    {
      "entropy": 5.4414092063903805,
      "epoch": 1.3924165457184325,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048110969781773173,
      "loss": 5.0648,
      "mean_token_accuracy": 0.20993245840072633,
      "num_tokens": 31862170.0,
      "step": 15350
    },
    {
      "entropy": 5.513597440719605,
      "epoch": 1.3928701015965168,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048109671817896614,
      "loss": 5.1018,
      "mean_token_accuracy": 0.20092002004384996,
      "num_tokens": 31872081.0,
      "step": 15355
    },
    {
      "entropy": 5.440402698516846,
      "epoch": 1.3933236574746009,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048108373427806797,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20320018082857133,
      "num_tokens": 31882715.0,
      "step": 15360
    },
    {
      "entropy": 5.491623115539551,
      "epoch": 1.3937772133526851,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048107074611530555,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20696883350610734,
      "num_tokens": 31891925.0,
      "step": 15365
    },
    {
      "entropy": 5.507652521133423,
      "epoch": 1.3942307692307692,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048105775369094756,
      "loss": 5.027,
      "mean_token_accuracy": 0.20350237488746642,
      "num_tokens": 31902320.0,
      "step": 15370
    },
    {
      "entropy": 5.429424095153808,
      "epoch": 1.3946843251088534,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004810447570052628,
      "loss": 4.9456,
      "mean_token_accuracy": 0.2057958498597145,
      "num_tokens": 31912726.0,
      "step": 15375
    },
    {
      "entropy": 5.456639051437378,
      "epoch": 1.3951378809869377,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048103175605851997,
      "loss": 5.0565,
      "mean_token_accuracy": 0.199539977312088,
      "num_tokens": 31924039.0,
      "step": 15380
    },
    {
      "entropy": 5.400138235092163,
      "epoch": 1.3955914368650217,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004810187508509879,
      "loss": 4.9298,
      "mean_token_accuracy": 0.2129800632596016,
      "num_tokens": 31933280.0,
      "step": 15385
    },
    {
      "entropy": 5.456555652618408,
      "epoch": 1.396044992743106,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004810057413829355,
      "loss": 5.139,
      "mean_token_accuracy": 0.1994667336344719,
      "num_tokens": 31943182.0,
      "step": 15390
    },
    {
      "entropy": 5.525270462036133,
      "epoch": 1.39649854862119,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048099272765463193,
      "loss": 5.0692,
      "mean_token_accuracy": 0.2036070257425308,
      "num_tokens": 31953256.0,
      "step": 15395
    },
    {
      "entropy": 5.555541372299194,
      "epoch": 1.3969521044992743,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004809797096663463,
      "loss": 5.1521,
      "mean_token_accuracy": 0.19529494643211365,
      "num_tokens": 31964231.0,
      "step": 15400
    },
    {
      "entropy": 5.475796127319336,
      "epoch": 1.3974056603773586,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004809666874183477,
      "loss": 5.006,
      "mean_token_accuracy": 0.20255526006221772,
      "num_tokens": 31974322.0,
      "step": 15405
    },
    {
      "entropy": 5.4569597244262695,
      "epoch": 1.3978592162554426,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004809536609109056,
      "loss": 5.0664,
      "mean_token_accuracy": 0.20720462054014205,
      "num_tokens": 31984239.0,
      "step": 15410
    },
    {
      "entropy": 5.445779752731323,
      "epoch": 1.398312772133527,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048094063014428925,
      "loss": 5.0569,
      "mean_token_accuracy": 0.19856468290090562,
      "num_tokens": 31994366.0,
      "step": 15415
    },
    {
      "entropy": 5.515515851974487,
      "epoch": 1.398766328011611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004809275951187682,
      "loss": 5.1372,
      "mean_token_accuracy": 0.198408405482769,
      "num_tokens": 32004921.0,
      "step": 15420
    },
    {
      "entropy": 5.443058919906616,
      "epoch": 1.3992198838896952,
      "grad_norm": 1.0,
      "learning_rate": 0.00048091455583461204,
      "loss": 5.0648,
      "mean_token_accuracy": 0.20184063017368317,
      "num_tokens": 32016385.0,
      "step": 15425
    },
    {
      "entropy": 5.46964020729065,
      "epoch": 1.3996734397677795,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048090151229209036,
      "loss": 5.0434,
      "mean_token_accuracy": 0.19756338894367217,
      "num_tokens": 32027398.0,
      "step": 15430
    },
    {
      "entropy": 5.508984088897705,
      "epoch": 1.4001269956458635,
      "grad_norm": 1.0,
      "learning_rate": 0.0004808884644914729,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20654245913028718,
      "num_tokens": 32038326.0,
      "step": 15435
    },
    {
      "entropy": 5.473737478256226,
      "epoch": 1.4005805515239478,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004808754124330296,
      "loss": 5.1155,
      "mean_token_accuracy": 0.19969897866249084,
      "num_tokens": 32048744.0,
      "step": 15440
    },
    {
      "entropy": 5.478224563598633,
      "epoch": 1.4010341074020318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004808623561170302,
      "loss": 4.9805,
      "mean_token_accuracy": 0.21029338836669922,
      "num_tokens": 32059133.0,
      "step": 15445
    },
    {
      "entropy": 5.460933017730713,
      "epoch": 1.401487663280116,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004808492955437449,
      "loss": 5.0243,
      "mean_token_accuracy": 0.2009618729352951,
      "num_tokens": 32069277.0,
      "step": 15450
    },
    {
      "entropy": 5.456944704055786,
      "epoch": 1.4019412191582004,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048083623071344374,
      "loss": 4.9734,
      "mean_token_accuracy": 0.2077607974410057,
      "num_tokens": 32079802.0,
      "step": 15455
    },
    {
      "entropy": 5.473855209350586,
      "epoch": 1.4023947750362844,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048082316162639685,
      "loss": 5.0919,
      "mean_token_accuracy": 0.19972393661737442,
      "num_tokens": 32090962.0,
      "step": 15460
    },
    {
      "entropy": 5.462399005889893,
      "epoch": 1.4028483309143687,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048081008828287447,
      "loss": 5.0661,
      "mean_token_accuracy": 0.19965442270040512,
      "num_tokens": 32101058.0,
      "step": 15465
    },
    {
      "entropy": 5.5409893035888675,
      "epoch": 1.4033018867924527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048079701068314706,
      "loss": 5.0458,
      "mean_token_accuracy": 0.20538726300001145,
      "num_tokens": 32111763.0,
      "step": 15470
    },
    {
      "entropy": 5.487426519393921,
      "epoch": 1.403755442670537,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048078392882748507,
      "loss": 5.1161,
      "mean_token_accuracy": 0.20131370276212693,
      "num_tokens": 32123556.0,
      "step": 15475
    },
    {
      "entropy": 5.452586889266968,
      "epoch": 1.4042089985486212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048077084271615895,
      "loss": 4.997,
      "mean_token_accuracy": 0.20365253686904908,
      "num_tokens": 32134012.0,
      "step": 15480
    },
    {
      "entropy": 5.4020476818084715,
      "epoch": 1.4046625544267053,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004807577523494394,
      "loss": 4.9798,
      "mean_token_accuracy": 0.21200866550207137,
      "num_tokens": 32144540.0,
      "step": 15485
    },
    {
      "entropy": 5.4838134765625,
      "epoch": 1.4051161103047896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048074465772759707,
      "loss": 5.0833,
      "mean_token_accuracy": 0.20107725113630295,
      "num_tokens": 32154655.0,
      "step": 15490
    },
    {
      "entropy": 5.434706163406372,
      "epoch": 1.4055696661828736,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048073155885090276,
      "loss": 4.9749,
      "mean_token_accuracy": 0.205027037858963,
      "num_tokens": 32165191.0,
      "step": 15495
    },
    {
      "entropy": 5.492273426055908,
      "epoch": 1.4060232220609579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048071845571962733,
      "loss": 5.0889,
      "mean_token_accuracy": 0.1927748829126358,
      "num_tokens": 32175961.0,
      "step": 15500
    },
    {
      "entropy": 5.532474040985107,
      "epoch": 1.4064767779390421,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004807053483340419,
      "loss": 5.0478,
      "mean_token_accuracy": 0.19928376972675324,
      "num_tokens": 32187073.0,
      "step": 15505
    },
    {
      "entropy": 5.467940092086792,
      "epoch": 1.4069303338171264,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004806922366944175,
      "loss": 5.0375,
      "mean_token_accuracy": 0.20543879121541977,
      "num_tokens": 32197426.0,
      "step": 15510
    },
    {
      "entropy": 5.532773065567016,
      "epoch": 1.4073838896952104,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048067912080102513,
      "loss": 5.0102,
      "mean_token_accuracy": 0.20775387585163116,
      "num_tokens": 32206847.0,
      "step": 15515
    },
    {
      "entropy": 5.436613607406616,
      "epoch": 1.4078374455732947,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004806660006541362,
      "loss": 5.0496,
      "mean_token_accuracy": 0.19538942873477935,
      "num_tokens": 32217647.0,
      "step": 15520
    },
    {
      "entropy": 5.389906501770019,
      "epoch": 1.4082910014513788,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004806528762540219,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21112460196018218,
      "num_tokens": 32227133.0,
      "step": 15525
    },
    {
      "entropy": 5.488668537139892,
      "epoch": 1.408744557329463,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048063974760095374,
      "loss": 5.0988,
      "mean_token_accuracy": 0.1972224622964859,
      "num_tokens": 32237378.0,
      "step": 15530
    },
    {
      "entropy": 5.419837760925293,
      "epoch": 1.4091981132075473,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048062661469520327,
      "loss": 4.987,
      "mean_token_accuracy": 0.20672549307346344,
      "num_tokens": 32247213.0,
      "step": 15535
    },
    {
      "entropy": 5.521755170822144,
      "epoch": 1.4096516690856313,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048061347753704193,
      "loss": 5.065,
      "mean_token_accuracy": 0.20156003534793854,
      "num_tokens": 32257100.0,
      "step": 15540
    },
    {
      "entropy": 5.488410186767578,
      "epoch": 1.4101052249637156,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004806003361267415,
      "loss": 5.0251,
      "mean_token_accuracy": 0.1978863701224327,
      "num_tokens": 32267456.0,
      "step": 15545
    },
    {
      "entropy": 5.429026746749878,
      "epoch": 1.4105587808417996,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004805871904645737,
      "loss": 4.9634,
      "mean_token_accuracy": 0.20904991030693054,
      "num_tokens": 32278220.0,
      "step": 15550
    },
    {
      "entropy": 5.416950416564942,
      "epoch": 1.411012336719884,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004805740405508103,
      "loss": 5.0387,
      "mean_token_accuracy": 0.2059457629919052,
      "num_tokens": 32289036.0,
      "step": 15555
    },
    {
      "entropy": 5.425758647918701,
      "epoch": 1.4114658925979682,
      "grad_norm": 1.125,
      "learning_rate": 0.0004805608863857236,
      "loss": 4.9721,
      "mean_token_accuracy": 0.20169477611780168,
      "num_tokens": 32298909.0,
      "step": 15560
    },
    {
      "entropy": 5.475577926635742,
      "epoch": 1.4119194484760522,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048054772796958517,
      "loss": 4.9998,
      "mean_token_accuracy": 0.2035790354013443,
      "num_tokens": 32309851.0,
      "step": 15565
    },
    {
      "entropy": 5.4505894660949705,
      "epoch": 1.4123730043541365,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004805345653026675,
      "loss": 4.9809,
      "mean_token_accuracy": 0.2048426628112793,
      "num_tokens": 32319336.0,
      "step": 15570
    },
    {
      "entropy": 5.487655544281006,
      "epoch": 1.4128265602322205,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048052139838524263,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20487714558839798,
      "num_tokens": 32329898.0,
      "step": 15575
    },
    {
      "entropy": 5.4495281219482425,
      "epoch": 1.4132801161103048,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048050822721758274,
      "loss": 4.9262,
      "mean_token_accuracy": 0.2070409268140793,
      "num_tokens": 32340100.0,
      "step": 15580
    },
    {
      "entropy": 5.434144020080566,
      "epoch": 1.413733671988389,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004804950517999604,
      "loss": 5.0292,
      "mean_token_accuracy": 0.2006583109498024,
      "num_tokens": 32349663.0,
      "step": 15585
    },
    {
      "entropy": 5.507922792434693,
      "epoch": 1.414187227866473,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004804818721326481,
      "loss": 5.0723,
      "mean_token_accuracy": 0.20222700536251068,
      "num_tokens": 32360461.0,
      "step": 15590
    },
    {
      "entropy": 5.479337215423584,
      "epoch": 1.4146407837445574,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004804686882159183,
      "loss": 5.077,
      "mean_token_accuracy": 0.2038666307926178,
      "num_tokens": 32370910.0,
      "step": 15595
    },
    {
      "entropy": 5.5347740173339846,
      "epoch": 1.4150943396226414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048045550005004355,
      "loss": 5.0633,
      "mean_token_accuracy": 0.20874321311712266,
      "num_tokens": 32381669.0,
      "step": 15600
    },
    {
      "entropy": 5.436280012130737,
      "epoch": 1.4155478955007257,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004804423076352968,
      "loss": 4.9954,
      "mean_token_accuracy": 0.20037585347890854,
      "num_tokens": 32392024.0,
      "step": 15605
    },
    {
      "entropy": 5.368772029876709,
      "epoch": 1.41600145137881,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048042911097195075,
      "loss": 5.0122,
      "mean_token_accuracy": 0.20821220874786378,
      "num_tokens": 32401872.0,
      "step": 15610
    },
    {
      "entropy": 5.402716445922851,
      "epoch": 1.416455007256894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004804159100602784,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2087838977575302,
      "num_tokens": 32411749.0,
      "step": 15615
    },
    {
      "entropy": 5.559806823730469,
      "epoch": 1.4169085631349783,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048040270490055264,
      "loss": 5.1569,
      "mean_token_accuracy": 0.19089394211769103,
      "num_tokens": 32421358.0,
      "step": 15620
    },
    {
      "entropy": 5.496081209182739,
      "epoch": 1.4173621190130623,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004803894954930465,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20114330798387528,
      "num_tokens": 32432322.0,
      "step": 15625
    },
    {
      "entropy": 5.460557270050049,
      "epoch": 1.4178156748911466,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004803762818380334,
      "loss": 5.0513,
      "mean_token_accuracy": 0.19781472980976106,
      "num_tokens": 32440930.0,
      "step": 15630
    },
    {
      "entropy": 5.4865857601165775,
      "epoch": 1.4182692307692308,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048036306393578644,
      "loss": 5.0588,
      "mean_token_accuracy": 0.19708771705627443,
      "num_tokens": 32451124.0,
      "step": 15635
    },
    {
      "entropy": 5.523217868804932,
      "epoch": 1.4187227866473149,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004803498417865789,
      "loss": 5.0303,
      "mean_token_accuracy": 0.20069751739501954,
      "num_tokens": 32460636.0,
      "step": 15640
    },
    {
      "entropy": 5.494079637527466,
      "epoch": 1.4191763425253991,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004803366153906844,
      "loss": 5.078,
      "mean_token_accuracy": 0.20009166449308396,
      "num_tokens": 32470501.0,
      "step": 15645
    },
    {
      "entropy": 5.4987688064575195,
      "epoch": 1.4196298984034832,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004803233847483763,
      "loss": 5.1395,
      "mean_token_accuracy": 0.19809164553880693,
      "num_tokens": 32481559.0,
      "step": 15650
    },
    {
      "entropy": 5.607738590240478,
      "epoch": 1.4200834542815675,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048031014985992825,
      "loss": 5.2143,
      "mean_token_accuracy": 0.1941840723156929,
      "num_tokens": 32493000.0,
      "step": 15655
    },
    {
      "entropy": 5.491443681716919,
      "epoch": 1.4205370101596517,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000480296910725614,
      "loss": 4.9954,
      "mean_token_accuracy": 0.21452815383672713,
      "num_tokens": 32502773.0,
      "step": 15660
    },
    {
      "entropy": 5.449616861343384,
      "epoch": 1.4209905660377358,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004802836673457074,
      "loss": 5.1041,
      "mean_token_accuracy": 0.19487740099430084,
      "num_tokens": 32513430.0,
      "step": 15665
    },
    {
      "entropy": 5.401146984100341,
      "epoch": 1.42144412191582,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048027041972048225,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20312780737876893,
      "num_tokens": 32524409.0,
      "step": 15670
    },
    {
      "entropy": 5.512161064147949,
      "epoch": 1.421897677793904,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004802571678502124,
      "loss": 5.0597,
      "mean_token_accuracy": 0.1949110746383667,
      "num_tokens": 32534696.0,
      "step": 15675
    },
    {
      "entropy": 5.559273672103882,
      "epoch": 1.4223512336719883,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000480243911735172,
      "loss": 5.0457,
      "mean_token_accuracy": 0.20404788851737976,
      "num_tokens": 32544782.0,
      "step": 15680
    },
    {
      "entropy": 5.447524356842041,
      "epoch": 1.4228047895500726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048023065137563534,
      "loss": 5.0465,
      "mean_token_accuracy": 0.19569372087717057,
      "num_tokens": 32554818.0,
      "step": 15685
    },
    {
      "entropy": 5.436730623245239,
      "epoch": 1.4232583454281569,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004802173867718764,
      "loss": 5.0747,
      "mean_token_accuracy": 0.20380208194255828,
      "num_tokens": 32565007.0,
      "step": 15690
    },
    {
      "entropy": 5.4994275093078615,
      "epoch": 1.423711901306241,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004802041179241697,
      "loss": 5.0216,
      "mean_token_accuracy": 0.20204194337129594,
      "num_tokens": 32575657.0,
      "step": 15695
    },
    {
      "entropy": 5.461858129501342,
      "epoch": 1.4241654571843252,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004801908448327895,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19369821548461913,
      "num_tokens": 32586676.0,
      "step": 15700
    },
    {
      "entropy": 5.466211318969727,
      "epoch": 1.4246190130624092,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004801775674980103,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20252088010311126,
      "num_tokens": 32596191.0,
      "step": 15705
    },
    {
      "entropy": 5.533839464187622,
      "epoch": 1.4250725689404935,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048016428592010676,
      "loss": 5.0802,
      "mean_token_accuracy": 0.19708123356103896,
      "num_tokens": 32607178.0,
      "step": 15710
    },
    {
      "entropy": 5.522767639160156,
      "epoch": 1.4255261248185778,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048015100009935347,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20005394518375397,
      "num_tokens": 32617301.0,
      "step": 15715
    },
    {
      "entropy": 5.391142988204956,
      "epoch": 1.4259796806966618,
      "grad_norm": 1.0,
      "learning_rate": 0.0004801377100360253,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2088123455643654,
      "num_tokens": 32627395.0,
      "step": 15720
    },
    {
      "entropy": 5.446273422241211,
      "epoch": 1.426433236574746,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000480124415730397,
      "loss": 5.016,
      "mean_token_accuracy": 0.20472394526004792,
      "num_tokens": 32638460.0,
      "step": 15725
    },
    {
      "entropy": 5.551666641235352,
      "epoch": 1.4268867924528301,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004801111171827434,
      "loss": 5.0809,
      "mean_token_accuracy": 0.19856278747320175,
      "num_tokens": 32649319.0,
      "step": 15730
    },
    {
      "entropy": 5.487170839309693,
      "epoch": 1.4273403483309144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004800978143933398,
      "loss": 5.0229,
      "mean_token_accuracy": 0.2052985429763794,
      "num_tokens": 32659976.0,
      "step": 15735
    },
    {
      "entropy": 5.418977212905884,
      "epoch": 1.4277939042089987,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048008450736246094,
      "loss": 4.9536,
      "mean_token_accuracy": 0.2020311713218689,
      "num_tokens": 32670772.0,
      "step": 15740
    },
    {
      "entropy": 5.513687801361084,
      "epoch": 1.4282474600870827,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048007119609038237,
      "loss": 5.0967,
      "mean_token_accuracy": 0.19420246332883834,
      "num_tokens": 32682368.0,
      "step": 15745
    },
    {
      "entropy": 5.397270441055298,
      "epoch": 1.428701015965167,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048005788057737907,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2074650853872299,
      "num_tokens": 32692649.0,
      "step": 15750
    },
    {
      "entropy": 5.482013940811157,
      "epoch": 1.429154571843251,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004800445608237266,
      "loss": 5.1401,
      "mean_token_accuracy": 0.19943402111530303,
      "num_tokens": 32704075.0,
      "step": 15755
    },
    {
      "entropy": 5.5202741622924805,
      "epoch": 1.4296081277213353,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048003123682970045,
      "loss": 5.1601,
      "mean_token_accuracy": 0.19761826246976852,
      "num_tokens": 32714670.0,
      "step": 15760
    },
    {
      "entropy": 5.463421058654785,
      "epoch": 1.4300616835994195,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048001790859557596,
      "loss": 4.9785,
      "mean_token_accuracy": 0.20732996463775635,
      "num_tokens": 32724601.0,
      "step": 15765
    },
    {
      "entropy": 5.4143373489379885,
      "epoch": 1.4305152394775036,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004800045761216289,
      "loss": 4.9436,
      "mean_token_accuracy": 0.20634355694055556,
      "num_tokens": 32735015.0,
      "step": 15770
    },
    {
      "entropy": 5.4462672710418705,
      "epoch": 1.4309687953555879,
      "grad_norm": 1.03125,
      "learning_rate": 0.000479991239408135,
      "loss": 5.0358,
      "mean_token_accuracy": 0.2087372660636902,
      "num_tokens": 32745899.0,
      "step": 15775
    },
    {
      "entropy": 5.463490533828735,
      "epoch": 1.431422351233672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047997789845537003,
      "loss": 5.0001,
      "mean_token_accuracy": 0.2086031198501587,
      "num_tokens": 32756052.0,
      "step": 15780
    },
    {
      "entropy": 5.416792964935302,
      "epoch": 1.4318759071117562,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004799645532636098,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20779118537902833,
      "num_tokens": 32767118.0,
      "step": 15785
    },
    {
      "entropy": 5.390763139724731,
      "epoch": 1.4323294629898404,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047995120383313044,
      "loss": 4.8893,
      "mean_token_accuracy": 0.214095738530159,
      "num_tokens": 32777069.0,
      "step": 15790
    },
    {
      "entropy": 5.460057067871094,
      "epoch": 1.4327830188679245,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047993785016420796,
      "loss": 5.1078,
      "mean_token_accuracy": 0.19337689578533174,
      "num_tokens": 32786967.0,
      "step": 15795
    },
    {
      "entropy": 5.468468284606933,
      "epoch": 1.4332365747460087,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047992449225711854,
      "loss": 4.9833,
      "mean_token_accuracy": 0.20572699904441832,
      "num_tokens": 32797563.0,
      "step": 15800
    },
    {
      "entropy": 5.476965618133545,
      "epoch": 1.4336901306240928,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047991113011213837,
      "loss": 4.9556,
      "mean_token_accuracy": 0.20637069642543793,
      "num_tokens": 32808730.0,
      "step": 15805
    },
    {
      "entropy": 5.433117055892945,
      "epoch": 1.434143686502177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004798977637295438,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20631701201200486,
      "num_tokens": 32818050.0,
      "step": 15810
    },
    {
      "entropy": 5.427420234680175,
      "epoch": 1.4345972423802613,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798843931096113,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20257741659879686,
      "num_tokens": 32828993.0,
      "step": 15815
    },
    {
      "entropy": 5.428350114822388,
      "epoch": 1.4350507982583454,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004798710182526173,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2052305147051811,
      "num_tokens": 32839331.0,
      "step": 15820
    },
    {
      "entropy": 5.46976866722107,
      "epoch": 1.4355043541364296,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047985763915883847,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20204367786645888,
      "num_tokens": 32849946.0,
      "step": 15825
    },
    {
      "entropy": 5.393970966339111,
      "epoch": 1.4359579100145137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004798442558285514,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20497775375843047,
      "num_tokens": 32859960.0,
      "step": 15830
    },
    {
      "entropy": 5.424088191986084,
      "epoch": 1.436411465892598,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004798308682620329,
      "loss": 4.9742,
      "mean_token_accuracy": 0.19887613654136657,
      "num_tokens": 32870317.0,
      "step": 15835
    },
    {
      "entropy": 5.53906774520874,
      "epoch": 1.4368650217706822,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047981747645955996,
      "loss": 5.151,
      "mean_token_accuracy": 0.19574028700590135,
      "num_tokens": 32880906.0,
      "step": 15840
    },
    {
      "entropy": 5.56572732925415,
      "epoch": 1.4373185776487662,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004798040804214094,
      "loss": 5.1879,
      "mean_token_accuracy": 0.1897825300693512,
      "num_tokens": 32891038.0,
      "step": 15845
    },
    {
      "entropy": 5.569432735443115,
      "epoch": 1.4377721335268505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004797906801478582,
      "loss": 5.1249,
      "mean_token_accuracy": 0.2030026987195015,
      "num_tokens": 32901013.0,
      "step": 15850
    },
    {
      "entropy": 5.516676425933838,
      "epoch": 1.4382256894049346,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004797772756391836,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20463033020496368,
      "num_tokens": 32911592.0,
      "step": 15855
    },
    {
      "entropy": 5.423782253265381,
      "epoch": 1.4386792452830188,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797638668956627,
      "loss": 4.9559,
      "mean_token_accuracy": 0.20585375279188156,
      "num_tokens": 32922087.0,
      "step": 15860
    },
    {
      "entropy": 5.45404200553894,
      "epoch": 1.439132801161103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004797504539175729,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20368013083934783,
      "num_tokens": 32932039.0,
      "step": 15865
    },
    {
      "entropy": 5.473904275894165,
      "epoch": 1.4395863570391874,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004797370367051915,
      "loss": 4.9911,
      "mean_token_accuracy": 0.20422232896089554,
      "num_tokens": 32941896.0,
      "step": 15870
    },
    {
      "entropy": 5.46373610496521,
      "epoch": 1.4400399129172714,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000479723615258796,
      "loss": 5.0187,
      "mean_token_accuracy": 0.1923507422208786,
      "num_tokens": 32951675.0,
      "step": 15875
    },
    {
      "entropy": 5.477837467193604,
      "epoch": 1.4404934687953557,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047971018957866396,
      "loss": 5.0921,
      "mean_token_accuracy": 0.20661236494779586,
      "num_tokens": 32961378.0,
      "step": 15880
    },
    {
      "entropy": 5.4550474166870115,
      "epoch": 1.4409470246734397,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047969675966507314,
      "loss": 4.9904,
      "mean_token_accuracy": 0.20333871394395828,
      "num_tokens": 32971770.0,
      "step": 15885
    },
    {
      "entropy": 5.438835620880127,
      "epoch": 1.441400580551524,
      "grad_norm": 1.0,
      "learning_rate": 0.000479683325518301,
      "loss": 5.002,
      "mean_token_accuracy": 0.2005586564540863,
      "num_tokens": 32982934.0,
      "step": 15890
    },
    {
      "entropy": 5.427374649047851,
      "epoch": 1.4418541364296082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004796698871386257,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20118418335914612,
      "num_tokens": 32993252.0,
      "step": 15895
    },
    {
      "entropy": 5.490049886703491,
      "epoch": 1.4423076923076923,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004796564445263249,
      "loss": 5.0508,
      "mean_token_accuracy": 0.20014355480670928,
      "num_tokens": 33003610.0,
      "step": 15900
    },
    {
      "entropy": 5.470461463928222,
      "epoch": 1.4427612481857766,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047964299768167665,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20554691404104233,
      "num_tokens": 33014037.0,
      "step": 15905
    },
    {
      "entropy": 5.460197305679321,
      "epoch": 1.4432148040638606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004796295466049591,
      "loss": 5.0176,
      "mean_token_accuracy": 0.19628096073865892,
      "num_tokens": 33023299.0,
      "step": 15910
    },
    {
      "entropy": 5.513853693008423,
      "epoch": 1.4436683599419449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047961609129645036,
      "loss": 5.0764,
      "mean_token_accuracy": 0.19446345269680024,
      "num_tokens": 33034194.0,
      "step": 15915
    },
    {
      "entropy": 5.45906252861023,
      "epoch": 1.4441219158200291,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047960263175642867,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2084976240992546,
      "num_tokens": 33044715.0,
      "step": 15920
    },
    {
      "entropy": 5.386568164825439,
      "epoch": 1.4445754716981132,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004795891679851724,
      "loss": 4.9661,
      "mean_token_accuracy": 0.2160312220454216,
      "num_tokens": 33055214.0,
      "step": 15925
    },
    {
      "entropy": 5.455825710296631,
      "epoch": 1.4450290275761974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004795756999829601,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20294891446828842,
      "num_tokens": 33065026.0,
      "step": 15930
    },
    {
      "entropy": 5.560147714614868,
      "epoch": 1.4454825834542815,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047956222775007016,
      "loss": 5.0568,
      "mean_token_accuracy": 0.1987060323357582,
      "num_tokens": 33075138.0,
      "step": 15935
    },
    {
      "entropy": 5.486181306838989,
      "epoch": 1.4459361393323658,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004795487512867812,
      "loss": 5.0379,
      "mean_token_accuracy": 0.20502393692731857,
      "num_tokens": 33086625.0,
      "step": 15940
    },
    {
      "entropy": 5.515254735946655,
      "epoch": 1.44638969521045,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047953527059337194,
      "loss": 5.0776,
      "mean_token_accuracy": 0.19539907425642014,
      "num_tokens": 33097076.0,
      "step": 15945
    },
    {
      "entropy": 5.470223426818848,
      "epoch": 1.446843251088534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004795217856701212,
      "loss": 5.0144,
      "mean_token_accuracy": 0.20341595709323884,
      "num_tokens": 33107168.0,
      "step": 15950
    },
    {
      "entropy": 5.506333351135254,
      "epoch": 1.4472968069666183,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047950829651730774,
      "loss": 5.0519,
      "mean_token_accuracy": 0.20027395188808442,
      "num_tokens": 33117862.0,
      "step": 15955
    },
    {
      "entropy": 5.489769983291626,
      "epoch": 1.4477503628447024,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004794948031352107,
      "loss": 4.9735,
      "mean_token_accuracy": 0.20904716104269028,
      "num_tokens": 33127144.0,
      "step": 15960
    },
    {
      "entropy": 5.535410833358765,
      "epoch": 1.4482039187227866,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047948130552410897,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19545748680830002,
      "num_tokens": 33139180.0,
      "step": 15965
    },
    {
      "entropy": 5.5159039974212645,
      "epoch": 1.448657474600871,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047946780368428175,
      "loss": 5.0065,
      "mean_token_accuracy": 0.19653345495462418,
      "num_tokens": 33149467.0,
      "step": 15970
    },
    {
      "entropy": 5.4328114032745365,
      "epoch": 1.449111030478955,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004794542976160081,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21548738479614257,
      "num_tokens": 33158752.0,
      "step": 15975
    },
    {
      "entropy": 5.412974882125854,
      "epoch": 1.4495645863570392,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047944078731956765,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20715339183807374,
      "num_tokens": 33168242.0,
      "step": 15980
    },
    {
      "entropy": 5.409467077255249,
      "epoch": 1.4500181422351233,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047942727279523957,
      "loss": 4.9854,
      "mean_token_accuracy": 0.210733562707901,
      "num_tokens": 33179134.0,
      "step": 15985
    },
    {
      "entropy": 5.485414934158325,
      "epoch": 1.4504716981132075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004794137540433034,
      "loss": 5.1127,
      "mean_token_accuracy": 0.1984705865383148,
      "num_tokens": 33188388.0,
      "step": 15990
    },
    {
      "entropy": 5.519913148880005,
      "epoch": 1.4509252539912918,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004794002310640386,
      "loss": 5.0083,
      "mean_token_accuracy": 0.2034293457865715,
      "num_tokens": 33198750.0,
      "step": 15995
    },
    {
      "entropy": 5.571501922607422,
      "epoch": 1.4513788098693758,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047938670385772503,
      "loss": 5.1255,
      "mean_token_accuracy": 0.1909438416361809,
      "num_tokens": 33209532.0,
      "step": 16000
    },
    {
      "entropy": 5.500415992736817,
      "epoch": 1.45183236574746,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047937317242464244,
      "loss": 4.977,
      "mean_token_accuracy": 0.2058201715350151,
      "num_tokens": 33219675.0,
      "step": 16005
    },
    {
      "entropy": 5.509267568588257,
      "epoch": 1.4522859216255442,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004793596367650704,
      "loss": 5.0847,
      "mean_token_accuracy": 0.19889651238918304,
      "num_tokens": 33230919.0,
      "step": 16010
    },
    {
      "entropy": 5.489873170852661,
      "epoch": 1.4527394775036284,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004793460968792891,
      "loss": 4.9844,
      "mean_token_accuracy": 0.20786553621292114,
      "num_tokens": 33241146.0,
      "step": 16015
    },
    {
      "entropy": 5.497219038009644,
      "epoch": 1.4531930333817127,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047933255276757836,
      "loss": 5.0404,
      "mean_token_accuracy": 0.19959067404270173,
      "num_tokens": 33251434.0,
      "step": 16020
    },
    {
      "entropy": 5.525360250473023,
      "epoch": 1.4536465892597967,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004793190044302184,
      "loss": 5.1201,
      "mean_token_accuracy": 0.1891581580042839,
      "num_tokens": 33261868.0,
      "step": 16025
    },
    {
      "entropy": 5.469266128540039,
      "epoch": 1.454100145137881,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047930545186748936,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20658151358366011,
      "num_tokens": 33271798.0,
      "step": 16030
    },
    {
      "entropy": 5.464654397964478,
      "epoch": 1.454553701015965,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047929189507967153,
      "loss": 5.0659,
      "mean_token_accuracy": 0.19962890148162843,
      "num_tokens": 33282614.0,
      "step": 16035
    },
    {
      "entropy": 5.420350074768066,
      "epoch": 1.4550072568940493,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047927833406704516,
      "loss": 4.975,
      "mean_token_accuracy": 0.20225861072540283,
      "num_tokens": 33292906.0,
      "step": 16040
    },
    {
      "entropy": 5.457213354110718,
      "epoch": 1.4554608127721336,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004792647688298909,
      "loss": 4.9827,
      "mean_token_accuracy": 0.20239474773406982,
      "num_tokens": 33302720.0,
      "step": 16045
    },
    {
      "entropy": 5.417613506317139,
      "epoch": 1.4559143686502178,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004792511993684891,
      "loss": 4.9798,
      "mean_token_accuracy": 0.20726395696401595,
      "num_tokens": 33312959.0,
      "step": 16050
    },
    {
      "entropy": 5.456071567535401,
      "epoch": 1.4563679245283019,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047923762568312053,
      "loss": 4.923,
      "mean_token_accuracy": 0.20960447639226915,
      "num_tokens": 33322737.0,
      "step": 16055
    },
    {
      "entropy": 5.432339668273926,
      "epoch": 1.4568214804063861,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004792240477740657,
      "loss": 4.9482,
      "mean_token_accuracy": 0.2113322213292122,
      "num_tokens": 33333185.0,
      "step": 16060
    },
    {
      "entropy": 5.486125040054321,
      "epoch": 1.4572750362844702,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004792104656416056,
      "loss": 5.0801,
      "mean_token_accuracy": 0.19482650607824326,
      "num_tokens": 33343414.0,
      "step": 16065
    },
    {
      "entropy": 5.483291673660278,
      "epoch": 1.4577285921625545,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047919687928602104,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20450005680322647,
      "num_tokens": 33352527.0,
      "step": 16070
    },
    {
      "entropy": 5.509945487976074,
      "epoch": 1.4581821480406387,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047918328870759293,
      "loss": 5.0535,
      "mean_token_accuracy": 0.19920112788677216,
      "num_tokens": 33363526.0,
      "step": 16075
    },
    {
      "entropy": 5.483455085754395,
      "epoch": 1.4586357039187228,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004791696939066024,
      "loss": 5.0827,
      "mean_token_accuracy": 0.19908895790576936,
      "num_tokens": 33374302.0,
      "step": 16080
    },
    {
      "entropy": 5.4868992328643795,
      "epoch": 1.459089259796807,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004791560948833306,
      "loss": 5.0481,
      "mean_token_accuracy": 0.20238014906644822,
      "num_tokens": 33384786.0,
      "step": 16085
    },
    {
      "entropy": 5.4957318782806395,
      "epoch": 1.459542815674891,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004791424916380586,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20975643545389175,
      "num_tokens": 33394464.0,
      "step": 16090
    },
    {
      "entropy": 5.490728235244751,
      "epoch": 1.4599963715529753,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479128884171068,
      "loss": 5.0508,
      "mean_token_accuracy": 0.20181618332862855,
      "num_tokens": 33405476.0,
      "step": 16095
    },
    {
      "entropy": 5.509813213348389,
      "epoch": 1.4604499274310596,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047911527248264003,
      "loss": 5.0846,
      "mean_token_accuracy": 0.2008212089538574,
      "num_tokens": 33415758.0,
      "step": 16100
    },
    {
      "entropy": 5.475230264663696,
      "epoch": 1.4609034833091437,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047910165657305613,
      "loss": 5.0385,
      "mean_token_accuracy": 0.2024817645549774,
      "num_tokens": 33424845.0,
      "step": 16105
    },
    {
      "entropy": 5.4918581485748295,
      "epoch": 1.461357039187228,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047908803644259804,
      "loss": 5.0799,
      "mean_token_accuracy": 0.20061122477054597,
      "num_tokens": 33435928.0,
      "step": 16110
    },
    {
      "entropy": 5.49398775100708,
      "epoch": 1.461810595065312,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047907441209154726,
      "loss": 4.9852,
      "mean_token_accuracy": 0.20125750750303267,
      "num_tokens": 33447307.0,
      "step": 16115
    },
    {
      "entropy": 5.486487150192261,
      "epoch": 1.4622641509433962,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00047906078352018575,
      "loss": 5.0436,
      "mean_token_accuracy": 0.1945362851023674,
      "num_tokens": 33458116.0,
      "step": 16120
    },
    {
      "entropy": 5.35523943901062,
      "epoch": 1.4627177068214805,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004790471507287952,
      "loss": 4.8962,
      "mean_token_accuracy": 0.20788645893335342,
      "num_tokens": 33469065.0,
      "step": 16125
    },
    {
      "entropy": 5.47483263015747,
      "epoch": 1.4631712626995645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047903351371765757,
      "loss": 5.0239,
      "mean_token_accuracy": 0.2058606430888176,
      "num_tokens": 33479118.0,
      "step": 16130
    },
    {
      "entropy": 5.518635892868042,
      "epoch": 1.4636248185776488,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047901987248705484,
      "loss": 5.0466,
      "mean_token_accuracy": 0.1991819068789482,
      "num_tokens": 33489642.0,
      "step": 16135
    },
    {
      "entropy": 5.476372289657593,
      "epoch": 1.4640783744557329,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004790062270372693,
      "loss": 5.0796,
      "mean_token_accuracy": 0.19231854677200316,
      "num_tokens": 33499771.0,
      "step": 16140
    },
    {
      "entropy": 5.460461521148682,
      "epoch": 1.4645319303338171,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004789925773685828,
      "loss": 5.0241,
      "mean_token_accuracy": 0.20152659863233566,
      "num_tokens": 33509975.0,
      "step": 16145
    },
    {
      "entropy": 5.505796718597412,
      "epoch": 1.4649854862119014,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004789789234812779,
      "loss": 5.063,
      "mean_token_accuracy": 0.19863290190696717,
      "num_tokens": 33520017.0,
      "step": 16150
    },
    {
      "entropy": 5.512185573577881,
      "epoch": 1.4654390420899854,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047896526537563686,
      "loss": 5.1206,
      "mean_token_accuracy": 0.19162429571151735,
      "num_tokens": 33531977.0,
      "step": 16155
    },
    {
      "entropy": 5.497498464584351,
      "epoch": 1.4658925979680697,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047895160305194223,
      "loss": 5.0822,
      "mean_token_accuracy": 0.1961600065231323,
      "num_tokens": 33543660.0,
      "step": 16160
    },
    {
      "entropy": 5.485444498062134,
      "epoch": 1.4663461538461537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789379365104764,
      "loss": 5.0205,
      "mean_token_accuracy": 0.2038754180073738,
      "num_tokens": 33553814.0,
      "step": 16165
    },
    {
      "entropy": 5.4929135799407955,
      "epoch": 1.466799709724238,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004789242657515221,
      "loss": 5.0254,
      "mean_token_accuracy": 0.20848118215799333,
      "num_tokens": 33564250.0,
      "step": 16170
    },
    {
      "entropy": 5.481130838394165,
      "epoch": 1.4672532656023223,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004789105907753621,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2035487875342369,
      "num_tokens": 33574755.0,
      "step": 16175
    },
    {
      "entropy": 5.451443719863891,
      "epoch": 1.4677068214804063,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000478896911582279,
      "loss": 4.9457,
      "mean_token_accuracy": 0.20966747105121614,
      "num_tokens": 33585268.0,
      "step": 16180
    },
    {
      "entropy": 5.468293523788452,
      "epoch": 1.4681603773584906,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004788832281725558,
      "loss": 5.0795,
      "mean_token_accuracy": 0.20301875174045564,
      "num_tokens": 33595753.0,
      "step": 16185
    },
    {
      "entropy": 5.443698072433472,
      "epoch": 1.4686139332365746,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004788695405464755,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20774186849594117,
      "num_tokens": 33605270.0,
      "step": 16190
    },
    {
      "entropy": 5.50835337638855,
      "epoch": 1.469067489114659,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047885584870432127,
      "loss": 5.056,
      "mean_token_accuracy": 0.20199033319950105,
      "num_tokens": 33615096.0,
      "step": 16195
    },
    {
      "entropy": 5.489577198028565,
      "epoch": 1.4695210449927432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047884215264637606,
      "loss": 4.956,
      "mean_token_accuracy": 0.20772962123155594,
      "num_tokens": 33624770.0,
      "step": 16200
    },
    {
      "entropy": 5.520768451690674,
      "epoch": 1.4699746008708272,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004788284523729232,
      "loss": 5.1114,
      "mean_token_accuracy": 0.1957830160856247,
      "num_tokens": 33634545.0,
      "step": 16205
    },
    {
      "entropy": 5.526898336410523,
      "epoch": 1.4704281567489115,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047881474788424595,
      "loss": 5.0473,
      "mean_token_accuracy": 0.1975664019584656,
      "num_tokens": 33644950.0,
      "step": 16210
    },
    {
      "entropy": 5.510557317733765,
      "epoch": 1.4708817126269955,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047880103918062776,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20191543102264403,
      "num_tokens": 33654369.0,
      "step": 16215
    },
    {
      "entropy": 5.36762638092041,
      "epoch": 1.4713352685050798,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047878732626235225,
      "loss": 5.0402,
      "mean_token_accuracy": 0.1993912026286125,
      "num_tokens": 33664687.0,
      "step": 16220
    },
    {
      "entropy": 5.4388457298278805,
      "epoch": 1.471788824383164,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004787736091297028,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20965267717838287,
      "num_tokens": 33675832.0,
      "step": 16225
    },
    {
      "entropy": 5.541381978988648,
      "epoch": 1.4722423802612483,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004787598877829633,
      "loss": 5.0774,
      "mean_token_accuracy": 0.19990268498659133,
      "num_tokens": 33686788.0,
      "step": 16230
    },
    {
      "entropy": 5.459230518341064,
      "epoch": 1.4726959361393324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004787461622224173,
      "loss": 4.9728,
      "mean_token_accuracy": 0.2072235971689224,
      "num_tokens": 33696384.0,
      "step": 16235
    },
    {
      "entropy": 5.405389881134033,
      "epoch": 1.4731494920174166,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004787324324483488,
      "loss": 5.0039,
      "mean_token_accuracy": 0.19962379932403565,
      "num_tokens": 33707758.0,
      "step": 16240
    },
    {
      "entropy": 5.445700073242188,
      "epoch": 1.4736030478955007,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004787186984610416,
      "loss": 4.9762,
      "mean_token_accuracy": 0.20738447904586793,
      "num_tokens": 33717747.0,
      "step": 16245
    },
    {
      "entropy": 5.458881664276123,
      "epoch": 1.474056603773585,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047870496026077984,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20439188182353973,
      "num_tokens": 33727960.0,
      "step": 16250
    },
    {
      "entropy": 5.409978675842285,
      "epoch": 1.4745101596516692,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786912178478476,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20595768839120865,
      "num_tokens": 33738874.0,
      "step": 16255
    },
    {
      "entropy": 5.426355361938477,
      "epoch": 1.4749637155297532,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004786774712225291,
      "loss": 4.961,
      "mean_token_accuracy": 0.2080053612589836,
      "num_tokens": 33749071.0,
      "step": 16260
    },
    {
      "entropy": 5.436343860626221,
      "epoch": 1.4754172714078375,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004786637203851086,
      "loss": 4.9874,
      "mean_token_accuracy": 0.20291364938020706,
      "num_tokens": 33759509.0,
      "step": 16265
    },
    {
      "entropy": 5.503203725814819,
      "epoch": 1.4758708272859216,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004786499653358705,
      "loss": 5.0265,
      "mean_token_accuracy": 0.20802777409553527,
      "num_tokens": 33769974.0,
      "step": 16270
    },
    {
      "entropy": 5.541727781295776,
      "epoch": 1.4763243831640058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004786362060750991,
      "loss": 5.0744,
      "mean_token_accuracy": 0.20014467760920523,
      "num_tokens": 33780403.0,
      "step": 16275
    },
    {
      "entropy": 5.495248937606812,
      "epoch": 1.47677793904209,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047862244260307913,
      "loss": 5.0695,
      "mean_token_accuracy": 0.20937622487545013,
      "num_tokens": 33791364.0,
      "step": 16280
    },
    {
      "entropy": 5.442668390274048,
      "epoch": 1.4772314949201741,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004786086749200951,
      "loss": 5.0002,
      "mean_token_accuracy": 0.2074051097035408,
      "num_tokens": 33801964.0,
      "step": 16285
    },
    {
      "entropy": 5.502958011627197,
      "epoch": 1.4776850507982584,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004785949030264319,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20473655313253403,
      "num_tokens": 33813132.0,
      "step": 16290
    },
    {
      "entropy": 5.482903337478637,
      "epoch": 1.4781386066763424,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004785811269223741,
      "loss": 5.0045,
      "mean_token_accuracy": 0.21504738479852675,
      "num_tokens": 33823069.0,
      "step": 16295
    },
    {
      "entropy": 5.526265478134155,
      "epoch": 1.4785921625544267,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004785673466082068,
      "loss": 5.0065,
      "mean_token_accuracy": 0.20451238751411438,
      "num_tokens": 33832626.0,
      "step": 16300
    },
    {
      "entropy": 5.423535013198853,
      "epoch": 1.479045718432511,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047855356208421484,
      "loss": 4.9846,
      "mean_token_accuracy": 0.20531942695379257,
      "num_tokens": 33843384.0,
      "step": 16305
    },
    {
      "entropy": 5.421808290481567,
      "epoch": 1.479499274310595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004785397733506834,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20820764154195787,
      "num_tokens": 33854286.0,
      "step": 16310
    },
    {
      "entropy": 5.415575790405273,
      "epoch": 1.4799528301886793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004785259804078976,
      "loss": 4.9538,
      "mean_token_accuracy": 0.20777007490396499,
      "num_tokens": 33863379.0,
      "step": 16315
    },
    {
      "entropy": 5.528962898254394,
      "epoch": 1.4804063860667633,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004785121832561426,
      "loss": 5.196,
      "mean_token_accuracy": 0.1923913463950157,
      "num_tokens": 33875641.0,
      "step": 16320
    },
    {
      "entropy": 5.500227355957032,
      "epoch": 1.4808599419448476,
      "grad_norm": 1.0,
      "learning_rate": 0.00047849838189570375,
      "loss": 4.9704,
      "mean_token_accuracy": 0.2062392935156822,
      "num_tokens": 33885023.0,
      "step": 16325
    },
    {
      "entropy": 5.584210300445557,
      "epoch": 1.4813134978229319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004784845763268666,
      "loss": 5.1143,
      "mean_token_accuracy": 0.19259014427661897,
      "num_tokens": 33895562.0,
      "step": 16330
    },
    {
      "entropy": 5.436470556259155,
      "epoch": 1.481767053701016,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004784707665499165,
      "loss": 5.0273,
      "mean_token_accuracy": 0.2102821946144104,
      "num_tokens": 33907095.0,
      "step": 16335
    },
    {
      "entropy": 5.387659215927124,
      "epoch": 1.4822206095791002,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004784569525651391,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2163789764046669,
      "num_tokens": 33917475.0,
      "step": 16340
    },
    {
      "entropy": 5.430476999282837,
      "epoch": 1.4826741654571842,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047844313437282005,
      "loss": 5.0141,
      "mean_token_accuracy": 0.2002805709838867,
      "num_tokens": 33927740.0,
      "step": 16345
    },
    {
      "entropy": 5.4267418384552,
      "epoch": 1.4831277213352685,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047842931197324525,
      "loss": 4.952,
      "mean_token_accuracy": 0.20934444218873977,
      "num_tokens": 33937416.0,
      "step": 16350
    },
    {
      "entropy": 5.425552940368652,
      "epoch": 1.4835812772133528,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047841548536670033,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21422414779663085,
      "num_tokens": 33946968.0,
      "step": 16355
    },
    {
      "entropy": 5.5388099193573,
      "epoch": 1.4840348330914368,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047840165455347136,
      "loss": 5.0852,
      "mean_token_accuracy": 0.19429281800985337,
      "num_tokens": 33957877.0,
      "step": 16360
    },
    {
      "entropy": 5.499372148513794,
      "epoch": 1.484488388969521,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047838781953384436,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19985658228397368,
      "num_tokens": 33967890.0,
      "step": 16365
    },
    {
      "entropy": 5.442298507690429,
      "epoch": 1.484941944847605,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004783739803081054,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20722893327474595,
      "num_tokens": 33977117.0,
      "step": 16370
    },
    {
      "entropy": 5.469358491897583,
      "epoch": 1.4853955007256894,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047836013687654075,
      "loss": 5.0944,
      "mean_token_accuracy": 0.197372367978096,
      "num_tokens": 33986943.0,
      "step": 16375
    },
    {
      "entropy": 5.4732842445373535,
      "epoch": 1.4858490566037736,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047834628923943654,
      "loss": 5.0004,
      "mean_token_accuracy": 0.20302316546440125,
      "num_tokens": 33996499.0,
      "step": 16380
    },
    {
      "entropy": 5.5385230541229244,
      "epoch": 1.4863026124818577,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004783324373970793,
      "loss": 5.1122,
      "mean_token_accuracy": 0.20174136459827424,
      "num_tokens": 34006749.0,
      "step": 16385
    },
    {
      "entropy": 5.5679679870605465,
      "epoch": 1.486756168359942,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004783185813497555,
      "loss": 5.1131,
      "mean_token_accuracy": 0.1920430764555931,
      "num_tokens": 34017194.0,
      "step": 16390
    },
    {
      "entropy": 5.541804218292237,
      "epoch": 1.487209724238026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004783047210977516,
      "loss": 5.0368,
      "mean_token_accuracy": 0.2090506985783577,
      "num_tokens": 34027386.0,
      "step": 16395
    },
    {
      "entropy": 5.541867780685425,
      "epoch": 1.4876632801161103,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047829085664135424,
      "loss": 5.0698,
      "mean_token_accuracy": 0.19776039570569992,
      "num_tokens": 34037540.0,
      "step": 16400
    },
    {
      "entropy": 5.447138357162475,
      "epoch": 1.4881168359941945,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047827698798085015,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20726717859506608,
      "num_tokens": 34048008.0,
      "step": 16405
    },
    {
      "entropy": 5.5058471202850345,
      "epoch": 1.4885703918722788,
      "grad_norm": 1.125,
      "learning_rate": 0.0004782631151165262,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20902367681264877,
      "num_tokens": 34058009.0,
      "step": 16410
    },
    {
      "entropy": 5.511841058731079,
      "epoch": 1.4890239477503628,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047824923804866914,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20172156542539596,
      "num_tokens": 34067625.0,
      "step": 16415
    },
    {
      "entropy": 5.414007949829101,
      "epoch": 1.4894775036284469,
      "grad_norm": 1.0,
      "learning_rate": 0.00047823535677756607,
      "loss": 5.0106,
      "mean_token_accuracy": 0.1978861317038536,
      "num_tokens": 34077349.0,
      "step": 16420
    },
    {
      "entropy": 5.532129621505737,
      "epoch": 1.4899310595065312,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004782214713035041,
      "loss": 5.1318,
      "mean_token_accuracy": 0.19498357325792312,
      "num_tokens": 34088636.0,
      "step": 16425
    },
    {
      "entropy": 5.522089290618896,
      "epoch": 1.4903846153846154,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047820758162677026,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20185174643993378,
      "num_tokens": 34098806.0,
      "step": 16430
    },
    {
      "entropy": 5.449494934082031,
      "epoch": 1.4908381712626997,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047819368774765187,
      "loss": 5.0152,
      "mean_token_accuracy": 0.20564057677984238,
      "num_tokens": 34109713.0,
      "step": 16435
    },
    {
      "entropy": 5.445164585113526,
      "epoch": 1.4912917271407837,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047817978966643626,
      "loss": 4.9892,
      "mean_token_accuracy": 0.2078770637512207,
      "num_tokens": 34119585.0,
      "step": 16440
    },
    {
      "entropy": 5.544268798828125,
      "epoch": 1.491745283018868,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004781658873834108,
      "loss": 5.0416,
      "mean_token_accuracy": 0.19779797941446303,
      "num_tokens": 34131245.0,
      "step": 16445
    },
    {
      "entropy": 5.49971981048584,
      "epoch": 1.492198838896952,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047815198089886296,
      "loss": 5.0118,
      "mean_token_accuracy": 0.2045518785715103,
      "num_tokens": 34141229.0,
      "step": 16450
    },
    {
      "entropy": 5.47653374671936,
      "epoch": 1.4926523947750363,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004781380702130805,
      "loss": 5.0486,
      "mean_token_accuracy": 0.19726551175117493,
      "num_tokens": 34151781.0,
      "step": 16455
    },
    {
      "entropy": 5.417407560348511,
      "epoch": 1.4931059506531206,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047812415532635087,
      "loss": 5.002,
      "mean_token_accuracy": 0.2048860549926758,
      "num_tokens": 34162305.0,
      "step": 16460
    },
    {
      "entropy": 5.498383903503418,
      "epoch": 1.4935595065312046,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000478110236238962,
      "loss": 5.087,
      "mean_token_accuracy": 0.1978554055094719,
      "num_tokens": 34173613.0,
      "step": 16465
    },
    {
      "entropy": 5.472800397872925,
      "epoch": 1.4940130624092889,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047809631295120164,
      "loss": 5.0069,
      "mean_token_accuracy": 0.20078982263803483,
      "num_tokens": 34183540.0,
      "step": 16470
    },
    {
      "entropy": 5.450933361053467,
      "epoch": 1.494466618287373,
      "grad_norm": 1.0,
      "learning_rate": 0.00047808238546335786,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20164911299943925,
      "num_tokens": 34193292.0,
      "step": 16475
    },
    {
      "entropy": 5.506211948394776,
      "epoch": 1.4949201741654572,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047806845377571856,
      "loss": 5.07,
      "mean_token_accuracy": 0.20894212722778321,
      "num_tokens": 34202697.0,
      "step": 16480
    },
    {
      "entropy": 5.462220191955566,
      "epoch": 1.4953737300435415,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047805451788857187,
      "loss": 4.965,
      "mean_token_accuracy": 0.20638453364372253,
      "num_tokens": 34213396.0,
      "step": 16485
    },
    {
      "entropy": 5.507080602645874,
      "epoch": 1.4958272859216255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000478040577802206,
      "loss": 4.9704,
      "mean_token_accuracy": 0.2080722451210022,
      "num_tokens": 34223020.0,
      "step": 16490
    },
    {
      "entropy": 5.410426235198974,
      "epoch": 1.4962808417997098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004780266335169093,
      "loss": 5.0043,
      "mean_token_accuracy": 0.20901964604854584,
      "num_tokens": 34233249.0,
      "step": 16495
    },
    {
      "entropy": 5.56065182685852,
      "epoch": 1.4967343976777938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047801268503297,
      "loss": 5.0666,
      "mean_token_accuracy": 0.19952577948570252,
      "num_tokens": 34244091.0,
      "step": 16500
    },
    {
      "entropy": 5.526365327835083,
      "epoch": 1.497187953555878,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047799873235067667,
      "loss": 5.0661,
      "mean_token_accuracy": 0.20329954028129577,
      "num_tokens": 34255060.0,
      "step": 16505
    },
    {
      "entropy": 5.435563945770264,
      "epoch": 1.4976415094339623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004779847754703178,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21159472316503525,
      "num_tokens": 34264361.0,
      "step": 16510
    },
    {
      "entropy": 5.440701484680176,
      "epoch": 1.4980950653120464,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004779708143921821,
      "loss": 5.0182,
      "mean_token_accuracy": 0.20453838855028153,
      "num_tokens": 34274480.0,
      "step": 16515
    },
    {
      "entropy": 5.460260534286499,
      "epoch": 1.4985486211901307,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047795684911655824,
      "loss": 5.0082,
      "mean_token_accuracy": 0.20553855001926422,
      "num_tokens": 34285951.0,
      "step": 16520
    },
    {
      "entropy": 5.507027626037598,
      "epoch": 1.4990021770682147,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477942879643735,
      "loss": 5.0732,
      "mean_token_accuracy": 0.1956965371966362,
      "num_tokens": 34296384.0,
      "step": 16525
    },
    {
      "entropy": 5.470954036712646,
      "epoch": 1.499455732946299,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047792890597400125,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20878826379776,
      "num_tokens": 34306776.0,
      "step": 16530
    },
    {
      "entropy": 5.514407634735107,
      "epoch": 1.4999092888243832,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047791492810764604,
      "loss": 5.013,
      "mean_token_accuracy": 0.2054852917790413,
      "num_tokens": 34317047.0,
      "step": 16535
    },
    {
      "entropy": 5.449463796615601,
      "epoch": 1.5003628447024675,
      "grad_norm": 1.0,
      "learning_rate": 0.0004779009460449584,
      "loss": 4.9674,
      "mean_token_accuracy": 0.2083308905363083,
      "num_tokens": 34326392.0,
      "step": 16540
    },
    {
      "entropy": 5.503531122207642,
      "epoch": 1.5008164005805515,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047788695978622756,
      "loss": 5.0126,
      "mean_token_accuracy": 0.21120216250419616,
      "num_tokens": 34336580.0,
      "step": 16545
    },
    {
      "entropy": 5.450557279586792,
      "epoch": 1.5012699564586356,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047787296933174265,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20305247008800506,
      "num_tokens": 34346792.0,
      "step": 16550
    },
    {
      "entropy": 5.5099790573120115,
      "epoch": 1.5017235123367199,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000477858974681793,
      "loss": 5.0496,
      "mean_token_accuracy": 0.2033532068133354,
      "num_tokens": 34356675.0,
      "step": 16555
    },
    {
      "entropy": 5.521383571624756,
      "epoch": 1.5021770682148041,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004778449758366681,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21447925865650178,
      "num_tokens": 34366357.0,
      "step": 16560
    },
    {
      "entropy": 5.44498348236084,
      "epoch": 1.5026306240928884,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004778309727966574,
      "loss": 5.0014,
      "mean_token_accuracy": 0.20686540603637696,
      "num_tokens": 34377229.0,
      "step": 16565
    },
    {
      "entropy": 5.43284330368042,
      "epoch": 1.5030841799709724,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047781696556205045,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2040479376912117,
      "num_tokens": 34387873.0,
      "step": 16570
    },
    {
      "entropy": 5.49597134590149,
      "epoch": 1.5035377358490565,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047780295413313696,
      "loss": 4.9691,
      "mean_token_accuracy": 0.20865806341171264,
      "num_tokens": 34398889.0,
      "step": 16575
    },
    {
      "entropy": 5.448739242553711,
      "epoch": 1.5039912917271407,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047778893851020676,
      "loss": 4.9186,
      "mean_token_accuracy": 0.20451200604438782,
      "num_tokens": 34408508.0,
      "step": 16580
    },
    {
      "entropy": 5.430432462692261,
      "epoch": 1.504444847605225,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004777749186935496,
      "loss": 5.0831,
      "mean_token_accuracy": 0.20057281851768494,
      "num_tokens": 34418928.0,
      "step": 16585
    },
    {
      "entropy": 5.500459241867065,
      "epoch": 1.5048984034833093,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004777608946834555,
      "loss": 5.0525,
      "mean_token_accuracy": 0.20214829593896866,
      "num_tokens": 34429789.0,
      "step": 16590
    },
    {
      "entropy": 5.56063404083252,
      "epoch": 1.5053519593613933,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004777468664802144,
      "loss": 5.1551,
      "mean_token_accuracy": 0.19416424483060837,
      "num_tokens": 34440795.0,
      "step": 16595
    },
    {
      "entropy": 5.541483449935913,
      "epoch": 1.5058055152394774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047773283408411636,
      "loss": 5.0441,
      "mean_token_accuracy": 0.19526468813419343,
      "num_tokens": 34450772.0,
      "step": 16600
    },
    {
      "entropy": 5.456420469284057,
      "epoch": 1.5062590711175616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047771879749545175,
      "loss": 5.003,
      "mean_token_accuracy": 0.20167505890130996,
      "num_tokens": 34461273.0,
      "step": 16605
    },
    {
      "entropy": 5.47718997001648,
      "epoch": 1.506712626995646,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047770475671451073,
      "loss": 5.0562,
      "mean_token_accuracy": 0.20287921130657197,
      "num_tokens": 34470942.0,
      "step": 16610
    },
    {
      "entropy": 5.515540313720703,
      "epoch": 1.5071661828737302,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004776907117415837,
      "loss": 4.9805,
      "mean_token_accuracy": 0.20255957096815108,
      "num_tokens": 34481117.0,
      "step": 16615
    },
    {
      "entropy": 5.4488598823547365,
      "epoch": 1.5076197387518142,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047767666257696103,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20787885636091233,
      "num_tokens": 34490953.0,
      "step": 16620
    },
    {
      "entropy": 5.443657064437867,
      "epoch": 1.5080732946298983,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047766260922093337,
      "loss": 4.916,
      "mean_token_accuracy": 0.205417500436306,
      "num_tokens": 34502765.0,
      "step": 16625
    },
    {
      "entropy": 5.56479024887085,
      "epoch": 1.5085268505079825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004776485516737913,
      "loss": 5.1003,
      "mean_token_accuracy": 0.19664451330900193,
      "num_tokens": 34514473.0,
      "step": 16630
    },
    {
      "entropy": 5.538726806640625,
      "epoch": 1.5089804063860668,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004776344899358256,
      "loss": 5.1208,
      "mean_token_accuracy": 0.1967710256576538,
      "num_tokens": 34524218.0,
      "step": 16635
    },
    {
      "entropy": 5.490627384185791,
      "epoch": 1.509433962264151,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004776204240073269,
      "loss": 5.0052,
      "mean_token_accuracy": 0.2014483705163002,
      "num_tokens": 34535697.0,
      "step": 16640
    },
    {
      "entropy": 5.509071969985962,
      "epoch": 1.509887518142235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047760635388858624,
      "loss": 5.0442,
      "mean_token_accuracy": 0.20891896039247512,
      "num_tokens": 34545130.0,
      "step": 16645
    },
    {
      "entropy": 5.454761981964111,
      "epoch": 1.5103410740203191,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047759227957989465,
      "loss": 4.9643,
      "mean_token_accuracy": 0.2079317569732666,
      "num_tokens": 34554974.0,
      "step": 16650
    },
    {
      "entropy": 5.485178279876709,
      "epoch": 1.5107946298984034,
      "grad_norm": 1.015625,
      "learning_rate": 0.000477578201081543,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20100889652967452,
      "num_tokens": 34565660.0,
      "step": 16655
    },
    {
      "entropy": 5.537592601776123,
      "epoch": 1.5112481857764877,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004775641183938225,
      "loss": 5.0433,
      "mean_token_accuracy": 0.20447785705327987,
      "num_tokens": 34576447.0,
      "step": 16660
    },
    {
      "entropy": 5.565694332122803,
      "epoch": 1.511701741654572,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047755003151702443,
      "loss": 5.173,
      "mean_token_accuracy": 0.19871526211500168,
      "num_tokens": 34588333.0,
      "step": 16665
    },
    {
      "entropy": 5.423282337188721,
      "epoch": 1.512155297532656,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047753594045144015,
      "loss": 4.9193,
      "mean_token_accuracy": 0.20331991463899612,
      "num_tokens": 34598166.0,
      "step": 16670
    },
    {
      "entropy": 5.45153112411499,
      "epoch": 1.5126088534107403,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004775218451973609,
      "loss": 4.9566,
      "mean_token_accuracy": 0.20150122046470642,
      "num_tokens": 34608615.0,
      "step": 16675
    },
    {
      "entropy": 5.474885368347168,
      "epoch": 1.5130624092888243,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004775077457550784,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2053156614303589,
      "num_tokens": 34618618.0,
      "step": 16680
    },
    {
      "entropy": 5.481243419647217,
      "epoch": 1.5135159651669086,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047749364212488403,
      "loss": 5.0388,
      "mean_token_accuracy": 0.19779873341321946,
      "num_tokens": 34629412.0,
      "step": 16685
    },
    {
      "entropy": 5.54996166229248,
      "epoch": 1.5139695210449928,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004774795343070696,
      "loss": 5.1042,
      "mean_token_accuracy": 0.20105478912591934,
      "num_tokens": 34640487.0,
      "step": 16690
    },
    {
      "entropy": 5.509821653366089,
      "epoch": 1.5144230769230769,
      "grad_norm": 1.0,
      "learning_rate": 0.0004774654223019268,
      "loss": 5.0338,
      "mean_token_accuracy": 0.20159787237644194,
      "num_tokens": 34650126.0,
      "step": 16695
    },
    {
      "entropy": 5.406566762924195,
      "epoch": 1.5148766328011611,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004774513061097474,
      "loss": 4.8821,
      "mean_token_accuracy": 0.21117860376834868,
      "num_tokens": 34659721.0,
      "step": 16700
    },
    {
      "entropy": 5.463156604766846,
      "epoch": 1.5153301886792452,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004774371857308235,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20736206471920013,
      "num_tokens": 34670611.0,
      "step": 16705
    },
    {
      "entropy": 5.461158514022827,
      "epoch": 1.5157837445573294,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047742306116544693,
      "loss": 4.9982,
      "mean_token_accuracy": 0.20645657032728196,
      "num_tokens": 34680173.0,
      "step": 16710
    },
    {
      "entropy": 5.4701244831085205,
      "epoch": 1.5162373004354137,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047740893241390994,
      "loss": 5.0449,
      "mean_token_accuracy": 0.20616694539785385,
      "num_tokens": 34690176.0,
      "step": 16715
    },
    {
      "entropy": 5.515377187728882,
      "epoch": 1.516690856313498,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004773947994765046,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2083370253443718,
      "num_tokens": 34700481.0,
      "step": 16720
    },
    {
      "entropy": 5.455897617340088,
      "epoch": 1.517144412191582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004773806623535232,
      "loss": 5.1347,
      "mean_token_accuracy": 0.19965612143278122,
      "num_tokens": 34710643.0,
      "step": 16725
    },
    {
      "entropy": 5.507606935501099,
      "epoch": 1.517597968069666,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047736652104525816,
      "loss": 5.0098,
      "mean_token_accuracy": 0.1999215304851532,
      "num_tokens": 34720998.0,
      "step": 16730
    },
    {
      "entropy": 5.544073629379272,
      "epoch": 1.5180515239477503,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004773523755520019,
      "loss": 5.0635,
      "mean_token_accuracy": 0.20022758543491365,
      "num_tokens": 34731071.0,
      "step": 16735
    },
    {
      "entropy": 5.468399381637573,
      "epoch": 1.5185050798258346,
      "grad_norm": 1.078125,
      "learning_rate": 0.000477338225874047,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20834554135799407,
      "num_tokens": 34741346.0,
      "step": 16740
    },
    {
      "entropy": 5.4984602451324465,
      "epoch": 1.5189586357039189,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047732407201168595,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2056797116994858,
      "num_tokens": 34751092.0,
      "step": 16745
    },
    {
      "entropy": 5.559841680526733,
      "epoch": 1.519412191582003,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004773099139652116,
      "loss": 5.0843,
      "mean_token_accuracy": 0.2010665148496628,
      "num_tokens": 34761477.0,
      "step": 16750
    },
    {
      "entropy": 5.448795461654663,
      "epoch": 1.519865747460087,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004772957517349167,
      "loss": 4.9997,
      "mean_token_accuracy": 0.20591279715299607,
      "num_tokens": 34772491.0,
      "step": 16755
    },
    {
      "entropy": 5.4718138694763185,
      "epoch": 1.5203193033381712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004772815853210939,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19933177679777145,
      "num_tokens": 34783192.0,
      "step": 16760
    },
    {
      "entropy": 5.4850438117980955,
      "epoch": 1.5207728592162555,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047726741472403655,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20672593712806703,
      "num_tokens": 34794164.0,
      "step": 16765
    },
    {
      "entropy": 5.508155679702758,
      "epoch": 1.5212264150943398,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047725323994403753,
      "loss": 5.0715,
      "mean_token_accuracy": 0.19614226073026658,
      "num_tokens": 34804390.0,
      "step": 16770
    },
    {
      "entropy": 5.594349098205567,
      "epoch": 1.5216799709724238,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047723906098138993,
      "loss": 5.155,
      "mean_token_accuracy": 0.1935633584856987,
      "num_tokens": 34814751.0,
      "step": 16775
    },
    {
      "entropy": 5.500635290145874,
      "epoch": 1.5221335268505078,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047722487783638697,
      "loss": 4.9794,
      "mean_token_accuracy": 0.20784967839717866,
      "num_tokens": 34824977.0,
      "step": 16780
    },
    {
      "entropy": 5.4684912204742435,
      "epoch": 1.5225870827285921,
      "grad_norm": 1.140625,
      "learning_rate": 0.000477210690509322,
      "loss": 5.0588,
      "mean_token_accuracy": 0.20255587100982667,
      "num_tokens": 34835540.0,
      "step": 16785
    },
    {
      "entropy": 5.525739097595215,
      "epoch": 1.5230406386066764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047719649900048843,
      "loss": 5.0292,
      "mean_token_accuracy": 0.20400345474481582,
      "num_tokens": 34845233.0,
      "step": 16790
    },
    {
      "entropy": 5.545043039321899,
      "epoch": 1.5234941944847606,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047718230331017975,
      "loss": 5.1223,
      "mean_token_accuracy": 0.19574718326330184,
      "num_tokens": 34855756.0,
      "step": 16795
    },
    {
      "entropy": 5.522501468658447,
      "epoch": 1.5239477503628447,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047716810343868956,
      "loss": 5.0383,
      "mean_token_accuracy": 0.191273033618927,
      "num_tokens": 34867869.0,
      "step": 16800
    },
    {
      "entropy": 5.5965899467468265,
      "epoch": 1.5244013062409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047715389938631143,
      "loss": 5.1004,
      "mean_token_accuracy": 0.1961804524064064,
      "num_tokens": 34878625.0,
      "step": 16805
    },
    {
      "entropy": 5.5731347560882565,
      "epoch": 1.524854862119013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047713969115333916,
      "loss": 5.0876,
      "mean_token_accuracy": 0.1929438978433609,
      "num_tokens": 34889408.0,
      "step": 16810
    },
    {
      "entropy": 5.465712594985962,
      "epoch": 1.5253084179970973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004771254787400666,
      "loss": 5.0956,
      "mean_token_accuracy": 0.19568723440170288,
      "num_tokens": 34899602.0,
      "step": 16815
    },
    {
      "entropy": 5.422050142288208,
      "epoch": 1.5257619738751815,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004771112621467876,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20706008970737458,
      "num_tokens": 34910215.0,
      "step": 16820
    },
    {
      "entropy": 5.453781223297119,
      "epoch": 1.5262155297532656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004770970413737961,
      "loss": 5.0117,
      "mean_token_accuracy": 0.20857041776180268,
      "num_tokens": 34919641.0,
      "step": 16825
    },
    {
      "entropy": 5.494102191925049,
      "epoch": 1.5266690856313496,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004770828164213865,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21438930034637452,
      "num_tokens": 34930243.0,
      "step": 16830
    },
    {
      "entropy": 5.524109840393066,
      "epoch": 1.5271226415094339,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004770685872898525,
      "loss": 5.0879,
      "mean_token_accuracy": 0.20063476711511613,
      "num_tokens": 34941057.0,
      "step": 16835
    },
    {
      "entropy": 5.489181804656982,
      "epoch": 1.5275761973875182,
      "grad_norm": 0.875,
      "learning_rate": 0.00047705435397948876,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2134463220834732,
      "num_tokens": 34951150.0,
      "step": 16840
    },
    {
      "entropy": 5.508066415786743,
      "epoch": 1.5280297532656024,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770401164905895,
      "loss": 5.0494,
      "mean_token_accuracy": 0.20280857831239701,
      "num_tokens": 34961648.0,
      "step": 16845
    },
    {
      "entropy": 5.499263668060303,
      "epoch": 1.5284833091436865,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047702587482344915,
      "loss": 5.0406,
      "mean_token_accuracy": 0.19851076304912568,
      "num_tokens": 34971573.0,
      "step": 16850
    },
    {
      "entropy": 5.529312515258789,
      "epoch": 1.5289368650217707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004770116289783622,
      "loss": 5.0283,
      "mean_token_accuracy": 0.19798604249954224,
      "num_tokens": 34981984.0,
      "step": 16855
    },
    {
      "entropy": 5.510770082473755,
      "epoch": 1.5293904208998548,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004769973789556233,
      "loss": 5.0463,
      "mean_token_accuracy": 0.20022655874490738,
      "num_tokens": 34990861.0,
      "step": 16860
    },
    {
      "entropy": 5.472228527069092,
      "epoch": 1.529843976777939,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047698312475552714,
      "loss": 5.0443,
      "mean_token_accuracy": 0.20059497505426407,
      "num_tokens": 35001504.0,
      "step": 16865
    },
    {
      "entropy": 5.514711380004883,
      "epoch": 1.5302975326560233,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004769688663783684,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21026831716299058,
      "num_tokens": 35012425.0,
      "step": 16870
    },
    {
      "entropy": 5.482202053070068,
      "epoch": 1.5307510885341074,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004769546038244421,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2071754142642021,
      "num_tokens": 35022854.0,
      "step": 16875
    },
    {
      "entropy": 5.526014947891236,
      "epoch": 1.5312046444121916,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047694033709404313,
      "loss": 5.0733,
      "mean_token_accuracy": 0.19844924956560134,
      "num_tokens": 35032694.0,
      "step": 16880
    },
    {
      "entropy": 5.488026809692383,
      "epoch": 1.5316582002902757,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047692606618746655,
      "loss": 4.9977,
      "mean_token_accuracy": 0.2060248002409935,
      "num_tokens": 35042217.0,
      "step": 16885
    },
    {
      "entropy": 5.480239391326904,
      "epoch": 1.53211175616836,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047691179110500735,
      "loss": 5.0745,
      "mean_token_accuracy": 0.19469793438911437,
      "num_tokens": 35052831.0,
      "step": 16890
    },
    {
      "entropy": 5.5425124168396,
      "epoch": 1.5325653120464442,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047689751184696097,
      "loss": 5.1061,
      "mean_token_accuracy": 0.19845710843801498,
      "num_tokens": 35061955.0,
      "step": 16895
    },
    {
      "entropy": 5.5257713317871096,
      "epoch": 1.5330188679245285,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004768832284136226,
      "loss": 5.005,
      "mean_token_accuracy": 0.20163582265377045,
      "num_tokens": 35072526.0,
      "step": 16900
    },
    {
      "entropy": 5.4757163524627686,
      "epoch": 1.5334724238026125,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004768689408052876,
      "loss": 4.9802,
      "mean_token_accuracy": 0.20861303061246872,
      "num_tokens": 35082415.0,
      "step": 16905
    },
    {
      "entropy": 5.436173772811889,
      "epoch": 1.5339259796806965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047685464902225144,
      "loss": 4.9634,
      "mean_token_accuracy": 0.2085571765899658,
      "num_tokens": 35092805.0,
      "step": 16910
    },
    {
      "entropy": 5.4842204570770265,
      "epoch": 1.5343795355587808,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004768403530648097,
      "loss": 5.0327,
      "mean_token_accuracy": 0.20818058848381044,
      "num_tokens": 35102963.0,
      "step": 16915
    },
    {
      "entropy": 5.521869802474976,
      "epoch": 1.534833091436865,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047682605293325806,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20084257870912553,
      "num_tokens": 35114140.0,
      "step": 16920
    },
    {
      "entropy": 5.5732200145721436,
      "epoch": 1.5352866473149493,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047681174862789215,
      "loss": 5.1061,
      "mean_token_accuracy": 0.2038816675543785,
      "num_tokens": 35125314.0,
      "step": 16925
    },
    {
      "entropy": 5.484570741653442,
      "epoch": 1.5357402031930334,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047679744014900795,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20243679136037826,
      "num_tokens": 35134934.0,
      "step": 16930
    },
    {
      "entropy": 5.436294937133789,
      "epoch": 1.5361937590711174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004767831274969011,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20469433665275574,
      "num_tokens": 35145341.0,
      "step": 16935
    },
    {
      "entropy": 5.412739849090576,
      "epoch": 1.5366473149492017,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047676881067186793,
      "loss": 4.9463,
      "mean_token_accuracy": 0.20421936362981796,
      "num_tokens": 35156046.0,
      "step": 16940
    },
    {
      "entropy": 5.539416217803955,
      "epoch": 1.537100870827286,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047675448967420426,
      "loss": 5.0975,
      "mean_token_accuracy": 0.19606997519731523,
      "num_tokens": 35168232.0,
      "step": 16945
    },
    {
      "entropy": 5.580508279800415,
      "epoch": 1.5375544267053702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004767401645042064,
      "loss": 5.0575,
      "mean_token_accuracy": 0.2020314961671829,
      "num_tokens": 35178585.0,
      "step": 16950
    },
    {
      "entropy": 5.472460174560547,
      "epoch": 1.5380079825834543,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047672583516217043,
      "loss": 4.9249,
      "mean_token_accuracy": 0.20863931775093078,
      "num_tokens": 35189417.0,
      "step": 16955
    },
    {
      "entropy": 5.45528507232666,
      "epoch": 1.5384615384615383,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004767115016483928,
      "loss": 4.983,
      "mean_token_accuracy": 0.2048747405409813,
      "num_tokens": 35200167.0,
      "step": 16960
    },
    {
      "entropy": 5.474616384506225,
      "epoch": 1.5389150943396226,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004766971639631699,
      "loss": 4.9924,
      "mean_token_accuracy": 0.20581676810979843,
      "num_tokens": 35210711.0,
      "step": 16965
    },
    {
      "entropy": 5.442872667312622,
      "epoch": 1.5393686502177069,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047668282210679825,
      "loss": 5.0311,
      "mean_token_accuracy": 0.19943596422672272,
      "num_tokens": 35220866.0,
      "step": 16970
    },
    {
      "entropy": 5.412336540222168,
      "epoch": 1.5398222060957911,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004766684760795745,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20932998210191728,
      "num_tokens": 35231478.0,
      "step": 16975
    },
    {
      "entropy": 5.415338563919067,
      "epoch": 1.5402757619738752,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047665412588179524,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21175016909837724,
      "num_tokens": 35242233.0,
      "step": 16980
    },
    {
      "entropy": 5.461297941207886,
      "epoch": 1.5407293178519592,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004766397715137573,
      "loss": 4.9799,
      "mean_token_accuracy": 0.1990339368581772,
      "num_tokens": 35253097.0,
      "step": 16985
    },
    {
      "entropy": 5.576567077636719,
      "epoch": 1.5411828737300435,
      "grad_norm": 1.0,
      "learning_rate": 0.00047662541297575743,
      "loss": 5.1428,
      "mean_token_accuracy": 0.19975366592407226,
      "num_tokens": 35264258.0,
      "step": 16990
    },
    {
      "entropy": 5.605187034606933,
      "epoch": 1.5416364296081277,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047661105026809274,
      "loss": 5.1254,
      "mean_token_accuracy": 0.19700509458780288,
      "num_tokens": 35273501.0,
      "step": 16995
    },
    {
      "entropy": 5.460659599304199,
      "epoch": 1.542089985486212,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004765966833910601,
      "loss": 5.0665,
      "mean_token_accuracy": 0.1999218925833702,
      "num_tokens": 35284125.0,
      "step": 17000
    },
    {
      "entropy": 5.564423942565918,
      "epoch": 1.542543541364296,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004765823123449567,
      "loss": 5.0637,
      "mean_token_accuracy": 0.19853903353214264,
      "num_tokens": 35296039.0,
      "step": 17005
    },
    {
      "entropy": 5.506070804595947,
      "epoch": 1.54299709724238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047656793713007965,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20298447459936142,
      "num_tokens": 35306178.0,
      "step": 17010
    },
    {
      "entropy": 5.462584400177002,
      "epoch": 1.5434506531204644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047655355774672636,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20627882033586503,
      "num_tokens": 35316445.0,
      "step": 17015
    },
    {
      "entropy": 5.519912672042847,
      "epoch": 1.5439042089985486,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047653917419519413,
      "loss": 5.0143,
      "mean_token_accuracy": 0.1988663762807846,
      "num_tokens": 35327274.0,
      "step": 17020
    },
    {
      "entropy": 5.478893756866455,
      "epoch": 1.544357764876633,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004765247864757804,
      "loss": 4.9896,
      "mean_token_accuracy": 0.20485960692167282,
      "num_tokens": 35339007.0,
      "step": 17025
    },
    {
      "entropy": 5.509318256378174,
      "epoch": 1.544811320754717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004765103945887827,
      "loss": 5.1123,
      "mean_token_accuracy": 0.19657397717237474,
      "num_tokens": 35349303.0,
      "step": 17030
    },
    {
      "entropy": 5.529847812652588,
      "epoch": 1.5452648766328012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004764959985344987,
      "loss": 4.9534,
      "mean_token_accuracy": 0.20820019841194154,
      "num_tokens": 35359556.0,
      "step": 17035
    },
    {
      "entropy": 5.45951886177063,
      "epoch": 1.5457184325108853,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004764815983132261,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20989398658275604,
      "num_tokens": 35370043.0,
      "step": 17040
    },
    {
      "entropy": 5.461466217041016,
      "epoch": 1.5461719883889695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004764671939252628,
      "loss": 4.968,
      "mean_token_accuracy": 0.20857801139354706,
      "num_tokens": 35380031.0,
      "step": 17045
    },
    {
      "entropy": 5.488165378570557,
      "epoch": 1.5466255442670538,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004764527853709064,
      "loss": 5.0661,
      "mean_token_accuracy": 0.199637970328331,
      "num_tokens": 35390481.0,
      "step": 17050
    },
    {
      "entropy": 5.500989580154419,
      "epoch": 1.5470791001451378,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004764383726504552,
      "loss": 5.0531,
      "mean_token_accuracy": 0.19923148900270463,
      "num_tokens": 35401756.0,
      "step": 17055
    },
    {
      "entropy": 5.526605939865112,
      "epoch": 1.547532656023222,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004764239557642071,
      "loss": 4.9325,
      "mean_token_accuracy": 0.21369479596614838,
      "num_tokens": 35411045.0,
      "step": 17060
    },
    {
      "entropy": 5.458116579055786,
      "epoch": 1.5479862119013061,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004764095347124603,
      "loss": 5.023,
      "mean_token_accuracy": 0.1952745273709297,
      "num_tokens": 35421256.0,
      "step": 17065
    },
    {
      "entropy": 5.468409061431885,
      "epoch": 1.5484397677793904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004763951094955129,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20428329855203628,
      "num_tokens": 35430798.0,
      "step": 17070
    },
    {
      "entropy": 5.4707554340362545,
      "epoch": 1.5488933236574747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004763806801136634,
      "loss": 4.9798,
      "mean_token_accuracy": 0.21311040967702866,
      "num_tokens": 35440459.0,
      "step": 17075
    },
    {
      "entropy": 5.475528955459595,
      "epoch": 1.549346879535559,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004763662465672101,
      "loss": 5.0243,
      "mean_token_accuracy": 0.20399243980646134,
      "num_tokens": 35451586.0,
      "step": 17080
    },
    {
      "entropy": 5.487248945236206,
      "epoch": 1.549800435413643,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047635180885645146,
      "loss": 5.0435,
      "mean_token_accuracy": 0.20108581632375716,
      "num_tokens": 35462197.0,
      "step": 17085
    },
    {
      "entropy": 5.514713144302368,
      "epoch": 1.550253991291727,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004763373669816862,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20966029018163682,
      "num_tokens": 35471737.0,
      "step": 17090
    },
    {
      "entropy": 5.527065849304199,
      "epoch": 1.5507075471698113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004763229209432128,
      "loss": 5.0571,
      "mean_token_accuracy": 0.1970732882618904,
      "num_tokens": 35482016.0,
      "step": 17095
    },
    {
      "entropy": 5.530784654617309,
      "epoch": 1.5511611030478956,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004763084707413301,
      "loss": 5.1574,
      "mean_token_accuracy": 0.19198521673679353,
      "num_tokens": 35492622.0,
      "step": 17100
    },
    {
      "entropy": 5.57908353805542,
      "epoch": 1.5516146589259798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047629401637633696,
      "loss": 5.1757,
      "mean_token_accuracy": 0.19465037435293198,
      "num_tokens": 35504468.0,
      "step": 17105
    },
    {
      "entropy": 5.486019611358643,
      "epoch": 1.5520682148040639,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047627955784853226,
      "loss": 4.9971,
      "mean_token_accuracy": 0.20591204166412352,
      "num_tokens": 35514272.0,
      "step": 17110
    },
    {
      "entropy": 5.4798156261444095,
      "epoch": 1.552521770682148,
      "grad_norm": 1.015625,
      "learning_rate": 0.000476265095158215,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20031564831733703,
      "num_tokens": 35524110.0,
      "step": 17115
    },
    {
      "entropy": 5.497633266448974,
      "epoch": 1.5529753265602322,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004762506283056843,
      "loss": 4.914,
      "mean_token_accuracy": 0.21007648408412932,
      "num_tokens": 35535504.0,
      "step": 17120
    },
    {
      "entropy": 5.5904745101928714,
      "epoch": 1.5534288824383164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004762361572912393,
      "loss": 5.117,
      "mean_token_accuracy": 0.19613465964794158,
      "num_tokens": 35544885.0,
      "step": 17125
    },
    {
      "entropy": 5.519411087036133,
      "epoch": 1.5538824383164007,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047622168211517926,
      "loss": 5.1224,
      "mean_token_accuracy": 0.2027273118495941,
      "num_tokens": 35555485.0,
      "step": 17130
    },
    {
      "entropy": 5.4349846839904785,
      "epoch": 1.5543359941944848,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047620720277780353,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2088460147380829,
      "num_tokens": 35567197.0,
      "step": 17135
    },
    {
      "entropy": 5.523997163772583,
      "epoch": 1.5547895500725688,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047619271927941165,
      "loss": 4.9829,
      "mean_token_accuracy": 0.2054537519812584,
      "num_tokens": 35577576.0,
      "step": 17140
    },
    {
      "entropy": 5.511078262329102,
      "epoch": 1.555243105950653,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000476178231620303,
      "loss": 5.0402,
      "mean_token_accuracy": 0.2045543149113655,
      "num_tokens": 35588012.0,
      "step": 17145
    },
    {
      "entropy": 5.454594993591309,
      "epoch": 1.5556966618287373,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004761637398007774,
      "loss": 5.0229,
      "mean_token_accuracy": 0.2018841803073883,
      "num_tokens": 35598076.0,
      "step": 17150
    },
    {
      "entropy": 5.394225978851319,
      "epoch": 1.5561502177068216,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004761492438211342,
      "loss": 4.8495,
      "mean_token_accuracy": 0.21599810868501662,
      "num_tokens": 35607170.0,
      "step": 17155
    },
    {
      "entropy": 5.468464374542236,
      "epoch": 1.5566037735849056,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047613474368167355,
      "loss": 4.9872,
      "mean_token_accuracy": 0.19680835902690888,
      "num_tokens": 35617865.0,
      "step": 17160
    },
    {
      "entropy": 5.472936391830444,
      "epoch": 1.5570573294629897,
      "grad_norm": 1.078125,
      "learning_rate": 0.000476120239382695,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20120879709720613,
      "num_tokens": 35629329.0,
      "step": 17165
    },
    {
      "entropy": 5.4649515628814695,
      "epoch": 1.557510885341074,
      "grad_norm": 1.0,
      "learning_rate": 0.00047610573092449877,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20129372775554658,
      "num_tokens": 35640044.0,
      "step": 17170
    },
    {
      "entropy": 5.522956418991089,
      "epoch": 1.5579644412191582,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004760912183073847,
      "loss": 5.0782,
      "mean_token_accuracy": 0.20332991182804108,
      "num_tokens": 35650435.0,
      "step": 17175
    },
    {
      "entropy": 5.491788578033447,
      "epoch": 1.5584179970972425,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000476076701531653,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20425090044736863,
      "num_tokens": 35659613.0,
      "step": 17180
    },
    {
      "entropy": 5.427504205703736,
      "epoch": 1.5588715529753265,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004760621805976039,
      "loss": 4.9856,
      "mean_token_accuracy": 0.20563696473836898,
      "num_tokens": 35670987.0,
      "step": 17185
    },
    {
      "entropy": 5.478841972351074,
      "epoch": 1.5593251088534106,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047604765550553766,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2048560231924057,
      "num_tokens": 35682505.0,
      "step": 17190
    },
    {
      "entropy": 5.522268867492675,
      "epoch": 1.5597786647314948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047603312625575466,
      "loss": 4.9462,
      "mean_token_accuracy": 0.20475015491247178,
      "num_tokens": 35693056.0,
      "step": 17195
    },
    {
      "entropy": 5.4625657081604,
      "epoch": 1.5602322206095791,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047601859284855544,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21257547438144683,
      "num_tokens": 35703659.0,
      "step": 17200
    },
    {
      "entropy": 5.438641500473023,
      "epoch": 1.5606857764876634,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004760040552842403,
      "loss": 5.0019,
      "mean_token_accuracy": 0.19803812056779863,
      "num_tokens": 35714367.0,
      "step": 17205
    },
    {
      "entropy": 5.425652980804443,
      "epoch": 1.5611393323657474,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047598951356311035,
      "loss": 4.9711,
      "mean_token_accuracy": 0.1990872249007225,
      "num_tokens": 35724773.0,
      "step": 17210
    },
    {
      "entropy": 5.519612550735474,
      "epoch": 1.5615928882438317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004759749676854658,
      "loss": 5.1162,
      "mean_token_accuracy": 0.20020296722650527,
      "num_tokens": 35735622.0,
      "step": 17215
    },
    {
      "entropy": 5.520531034469604,
      "epoch": 1.5620464441219157,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004759604176516079,
      "loss": 4.9725,
      "mean_token_accuracy": 0.2098139926791191,
      "num_tokens": 35744967.0,
      "step": 17220
    },
    {
      "entropy": 5.531719493865967,
      "epoch": 1.5625,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004759458634618372,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20392842590808868,
      "num_tokens": 35755647.0,
      "step": 17225
    },
    {
      "entropy": 5.503164672851563,
      "epoch": 1.5629535558780843,
      "grad_norm": 1.0,
      "learning_rate": 0.0004759313051164549,
      "loss": 5.0455,
      "mean_token_accuracy": 0.2000377207994461,
      "num_tokens": 35765595.0,
      "step": 17230
    },
    {
      "entropy": 5.501796817779541,
      "epoch": 1.5634071117561683,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000475916742615762,
      "loss": 5.0198,
      "mean_token_accuracy": 0.21368214190006257,
      "num_tokens": 35775804.0,
      "step": 17235
    },
    {
      "entropy": 5.46679916381836,
      "epoch": 1.5638606676342526,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004759021759600597,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2093169242143631,
      "num_tokens": 35786227.0,
      "step": 17240
    },
    {
      "entropy": 5.532622194290161,
      "epoch": 1.5643142235123366,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004758876051496492,
      "loss": 5.0919,
      "mean_token_accuracy": 0.20590397715568542,
      "num_tokens": 35797511.0,
      "step": 17245
    },
    {
      "entropy": 5.5053972721099855,
      "epoch": 1.5647677793904209,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047587303018483177,
      "loss": 5.0289,
      "mean_token_accuracy": 0.20084312409162522,
      "num_tokens": 35807248.0,
      "step": 17250
    },
    {
      "entropy": 5.453809404373169,
      "epoch": 1.5652213352685052,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047585845106590897,
      "loss": 5.0417,
      "mean_token_accuracy": 0.19742046445608138,
      "num_tokens": 35818164.0,
      "step": 17255
    },
    {
      "entropy": 5.467202234268188,
      "epoch": 1.5656748911465894,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047584386779318214,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2104337528347969,
      "num_tokens": 35828746.0,
      "step": 17260
    },
    {
      "entropy": 5.519303274154663,
      "epoch": 1.5661284470246735,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000475829280366953,
      "loss": 5.0675,
      "mean_token_accuracy": 0.19842927008867264,
      "num_tokens": 35839325.0,
      "step": 17265
    },
    {
      "entropy": 5.502063179016114,
      "epoch": 1.5665820029027575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047581468878752313,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20751725137233734,
      "num_tokens": 35849406.0,
      "step": 17270
    },
    {
      "entropy": 5.426310443878174,
      "epoch": 1.5670355587808418,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047580009305519427,
      "loss": 4.9205,
      "mean_token_accuracy": 0.20804960429668426,
      "num_tokens": 35858715.0,
      "step": 17275
    },
    {
      "entropy": 5.442573881149292,
      "epoch": 1.567489114658926,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004757854931702684,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20239466428756714,
      "num_tokens": 35869063.0,
      "step": 17280
    },
    {
      "entropy": 5.444556140899659,
      "epoch": 1.5679426705370103,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047577088913304733,
      "loss": 4.8804,
      "mean_token_accuracy": 0.20901139378547667,
      "num_tokens": 35878428.0,
      "step": 17285
    },
    {
      "entropy": 5.516884994506836,
      "epoch": 1.5683962264150944,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047575628094383317,
      "loss": 5.0248,
      "mean_token_accuracy": 0.20472852289676666,
      "num_tokens": 35888801.0,
      "step": 17290
    },
    {
      "entropy": 5.444619464874267,
      "epoch": 1.5688497822931784,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004757416686029279,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20317072570323944,
      "num_tokens": 35899381.0,
      "step": 17295
    },
    {
      "entropy": 5.528191423416137,
      "epoch": 1.5693033381712627,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004757270521106338,
      "loss": 5.0494,
      "mean_token_accuracy": 0.20813838094472886,
      "num_tokens": 35909546.0,
      "step": 17300
    },
    {
      "entropy": 5.501978206634521,
      "epoch": 1.569756894049347,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004757124314672531,
      "loss": 5.0418,
      "mean_token_accuracy": 0.2017609715461731,
      "num_tokens": 35919301.0,
      "step": 17305
    },
    {
      "entropy": 5.591321849822998,
      "epoch": 1.5702104499274312,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047569780667308816,
      "loss": 5.0762,
      "mean_token_accuracy": 0.1994296759366989,
      "num_tokens": 35928603.0,
      "step": 17310
    },
    {
      "entropy": 5.540690517425537,
      "epoch": 1.5706640058055152,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047568317772844147,
      "loss": 5.0568,
      "mean_token_accuracy": 0.20576718896627427,
      "num_tokens": 35939005.0,
      "step": 17315
    },
    {
      "entropy": 5.538856363296508,
      "epoch": 1.5711175616835993,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047566854463361544,
      "loss": 5.015,
      "mean_token_accuracy": 0.21285732537508012,
      "num_tokens": 35948471.0,
      "step": 17320
    },
    {
      "entropy": 5.437993288040161,
      "epoch": 1.5715711175616836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004756539073889128,
      "loss": 5.0453,
      "mean_token_accuracy": 0.2030189663171768,
      "num_tokens": 35958488.0,
      "step": 17325
    },
    {
      "entropy": 5.48625545501709,
      "epoch": 1.5720246734397678,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047563926599463623,
      "loss": 5.0906,
      "mean_token_accuracy": 0.19767525047063828,
      "num_tokens": 35968291.0,
      "step": 17330
    },
    {
      "entropy": 5.546093368530274,
      "epoch": 1.572478229317852,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004756246204510886,
      "loss": 5.0116,
      "mean_token_accuracy": 0.20795842111110688,
      "num_tokens": 35978565.0,
      "step": 17335
    },
    {
      "entropy": 5.52994179725647,
      "epoch": 1.5729317851959361,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047560997075857253,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20825816094875335,
      "num_tokens": 35987483.0,
      "step": 17340
    },
    {
      "entropy": 5.476396131515503,
      "epoch": 1.5733853410740202,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004755953169173913,
      "loss": 5.0499,
      "mean_token_accuracy": 0.20181325376033782,
      "num_tokens": 35998355.0,
      "step": 17345
    },
    {
      "entropy": 5.467070770263672,
      "epoch": 1.5738388969521044,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004755806589278477,
      "loss": 4.9872,
      "mean_token_accuracy": 0.21285985708236693,
      "num_tokens": 36008365.0,
      "step": 17350
    },
    {
      "entropy": 5.51667366027832,
      "epoch": 1.5742924528301887,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047556599679024497,
      "loss": 5.0824,
      "mean_token_accuracy": 0.19374739676713942,
      "num_tokens": 36018880.0,
      "step": 17355
    },
    {
      "entropy": 5.522989702224732,
      "epoch": 1.574746008708273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047555133050488637,
      "loss": 5.0883,
      "mean_token_accuracy": 0.19408987313508988,
      "num_tokens": 36029335.0,
      "step": 17360
    },
    {
      "entropy": 5.483440351486206,
      "epoch": 1.575199564586357,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047553666007207516,
      "loss": 4.9883,
      "mean_token_accuracy": 0.2119162768125534,
      "num_tokens": 36039774.0,
      "step": 17365
    },
    {
      "entropy": 5.55950517654419,
      "epoch": 1.575653120464441,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047552198549211463,
      "loss": 5.1255,
      "mean_token_accuracy": 0.20450095236301422,
      "num_tokens": 36049864.0,
      "step": 17370
    },
    {
      "entropy": 5.49333701133728,
      "epoch": 1.5761066763425253,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004755073067653084,
      "loss": 5.0493,
      "mean_token_accuracy": 0.19705418944358827,
      "num_tokens": 36059224.0,
      "step": 17375
    },
    {
      "entropy": 5.421458721160889,
      "epoch": 1.5765602322206096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004754926238919599,
      "loss": 4.9525,
      "mean_token_accuracy": 0.211106775701046,
      "num_tokens": 36069243.0,
      "step": 17380
    },
    {
      "entropy": 5.504794073104859,
      "epoch": 1.5770137880986939,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004754779368723729,
      "loss": 5.0186,
      "mean_token_accuracy": 0.2001550853252411,
      "num_tokens": 36079833.0,
      "step": 17385
    },
    {
      "entropy": 5.4966668605804445,
      "epoch": 1.577467343976778,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004754632457068511,
      "loss": 4.9962,
      "mean_token_accuracy": 0.2046745464205742,
      "num_tokens": 36090673.0,
      "step": 17390
    },
    {
      "entropy": 5.540458869934082,
      "epoch": 1.5779208998548622,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004754485503956982,
      "loss": 5.0358,
      "mean_token_accuracy": 0.20388307571411132,
      "num_tokens": 36101719.0,
      "step": 17395
    },
    {
      "entropy": 5.503682947158813,
      "epoch": 1.5783744557329462,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004754338509392183,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20253086239099502,
      "num_tokens": 36112352.0,
      "step": 17400
    },
    {
      "entropy": 5.374897909164429,
      "epoch": 1.5788280116110305,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004754191473377153,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20746100395917894,
      "num_tokens": 36121797.0,
      "step": 17405
    },
    {
      "entropy": 5.484429883956909,
      "epoch": 1.5792815674891147,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047540443959149316,
      "loss": 5.0087,
      "mean_token_accuracy": 0.21143339276313783,
      "num_tokens": 36132094.0,
      "step": 17410
    },
    {
      "entropy": 5.499725246429444,
      "epoch": 1.5797351233671988,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004753897277008562,
      "loss": 4.9994,
      "mean_token_accuracy": 0.203934970498085,
      "num_tokens": 36142860.0,
      "step": 17415
    },
    {
      "entropy": 5.508707189559937,
      "epoch": 1.580188679245283,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004753750116661086,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20103585422039033,
      "num_tokens": 36154555.0,
      "step": 17420
    },
    {
      "entropy": 5.576639652252197,
      "epoch": 1.580642235123367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047536029148755477,
      "loss": 5.1314,
      "mean_token_accuracy": 0.19167136400938034,
      "num_tokens": 36165479.0,
      "step": 17425
    },
    {
      "entropy": 5.5136706829071045,
      "epoch": 1.5810957910014514,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047534556716549893,
      "loss": 5.0417,
      "mean_token_accuracy": 0.2023715317249298,
      "num_tokens": 36175873.0,
      "step": 17430
    },
    {
      "entropy": 5.461419534683228,
      "epoch": 1.5815493468795356,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047533083870024575,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2108200564980507,
      "num_tokens": 36186792.0,
      "step": 17435
    },
    {
      "entropy": 5.6181721687316895,
      "epoch": 1.58200290275762,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004753161060920998,
      "loss": 5.1257,
      "mean_token_accuracy": 0.19894260317087173,
      "num_tokens": 36196691.0,
      "step": 17440
    },
    {
      "entropy": 5.569988870620728,
      "epoch": 1.582456458635704,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004753013693413657,
      "loss": 5.0235,
      "mean_token_accuracy": 0.2052525535225868,
      "num_tokens": 36207173.0,
      "step": 17445
    },
    {
      "entropy": 5.460013198852539,
      "epoch": 1.582910014513788,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047528662844834827,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20807620137929916,
      "num_tokens": 36217912.0,
      "step": 17450
    },
    {
      "entropy": 5.411827182769775,
      "epoch": 1.5833635703918723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047527188341335233,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2129110425710678,
      "num_tokens": 36227714.0,
      "step": 17455
    },
    {
      "entropy": 5.476075458526611,
      "epoch": 1.5838171262699565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004752571342366828,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20643447786569596,
      "num_tokens": 36239114.0,
      "step": 17460
    },
    {
      "entropy": 5.481729888916016,
      "epoch": 1.5842706821480408,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004752423809186446,
      "loss": 5.0203,
      "mean_token_accuracy": 0.20117921978235245,
      "num_tokens": 36249436.0,
      "step": 17465
    },
    {
      "entropy": 5.498609352111816,
      "epoch": 1.5847242380261248,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047522762345954305,
      "loss": 4.9713,
      "mean_token_accuracy": 0.2044916108250618,
      "num_tokens": 36259541.0,
      "step": 17470
    },
    {
      "entropy": 5.511799478530884,
      "epoch": 1.5851777939042089,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004752128618596832,
      "loss": 5.0275,
      "mean_token_accuracy": 0.19812653213739395,
      "num_tokens": 36270689.0,
      "step": 17475
    },
    {
      "entropy": 5.559856700897217,
      "epoch": 1.5856313497822931,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047519809611937036,
      "loss": 5.0809,
      "mean_token_accuracy": 0.20168314427137374,
      "num_tokens": 36281191.0,
      "step": 17480
    },
    {
      "entropy": 5.437845468521118,
      "epoch": 1.5860849056603774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004751833262389098,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21071845591068267,
      "num_tokens": 36291022.0,
      "step": 17485
    },
    {
      "entropy": 5.511733102798462,
      "epoch": 1.5865384615384617,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047516855221860715,
      "loss": 5.042,
      "mean_token_accuracy": 0.20490602403879166,
      "num_tokens": 36301237.0,
      "step": 17490
    },
    {
      "entropy": 5.544945573806762,
      "epoch": 1.5869920174165457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047515377405876774,
      "loss": 5.0164,
      "mean_token_accuracy": 0.19902748465538025,
      "num_tokens": 36311792.0,
      "step": 17495
    },
    {
      "entropy": 5.4585517883300785,
      "epoch": 1.5874455732946298,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004751389917596973,
      "loss": 4.988,
      "mean_token_accuracy": 0.2170154094696045,
      "num_tokens": 36321847.0,
      "step": 17500
    },
    {
      "entropy": 5.476059007644653,
      "epoch": 1.587899129172714,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047512420532170154,
      "loss": 5.0971,
      "mean_token_accuracy": 0.19473932832479476,
      "num_tokens": 36332735.0,
      "step": 17505
    },
    {
      "entropy": 5.569549465179444,
      "epoch": 1.5883526850507983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004751094147450862,
      "loss": 5.0814,
      "mean_token_accuracy": 0.20146389156579972,
      "num_tokens": 36343149.0,
      "step": 17510
    },
    {
      "entropy": 5.555501079559326,
      "epoch": 1.5888062409288826,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004750946200301571,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2055591121315956,
      "num_tokens": 36353081.0,
      "step": 17515
    },
    {
      "entropy": 5.50593376159668,
      "epoch": 1.5892597968069666,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047507982117722035,
      "loss": 5.0546,
      "mean_token_accuracy": 0.201623272895813,
      "num_tokens": 36363624.0,
      "step": 17520
    },
    {
      "entropy": 5.541685676574707,
      "epoch": 1.5897133526850507,
      "grad_norm": 1.03125,
      "learning_rate": 0.000475065018186582,
      "loss": 5.0761,
      "mean_token_accuracy": 0.19994409531354904,
      "num_tokens": 36374081.0,
      "step": 17525
    },
    {
      "entropy": 5.53523817062378,
      "epoch": 1.590166908563135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047505021105854794,
      "loss": 5.1406,
      "mean_token_accuracy": 0.1978500485420227,
      "num_tokens": 36384360.0,
      "step": 17530
    },
    {
      "entropy": 5.481768226623535,
      "epoch": 1.5906204644412192,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047503539979342456,
      "loss": 5.0594,
      "mean_token_accuracy": 0.2044894129037857,
      "num_tokens": 36395924.0,
      "step": 17535
    },
    {
      "entropy": 5.445667171478272,
      "epoch": 1.5910740203193035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004750205843915182,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21521376967430114,
      "num_tokens": 36405246.0,
      "step": 17540
    },
    {
      "entropy": 5.48762264251709,
      "epoch": 1.5915275761973875,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004750057648531351,
      "loss": 4.9761,
      "mean_token_accuracy": 0.2125127375125885,
      "num_tokens": 36414964.0,
      "step": 17545
    },
    {
      "entropy": 5.468229579925537,
      "epoch": 1.5919811320754715,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004749909411785819,
      "loss": 5.052,
      "mean_token_accuracy": 0.19814764261245726,
      "num_tokens": 36425011.0,
      "step": 17550
    },
    {
      "entropy": 5.423689365386963,
      "epoch": 1.5924346879535558,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000474976113368165,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21135228127241135,
      "num_tokens": 36434734.0,
      "step": 17555
    },
    {
      "entropy": 5.501957988739013,
      "epoch": 1.59288824383164,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004749612814221912,
      "loss": 5.038,
      "mean_token_accuracy": 0.20483655035495757,
      "num_tokens": 36444748.0,
      "step": 17560
    },
    {
      "entropy": 5.588495826721191,
      "epoch": 1.5933417997097243,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047494644534096703,
      "loss": 5.0479,
      "mean_token_accuracy": 0.19796278774738313,
      "num_tokens": 36455076.0,
      "step": 17565
    },
    {
      "entropy": 5.512688446044922,
      "epoch": 1.5937953555878084,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004749316051247995,
      "loss": 4.9676,
      "mean_token_accuracy": 0.2056950405240059,
      "num_tokens": 36465302.0,
      "step": 17570
    },
    {
      "entropy": 5.484334278106689,
      "epoch": 1.5942489114658926,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047491676077399543,
      "loss": 5.0783,
      "mean_token_accuracy": 0.20078891515731812,
      "num_tokens": 36476666.0,
      "step": 17575
    },
    {
      "entropy": 5.452722549438477,
      "epoch": 1.5947024673439767,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004749019122888618,
      "loss": 4.9335,
      "mean_token_accuracy": 0.208494570851326,
      "num_tokens": 36487379.0,
      "step": 17580
    },
    {
      "entropy": 5.5092161178588865,
      "epoch": 1.595156023222061,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047488705966970565,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21083510369062425,
      "num_tokens": 36497897.0,
      "step": 17585
    },
    {
      "entropy": 5.509284782409668,
      "epoch": 1.5956095791001452,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004748722029168342,
      "loss": 5.0687,
      "mean_token_accuracy": 0.20730422884225846,
      "num_tokens": 36508528.0,
      "step": 17590
    },
    {
      "entropy": 5.514298009872436,
      "epoch": 1.5960631349782293,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004748573420305546,
      "loss": 5.0609,
      "mean_token_accuracy": 0.19901353120803833,
      "num_tokens": 36519196.0,
      "step": 17595
    },
    {
      "entropy": 5.472837448120117,
      "epoch": 1.5965166908563135,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047484247701117427,
      "loss": 5.0332,
      "mean_token_accuracy": 0.2019036278128624,
      "num_tokens": 36529871.0,
      "step": 17600
    },
    {
      "entropy": 5.600695705413818,
      "epoch": 1.5969702467343976,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004748276078590006,
      "loss": 5.1166,
      "mean_token_accuracy": 0.1906803548336029,
      "num_tokens": 36541188.0,
      "step": 17605
    },
    {
      "entropy": 5.5623010158538815,
      "epoch": 1.5974238026124818,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004748127345743411,
      "loss": 5.0861,
      "mean_token_accuracy": 0.1991748109459877,
      "num_tokens": 36551968.0,
      "step": 17610
    },
    {
      "entropy": 5.518520832061768,
      "epoch": 1.5978773584905661,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004747978571575033,
      "loss": 5.0224,
      "mean_token_accuracy": 0.20348055064678192,
      "num_tokens": 36562870.0,
      "step": 17615
    },
    {
      "entropy": 5.4489270687103275,
      "epoch": 1.5983309143686504,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004747829756087949,
      "loss": 4.9533,
      "mean_token_accuracy": 0.21031002700328827,
      "num_tokens": 36573295.0,
      "step": 17620
    },
    {
      "entropy": 5.445514678955078,
      "epoch": 1.5987844702467344,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047476808992852364,
      "loss": 4.9519,
      "mean_token_accuracy": 0.2091691941022873,
      "num_tokens": 36583830.0,
      "step": 17625
    },
    {
      "entropy": 5.48184494972229,
      "epoch": 1.5992380261248185,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004747532001169974,
      "loss": 5.0311,
      "mean_token_accuracy": 0.2106089934706688,
      "num_tokens": 36594052.0,
      "step": 17630
    },
    {
      "entropy": 5.509792280197144,
      "epoch": 1.5996915820029027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047473830617452415,
      "loss": 5.0609,
      "mean_token_accuracy": 0.20581653863191604,
      "num_tokens": 36604378.0,
      "step": 17635
    },
    {
      "entropy": 5.556970739364624,
      "epoch": 1.600145137880987,
      "grad_norm": 1.0,
      "learning_rate": 0.00047472340810141163,
      "loss": 5.147,
      "mean_token_accuracy": 0.19290078580379486,
      "num_tokens": 36615588.0,
      "step": 17640
    },
    {
      "entropy": 5.588248300552368,
      "epoch": 1.6005986937590713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047470850589796833,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20253516882658004,
      "num_tokens": 36626369.0,
      "step": 17645
    },
    {
      "entropy": 5.546917581558228,
      "epoch": 1.6010522496371553,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047469359956450215,
      "loss": 5.0682,
      "mean_token_accuracy": 0.20150643587112427,
      "num_tokens": 36636565.0,
      "step": 17650
    },
    {
      "entropy": 5.577430868148804,
      "epoch": 1.6015058055152394,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047467868910132144,
      "loss": 5.1484,
      "mean_token_accuracy": 0.1941498950123787,
      "num_tokens": 36647721.0,
      "step": 17655
    },
    {
      "entropy": 5.435346221923828,
      "epoch": 1.6019593613933236,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004746637745087346,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2122989535331726,
      "num_tokens": 36657887.0,
      "step": 17660
    },
    {
      "entropy": 5.555538558959961,
      "epoch": 1.6024129172714079,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047464885578705,
      "loss": 5.0351,
      "mean_token_accuracy": 0.20261109620332718,
      "num_tokens": 36667918.0,
      "step": 17665
    },
    {
      "entropy": 5.448113107681275,
      "epoch": 1.6028664731494922,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047463393293657616,
      "loss": 4.9891,
      "mean_token_accuracy": 0.21008675992488862,
      "num_tokens": 36678491.0,
      "step": 17670
    },
    {
      "entropy": 5.493792295455933,
      "epoch": 1.6033200290275762,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004746190059576217,
      "loss": 5.052,
      "mean_token_accuracy": 0.20171189904212952,
      "num_tokens": 36688638.0,
      "step": 17675
    },
    {
      "entropy": 5.520508289337158,
      "epoch": 1.6037735849056602,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004746040748504954,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20498359203338623,
      "num_tokens": 36699344.0,
      "step": 17680
    },
    {
      "entropy": 5.533679056167602,
      "epoch": 1.6042271407837445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047458913961550593,
      "loss": 4.9993,
      "mean_token_accuracy": 0.20418418049812317,
      "num_tokens": 36709606.0,
      "step": 17685
    },
    {
      "entropy": 5.444143676757813,
      "epoch": 1.6046806966618288,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004745742002529622,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20646962225437165,
      "num_tokens": 36718740.0,
      "step": 17690
    },
    {
      "entropy": 5.486650180816651,
      "epoch": 1.605134252539913,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004745592567631731,
      "loss": 5.0395,
      "mean_token_accuracy": 0.19889770448207855,
      "num_tokens": 36729960.0,
      "step": 17695
    },
    {
      "entropy": 5.449067878723144,
      "epoch": 1.605587808417997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047454430914644773,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21313123553991317,
      "num_tokens": 36740221.0,
      "step": 17700
    },
    {
      "entropy": 5.396981620788575,
      "epoch": 1.6060413642960811,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004745293574030952,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21185587346553802,
      "num_tokens": 36750841.0,
      "step": 17705
    },
    {
      "entropy": 5.435892915725708,
      "epoch": 1.6064949201741654,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047451440153342477,
      "loss": 5.026,
      "mean_token_accuracy": 0.19739796966314316,
      "num_tokens": 36760944.0,
      "step": 17710
    },
    {
      "entropy": 5.519886016845703,
      "epoch": 1.6069484760522497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004744994415377457,
      "loss": 5.0015,
      "mean_token_accuracy": 0.207135434448719,
      "num_tokens": 36770922.0,
      "step": 17715
    },
    {
      "entropy": 5.558530521392822,
      "epoch": 1.607402031930334,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047448447741636724,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20146499425172806,
      "num_tokens": 36780702.0,
      "step": 17720
    },
    {
      "entropy": 5.543065357208252,
      "epoch": 1.607855587808418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047446950916959905,
      "loss": 5.08,
      "mean_token_accuracy": 0.19724457412958146,
      "num_tokens": 36791868.0,
      "step": 17725
    },
    {
      "entropy": 5.539865970611572,
      "epoch": 1.608309143686502,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004744545367977505,
      "loss": 5.0413,
      "mean_token_accuracy": 0.20392582416534424,
      "num_tokens": 36802330.0,
      "step": 17730
    },
    {
      "entropy": 5.506350994110107,
      "epoch": 1.6087626995645863,
      "grad_norm": 1.125,
      "learning_rate": 0.00047443956030113134,
      "loss": 5.0658,
      "mean_token_accuracy": 0.20375836491584778,
      "num_tokens": 36812936.0,
      "step": 17735
    },
    {
      "entropy": 5.549493503570557,
      "epoch": 1.6092162554426706,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047442457968005123,
      "loss": 5.0391,
      "mean_token_accuracy": 0.20139739960432052,
      "num_tokens": 36823123.0,
      "step": 17740
    },
    {
      "entropy": 5.447511148452759,
      "epoch": 1.6096698113207548,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047440959493481996,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21495402604341507,
      "num_tokens": 36833050.0,
      "step": 17745
    },
    {
      "entropy": 5.500833415985108,
      "epoch": 1.6101233671988389,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047439460606574745,
      "loss": 5.0448,
      "mean_token_accuracy": 0.20559147894382476,
      "num_tokens": 36842520.0,
      "step": 17750
    },
    {
      "entropy": 5.4890522956848145,
      "epoch": 1.6105769230769231,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004743796130731437,
      "loss": 5.0053,
      "mean_token_accuracy": 0.19978885352611542,
      "num_tokens": 36853024.0,
      "step": 17755
    },
    {
      "entropy": 5.503660249710083,
      "epoch": 1.6110304789550072,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004743646159573188,
      "loss": 4.9619,
      "mean_token_accuracy": 0.21103161871433257,
      "num_tokens": 36862534.0,
      "step": 17760
    },
    {
      "entropy": 5.4798894882202145,
      "epoch": 1.6114840348330914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004743496147185828,
      "loss": 4.9568,
      "mean_token_accuracy": 0.21533622294664384,
      "num_tokens": 36872356.0,
      "step": 17765
    },
    {
      "entropy": 5.476791620254517,
      "epoch": 1.6119375907111757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047433460935724595,
      "loss": 5.009,
      "mean_token_accuracy": 0.2075714036822319,
      "num_tokens": 36882643.0,
      "step": 17770
    },
    {
      "entropy": 5.501908922195435,
      "epoch": 1.6123911465892597,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004743195998736186,
      "loss": 5.0018,
      "mean_token_accuracy": 0.2041184648871422,
      "num_tokens": 36893349.0,
      "step": 17775
    },
    {
      "entropy": 5.556226587295532,
      "epoch": 1.612844702467344,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047430458626801107,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20778820514678956,
      "num_tokens": 36903713.0,
      "step": 17780
    },
    {
      "entropy": 5.4703563213348385,
      "epoch": 1.613298258345428,
      "grad_norm": 1.125,
      "learning_rate": 0.000474289568540734,
      "loss": 4.932,
      "mean_token_accuracy": 0.20787305682897567,
      "num_tokens": 36912999.0,
      "step": 17785
    },
    {
      "entropy": 5.516793489456177,
      "epoch": 1.6137518142235123,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047427454669209785,
      "loss": 5.0537,
      "mean_token_accuracy": 0.20724982619285584,
      "num_tokens": 36923598.0,
      "step": 17790
    },
    {
      "entropy": 5.459460878372193,
      "epoch": 1.6142053701015966,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047425952072241326,
      "loss": 4.972,
      "mean_token_accuracy": 0.20285503268241883,
      "num_tokens": 36933036.0,
      "step": 17795
    },
    {
      "entropy": 5.49089207649231,
      "epoch": 1.6146589259796809,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000474244490631991,
      "loss": 5.0686,
      "mean_token_accuracy": 0.19884790927171708,
      "num_tokens": 36943137.0,
      "step": 17800
    },
    {
      "entropy": 5.613918256759644,
      "epoch": 1.615112481857765,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004742294564211419,
      "loss": 5.1893,
      "mean_token_accuracy": 0.19501156806945802,
      "num_tokens": 36955619.0,
      "step": 17805
    },
    {
      "entropy": 5.510789442062378,
      "epoch": 1.615566037735849,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004742144180901769,
      "loss": 4.9562,
      "mean_token_accuracy": 0.21337293237447738,
      "num_tokens": 36966377.0,
      "step": 17810
    },
    {
      "entropy": 5.531083250045777,
      "epoch": 1.6160195936139332,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047419937563940697,
      "loss": 5.021,
      "mean_token_accuracy": 0.2078086480498314,
      "num_tokens": 36977100.0,
      "step": 17815
    },
    {
      "entropy": 5.509808015823364,
      "epoch": 1.6164731494920175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047418432906914314,
      "loss": 5.0857,
      "mean_token_accuracy": 0.20011530071496964,
      "num_tokens": 36987120.0,
      "step": 17820
    },
    {
      "entropy": 5.542492389678955,
      "epoch": 1.6169267053701017,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004741692783796967,
      "loss": 5.0392,
      "mean_token_accuracy": 0.20618803203105926,
      "num_tokens": 36996885.0,
      "step": 17825
    },
    {
      "entropy": 5.513024425506591,
      "epoch": 1.6173802612481858,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047415422357137876,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21342350989580156,
      "num_tokens": 37006570.0,
      "step": 17830
    },
    {
      "entropy": 5.553627634048462,
      "epoch": 1.6178338171262698,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004741391646445008,
      "loss": 5.0692,
      "mean_token_accuracy": 0.19652530997991563,
      "num_tokens": 37017185.0,
      "step": 17835
    },
    {
      "entropy": 5.521615839004516,
      "epoch": 1.618287373004354,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004741241015993741,
      "loss": 5.0276,
      "mean_token_accuracy": 0.20495287030935289,
      "num_tokens": 37028386.0,
      "step": 17840
    },
    {
      "entropy": 5.459984064102173,
      "epoch": 1.6187409288824384,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047410903443631026,
      "loss": 4.9388,
      "mean_token_accuracy": 0.20692460536956786,
      "num_tokens": 37038326.0,
      "step": 17845
    },
    {
      "entropy": 5.5151317596435545,
      "epoch": 1.6191944847605226,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004740939631556209,
      "loss": 5.0365,
      "mean_token_accuracy": 0.20376701205968856,
      "num_tokens": 37048570.0,
      "step": 17850
    },
    {
      "entropy": 5.456529808044434,
      "epoch": 1.6196480406386067,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004740788877576176,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2112232819199562,
      "num_tokens": 37058909.0,
      "step": 17855
    },
    {
      "entropy": 5.495297050476074,
      "epoch": 1.6201015965166907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004740638082426121,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21210072934627533,
      "num_tokens": 37068957.0,
      "step": 17860
    },
    {
      "entropy": 5.490260887145996,
      "epoch": 1.620555152394775,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047404872461091643,
      "loss": 4.9561,
      "mean_token_accuracy": 0.20613258481025695,
      "num_tokens": 37080187.0,
      "step": 17865
    },
    {
      "entropy": 5.507491159439087,
      "epoch": 1.6210087082728593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047403363686284236,
      "loss": 5.0318,
      "mean_token_accuracy": 0.20411915332078934,
      "num_tokens": 37090314.0,
      "step": 17870
    },
    {
      "entropy": 5.543750429153443,
      "epoch": 1.6214622641509435,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047401854499870197,
      "loss": 5.074,
      "mean_token_accuracy": 0.20306196957826614,
      "num_tokens": 37099907.0,
      "step": 17875
    },
    {
      "entropy": 5.505069637298584,
      "epoch": 1.6219158200290276,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047400344901880735,
      "loss": 5.0854,
      "mean_token_accuracy": 0.19933670312166213,
      "num_tokens": 37110495.0,
      "step": 17880
    },
    {
      "entropy": 5.524602603912354,
      "epoch": 1.6223693759071116,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004739883489234707,
      "loss": 5.0635,
      "mean_token_accuracy": 0.20316663831472398,
      "num_tokens": 37121957.0,
      "step": 17885
    },
    {
      "entropy": 5.5381800651550295,
      "epoch": 1.6228229317851959,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047397324471300424,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2094401389360428,
      "num_tokens": 37132743.0,
      "step": 17890
    },
    {
      "entropy": 5.520811605453491,
      "epoch": 1.6232764876632801,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047395813638772047,
      "loss": 5.0649,
      "mean_token_accuracy": 0.20002538412809373,
      "num_tokens": 37143947.0,
      "step": 17895
    },
    {
      "entropy": 5.606742572784424,
      "epoch": 1.6237300435413644,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047394302394793164,
      "loss": 5.1701,
      "mean_token_accuracy": 0.1896388277411461,
      "num_tokens": 37154316.0,
      "step": 17900
    },
    {
      "entropy": 5.546631574630737,
      "epoch": 1.6241835994194485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004739279073939504,
      "loss": 5.0659,
      "mean_token_accuracy": 0.19914041161537172,
      "num_tokens": 37164999.0,
      "step": 17905
    },
    {
      "entropy": 5.53722620010376,
      "epoch": 1.6246371552975325,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004739127867260894,
      "loss": 5.0524,
      "mean_token_accuracy": 0.2010467141866684,
      "num_tokens": 37174392.0,
      "step": 17910
    },
    {
      "entropy": 5.467570638656616,
      "epoch": 1.6250907111756168,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004738976619446613,
      "loss": 5.0475,
      "mean_token_accuracy": 0.20052966475486755,
      "num_tokens": 37185341.0,
      "step": 17915
    },
    {
      "entropy": 5.5671929836273195,
      "epoch": 1.625544267053701,
      "grad_norm": 0.890625,
      "learning_rate": 0.00047388253304997874,
      "loss": 5.0515,
      "mean_token_accuracy": 0.19656094610691072,
      "num_tokens": 37196725.0,
      "step": 17920
    },
    {
      "entropy": 5.5232336044311525,
      "epoch": 1.6259978229317853,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004738674000423548,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2106102377176285,
      "num_tokens": 37206877.0,
      "step": 17925
    },
    {
      "entropy": 5.40243763923645,
      "epoch": 1.6264513788098693,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004738522629221023,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2058248147368431,
      "num_tokens": 37217277.0,
      "step": 17930
    },
    {
      "entropy": 5.4153111457824705,
      "epoch": 1.6269049346879536,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004738371216895344,
      "loss": 4.925,
      "mean_token_accuracy": 0.21081706881523132,
      "num_tokens": 37227716.0,
      "step": 17935
    },
    {
      "entropy": 5.469329214096069,
      "epoch": 1.6273584905660377,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004738219763449641,
      "loss": 5.034,
      "mean_token_accuracy": 0.20435808897018432,
      "num_tokens": 37238178.0,
      "step": 17940
    },
    {
      "entropy": 5.489963245391846,
      "epoch": 1.627812046444122,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004738068268887047,
      "loss": 5.07,
      "mean_token_accuracy": 0.20421854108572007,
      "num_tokens": 37248408.0,
      "step": 17945
    },
    {
      "entropy": 5.5134241580963135,
      "epoch": 1.6282656023222062,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004737916733210694,
      "loss": 5.1703,
      "mean_token_accuracy": 0.1944072127342224,
      "num_tokens": 37259176.0,
      "step": 17950
    },
    {
      "entropy": 5.532334995269776,
      "epoch": 1.6287191582002902,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047377651564237165,
      "loss": 4.9234,
      "mean_token_accuracy": 0.2068418338894844,
      "num_tokens": 37269205.0,
      "step": 17955
    },
    {
      "entropy": 5.529959297180175,
      "epoch": 1.6291727140783745,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047376135385292494,
      "loss": 5.0478,
      "mean_token_accuracy": 0.20907308608293534,
      "num_tokens": 37278841.0,
      "step": 17960
    },
    {
      "entropy": 5.490876150131226,
      "epoch": 1.6296262699564585,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004737461879530428,
      "loss": 5.0788,
      "mean_token_accuracy": 0.2029963567852974,
      "num_tokens": 37288487.0,
      "step": 17965
    },
    {
      "entropy": 5.548338794708252,
      "epoch": 1.6300798258345428,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047373101794303884,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20348800867795944,
      "num_tokens": 37300086.0,
      "step": 17970
    },
    {
      "entropy": 5.528116893768311,
      "epoch": 1.630533381712627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004737158438232267,
      "loss": 5.0896,
      "mean_token_accuracy": 0.19577748477458953,
      "num_tokens": 37310072.0,
      "step": 17975
    },
    {
      "entropy": 5.51202883720398,
      "epoch": 1.6309869375907113,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047370066559392033,
      "loss": 5.0106,
      "mean_token_accuracy": 0.21039992719888687,
      "num_tokens": 37320278.0,
      "step": 17980
    },
    {
      "entropy": 5.513598346710205,
      "epoch": 1.6314404934687954,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047368548325543366,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2022240102291107,
      "num_tokens": 37330741.0,
      "step": 17985
    },
    {
      "entropy": 5.536466026306153,
      "epoch": 1.6318940493468794,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004736702968080804,
      "loss": 5.0565,
      "mean_token_accuracy": 0.1967114433646202,
      "num_tokens": 37341213.0,
      "step": 17990
    },
    {
      "entropy": 5.467316484451294,
      "epoch": 1.6323476052249637,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047365510625217483,
      "loss": 4.923,
      "mean_token_accuracy": 0.20637372881174088,
      "num_tokens": 37350528.0,
      "step": 17995
    },
    {
      "entropy": 5.476673173904419,
      "epoch": 1.632801161103048,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047363991158803117,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20116400420665742,
      "num_tokens": 37359938.0,
      "step": 18000
    },
    {
      "epoch": 1.632801161103048,
      "eval_entropy": 5.256809419744155,
      "eval_loss": 5.060100555419922,
      "eval_mean_token_accuracy": 0.21065192328706991,
      "eval_num_tokens": 37359938.0,
      "eval_runtime": 20.499,
      "eval_samples_per_second": 1724.964,
      "eval_steps_per_second": 215.62,
      "step": 18000
    },
    {
      "entropy": 5.511171817779541,
      "epoch": 1.6332547169811322,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047362471281596334,
      "loss": 5.0259,
      "mean_token_accuracy": 0.20339687019586564,
      "num_tokens": 37369744.0,
      "step": 18005
    },
    {
      "entropy": 5.547972631454468,
      "epoch": 1.6337082728592163,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047360950993628585,
      "loss": 5.107,
      "mean_token_accuracy": 0.1997305393218994,
      "num_tokens": 37380578.0,
      "step": 18010
    },
    {
      "entropy": 5.515058755874634,
      "epoch": 1.6341618287373003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004735943029493132,
      "loss": 5.016,
      "mean_token_accuracy": 0.20206762105226517,
      "num_tokens": 37390237.0,
      "step": 18015
    },
    {
      "entropy": 5.5362725257873535,
      "epoch": 1.6346153846153846,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047357909185535965,
      "loss": 5.061,
      "mean_token_accuracy": 0.20199340134859084,
      "num_tokens": 37400261.0,
      "step": 18020
    },
    {
      "entropy": 5.530751276016235,
      "epoch": 1.6350689404934688,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004735638766547399,
      "loss": 5.0118,
      "mean_token_accuracy": 0.20671603828668594,
      "num_tokens": 37410012.0,
      "step": 18025
    },
    {
      "entropy": 5.550591659545899,
      "epoch": 1.6355224963715531,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047354865734776857,
      "loss": 5.1259,
      "mean_token_accuracy": 0.1921493247151375,
      "num_tokens": 37420530.0,
      "step": 18030
    },
    {
      "entropy": 5.527346992492676,
      "epoch": 1.6359760522496372,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004735334339347605,
      "loss": 4.9866,
      "mean_token_accuracy": 0.21734682321548462,
      "num_tokens": 37431146.0,
      "step": 18035
    },
    {
      "entropy": 5.5188727378845215,
      "epoch": 1.6364296081277212,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047351820641603035,
      "loss": 4.9456,
      "mean_token_accuracy": 0.20577428489923477,
      "num_tokens": 37441253.0,
      "step": 18040
    },
    {
      "entropy": 5.41828088760376,
      "epoch": 1.6368831640058055,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004735029747918931,
      "loss": 5.0074,
      "mean_token_accuracy": 0.1972391352057457,
      "num_tokens": 37451326.0,
      "step": 18045
    },
    {
      "entropy": 5.448575639724732,
      "epoch": 1.6373367198838897,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047348773906266374,
      "loss": 4.9642,
      "mean_token_accuracy": 0.20745179802179337,
      "num_tokens": 37462462.0,
      "step": 18050
    },
    {
      "entropy": 5.457478761672974,
      "epoch": 1.637790275761974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004734724992286574,
      "loss": 4.9649,
      "mean_token_accuracy": 0.20242210179567338,
      "num_tokens": 37473016.0,
      "step": 18055
    },
    {
      "entropy": 5.5375182151794435,
      "epoch": 1.638243831640058,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004734572552901892,
      "loss": 5.0898,
      "mean_token_accuracy": 0.20301338881254197,
      "num_tokens": 37483737.0,
      "step": 18060
    },
    {
      "entropy": 5.480352878570557,
      "epoch": 1.638697387518142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004734420072475743,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20475003719329835,
      "num_tokens": 37493070.0,
      "step": 18065
    },
    {
      "entropy": 5.500450086593628,
      "epoch": 1.6391509433962264,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004734267551011282,
      "loss": 4.9742,
      "mean_token_accuracy": 0.20884710401296616,
      "num_tokens": 37503696.0,
      "step": 18070
    },
    {
      "entropy": 5.467238903045654,
      "epoch": 1.6396044992743106,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004734114988511662,
      "loss": 4.9936,
      "mean_token_accuracy": 0.20570732206106185,
      "num_tokens": 37513468.0,
      "step": 18075
    },
    {
      "entropy": 5.403902006149292,
      "epoch": 1.640058055152395,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004733962384980039,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21198950707912445,
      "num_tokens": 37523631.0,
      "step": 18080
    },
    {
      "entropy": 5.525754356384278,
      "epoch": 1.640511611030479,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004733809740419568,
      "loss": 5.0658,
      "mean_token_accuracy": 0.20697355270385742,
      "num_tokens": 37534602.0,
      "step": 18085
    },
    {
      "entropy": 5.4536527633667,
      "epoch": 1.640965166908563,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047336570548334053,
      "loss": 4.9872,
      "mean_token_accuracy": 0.2045159474015236,
      "num_tokens": 37544547.0,
      "step": 18090
    },
    {
      "entropy": 5.553745985031128,
      "epoch": 1.6414187227866472,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047335043282247105,
      "loss": 5.0221,
      "mean_token_accuracy": 0.20612505972385406,
      "num_tokens": 37554903.0,
      "step": 18095
    },
    {
      "entropy": 5.479746913909912,
      "epoch": 1.6418722786647315,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047333515605966396,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2023545578122139,
      "num_tokens": 37565586.0,
      "step": 18100
    },
    {
      "entropy": 5.466207647323609,
      "epoch": 1.6423258345428158,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047331987519523533,
      "loss": 5.0326,
      "mean_token_accuracy": 0.2078264221549034,
      "num_tokens": 37575996.0,
      "step": 18105
    },
    {
      "entropy": 5.462323617935181,
      "epoch": 1.6427793904208998,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004733045902295012,
      "loss": 4.9398,
      "mean_token_accuracy": 0.19835886359214783,
      "num_tokens": 37586441.0,
      "step": 18110
    },
    {
      "entropy": 5.536691761016845,
      "epoch": 1.643232946298984,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047328930116277755,
      "loss": 5.0435,
      "mean_token_accuracy": 0.20475217401981355,
      "num_tokens": 37598226.0,
      "step": 18115
    },
    {
      "entropy": 5.523672294616699,
      "epoch": 1.6436865021770681,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004732740079953806,
      "loss": 5.0482,
      "mean_token_accuracy": 0.20536264032125473,
      "num_tokens": 37609102.0,
      "step": 18120
    },
    {
      "entropy": 5.508391284942627,
      "epoch": 1.6441400580551524,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004732587107276267,
      "loss": 5.0472,
      "mean_token_accuracy": 0.20113572627305984,
      "num_tokens": 37619349.0,
      "step": 18125
    },
    {
      "entropy": 5.4941630363464355,
      "epoch": 1.6445936139332367,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047324340935983225,
      "loss": 5.0289,
      "mean_token_accuracy": 0.21285420358181,
      "num_tokens": 37629844.0,
      "step": 18130
    },
    {
      "entropy": 5.443892908096314,
      "epoch": 1.6450471698113207,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047322810389231333,
      "loss": 4.9291,
      "mean_token_accuracy": 0.2090269923210144,
      "num_tokens": 37639339.0,
      "step": 18135
    },
    {
      "entropy": 5.4682563781738285,
      "epoch": 1.645500725689405,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004732127943253868,
      "loss": 4.9682,
      "mean_token_accuracy": 0.20977348536252977,
      "num_tokens": 37648836.0,
      "step": 18140
    },
    {
      "entropy": 5.521398591995239,
      "epoch": 1.645954281567489,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004731974806593693,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21081233620643616,
      "num_tokens": 37658424.0,
      "step": 18145
    },
    {
      "entropy": 5.499155282974243,
      "epoch": 1.6464078374455733,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047318216289457727,
      "loss": 4.9823,
      "mean_token_accuracy": 0.206360924243927,
      "num_tokens": 37669849.0,
      "step": 18150
    },
    {
      "entropy": 5.478052806854248,
      "epoch": 1.6468613933236576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047316684103132766,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20385076254606246,
      "num_tokens": 37678987.0,
      "step": 18155
    },
    {
      "entropy": 5.540914392471313,
      "epoch": 1.6473149492017418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004731515150699372,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20924765914678572,
      "num_tokens": 37689692.0,
      "step": 18160
    },
    {
      "entropy": 5.549307727813721,
      "epoch": 1.6477685050798259,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047313618501072303,
      "loss": 5.1513,
      "mean_token_accuracy": 0.19636880308389665,
      "num_tokens": 37700092.0,
      "step": 18165
    },
    {
      "entropy": 5.50684289932251,
      "epoch": 1.64822206095791,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047312085085400194,
      "loss": 5.049,
      "mean_token_accuracy": 0.2032184660434723,
      "num_tokens": 37710305.0,
      "step": 18170
    },
    {
      "entropy": 5.629374217987061,
      "epoch": 1.6486756168359942,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047310551260009126,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19262295216321945,
      "num_tokens": 37720451.0,
      "step": 18175
    },
    {
      "entropy": 5.553662490844727,
      "epoch": 1.6491291727140784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000473090170249308,
      "loss": 5.024,
      "mean_token_accuracy": 0.2077767089009285,
      "num_tokens": 37731952.0,
      "step": 18180
    },
    {
      "entropy": 5.46588020324707,
      "epoch": 1.6495827285921627,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004730748238019696,
      "loss": 4.9644,
      "mean_token_accuracy": 0.1991961494088173,
      "num_tokens": 37742201.0,
      "step": 18185
    },
    {
      "entropy": 5.4644073963165285,
      "epoch": 1.6500362844702468,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047305947325839337,
      "loss": 4.981,
      "mean_token_accuracy": 0.20674336403608323,
      "num_tokens": 37753317.0,
      "step": 18190
    },
    {
      "entropy": 5.529191160202027,
      "epoch": 1.6504898403483308,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004730441186188968,
      "loss": 5.0391,
      "mean_token_accuracy": 0.202471961081028,
      "num_tokens": 37763452.0,
      "step": 18195
    },
    {
      "entropy": 5.50238790512085,
      "epoch": 1.650943396226415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047302875988379734,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2040795624256134,
      "num_tokens": 37772804.0,
      "step": 18200
    },
    {
      "entropy": 5.5216504573822025,
      "epoch": 1.6513969521044993,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047301339705341266,
      "loss": 5.0328,
      "mean_token_accuracy": 0.1995303675532341,
      "num_tokens": 37784441.0,
      "step": 18205
    },
    {
      "entropy": 5.498663234710693,
      "epoch": 1.6518505079825836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047299803012806056,
      "loss": 5.037,
      "mean_token_accuracy": 0.20703674852848053,
      "num_tokens": 37795103.0,
      "step": 18210
    },
    {
      "entropy": 5.489672660827637,
      "epoch": 1.6523040638606676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004729826591080586,
      "loss": 5.0295,
      "mean_token_accuracy": 0.20495232492685317,
      "num_tokens": 37806654.0,
      "step": 18215
    },
    {
      "entropy": 5.523226261138916,
      "epoch": 1.6527576197387517,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047296728399372483,
      "loss": 5.0165,
      "mean_token_accuracy": 0.2009401485323906,
      "num_tokens": 37816673.0,
      "step": 18220
    },
    {
      "entropy": 5.539667224884033,
      "epoch": 1.653211175616836,
      "grad_norm": 1.0,
      "learning_rate": 0.00047295190478537725,
      "loss": 5.0383,
      "mean_token_accuracy": 0.19783694297075272,
      "num_tokens": 37827389.0,
      "step": 18225
    },
    {
      "entropy": 5.600225400924683,
      "epoch": 1.6536647314949202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047293652148333376,
      "loss": 5.0507,
      "mean_token_accuracy": 0.2023376613855362,
      "num_tokens": 37838087.0,
      "step": 18230
    },
    {
      "entropy": 5.41401104927063,
      "epoch": 1.6541182873730045,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047292113408791266,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2029065489768982,
      "num_tokens": 37848630.0,
      "step": 18235
    },
    {
      "entropy": 5.410132217407226,
      "epoch": 1.6545718432510885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047290574259943195,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21658093780279158,
      "num_tokens": 37858689.0,
      "step": 18240
    },
    {
      "entropy": 5.547229766845703,
      "epoch": 1.6550253991291726,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047289034701821013,
      "loss": 5.0856,
      "mean_token_accuracy": 0.2034848541021347,
      "num_tokens": 37869049.0,
      "step": 18245
    },
    {
      "entropy": 5.473025465011597,
      "epoch": 1.6554789550072568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047287494734456545,
      "loss": 4.9809,
      "mean_token_accuracy": 0.20671870857477187,
      "num_tokens": 37879495.0,
      "step": 18250
    },
    {
      "entropy": 5.523059368133545,
      "epoch": 1.655932510885341,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047285954357881646,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20215840339660646,
      "num_tokens": 37889442.0,
      "step": 18255
    },
    {
      "entropy": 5.5659104347229,
      "epoch": 1.6563860667634254,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004728441357212817,
      "loss": 5.17,
      "mean_token_accuracy": 0.18997660279273987,
      "num_tokens": 37900147.0,
      "step": 18260
    },
    {
      "entropy": 5.541626787185669,
      "epoch": 1.6568396226415094,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004728287237722798,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20387801378965378,
      "num_tokens": 37910307.0,
      "step": 18265
    },
    {
      "entropy": 5.4466005802154545,
      "epoch": 1.6572931785195935,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004728133077321295,
      "loss": 4.96,
      "mean_token_accuracy": 0.20442774891853333,
      "num_tokens": 37920838.0,
      "step": 18270
    },
    {
      "entropy": 5.528741884231567,
      "epoch": 1.6577467343976777,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047279788760114954,
      "loss": 5.0428,
      "mean_token_accuracy": 0.20524655878543854,
      "num_tokens": 37930717.0,
      "step": 18275
    },
    {
      "entropy": 5.556101369857788,
      "epoch": 1.658200290275762,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004727824633796589,
      "loss": 5.0459,
      "mean_token_accuracy": 0.20111787021160127,
      "num_tokens": 37941254.0,
      "step": 18280
    },
    {
      "entropy": 5.474472284317017,
      "epoch": 1.6586538461538463,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004727670350679765,
      "loss": 5.0176,
      "mean_token_accuracy": 0.2114999607205391,
      "num_tokens": 37952132.0,
      "step": 18285
    },
    {
      "entropy": 5.490719175338745,
      "epoch": 1.6591074020319303,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004727516026664214,
      "loss": 5.0252,
      "mean_token_accuracy": 0.2099560484290123,
      "num_tokens": 37962026.0,
      "step": 18290
    },
    {
      "entropy": 5.494945859909057,
      "epoch": 1.6595609579100146,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004727361661753128,
      "loss": 4.9438,
      "mean_token_accuracy": 0.2045206382870674,
      "num_tokens": 37972911.0,
      "step": 18295
    },
    {
      "entropy": 5.5300699234008786,
      "epoch": 1.6600145137880986,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004727207255949699,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20639899522066116,
      "num_tokens": 37983577.0,
      "step": 18300
    },
    {
      "entropy": 5.53853759765625,
      "epoch": 1.6604680696661829,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047270528092571197,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20507221221923827,
      "num_tokens": 37994294.0,
      "step": 18305
    },
    {
      "entropy": 5.513009262084961,
      "epoch": 1.6609216255442671,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004726898321678584,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20539045631885527,
      "num_tokens": 38005144.0,
      "step": 18310
    },
    {
      "entropy": 5.481329345703125,
      "epoch": 1.6613751814223512,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004726743793217288,
      "loss": 5.0491,
      "mean_token_accuracy": 0.1994861990213394,
      "num_tokens": 38015277.0,
      "step": 18315
    },
    {
      "entropy": 5.52866039276123,
      "epoch": 1.6618287373004355,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047265892238764265,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20707998871803285,
      "num_tokens": 38027530.0,
      "step": 18320
    },
    {
      "entropy": 5.577838993072509,
      "epoch": 1.6622822931785195,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004726434613659196,
      "loss": 5.0722,
      "mean_token_accuracy": 0.20743615478277205,
      "num_tokens": 38036895.0,
      "step": 18325
    },
    {
      "entropy": 5.547301959991455,
      "epoch": 1.6627358490566038,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004726279962568794,
      "loss": 5.0905,
      "mean_token_accuracy": 0.20268693417310715,
      "num_tokens": 38047001.0,
      "step": 18330
    },
    {
      "entropy": 5.52684211730957,
      "epoch": 1.663189404934688,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004726125270608419,
      "loss": 4.9652,
      "mean_token_accuracy": 0.2136825889348984,
      "num_tokens": 38056815.0,
      "step": 18335
    },
    {
      "entropy": 5.51765546798706,
      "epoch": 1.6636429608127723,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047259705377812703,
      "loss": 5.0644,
      "mean_token_accuracy": 0.20229309499263765,
      "num_tokens": 38067139.0,
      "step": 18340
    },
    {
      "entropy": 5.5113856315612795,
      "epoch": 1.6640965166908563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004725815764090546,
      "loss": 4.979,
      "mean_token_accuracy": 0.21099628508090973,
      "num_tokens": 38076824.0,
      "step": 18345
    },
    {
      "entropy": 5.525916671752929,
      "epoch": 1.6645500725689404,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047256609495394494,
      "loss": 5.0265,
      "mean_token_accuracy": 0.2043920114636421,
      "num_tokens": 38087517.0,
      "step": 18350
    },
    {
      "entropy": 5.540383386611938,
      "epoch": 1.6650036284470247,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472550609413118,
      "loss": 5.0114,
      "mean_token_accuracy": 0.20258957743644715,
      "num_tokens": 38098499.0,
      "step": 18355
    },
    {
      "entropy": 5.475964975357056,
      "epoch": 1.665457184325109,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004725351197868942,
      "loss": 5.002,
      "mean_token_accuracy": 0.21008841395378114,
      "num_tokens": 38108327.0,
      "step": 18360
    },
    {
      "entropy": 5.543871355056763,
      "epoch": 1.6659107402031932,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047251962607559373,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20743479281663896,
      "num_tokens": 38118667.0,
      "step": 18365
    },
    {
      "entropy": 5.459502553939819,
      "epoch": 1.6663642960812772,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047250412827953717,
      "loss": 4.9724,
      "mean_token_accuracy": 0.20886190384626388,
      "num_tokens": 38128571.0,
      "step": 18370
    },
    {
      "entropy": 5.55834002494812,
      "epoch": 1.6668178519593613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004724886263990448,
      "loss": 5.0984,
      "mean_token_accuracy": 0.19585662037134172,
      "num_tokens": 38138921.0,
      "step": 18375
    },
    {
      "entropy": 5.57820463180542,
      "epoch": 1.6672714078374455,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004724731204344372,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20742853879928588,
      "num_tokens": 38148858.0,
      "step": 18380
    },
    {
      "entropy": 5.438315725326538,
      "epoch": 1.6677249637155298,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047245761038603536,
      "loss": 4.95,
      "mean_token_accuracy": 0.21081405878067017,
      "num_tokens": 38159107.0,
      "step": 18385
    },
    {
      "entropy": 5.578572463989258,
      "epoch": 1.668178519593614,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047244209625415975,
      "loss": 5.0768,
      "mean_token_accuracy": 0.2027849793434143,
      "num_tokens": 38169228.0,
      "step": 18390
    },
    {
      "entropy": 5.588312578201294,
      "epoch": 1.6686320754716981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004724265780391312,
      "loss": 5.0124,
      "mean_token_accuracy": 0.20619215816259384,
      "num_tokens": 38179175.0,
      "step": 18395
    },
    {
      "entropy": 5.521461391448975,
      "epoch": 1.6690856313497822,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004724110557412708,
      "loss": 5.0323,
      "mean_token_accuracy": 0.2078677698969841,
      "num_tokens": 38188537.0,
      "step": 18400
    },
    {
      "entropy": 5.499007797241211,
      "epoch": 1.6695391872278664,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047239552936089943,
      "loss": 5.0447,
      "mean_token_accuracy": 0.20057985186576843,
      "num_tokens": 38198972.0,
      "step": 18405
    },
    {
      "entropy": 5.501136779785156,
      "epoch": 1.6699927431059507,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047237999889833816,
      "loss": 4.9339,
      "mean_token_accuracy": 0.20863931775093078,
      "num_tokens": 38209472.0,
      "step": 18410
    },
    {
      "entropy": 5.422840404510498,
      "epoch": 1.670446298984035,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047236446435390826,
      "loss": 4.9203,
      "mean_token_accuracy": 0.2069828063249588,
      "num_tokens": 38219977.0,
      "step": 18415
    },
    {
      "entropy": 5.473114490509033,
      "epoch": 1.670899854862119,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047234892572793093,
      "loss": 5.0492,
      "mean_token_accuracy": 0.20733280777931212,
      "num_tokens": 38230492.0,
      "step": 18420
    },
    {
      "entropy": 5.513241386413574,
      "epoch": 1.671353410740203,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004723333830207275,
      "loss": 5.0012,
      "mean_token_accuracy": 0.19537861049175262,
      "num_tokens": 38240799.0,
      "step": 18425
    },
    {
      "entropy": 5.512304449081421,
      "epoch": 1.6718069666182873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047231783623261954,
      "loss": 4.9714,
      "mean_token_accuracy": 0.20044245421886445,
      "num_tokens": 38251439.0,
      "step": 18430
    },
    {
      "entropy": 5.5602222919464115,
      "epoch": 1.6722605224963716,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004723022853639283,
      "loss": 5.1113,
      "mean_token_accuracy": 0.19738350361585616,
      "num_tokens": 38261795.0,
      "step": 18435
    },
    {
      "entropy": 5.462391996383667,
      "epoch": 1.6727140783744558,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004722867304149756,
      "loss": 4.9804,
      "mean_token_accuracy": 0.20544118583202362,
      "num_tokens": 38272172.0,
      "step": 18440
    },
    {
      "entropy": 5.519338464736938,
      "epoch": 1.67316763425254,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004722711713860831,
      "loss": 5.0674,
      "mean_token_accuracy": 0.19758394062519075,
      "num_tokens": 38282829.0,
      "step": 18445
    },
    {
      "entropy": 5.564418315887451,
      "epoch": 1.673621190130624,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004722556082775724,
      "loss": 5.0702,
      "mean_token_accuracy": 0.20760981738567352,
      "num_tokens": 38293793.0,
      "step": 18450
    },
    {
      "entropy": 5.554120635986328,
      "epoch": 1.6740747460087082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004722400410897655,
      "loss": 5.0687,
      "mean_token_accuracy": 0.20341993421316146,
      "num_tokens": 38302577.0,
      "step": 18455
    },
    {
      "entropy": 5.5128473281860355,
      "epoch": 1.6745283018867925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004722244698229842,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20554074943065642,
      "num_tokens": 38312571.0,
      "step": 18460
    },
    {
      "entropy": 5.5175055980682375,
      "epoch": 1.6749818577648767,
      "grad_norm": 1.0,
      "learning_rate": 0.00047220889447755073,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21119394749403,
      "num_tokens": 38323544.0,
      "step": 18465
    },
    {
      "entropy": 5.557501602172851,
      "epoch": 1.6754354136429608,
      "grad_norm": 1.125,
      "learning_rate": 0.0004721933150537869,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20210747569799423,
      "num_tokens": 38334509.0,
      "step": 18470
    },
    {
      "entropy": 5.453472852706909,
      "epoch": 1.675888969521045,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047217773155201526,
      "loss": 4.9549,
      "mean_token_accuracy": 0.20459824055433273,
      "num_tokens": 38344747.0,
      "step": 18475
    },
    {
      "entropy": 5.53552279472351,
      "epoch": 1.676342525399129,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047216214397255776,
      "loss": 5.0651,
      "mean_token_accuracy": 0.19655898809432984,
      "num_tokens": 38355671.0,
      "step": 18480
    },
    {
      "entropy": 5.517773103713989,
      "epoch": 1.6767960812772134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047214655231573693,
      "loss": 5.0369,
      "mean_token_accuracy": 0.2041994571685791,
      "num_tokens": 38366368.0,
      "step": 18485
    },
    {
      "entropy": 5.513717985153198,
      "epoch": 1.6772496371552976,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004721309565818752,
      "loss": 5.0894,
      "mean_token_accuracy": 0.20049837082624436,
      "num_tokens": 38377322.0,
      "step": 18490
    },
    {
      "entropy": 5.584578275680542,
      "epoch": 1.6777031930333817,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047211535677129496,
      "loss": 5.087,
      "mean_token_accuracy": 0.20294539034366607,
      "num_tokens": 38389204.0,
      "step": 18495
    },
    {
      "entropy": 5.510067462921143,
      "epoch": 1.678156748911466,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047209975288431893,
      "loss": 5.0408,
      "mean_token_accuracy": 0.20719282180070878,
      "num_tokens": 38399117.0,
      "step": 18500
    },
    {
      "entropy": 5.522906494140625,
      "epoch": 1.67861030478955,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047208414492126986,
      "loss": 4.9987,
      "mean_token_accuracy": 0.19984093606472014,
      "num_tokens": 38410341.0,
      "step": 18505
    },
    {
      "entropy": 5.558302021026611,
      "epoch": 1.6790638606676342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047206853288247043,
      "loss": 5.0797,
      "mean_token_accuracy": 0.19582510143518447,
      "num_tokens": 38422742.0,
      "step": 18510
    },
    {
      "entropy": 5.5316057205200195,
      "epoch": 1.6795174165457185,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047205291676824345,
      "loss": 5.0714,
      "mean_token_accuracy": 0.197774276137352,
      "num_tokens": 38433022.0,
      "step": 18515
    },
    {
      "entropy": 5.5605991840362545,
      "epoch": 1.6799709724238028,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472037296578912,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20111667215824128,
      "num_tokens": 38444325.0,
      "step": 18520
    },
    {
      "entropy": 5.5038533210754395,
      "epoch": 1.6804245283018868,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000472021672314799,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21152271330356598,
      "num_tokens": 38454576.0,
      "step": 18525
    },
    {
      "entropy": 5.488370323181153,
      "epoch": 1.6808780841799709,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047200604397622763,
      "loss": 4.9664,
      "mean_token_accuracy": 0.209333173930645,
      "num_tokens": 38465090.0,
      "step": 18530
    },
    {
      "entropy": 5.56174955368042,
      "epoch": 1.6813316400580551,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004719904115635211,
      "loss": 5.2117,
      "mean_token_accuracy": 0.1902170941233635,
      "num_tokens": 38477253.0,
      "step": 18535
    },
    {
      "entropy": 5.581843423843384,
      "epoch": 1.6817851959361394,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004719747750770026,
      "loss": 5.1157,
      "mean_token_accuracy": 0.20014643520116807,
      "num_tokens": 38488040.0,
      "step": 18540
    },
    {
      "entropy": 5.487832021713257,
      "epoch": 1.6822387518142237,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004719591345169956,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2080007091164589,
      "num_tokens": 38498016.0,
      "step": 18545
    },
    {
      "entropy": 5.582905292510986,
      "epoch": 1.6826923076923077,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047194348988382345,
      "loss": 5.1101,
      "mean_token_accuracy": 0.20082142800092698,
      "num_tokens": 38508996.0,
      "step": 18550
    },
    {
      "entropy": 5.51684536933899,
      "epoch": 1.6831458635703918,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004719278411778097,
      "loss": 5.0648,
      "mean_token_accuracy": 0.20538436770439147,
      "num_tokens": 38519608.0,
      "step": 18555
    },
    {
      "entropy": 5.5028585433959964,
      "epoch": 1.683599419448476,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047191218839927817,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20028443336486818,
      "num_tokens": 38530737.0,
      "step": 18560
    },
    {
      "entropy": 5.497863245010376,
      "epoch": 1.6840529753265603,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004718965315485523,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2058178350329399,
      "num_tokens": 38540925.0,
      "step": 18565
    },
    {
      "entropy": 5.513472843170166,
      "epoch": 1.6845065312046446,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004718808706259561,
      "loss": 5.0513,
      "mean_token_accuracy": 0.198410527408123,
      "num_tokens": 38553176.0,
      "step": 18570
    },
    {
      "entropy": 5.547786235809326,
      "epoch": 1.6849600870827286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047186520563181325,
      "loss": 5.0853,
      "mean_token_accuracy": 0.1984600692987442,
      "num_tokens": 38563394.0,
      "step": 18575
    },
    {
      "entropy": 5.571151638031006,
      "epoch": 1.6854136429608126,
      "grad_norm": 1.0,
      "learning_rate": 0.0004718495365664477,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20615155547857283,
      "num_tokens": 38573243.0,
      "step": 18580
    },
    {
      "entropy": 5.506847476959228,
      "epoch": 1.685867198838897,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047183386343018364,
      "loss": 5.0534,
      "mean_token_accuracy": 0.20158745348453522,
      "num_tokens": 38584014.0,
      "step": 18585
    },
    {
      "entropy": 5.50274543762207,
      "epoch": 1.6863207547169812,
      "grad_norm": 1.0,
      "learning_rate": 0.0004718181862233451,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2084433391690254,
      "num_tokens": 38594467.0,
      "step": 18590
    },
    {
      "entropy": 5.584379100799561,
      "epoch": 1.6867743105950654,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004718025049462563,
      "loss": 5.0571,
      "mean_token_accuracy": 0.20305939614772797,
      "num_tokens": 38604407.0,
      "step": 18595
    },
    {
      "entropy": 5.560861873626709,
      "epoch": 1.6872278664731495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004717868195992416,
      "loss": 5.0528,
      "mean_token_accuracy": 0.2088589459657669,
      "num_tokens": 38614508.0,
      "step": 18600
    },
    {
      "entropy": 5.575065040588379,
      "epoch": 1.6876814223512335,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004717711301826253,
      "loss": 5.0651,
      "mean_token_accuracy": 0.20356738418340684,
      "num_tokens": 38626244.0,
      "step": 18605
    },
    {
      "entropy": 5.477683877944946,
      "epoch": 1.6881349782293178,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047175543669673186,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20338715761899948,
      "num_tokens": 38636694.0,
      "step": 18610
    },
    {
      "entropy": 5.592471599578857,
      "epoch": 1.688588534107402,
      "grad_norm": 1.0,
      "learning_rate": 0.00047173973914188587,
      "loss": 5.1045,
      "mean_token_accuracy": 0.19840389341115952,
      "num_tokens": 38647414.0,
      "step": 18615
    },
    {
      "entropy": 5.522599124908448,
      "epoch": 1.6890420899854863,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047172403751841196,
      "loss": 5.0201,
      "mean_token_accuracy": 0.2026630625128746,
      "num_tokens": 38657425.0,
      "step": 18620
    },
    {
      "entropy": 5.5504255294799805,
      "epoch": 1.6894956458635704,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047170833182663485,
      "loss": 5.0395,
      "mean_token_accuracy": 0.2048941120505333,
      "num_tokens": 38667751.0,
      "step": 18625
    },
    {
      "entropy": 5.4390911102294925,
      "epoch": 1.6899492017416544,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004716926220668793,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2103493943810463,
      "num_tokens": 38676918.0,
      "step": 18630
    },
    {
      "entropy": 5.449511766433716,
      "epoch": 1.6904027576197387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047167690823947015,
      "loss": 5.0069,
      "mean_token_accuracy": 0.21317176818847655,
      "num_tokens": 38685887.0,
      "step": 18635
    },
    {
      "entropy": 5.46210298538208,
      "epoch": 1.690856313497823,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004716611903447325,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20897857546806337,
      "num_tokens": 38696179.0,
      "step": 18640
    },
    {
      "entropy": 5.506293821334839,
      "epoch": 1.6913098693759072,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004716454683829912,
      "loss": 4.9863,
      "mean_token_accuracy": 0.20195146650075912,
      "num_tokens": 38705777.0,
      "step": 18645
    },
    {
      "entropy": 5.5022331237792965,
      "epoch": 1.6917634252539913,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047162974235457163,
      "loss": 5.0194,
      "mean_token_accuracy": 0.2052154392004013,
      "num_tokens": 38717709.0,
      "step": 18650
    },
    {
      "entropy": 5.513932704925537,
      "epoch": 1.6922169811320755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004716140122597988,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20939967036247253,
      "num_tokens": 38729517.0,
      "step": 18655
    },
    {
      "entropy": 5.5043616771698,
      "epoch": 1.6926705370101596,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004715982780989981,
      "loss": 5.0192,
      "mean_token_accuracy": 0.2061331108212471,
      "num_tokens": 38740878.0,
      "step": 18660
    },
    {
      "entropy": 5.506402587890625,
      "epoch": 1.6931240928882438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047158253987249487,
      "loss": 4.9572,
      "mean_token_accuracy": 0.20769711285829545,
      "num_tokens": 38751291.0,
      "step": 18665
    },
    {
      "entropy": 5.484677314758301,
      "epoch": 1.693577648766328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047156679758061465,
      "loss": 4.948,
      "mean_token_accuracy": 0.21552773416042328,
      "num_tokens": 38760940.0,
      "step": 18670
    },
    {
      "entropy": 5.498112201690674,
      "epoch": 1.6940312046444121,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047155105122368297,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21133071333169937,
      "num_tokens": 38770240.0,
      "step": 18675
    },
    {
      "entropy": 5.448402261734008,
      "epoch": 1.6944847605224964,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004715353008020255,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20739692449569702,
      "num_tokens": 38780452.0,
      "step": 18680
    },
    {
      "entropy": 5.557797908782959,
      "epoch": 1.6949383164005805,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047151954631596794,
      "loss": 5.0736,
      "mean_token_accuracy": 0.2015103355050087,
      "num_tokens": 38790763.0,
      "step": 18685
    },
    {
      "entropy": 5.506512022018432,
      "epoch": 1.6953918722786647,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047150378776583603,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21901689022779464,
      "num_tokens": 38802164.0,
      "step": 18690
    },
    {
      "entropy": 5.507416725158691,
      "epoch": 1.695845428156749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004714880251519557,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20090519338846208,
      "num_tokens": 38812901.0,
      "step": 18695
    },
    {
      "entropy": 5.484809684753418,
      "epoch": 1.6962989840348333,
      "grad_norm": 0.953125,
      "learning_rate": 0.000471472258474653,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20857700258493422,
      "num_tokens": 38823305.0,
      "step": 18700
    },
    {
      "entropy": 5.495634984970093,
      "epoch": 1.6967525399129173,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047145648773425384,
      "loss": 4.9574,
      "mean_token_accuracy": 0.2041999951004982,
      "num_tokens": 38833602.0,
      "step": 18705
    },
    {
      "entropy": 5.514457893371582,
      "epoch": 1.6972060957910013,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047144071293108456,
      "loss": 5.0578,
      "mean_token_accuracy": 0.20269525349140166,
      "num_tokens": 38843950.0,
      "step": 18710
    },
    {
      "entropy": 5.519618082046509,
      "epoch": 1.6976596516690856,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004714249340654712,
      "loss": 4.9856,
      "mean_token_accuracy": 0.2041005238890648,
      "num_tokens": 38853753.0,
      "step": 18715
    },
    {
      "entropy": 5.464105033874512,
      "epoch": 1.6981132075471699,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004714091511377402,
      "loss": 4.9127,
      "mean_token_accuracy": 0.2089301362633705,
      "num_tokens": 38863394.0,
      "step": 18720
    },
    {
      "entropy": 5.477627611160278,
      "epoch": 1.6985667634252541,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047139336414821784,
      "loss": 5.0084,
      "mean_token_accuracy": 0.19901751428842546,
      "num_tokens": 38872972.0,
      "step": 18725
    },
    {
      "entropy": 5.461693048477173,
      "epoch": 1.6990203193033382,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047137757309723073,
      "loss": 4.9919,
      "mean_token_accuracy": 0.21485331356525422,
      "num_tokens": 38883251.0,
      "step": 18730
    },
    {
      "entropy": 5.4876368045806885,
      "epoch": 1.6994738751814222,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004713617779851053,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20316289067268373,
      "num_tokens": 38893274.0,
      "step": 18735
    },
    {
      "entropy": 5.559378385543823,
      "epoch": 1.6999274310595065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004713459788121683,
      "loss": 5.0994,
      "mean_token_accuracy": 0.19765110164880753,
      "num_tokens": 38903870.0,
      "step": 18740
    },
    {
      "entropy": 5.5256421089172365,
      "epoch": 1.7003809869375908,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004713301755787464,
      "loss": 5.0435,
      "mean_token_accuracy": 0.2014593616127968,
      "num_tokens": 38914921.0,
      "step": 18745
    },
    {
      "entropy": 5.546377801895142,
      "epoch": 1.700834542815675,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004713143682851665,
      "loss": 4.9999,
      "mean_token_accuracy": 0.20599546134471894,
      "num_tokens": 38924630.0,
      "step": 18750
    },
    {
      "entropy": 5.454897737503051,
      "epoch": 1.701288098693759,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004712985569317553,
      "loss": 4.9742,
      "mean_token_accuracy": 0.204600790143013,
      "num_tokens": 38934926.0,
      "step": 18755
    },
    {
      "entropy": 5.56922607421875,
      "epoch": 1.7017416545718431,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004712827415188401,
      "loss": 5.1371,
      "mean_token_accuracy": 0.1945800855755806,
      "num_tokens": 38944513.0,
      "step": 18760
    },
    {
      "entropy": 5.561591768264771,
      "epoch": 1.7021952104499274,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004712669220467476,
      "loss": 4.9947,
      "mean_token_accuracy": 0.2074137046933174,
      "num_tokens": 38955918.0,
      "step": 18765
    },
    {
      "entropy": 5.548800992965698,
      "epoch": 1.7026487663280117,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047125109851580533,
      "loss": 5.0131,
      "mean_token_accuracy": 0.20577421486377717,
      "num_tokens": 38966221.0,
      "step": 18770
    },
    {
      "entropy": 5.49301290512085,
      "epoch": 1.703102322206096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047123527092634025,
      "loss": 4.9745,
      "mean_token_accuracy": 0.20071849822998047,
      "num_tokens": 38976463.0,
      "step": 18775
    },
    {
      "entropy": 5.4543568134307865,
      "epoch": 1.70355587808418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004712194392786797,
      "loss": 4.9074,
      "mean_token_accuracy": 0.20934713929891585,
      "num_tokens": 38986375.0,
      "step": 18780
    },
    {
      "entropy": 5.595263910293579,
      "epoch": 1.704009433962264,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047120360357315127,
      "loss": 5.0536,
      "mean_token_accuracy": 0.20466777235269545,
      "num_tokens": 38996116.0,
      "step": 18785
    },
    {
      "entropy": 5.468366289138794,
      "epoch": 1.7044629898403483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047118776381008216,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20442012697458267,
      "num_tokens": 39005846.0,
      "step": 18790
    },
    {
      "entropy": 5.541969347000122,
      "epoch": 1.7049165457184325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047117191998980026,
      "loss": 5.0147,
      "mean_token_accuracy": 0.19670799374580383,
      "num_tokens": 39016744.0,
      "step": 18795
    },
    {
      "entropy": 5.525983285903931,
      "epoch": 1.7053701015965168,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047115607211263295,
      "loss": 5.0568,
      "mean_token_accuracy": 0.19816389679908752,
      "num_tokens": 39026725.0,
      "step": 18800
    },
    {
      "entropy": 5.554859399795532,
      "epoch": 1.7058236574746009,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047114022017890814,
      "loss": 5.0195,
      "mean_token_accuracy": 0.20423724949359895,
      "num_tokens": 39036851.0,
      "step": 18805
    },
    {
      "entropy": 5.502058696746826,
      "epoch": 1.706277213352685,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004711243641889536,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2052324578166008,
      "num_tokens": 39047179.0,
      "step": 18810
    },
    {
      "entropy": 5.449604272842407,
      "epoch": 1.7067307692307692,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047110850414309714,
      "loss": 4.8444,
      "mean_token_accuracy": 0.21609296649694443,
      "num_tokens": 39057758.0,
      "step": 18815
    },
    {
      "entropy": 5.511652517318725,
      "epoch": 1.7071843251088534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047109264004166696,
      "loss": 5.0617,
      "mean_token_accuracy": 0.19992961287498473,
      "num_tokens": 39068142.0,
      "step": 18820
    },
    {
      "entropy": 5.4653867244720455,
      "epoch": 1.7076378809869377,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004710767718849909,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2027351438999176,
      "num_tokens": 39078157.0,
      "step": 18825
    },
    {
      "entropy": 5.488234376907348,
      "epoch": 1.7080914368650217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004710608996733973,
      "loss": 5.0097,
      "mean_token_accuracy": 0.211202010512352,
      "num_tokens": 39089090.0,
      "step": 18830
    },
    {
      "entropy": 5.56004695892334,
      "epoch": 1.708544992743106,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047104502340721433,
      "loss": 5.0614,
      "mean_token_accuracy": 0.19800453633069992,
      "num_tokens": 39099079.0,
      "step": 18835
    },
    {
      "entropy": 5.526583051681518,
      "epoch": 1.70899854862119,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047102914308677033,
      "loss": 4.9921,
      "mean_token_accuracy": 0.2017642602324486,
      "num_tokens": 39109927.0,
      "step": 18840
    },
    {
      "entropy": 5.538265323638916,
      "epoch": 1.7094521044992743,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047101325871239363,
      "loss": 4.9951,
      "mean_token_accuracy": 0.2031232237815857,
      "num_tokens": 39119806.0,
      "step": 18845
    },
    {
      "entropy": 5.558316230773926,
      "epoch": 1.7099056603773586,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709973702844128,
      "loss": 5.0771,
      "mean_token_accuracy": 0.20081631392240523,
      "num_tokens": 39130147.0,
      "step": 18850
    },
    {
      "entropy": 5.448058223724365,
      "epoch": 1.7103592162554426,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047098147780315645,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21496572643518447,
      "num_tokens": 39139937.0,
      "step": 18855
    },
    {
      "entropy": 5.492663812637329,
      "epoch": 1.710812772133527,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004709655812689532,
      "loss": 5.0173,
      "mean_token_accuracy": 0.20803554952144623,
      "num_tokens": 39149977.0,
      "step": 18860
    },
    {
      "entropy": 5.475897932052613,
      "epoch": 1.711266328011611,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004709496806821318,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20517557114362717,
      "num_tokens": 39161076.0,
      "step": 18865
    },
    {
      "entropy": 5.541501426696778,
      "epoch": 1.7117198838896952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047093377604302106,
      "loss": 5.0802,
      "mean_token_accuracy": 0.19836925715208054,
      "num_tokens": 39170776.0,
      "step": 18870
    },
    {
      "entropy": 5.562673425674438,
      "epoch": 1.7121734397677795,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004709178673519499,
      "loss": 5.0397,
      "mean_token_accuracy": 0.19924242794513702,
      "num_tokens": 39181134.0,
      "step": 18875
    },
    {
      "entropy": 5.544519472122192,
      "epoch": 1.7126269956458637,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004709019546092473,
      "loss": 5.1125,
      "mean_token_accuracy": 0.19395548850297928,
      "num_tokens": 39193331.0,
      "step": 18880
    },
    {
      "entropy": 5.582593297958374,
      "epoch": 1.7130805515239478,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047088603781524237,
      "loss": 5.0753,
      "mean_token_accuracy": 0.19931646585464477,
      "num_tokens": 39204372.0,
      "step": 18885
    },
    {
      "entropy": 5.5269989490509035,
      "epoch": 1.7135341074020318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004708701169702642,
      "loss": 4.988,
      "mean_token_accuracy": 0.2095421716570854,
      "num_tokens": 39214725.0,
      "step": 18890
    },
    {
      "entropy": 5.4547687530517575,
      "epoch": 1.713987663280116,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004708541920746422,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2109682485461235,
      "num_tokens": 39225119.0,
      "step": 18895
    },
    {
      "entropy": 5.483250904083252,
      "epoch": 1.7144412191582004,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004708382631287055,
      "loss": 4.9626,
      "mean_token_accuracy": 0.20068428963422774,
      "num_tokens": 39235021.0,
      "step": 18900
    },
    {
      "entropy": 5.536379957199097,
      "epoch": 1.7148947750362846,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004708223301327837,
      "loss": 5.1335,
      "mean_token_accuracy": 0.19601042419672013,
      "num_tokens": 39245830.0,
      "step": 18905
    },
    {
      "entropy": 5.585514068603516,
      "epoch": 1.7153483309143687,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047080639308720617,
      "loss": 5.1009,
      "mean_token_accuracy": 0.20642118602991105,
      "num_tokens": 39255264.0,
      "step": 18910
    },
    {
      "entropy": 5.548873472213745,
      "epoch": 1.7158018867924527,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047079045199230254,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20718229115009307,
      "num_tokens": 39264966.0,
      "step": 18915
    },
    {
      "entropy": 5.505400085449219,
      "epoch": 1.716255442670537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047077450684840247,
      "loss": 5.0832,
      "mean_token_accuracy": 0.1983361214399338,
      "num_tokens": 39275082.0,
      "step": 18920
    },
    {
      "entropy": 5.57161955833435,
      "epoch": 1.7167089985486212,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004707585576558357,
      "loss": 5.1595,
      "mean_token_accuracy": 0.1953936904668808,
      "num_tokens": 39285426.0,
      "step": 18925
    },
    {
      "entropy": 5.6086994171142575,
      "epoch": 1.7171625544267055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004707426044149321,
      "loss": 5.0995,
      "mean_token_accuracy": 0.19788185954093934,
      "num_tokens": 39296633.0,
      "step": 18930
    },
    {
      "entropy": 5.510032510757446,
      "epoch": 1.7176161103047896,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047072664712602153,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2131096675992012,
      "num_tokens": 39306775.0,
      "step": 18935
    },
    {
      "entropy": 5.508192873001098,
      "epoch": 1.7180696661828736,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000470710685789434,
      "loss": 5.035,
      "mean_token_accuracy": 0.19883523732423783,
      "num_tokens": 39317917.0,
      "step": 18940
    },
    {
      "entropy": 5.535958242416382,
      "epoch": 1.7185232220609579,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047069472040549965,
      "loss": 5.036,
      "mean_token_accuracy": 0.2056172490119934,
      "num_tokens": 39328387.0,
      "step": 18945
    },
    {
      "entropy": 5.425024223327637,
      "epoch": 1.7189767779390421,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004706787509745486,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2168424606323242,
      "num_tokens": 39337639.0,
      "step": 18950
    },
    {
      "entropy": 5.514598035812378,
      "epoch": 1.7194303338171264,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047066277749691115,
      "loss": 5.0271,
      "mean_token_accuracy": 0.20351004004478454,
      "num_tokens": 39347971.0,
      "step": 18955
    },
    {
      "entropy": 5.563658571243286,
      "epoch": 1.7198838896952104,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004706467999729175,
      "loss": 5.0609,
      "mean_token_accuracy": 0.20269916355609893,
      "num_tokens": 39358038.0,
      "step": 18960
    },
    {
      "entropy": 5.555438899993897,
      "epoch": 1.7203374455732945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004706308184028983,
      "loss": 5.0489,
      "mean_token_accuracy": 0.202947998046875,
      "num_tokens": 39367834.0,
      "step": 18965
    },
    {
      "entropy": 5.551628732681275,
      "epoch": 1.7207910014513788,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004706148327871839,
      "loss": 5.0263,
      "mean_token_accuracy": 0.20801008641719818,
      "num_tokens": 39378924.0,
      "step": 18970
    },
    {
      "entropy": 5.583839321136475,
      "epoch": 1.721244557329463,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047059884312610487,
      "loss": 5.1352,
      "mean_token_accuracy": 0.19632723778486252,
      "num_tokens": 39390082.0,
      "step": 18975
    },
    {
      "entropy": 5.522844743728638,
      "epoch": 1.7216981132075473,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047058284941999185,
      "loss": 5.0653,
      "mean_token_accuracy": 0.20150293707847594,
      "num_tokens": 39400485.0,
      "step": 18980
    },
    {
      "entropy": 5.5299999713897705,
      "epoch": 1.7221516690856313,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004705668516691758,
      "loss": 5.0335,
      "mean_token_accuracy": 0.20307649224996566,
      "num_tokens": 39411601.0,
      "step": 18985
    },
    {
      "entropy": 5.523506689071655,
      "epoch": 1.7226052249637154,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047055084987398735,
      "loss": 4.9647,
      "mean_token_accuracy": 0.2111507087945938,
      "num_tokens": 39421625.0,
      "step": 18990
    },
    {
      "entropy": 5.503273916244507,
      "epoch": 1.7230587808417996,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004705348440347575,
      "loss": 5.001,
      "mean_token_accuracy": 0.20387445539236068,
      "num_tokens": 39432033.0,
      "step": 18995
    },
    {
      "entropy": 5.509365177154541,
      "epoch": 1.723512336719884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047051883415181724,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20731743574142455,
      "num_tokens": 39442514.0,
      "step": 19000
    },
    {
      "entropy": 5.508768844604492,
      "epoch": 1.7239658925979682,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004705028202254977,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2113283485174179,
      "num_tokens": 39452497.0,
      "step": 19005
    },
    {
      "entropy": 5.4832415103912355,
      "epoch": 1.7244194484760522,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047048680225612996,
      "loss": 5.0177,
      "mean_token_accuracy": 0.20899880230426787,
      "num_tokens": 39462173.0,
      "step": 19010
    },
    {
      "entropy": 5.535536670684815,
      "epoch": 1.7248730043541363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004704707802440454,
      "loss": 5.0631,
      "mean_token_accuracy": 0.21252347975969316,
      "num_tokens": 39473123.0,
      "step": 19015
    },
    {
      "entropy": 5.562990760803222,
      "epoch": 1.7253265602322205,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047045475418957524,
      "loss": 5.0343,
      "mean_token_accuracy": 0.19513361603021623,
      "num_tokens": 39484142.0,
      "step": 19020
    },
    {
      "entropy": 5.549053430557251,
      "epoch": 1.7257801161103048,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004704387240930509,
      "loss": 5.0436,
      "mean_token_accuracy": 0.20033205300569534,
      "num_tokens": 39494926.0,
      "step": 19025
    },
    {
      "entropy": 5.492986488342285,
      "epoch": 1.726233671988389,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004704226899548041,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2030912771821022,
      "num_tokens": 39505153.0,
      "step": 19030
    },
    {
      "entropy": 5.480626821517944,
      "epoch": 1.726687227866473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047040665177516616,
      "loss": 4.9727,
      "mean_token_accuracy": 0.2088142678141594,
      "num_tokens": 39515905.0,
      "step": 19035
    },
    {
      "entropy": 5.483314895629883,
      "epoch": 1.7271407837445574,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047039060955446886,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21209943741559983,
      "num_tokens": 39526762.0,
      "step": 19040
    },
    {
      "entropy": 5.484533214569092,
      "epoch": 1.7275943396226414,
      "grad_norm": 1.0625,
      "learning_rate": 0.000470374563293044,
      "loss": 4.9748,
      "mean_token_accuracy": 0.20307250767946244,
      "num_tokens": 39537340.0,
      "step": 19045
    },
    {
      "entropy": 5.537242794036866,
      "epoch": 1.7280478955007257,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004703585129912234,
      "loss": 4.991,
      "mean_token_accuracy": 0.20512688606977464,
      "num_tokens": 39547144.0,
      "step": 19050
    },
    {
      "entropy": 5.618268918991089,
      "epoch": 1.72850145137881,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047034245864933894,
      "loss": 5.1177,
      "mean_token_accuracy": 0.1946992427110672,
      "num_tokens": 39557707.0,
      "step": 19055
    },
    {
      "entropy": 5.5895966529846195,
      "epoch": 1.728955007256894,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047032640026772274,
      "loss": 5.0834,
      "mean_token_accuracy": 0.19950711131095886,
      "num_tokens": 39567629.0,
      "step": 19060
    },
    {
      "entropy": 5.534259033203125,
      "epoch": 1.7294085631349783,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004703103378467067,
      "loss": 4.9866,
      "mean_token_accuracy": 0.20405916571617128,
      "num_tokens": 39577918.0,
      "step": 19065
    },
    {
      "entropy": 5.545049476623535,
      "epoch": 1.7298621190130623,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047029427138662314,
      "loss": 5.0353,
      "mean_token_accuracy": 0.20991769134998323,
      "num_tokens": 39587838.0,
      "step": 19070
    },
    {
      "entropy": 5.4847307205200195,
      "epoch": 1.7303156748911466,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004702782008878042,
      "loss": 4.9261,
      "mean_token_accuracy": 0.2125234305858612,
      "num_tokens": 39597353.0,
      "step": 19075
    },
    {
      "entropy": 5.498964595794678,
      "epoch": 1.7307692307692308,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004702621263505824,
      "loss": 4.967,
      "mean_token_accuracy": 0.20899804085493087,
      "num_tokens": 39607349.0,
      "step": 19080
    },
    {
      "entropy": 5.524208688735962,
      "epoch": 1.731222786647315,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004702460477752901,
      "loss": 5.0516,
      "mean_token_accuracy": 0.20135973393917084,
      "num_tokens": 39617810.0,
      "step": 19085
    },
    {
      "entropy": 5.498603820800781,
      "epoch": 1.7316763425253991,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004702299651622597,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21558779627084732,
      "num_tokens": 39626805.0,
      "step": 19090
    },
    {
      "entropy": 5.4490362167358395,
      "epoch": 1.7321298984034832,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004702138785118239,
      "loss": 4.9573,
      "mean_token_accuracy": 0.20946362912654876,
      "num_tokens": 39637839.0,
      "step": 19095
    },
    {
      "entropy": 5.4664324760437015,
      "epoch": 1.7325834542815675,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004701977878243153,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20754453390836716,
      "num_tokens": 39648697.0,
      "step": 19100
    },
    {
      "entropy": 5.464032983779907,
      "epoch": 1.7330370101596517,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004701816931000668,
      "loss": 4.9644,
      "mean_token_accuracy": 0.20650705695152283,
      "num_tokens": 39658154.0,
      "step": 19105
    },
    {
      "entropy": 5.502857494354248,
      "epoch": 1.733490566037736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047016559433941106,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20509715080261232,
      "num_tokens": 39667939.0,
      "step": 19110
    },
    {
      "entropy": 5.482723522186279,
      "epoch": 1.73394412191582,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004701494915426812,
      "loss": 4.9316,
      "mean_token_accuracy": 0.2047874704003334,
      "num_tokens": 39678289.0,
      "step": 19115
    },
    {
      "entropy": 5.546759462356567,
      "epoch": 1.734397677793904,
      "grad_norm": 1.125,
      "learning_rate": 0.00047013338471021,
      "loss": 4.9937,
      "mean_token_accuracy": 0.20536402910947799,
      "num_tokens": 39689901.0,
      "step": 19120
    },
    {
      "entropy": 5.515721130371094,
      "epoch": 1.7348512336719883,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047011727384233074,
      "loss": 5.0389,
      "mean_token_accuracy": 0.19690279215574263,
      "num_tokens": 39700831.0,
      "step": 19125
    },
    {
      "entropy": 5.5482320308685305,
      "epoch": 1.7353047895500726,
      "grad_norm": 1.0,
      "learning_rate": 0.0004701011589393765,
      "loss": 5.033,
      "mean_token_accuracy": 0.2001452073454857,
      "num_tokens": 39711794.0,
      "step": 19130
    },
    {
      "entropy": 5.52657151222229,
      "epoch": 1.7357583454281569,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047008504000168054,
      "loss": 4.9726,
      "mean_token_accuracy": 0.20627530217170714,
      "num_tokens": 39721889.0,
      "step": 19135
    },
    {
      "entropy": 5.483045530319214,
      "epoch": 1.736211901306241,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004700689170295763,
      "loss": 5.0579,
      "mean_token_accuracy": 0.20222830325365065,
      "num_tokens": 39732562.0,
      "step": 19140
    },
    {
      "entropy": 5.572683954238892,
      "epoch": 1.736665457184325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047005279002339717,
      "loss": 5.018,
      "mean_token_accuracy": 0.20817613154649733,
      "num_tokens": 39742286.0,
      "step": 19145
    },
    {
      "entropy": 5.538823175430298,
      "epoch": 1.7371190130624092,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047003665898347663,
      "loss": 5.0252,
      "mean_token_accuracy": 0.2073296308517456,
      "num_tokens": 39754336.0,
      "step": 19150
    },
    {
      "entropy": 5.495729303359985,
      "epoch": 1.7375725689404935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004700205239101483,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21193399727344514,
      "num_tokens": 39763545.0,
      "step": 19155
    },
    {
      "entropy": 5.52259202003479,
      "epoch": 1.7380261248185778,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004700043848037458,
      "loss": 5.0854,
      "mean_token_accuracy": 0.19503413289785385,
      "num_tokens": 39774536.0,
      "step": 19160
    },
    {
      "entropy": 5.509861278533935,
      "epoch": 1.7384796806966618,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004699882416646029,
      "loss": 4.9732,
      "mean_token_accuracy": 0.21053527742624284,
      "num_tokens": 39785377.0,
      "step": 19165
    },
    {
      "entropy": 5.5219244956970215,
      "epoch": 1.7389332365747459,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004699720944930535,
      "loss": 5.0905,
      "mean_token_accuracy": 0.19307584017515184,
      "num_tokens": 39796692.0,
      "step": 19170
    },
    {
      "entropy": 5.461316347122192,
      "epoch": 1.7393867924528301,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046995594328943147,
      "loss": 4.9871,
      "mean_token_accuracy": 0.20763515830039977,
      "num_tokens": 39807927.0,
      "step": 19175
    },
    {
      "entropy": 5.656646585464477,
      "epoch": 1.7398403483309144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004699397880540709,
      "loss": 5.1689,
      "mean_token_accuracy": 0.20154161900281906,
      "num_tokens": 39817617.0,
      "step": 19180
    },
    {
      "entropy": 5.548462867736816,
      "epoch": 1.7402939042089987,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004699236287873058,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21022000163793564,
      "num_tokens": 39827092.0,
      "step": 19185
    },
    {
      "entropy": 5.495359420776367,
      "epoch": 1.7407474600870827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004699074654894705,
      "loss": 4.9617,
      "mean_token_accuracy": 0.20519209057092666,
      "num_tokens": 39837378.0,
      "step": 19190
    },
    {
      "entropy": 5.51698932647705,
      "epoch": 1.7412010159651667,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000469891298160899,
      "loss": 4.9807,
      "mean_token_accuracy": 0.1958746701478958,
      "num_tokens": 39848191.0,
      "step": 19195
    },
    {
      "entropy": 5.494634103775025,
      "epoch": 1.741654571843251,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046987512680192586,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2061051085591316,
      "num_tokens": 39857886.0,
      "step": 19200
    },
    {
      "entropy": 5.560719919204712,
      "epoch": 1.7421081277213353,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046985895141288546,
      "loss": 5.0745,
      "mean_token_accuracy": 0.2028934597969055,
      "num_tokens": 39868348.0,
      "step": 19205
    },
    {
      "entropy": 5.568835687637329,
      "epoch": 1.7425616835994195,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046984277199411224,
      "loss": 5.0827,
      "mean_token_accuracy": 0.2048014745116234,
      "num_tokens": 39879129.0,
      "step": 19210
    },
    {
      "entropy": 5.479235744476318,
      "epoch": 1.7430152394775036,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046982658854594083,
      "loss": 4.9735,
      "mean_token_accuracy": 0.2069411188364029,
      "num_tokens": 39890239.0,
      "step": 19215
    },
    {
      "entropy": 5.446046924591064,
      "epoch": 1.7434687953555879,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046981040106870595,
      "loss": 4.8868,
      "mean_token_accuracy": 0.21154987961053848,
      "num_tokens": 39900907.0,
      "step": 19220
    },
    {
      "entropy": 5.530134630203247,
      "epoch": 1.743922351233672,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004697942095627423,
      "loss": 4.9989,
      "mean_token_accuracy": 0.209362630546093,
      "num_tokens": 39911564.0,
      "step": 19225
    },
    {
      "entropy": 5.51063084602356,
      "epoch": 1.7443759071117562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004697780140283848,
      "loss": 4.9491,
      "mean_token_accuracy": 0.20574220269918442,
      "num_tokens": 39922168.0,
      "step": 19230
    },
    {
      "entropy": 5.540814733505249,
      "epoch": 1.7448294629898404,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004697618144659683,
      "loss": 4.9865,
      "mean_token_accuracy": 0.20843921601772308,
      "num_tokens": 39932690.0,
      "step": 19235
    },
    {
      "entropy": 5.622589206695556,
      "epoch": 1.7452830188679245,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004697456108758278,
      "loss": 5.1385,
      "mean_token_accuracy": 0.2018663093447685,
      "num_tokens": 39942970.0,
      "step": 19240
    },
    {
      "entropy": 5.485025405883789,
      "epoch": 1.7457365747460087,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004697294032582986,
      "loss": 4.9738,
      "mean_token_accuracy": 0.20647140443325043,
      "num_tokens": 39953255.0,
      "step": 19245
    },
    {
      "entropy": 5.441250514984131,
      "epoch": 1.7461901306240928,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046971319161371564,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21407563537359237,
      "num_tokens": 39963521.0,
      "step": 19250
    },
    {
      "entropy": 5.515857315063476,
      "epoch": 1.746643686502177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046969697594241426,
      "loss": 4.9787,
      "mean_token_accuracy": 0.20265513360500337,
      "num_tokens": 39973494.0,
      "step": 19255
    },
    {
      "entropy": 5.522301149368286,
      "epoch": 1.7470972423802613,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004696807562447298,
      "loss": 5.0976,
      "mean_token_accuracy": 0.20003973692655563,
      "num_tokens": 39984652.0,
      "step": 19260
    },
    {
      "entropy": 5.5341369152069095,
      "epoch": 1.7475507982583456,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046966453252099763,
      "loss": 5.0379,
      "mean_token_accuracy": 0.2052340716123581,
      "num_tokens": 39995490.0,
      "step": 19265
    },
    {
      "entropy": 5.476441144943237,
      "epoch": 1.7480043541364296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046964830477155335,
      "loss": 4.8282,
      "mean_token_accuracy": 0.22589887380599977,
      "num_tokens": 40006447.0,
      "step": 19270
    },
    {
      "entropy": 5.564656352996826,
      "epoch": 1.7484579100145137,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046963207299673257,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20302298814058303,
      "num_tokens": 40015733.0,
      "step": 19275
    },
    {
      "entropy": 5.557398319244385,
      "epoch": 1.748911465892598,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046961583719687085,
      "loss": 4.9373,
      "mean_token_accuracy": 0.20754121243953705,
      "num_tokens": 40025844.0,
      "step": 19280
    },
    {
      "entropy": 5.514825487136841,
      "epoch": 1.7493650217706822,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046959959737230414,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21481667459011078,
      "num_tokens": 40035942.0,
      "step": 19285
    },
    {
      "entropy": 5.42282395362854,
      "epoch": 1.7498185776487665,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046958335352336806,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21713285148143768,
      "num_tokens": 40045551.0,
      "step": 19290
    },
    {
      "entropy": 5.51376371383667,
      "epoch": 1.7502721335268505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004695671056503987,
      "loss": 5.0317,
      "mean_token_accuracy": 0.20856801122426988,
      "num_tokens": 40055803.0,
      "step": 19295
    },
    {
      "entropy": 5.5142841815948485,
      "epoch": 1.7507256894049346,
      "grad_norm": 1.03125,
      "learning_rate": 0.000469550853753732,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20099260956048964,
      "num_tokens": 40066004.0,
      "step": 19300
    },
    {
      "entropy": 5.501036930084228,
      "epoch": 1.7511792452830188,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000469534597833704,
      "loss": 4.9929,
      "mean_token_accuracy": 0.2118948593735695,
      "num_tokens": 40075522.0,
      "step": 19305
    },
    {
      "entropy": 5.496591711044312,
      "epoch": 1.751632801161103,
      "grad_norm": 1.078125,
      "learning_rate": 0.000469518337890651,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2103877067565918,
      "num_tokens": 40084839.0,
      "step": 19310
    },
    {
      "entropy": 5.559009075164795,
      "epoch": 1.7520863570391874,
      "grad_norm": 1.0,
      "learning_rate": 0.00046950207392490917,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20451847165822984,
      "num_tokens": 40094392.0,
      "step": 19315
    },
    {
      "entropy": 5.438671445846557,
      "epoch": 1.7525399129172714,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046948580593681496,
      "loss": 4.9834,
      "mean_token_accuracy": 0.20572173595428467,
      "num_tokens": 40105990.0,
      "step": 19320
    },
    {
      "entropy": 5.4502801418304445,
      "epoch": 1.7529934687953554,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004694695339267047,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21390351206064223,
      "num_tokens": 40116143.0,
      "step": 19325
    },
    {
      "entropy": 5.614202833175659,
      "epoch": 1.7534470246734397,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046945325789491483,
      "loss": 5.0659,
      "mean_token_accuracy": 0.20463183969259263,
      "num_tokens": 40126849.0,
      "step": 19330
    },
    {
      "entropy": 5.510346460342407,
      "epoch": 1.753900580551524,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004694369778417822,
      "loss": 5.0399,
      "mean_token_accuracy": 0.2019796848297119,
      "num_tokens": 40137259.0,
      "step": 19335
    },
    {
      "entropy": 5.414547252655029,
      "epoch": 1.7543541364296082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004694206937676431,
      "loss": 4.8966,
      "mean_token_accuracy": 0.20899513810873033,
      "num_tokens": 40148578.0,
      "step": 19340
    },
    {
      "entropy": 5.565746641159057,
      "epoch": 1.7548076923076923,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004694044056728347,
      "loss": 5.0682,
      "mean_token_accuracy": 0.1964053377509117,
      "num_tokens": 40159754.0,
      "step": 19345
    },
    {
      "entropy": 5.538667440414429,
      "epoch": 1.7552612481857763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046938811355769363,
      "loss": 5.0042,
      "mean_token_accuracy": 0.2027725711464882,
      "num_tokens": 40169388.0,
      "step": 19350
    },
    {
      "entropy": 5.502602910995483,
      "epoch": 1.7557148040638606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046937181742255673,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20214175283908845,
      "num_tokens": 40179271.0,
      "step": 19355
    },
    {
      "entropy": 5.557893991470337,
      "epoch": 1.7561683599419449,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004693555172677612,
      "loss": 5.0517,
      "mean_token_accuracy": 0.20480775088071823,
      "num_tokens": 40189292.0,
      "step": 19360
    },
    {
      "entropy": 5.472114896774292,
      "epoch": 1.7566219158200291,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000469339213093644,
      "loss": 5.009,
      "mean_token_accuracy": 0.20178278386592866,
      "num_tokens": 40199271.0,
      "step": 19365
    },
    {
      "entropy": 5.507282257080078,
      "epoch": 1.7570754716981132,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046932290490054235,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20328470319509506,
      "num_tokens": 40210386.0,
      "step": 19370
    },
    {
      "entropy": 5.564381408691406,
      "epoch": 1.7575290275761972,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004693065926887936,
      "loss": 4.9976,
      "mean_token_accuracy": 0.20100197196006775,
      "num_tokens": 40220892.0,
      "step": 19375
    },
    {
      "entropy": 5.565685033798218,
      "epoch": 1.7579825834542815,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046929027645873484,
      "loss": 4.9877,
      "mean_token_accuracy": 0.20461044311523438,
      "num_tokens": 40231737.0,
      "step": 19380
    },
    {
      "entropy": 5.487754487991333,
      "epoch": 1.7584361393323658,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004692739562107038,
      "loss": 5.0377,
      "mean_token_accuracy": 0.20050707459449768,
      "num_tokens": 40242218.0,
      "step": 19385
    },
    {
      "entropy": 5.499060773849488,
      "epoch": 1.75888969521045,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046925763194503775,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20986446738243103,
      "num_tokens": 40253659.0,
      "step": 19390
    },
    {
      "entropy": 5.450934648513794,
      "epoch": 1.759343251088534,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004692413036620744,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2149718463420868,
      "num_tokens": 40264328.0,
      "step": 19395
    },
    {
      "entropy": 5.464368915557861,
      "epoch": 1.7597968069666183,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004692249713621514,
      "loss": 4.946,
      "mean_token_accuracy": 0.2071563258767128,
      "num_tokens": 40275188.0,
      "step": 19400
    },
    {
      "entropy": 5.575522470474243,
      "epoch": 1.7602503628447024,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046920863504560647,
      "loss": 5.1228,
      "mean_token_accuracy": 0.19950825870037078,
      "num_tokens": 40287094.0,
      "step": 19405
    },
    {
      "entropy": 5.535629320144653,
      "epoch": 1.7607039187227866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004691922947127775,
      "loss": 4.9732,
      "mean_token_accuracy": 0.20288341492414474,
      "num_tokens": 40296551.0,
      "step": 19410
    },
    {
      "entropy": 5.493284082412719,
      "epoch": 1.761157474600871,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004691759503640024,
      "loss": 4.9854,
      "mean_token_accuracy": 0.20166113674640657,
      "num_tokens": 40306920.0,
      "step": 19415
    },
    {
      "entropy": 5.486135101318359,
      "epoch": 1.761611030478955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004691596019996191,
      "loss": 4.8891,
      "mean_token_accuracy": 0.2151423826813698,
      "num_tokens": 40316797.0,
      "step": 19420
    },
    {
      "entropy": 5.435636186599732,
      "epoch": 1.7620645863570392,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004691432496199658,
      "loss": 4.945,
      "mean_token_accuracy": 0.203757406771183,
      "num_tokens": 40327002.0,
      "step": 19425
    },
    {
      "entropy": 5.617444562911987,
      "epoch": 1.7625181422351233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046912689322538064,
      "loss": 5.084,
      "mean_token_accuracy": 0.20587912946939468,
      "num_tokens": 40337247.0,
      "step": 19430
    },
    {
      "entropy": 5.523028898239136,
      "epoch": 1.7629716981132075,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046911053281620187,
      "loss": 5.0319,
      "mean_token_accuracy": 0.20531492978334426,
      "num_tokens": 40346960.0,
      "step": 19435
    },
    {
      "entropy": 5.554391193389892,
      "epoch": 1.7634252539912918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046909416839276775,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2060078427195549,
      "num_tokens": 40357169.0,
      "step": 19440
    },
    {
      "entropy": 5.494356679916382,
      "epoch": 1.763878809869376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004690777999554168,
      "loss": 4.871,
      "mean_token_accuracy": 0.21701277941465377,
      "num_tokens": 40367660.0,
      "step": 19445
    },
    {
      "entropy": 5.48631534576416,
      "epoch": 1.76433236574746,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004690614275044875,
      "loss": 4.8981,
      "mean_token_accuracy": 0.2141926497220993,
      "num_tokens": 40378217.0,
      "step": 19450
    },
    {
      "entropy": 5.511997699737549,
      "epoch": 1.7647859216255442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004690450510403184,
      "loss": 5.0085,
      "mean_token_accuracy": 0.20661132037639618,
      "num_tokens": 40388249.0,
      "step": 19455
    },
    {
      "entropy": 5.5403870105743405,
      "epoch": 1.7652394775036284,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004690286705632483,
      "loss": 5.0429,
      "mean_token_accuracy": 0.2022801861166954,
      "num_tokens": 40398239.0,
      "step": 19460
    },
    {
      "entropy": 5.527664661407471,
      "epoch": 1.7656930333817127,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004690122860736157,
      "loss": 5.0476,
      "mean_token_accuracy": 0.20684996545314788,
      "num_tokens": 40408473.0,
      "step": 19465
    },
    {
      "entropy": 5.639688348770141,
      "epoch": 1.766146589259797,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046899589757175964,
      "loss": 5.128,
      "mean_token_accuracy": 0.19599819481372832,
      "num_tokens": 40418674.0,
      "step": 19470
    },
    {
      "entropy": 5.489195823669434,
      "epoch": 1.766600145137881,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000468979505058019,
      "loss": 4.9589,
      "mean_token_accuracy": 0.2068178355693817,
      "num_tokens": 40429815.0,
      "step": 19475
    },
    {
      "entropy": 5.607937955856324,
      "epoch": 1.767053701015965,
      "grad_norm": 1.125,
      "learning_rate": 0.00046896310853273274,
      "loss": 5.0752,
      "mean_token_accuracy": 0.20456432402133942,
      "num_tokens": 40439809.0,
      "step": 19480
    },
    {
      "entropy": 5.617741441726684,
      "epoch": 1.7675072568940493,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046894670799624,
      "loss": 5.0626,
      "mean_token_accuracy": 0.19883812963962555,
      "num_tokens": 40449503.0,
      "step": 19485
    },
    {
      "entropy": 5.5056475639343265,
      "epoch": 1.7679608127721336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046893030344887995,
      "loss": 5.0678,
      "mean_token_accuracy": 0.19412409514188766,
      "num_tokens": 40459794.0,
      "step": 19490
    },
    {
      "entropy": 5.484907579421997,
      "epoch": 1.7684143686502178,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046891389489099166,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21248606443405152,
      "num_tokens": 40470044.0,
      "step": 19495
    },
    {
      "entropy": 5.530512571334839,
      "epoch": 1.7688679245283019,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046889748232291476,
      "loss": 4.9799,
      "mean_token_accuracy": 0.2080856442451477,
      "num_tokens": 40480709.0,
      "step": 19500
    },
    {
      "entropy": 5.438140058517456,
      "epoch": 1.769321480406386,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046888106574498843,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2188742369413376,
      "num_tokens": 40490906.0,
      "step": 19505
    },
    {
      "entropy": 5.464437818527221,
      "epoch": 1.7697750362844702,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004688646451575522,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2143838033080101,
      "num_tokens": 40501971.0,
      "step": 19510
    },
    {
      "entropy": 5.481470727920533,
      "epoch": 1.7702285921625545,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046884822056094587,
      "loss": 5.0354,
      "mean_token_accuracy": 0.19920152574777603,
      "num_tokens": 40512685.0,
      "step": 19515
    },
    {
      "entropy": 5.446528577804566,
      "epoch": 1.7706821480406387,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046883179195550883,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20782757103443145,
      "num_tokens": 40523460.0,
      "step": 19520
    },
    {
      "entropy": 5.5545446395874025,
      "epoch": 1.7711357039187228,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046881535934158094,
      "loss": 4.9608,
      "mean_token_accuracy": 0.2085803419351578,
      "num_tokens": 40533594.0,
      "step": 19525
    },
    {
      "entropy": 5.516283655166626,
      "epoch": 1.7715892597968068,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004687989227195021,
      "loss": 5.0036,
      "mean_token_accuracy": 0.20703327655792236,
      "num_tokens": 40543504.0,
      "step": 19530
    },
    {
      "entropy": 5.50254077911377,
      "epoch": 1.772042815674891,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046878248208961204,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20424250811338424,
      "num_tokens": 40553711.0,
      "step": 19535
    },
    {
      "entropy": 5.4620030403137205,
      "epoch": 1.7724963715529753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000468766037452251,
      "loss": 5.0107,
      "mean_token_accuracy": 0.2051798552274704,
      "num_tokens": 40563731.0,
      "step": 19540
    },
    {
      "entropy": 5.559663248062134,
      "epoch": 1.7729499274310596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046874958880775886,
      "loss": 5.0256,
      "mean_token_accuracy": 0.20399201661348343,
      "num_tokens": 40574235.0,
      "step": 19545
    },
    {
      "entropy": 5.561195611953735,
      "epoch": 1.7734034833091437,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004687331361564758,
      "loss": 4.9539,
      "mean_token_accuracy": 0.20946568101644517,
      "num_tokens": 40584459.0,
      "step": 19550
    },
    {
      "entropy": 5.494787502288818,
      "epoch": 1.7738570391872277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004687166794987422,
      "loss": 4.9969,
      "mean_token_accuracy": 0.20427904427051544,
      "num_tokens": 40594268.0,
      "step": 19555
    },
    {
      "entropy": 5.467495536804199,
      "epoch": 1.774310595065312,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004687002188348983,
      "loss": 4.9625,
      "mean_token_accuracy": 0.2154368594288826,
      "num_tokens": 40604673.0,
      "step": 19560
    },
    {
      "entropy": 5.448693799972534,
      "epoch": 1.7747641509433962,
      "grad_norm": 1.125,
      "learning_rate": 0.00046868375416528443,
      "loss": 4.9493,
      "mean_token_accuracy": 0.21056706607341766,
      "num_tokens": 40614998.0,
      "step": 19565
    },
    {
      "entropy": 5.502387046813965,
      "epoch": 1.7752177068214805,
      "grad_norm": 1.125,
      "learning_rate": 0.00046866728549024123,
      "loss": 5.0313,
      "mean_token_accuracy": 0.20484382063150405,
      "num_tokens": 40624729.0,
      "step": 19570
    },
    {
      "entropy": 5.545193481445312,
      "epoch": 1.7756712626995645,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004686508128101092,
      "loss": 5.0859,
      "mean_token_accuracy": 0.20403294563293456,
      "num_tokens": 40635673.0,
      "step": 19575
    },
    {
      "entropy": 5.559050893783569,
      "epoch": 1.7761248185776488,
      "grad_norm": 1.015625,
      "learning_rate": 0.000468634336125229,
      "loss": 5.058,
      "mean_token_accuracy": 0.20020204931497573,
      "num_tokens": 40645990.0,
      "step": 19580
    },
    {
      "entropy": 5.572917890548706,
      "epoch": 1.7765783744557329,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046861785543594143,
      "loss": 4.9852,
      "mean_token_accuracy": 0.21186477690935135,
      "num_tokens": 40656339.0,
      "step": 19585
    },
    {
      "entropy": 5.50078330039978,
      "epoch": 1.7770319303338171,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004686013707425873,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21024974882602693,
      "num_tokens": 40666371.0,
      "step": 19590
    },
    {
      "entropy": 5.512684726715088,
      "epoch": 1.7774854862119014,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046858488204550735,
      "loss": 4.957,
      "mean_token_accuracy": 0.20563161373138428,
      "num_tokens": 40676423.0,
      "step": 19595
    },
    {
      "entropy": 5.495868921279907,
      "epoch": 1.7779390420899854,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004685683893450428,
      "loss": 4.916,
      "mean_token_accuracy": 0.2111999660730362,
      "num_tokens": 40686331.0,
      "step": 19600
    },
    {
      "entropy": 5.439804172515869,
      "epoch": 1.7783925979680697,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004685518926415346,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2131078854203224,
      "num_tokens": 40697408.0,
      "step": 19605
    },
    {
      "entropy": 5.423649454116822,
      "epoch": 1.7788461538461537,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000468535391935324,
      "loss": 4.856,
      "mean_token_accuracy": 0.22234155088663102,
      "num_tokens": 40707078.0,
      "step": 19610
    },
    {
      "entropy": 5.535253810882568,
      "epoch": 1.779299709724238,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004685188872267521,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20636095255613326,
      "num_tokens": 40718116.0,
      "step": 19615
    },
    {
      "entropy": 5.523155736923218,
      "epoch": 1.7797532656023223,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004685023785161603,
      "loss": 5.0187,
      "mean_token_accuracy": 0.20456632524728774,
      "num_tokens": 40727737.0,
      "step": 19620
    },
    {
      "entropy": 5.5154128074646,
      "epoch": 1.7802068214804065,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004684858658038901,
      "loss": 4.9649,
      "mean_token_accuracy": 0.202775239944458,
      "num_tokens": 40737455.0,
      "step": 19625
    },
    {
      "entropy": 5.590185356140137,
      "epoch": 1.7806603773584906,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004684693490902828,
      "loss": 4.9936,
      "mean_token_accuracy": 0.21528024077415467,
      "num_tokens": 40746715.0,
      "step": 19630
    },
    {
      "entropy": 5.465564012527466,
      "epoch": 1.7811139332365746,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046845282837568,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21505769342184067,
      "num_tokens": 40756432.0,
      "step": 19635
    },
    {
      "entropy": 5.411591958999634,
      "epoch": 1.781567489114659,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046843630366042355,
      "loss": 4.9553,
      "mean_token_accuracy": 0.2048589125275612,
      "num_tokens": 40766629.0,
      "step": 19640
    },
    {
      "entropy": 5.582585382461548,
      "epoch": 1.7820210449927432,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046841977494485506,
      "loss": 5.0881,
      "mean_token_accuracy": 0.1974947214126587,
      "num_tokens": 40776650.0,
      "step": 19645
    },
    {
      "entropy": 5.601768112182617,
      "epoch": 1.7824746008708274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004684032422293162,
      "loss": 4.9751,
      "mean_token_accuracy": 0.20934145748615265,
      "num_tokens": 40786722.0,
      "step": 19650
    },
    {
      "entropy": 5.398625421524048,
      "epoch": 1.7829281567489115,
      "grad_norm": 1.015625,
      "learning_rate": 0.000468386705514149,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2224738508462906,
      "num_tokens": 40797058.0,
      "step": 19655
    },
    {
      "entropy": 5.437365961074829,
      "epoch": 1.7833817126269955,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004683701647996956,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20820920020341874,
      "num_tokens": 40808207.0,
      "step": 19660
    },
    {
      "entropy": 5.570583391189575,
      "epoch": 1.7838352685050798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004683536200862978,
      "loss": 5.089,
      "mean_token_accuracy": 0.20439439564943312,
      "num_tokens": 40819655.0,
      "step": 19665
    },
    {
      "entropy": 5.5647047996521,
      "epoch": 1.784288824383164,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004683370713742979,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20476250648498534,
      "num_tokens": 40831015.0,
      "step": 19670
    },
    {
      "entropy": 5.48437008857727,
      "epoch": 1.7847423802612483,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046832051866403814,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21035304218530654,
      "num_tokens": 40841763.0,
      "step": 19675
    },
    {
      "entropy": 5.4942773342132565,
      "epoch": 1.7851959361393324,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004683039619558607,
      "loss": 4.9702,
      "mean_token_accuracy": 0.20394841432571412,
      "num_tokens": 40852479.0,
      "step": 19680
    },
    {
      "entropy": 5.536971139907837,
      "epoch": 1.7856494920174164,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046828740125010813,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20805347412824632,
      "num_tokens": 40862486.0,
      "step": 19685
    },
    {
      "entropy": 5.5686804294586185,
      "epoch": 1.7861030478955007,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004682708365471228,
      "loss": 5.0755,
      "mean_token_accuracy": 0.20404157787561417,
      "num_tokens": 40872182.0,
      "step": 19690
    },
    {
      "entropy": 5.5247588634490965,
      "epoch": 1.786556603773585,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046825426784724734,
      "loss": 5.0519,
      "mean_token_accuracy": 0.20590101182460785,
      "num_tokens": 40883036.0,
      "step": 19695
    },
    {
      "entropy": 5.5683763980865475,
      "epoch": 1.7870101596516692,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004682376951508244,
      "loss": 5.0895,
      "mean_token_accuracy": 0.19138999432325363,
      "num_tokens": 40893684.0,
      "step": 19700
    },
    {
      "entropy": 5.57241907119751,
      "epoch": 1.7874637155297532,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046822111845819665,
      "loss": 5.0727,
      "mean_token_accuracy": 0.20308079421520234,
      "num_tokens": 40904540.0,
      "step": 19705
    },
    {
      "entropy": 5.557100105285644,
      "epoch": 1.7879172714078373,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046820453776970697,
      "loss": 5.054,
      "mean_token_accuracy": 0.205801622569561,
      "num_tokens": 40915269.0,
      "step": 19710
    },
    {
      "entropy": 5.52552809715271,
      "epoch": 1.7883708272859216,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046818795308569815,
      "loss": 5.0581,
      "mean_token_accuracy": 0.20193398147821426,
      "num_tokens": 40925987.0,
      "step": 19715
    },
    {
      "entropy": 5.469293165206909,
      "epoch": 1.7888243831640058,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004681713644065133,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2131742060184479,
      "num_tokens": 40936130.0,
      "step": 19720
    },
    {
      "entropy": 5.439738702774048,
      "epoch": 1.78927793904209,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004681547717324954,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2143712431192398,
      "num_tokens": 40945476.0,
      "step": 19725
    },
    {
      "entropy": 5.4216382026672365,
      "epoch": 1.7897314949201741,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046813817506398755,
      "loss": 4.8598,
      "mean_token_accuracy": 0.21695656776428224,
      "num_tokens": 40956492.0,
      "step": 19730
    },
    {
      "entropy": 5.554205465316772,
      "epoch": 1.7901850507982582,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046812157440133303,
      "loss": 5.0703,
      "mean_token_accuracy": 0.204265259206295,
      "num_tokens": 40966990.0,
      "step": 19735
    },
    {
      "entropy": 5.551835060119629,
      "epoch": 1.7906386066763424,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004681049697448752,
      "loss": 4.9394,
      "mean_token_accuracy": 0.20988238602876663,
      "num_tokens": 40977065.0,
      "step": 19740
    },
    {
      "entropy": 5.4898449897766115,
      "epoch": 1.7910921625544267,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046808836109495734,
      "loss": 4.9952,
      "mean_token_accuracy": 0.20649953335523605,
      "num_tokens": 40987207.0,
      "step": 19745
    },
    {
      "entropy": 5.484203433990478,
      "epoch": 1.791545718432511,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004680717484519229,
      "loss": 4.9492,
      "mean_token_accuracy": 0.20586683303117753,
      "num_tokens": 40997541.0,
      "step": 19750
    },
    {
      "entropy": 5.488790988922119,
      "epoch": 1.791999274310595,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004680551318161156,
      "loss": 4.975,
      "mean_token_accuracy": 0.21097332686185838,
      "num_tokens": 41009044.0,
      "step": 19755
    },
    {
      "entropy": 5.525434255599976,
      "epoch": 1.7924528301886793,
      "grad_norm": 1.0,
      "learning_rate": 0.00046803851118787897,
      "loss": 4.9968,
      "mean_token_accuracy": 0.20461712777614594,
      "num_tokens": 41018165.0,
      "step": 19760
    },
    {
      "entropy": 5.474305772781372,
      "epoch": 1.7929063860667633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004680218865675567,
      "loss": 4.9433,
      "mean_token_accuracy": 0.2071532815694809,
      "num_tokens": 41028972.0,
      "step": 19765
    },
    {
      "entropy": 5.486135721206665,
      "epoch": 1.7933599419448476,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004680052579554926,
      "loss": 4.9715,
      "mean_token_accuracy": 0.20814012885093688,
      "num_tokens": 41039552.0,
      "step": 19770
    },
    {
      "entropy": 5.596327066421509,
      "epoch": 1.7938134978229319,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004679886253520307,
      "loss": 5.058,
      "mean_token_accuracy": 0.19694051742553711,
      "num_tokens": 41049808.0,
      "step": 19775
    },
    {
      "entropy": 5.517556190490723,
      "epoch": 1.794267053701016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046797198875751474,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2051583707332611,
      "num_tokens": 41060574.0,
      "step": 19780
    },
    {
      "entropy": 5.589335012435913,
      "epoch": 1.7947206095791002,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004679553481722889,
      "loss": 5.1199,
      "mean_token_accuracy": 0.1991354376077652,
      "num_tokens": 41071239.0,
      "step": 19785
    },
    {
      "entropy": 5.57397494316101,
      "epoch": 1.7951741654571842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046793870359669733,
      "loss": 5.1171,
      "mean_token_accuracy": 0.2064829632639885,
      "num_tokens": 41080819.0,
      "step": 19790
    },
    {
      "entropy": 5.55178394317627,
      "epoch": 1.7956277213352685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004679220550310842,
      "loss": 4.9265,
      "mean_token_accuracy": 0.2165278434753418,
      "num_tokens": 41090457.0,
      "step": 19795
    },
    {
      "entropy": 5.514302015304565,
      "epoch": 1.7960812772133528,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004679054024757937,
      "loss": 4.9891,
      "mean_token_accuracy": 0.20996480882167817,
      "num_tokens": 41101210.0,
      "step": 19800
    },
    {
      "entropy": 5.510220861434936,
      "epoch": 1.796534833091437,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004678887459311704,
      "loss": 4.9955,
      "mean_token_accuracy": 0.20979314297437668,
      "num_tokens": 41110566.0,
      "step": 19805
    },
    {
      "entropy": 5.441089677810669,
      "epoch": 1.796988388969521,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004678720853975587,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2132904589176178,
      "num_tokens": 41120000.0,
      "step": 19810
    },
    {
      "entropy": 5.455682373046875,
      "epoch": 1.797441944847605,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046785542087530307,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2143726408481598,
      "num_tokens": 41129944.0,
      "step": 19815
    },
    {
      "entropy": 5.472288274765015,
      "epoch": 1.7978955007256894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004678387523647483,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2070723608136177,
      "num_tokens": 41138803.0,
      "step": 19820
    },
    {
      "entropy": 5.520013809204102,
      "epoch": 1.7983490566037736,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046782207986623885,
      "loss": 5.0615,
      "mean_token_accuracy": 0.19517021626234055,
      "num_tokens": 41150365.0,
      "step": 19825
    },
    {
      "entropy": 5.610871505737305,
      "epoch": 1.798802612481858,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046780540338011974,
      "loss": 5.02,
      "mean_token_accuracy": 0.2008575588464737,
      "num_tokens": 41160310.0,
      "step": 19830
    },
    {
      "entropy": 5.503529834747314,
      "epoch": 1.799256168359942,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004677887229067358,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2091690182685852,
      "num_tokens": 41170506.0,
      "step": 19835
    },
    {
      "entropy": 5.534892129898071,
      "epoch": 1.799709724238026,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046777203844643184,
      "loss": 5.0202,
      "mean_token_accuracy": 0.19795064479112626,
      "num_tokens": 41181293.0,
      "step": 19840
    },
    {
      "entropy": 5.4575417041778564,
      "epoch": 1.8001632801161103,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046775534999955303,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2056989386677742,
      "num_tokens": 41191637.0,
      "step": 19845
    },
    {
      "entropy": 5.48751482963562,
      "epoch": 1.8006168359941945,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004677386575664445,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21417139172554017,
      "num_tokens": 41201833.0,
      "step": 19850
    },
    {
      "entropy": 5.5553830623626705,
      "epoch": 1.8010703918722788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046772196114745133,
      "loss": 5.0088,
      "mean_token_accuracy": 0.21154498308897018,
      "num_tokens": 41211243.0,
      "step": 19855
    },
    {
      "entropy": 5.537686347961426,
      "epoch": 1.8015239477503628,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004677052607429189,
      "loss": 4.972,
      "mean_token_accuracy": 0.21050416827201843,
      "num_tokens": 41221111.0,
      "step": 19860
    },
    {
      "entropy": 5.473995018005371,
      "epoch": 1.8019775036284469,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004676885563531927,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20612288415431976,
      "num_tokens": 41231578.0,
      "step": 19865
    },
    {
      "entropy": 5.500777292251587,
      "epoch": 1.8024310595065312,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046767184797861795,
      "loss": 5.0616,
      "mean_token_accuracy": 0.20313949137926102,
      "num_tokens": 41242130.0,
      "step": 19870
    },
    {
      "entropy": 5.556730842590332,
      "epoch": 1.8028846153846154,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046765513561954026,
      "loss": 5.0084,
      "mean_token_accuracy": 0.2023453250527382,
      "num_tokens": 41252885.0,
      "step": 19875
    },
    {
      "entropy": 5.555192470550537,
      "epoch": 1.8033381712626997,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046763841927630534,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20067857652902604,
      "num_tokens": 41263532.0,
      "step": 19880
    },
    {
      "entropy": 5.549077892303467,
      "epoch": 1.8037917271407837,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046762169894925876,
      "loss": 4.9968,
      "mean_token_accuracy": 0.2067990019917488,
      "num_tokens": 41273469.0,
      "step": 19885
    },
    {
      "entropy": 5.524668884277344,
      "epoch": 1.8042452830188678,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004676049746387464,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2050783470273018,
      "num_tokens": 41283663.0,
      "step": 19890
    },
    {
      "entropy": 5.477745485305786,
      "epoch": 1.804698838896952,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004675882463451141,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2139314129948616,
      "num_tokens": 41294657.0,
      "step": 19895
    },
    {
      "entropy": 5.536194944381714,
      "epoch": 1.8051523947750363,
      "grad_norm": 1.125,
      "learning_rate": 0.0004675715140687078,
      "loss": 4.9974,
      "mean_token_accuracy": 0.21073366701602936,
      "num_tokens": 41305124.0,
      "step": 19900
    },
    {
      "entropy": 5.465384435653687,
      "epoch": 1.8056059506531206,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046755477780987344,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21205850392580033,
      "num_tokens": 41315839.0,
      "step": 19905
    },
    {
      "entropy": 5.5487367630004885,
      "epoch": 1.8060595065312046,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046753803756895726,
      "loss": 5.0322,
      "mean_token_accuracy": 0.20375055372714995,
      "num_tokens": 41326388.0,
      "step": 19910
    },
    {
      "entropy": 5.518201732635498,
      "epoch": 1.8065130624092887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004675212933463054,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19393086135387422,
      "num_tokens": 41337333.0,
      "step": 19915
    },
    {
      "entropy": 5.582690715789795,
      "epoch": 1.806966618287373,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004675045451422641,
      "loss": 5.0606,
      "mean_token_accuracy": 0.19874513745307923,
      "num_tokens": 41348353.0,
      "step": 19920
    },
    {
      "entropy": 5.561363458633423,
      "epoch": 1.8074201741654572,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046748779295717973,
      "loss": 5.0298,
      "mean_token_accuracy": 0.21072236597537994,
      "num_tokens": 41358797.0,
      "step": 19925
    },
    {
      "entropy": 5.538416910171509,
      "epoch": 1.8078737300435415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004674710367913989,
      "loss": 5.015,
      "mean_token_accuracy": 0.20656365007162095,
      "num_tokens": 41368818.0,
      "step": 19930
    },
    {
      "entropy": 5.585899019241333,
      "epoch": 1.8083272859216255,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004674542766452678,
      "loss": 4.9961,
      "mean_token_accuracy": 0.20926343500614167,
      "num_tokens": 41378949.0,
      "step": 19935
    },
    {
      "entropy": 5.547982454299927,
      "epoch": 1.8087808417997098,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046743751251913327,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2045910596847534,
      "num_tokens": 41388892.0,
      "step": 19940
    },
    {
      "entropy": 5.550196695327759,
      "epoch": 1.8092343976777938,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046742074441334197,
      "loss": 4.9859,
      "mean_token_accuracy": 0.2078236907720566,
      "num_tokens": 41399197.0,
      "step": 19945
    },
    {
      "entropy": 5.521719932556152,
      "epoch": 1.809687953555878,
      "grad_norm": 1.125,
      "learning_rate": 0.0004674039723282406,
      "loss": 5.0266,
      "mean_token_accuracy": 0.20002949237823486,
      "num_tokens": 41409551.0,
      "step": 19950
    },
    {
      "entropy": 5.481037569046021,
      "epoch": 1.8101415094339623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046738719626417607,
      "loss": 4.9289,
      "mean_token_accuracy": 0.2104635313153267,
      "num_tokens": 41418900.0,
      "step": 19955
    },
    {
      "entropy": 5.483044624328613,
      "epoch": 1.8105950653120464,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046737041622149536,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20926041752099991,
      "num_tokens": 41428204.0,
      "step": 19960
    },
    {
      "entropy": 5.52245979309082,
      "epoch": 1.8110486211901307,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004673536322005454,
      "loss": 5.017,
      "mean_token_accuracy": 0.20543889254331588,
      "num_tokens": 41437922.0,
      "step": 19965
    },
    {
      "entropy": 5.540520906448364,
      "epoch": 1.8115021770682147,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004673368442016732,
      "loss": 5.0033,
      "mean_token_accuracy": 0.20341960489749908,
      "num_tokens": 41447411.0,
      "step": 19970
    },
    {
      "entropy": 5.578132677078247,
      "epoch": 1.811955732946299,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004673200522252261,
      "loss": 5.0826,
      "mean_token_accuracy": 0.20498971045017242,
      "num_tokens": 41458756.0,
      "step": 19975
    },
    {
      "entropy": 5.543388175964355,
      "epoch": 1.8124092888243832,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004673032562715514,
      "loss": 4.9879,
      "mean_token_accuracy": 0.20887200832366942,
      "num_tokens": 41468610.0,
      "step": 19980
    },
    {
      "entropy": 5.527509593963623,
      "epoch": 1.8128628447024675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046728645634099627,
      "loss": 5.0901,
      "mean_token_accuracy": 0.1987825319170952,
      "num_tokens": 41479849.0,
      "step": 19985
    },
    {
      "entropy": 5.515531539916992,
      "epoch": 1.8133164005805515,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046726965243390825,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20240487307310104,
      "num_tokens": 41490945.0,
      "step": 19990
    },
    {
      "entropy": 5.470834827423095,
      "epoch": 1.8137699564586356,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004672528445506348,
      "loss": 4.9543,
      "mean_token_accuracy": 0.20127549171447753,
      "num_tokens": 41501479.0,
      "step": 19995
    },
    {
      "entropy": 5.51609525680542,
      "epoch": 1.8142235123367199,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004672360326915236,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21051958948373795,
      "num_tokens": 41512447.0,
      "step": 20000
    },
    {
      "entropy": 5.462694454193115,
      "epoch": 1.8146770682148041,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004672192168569222,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21520113497972487,
      "num_tokens": 41521905.0,
      "step": 20005
    },
    {
      "entropy": 5.471878480911255,
      "epoch": 1.8151306240928884,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046720239704717845,
      "loss": 5.0065,
      "mean_token_accuracy": 0.21198217272758485,
      "num_tokens": 41533143.0,
      "step": 20010
    },
    {
      "entropy": 5.482673501968383,
      "epoch": 1.8155841799709724,
      "grad_norm": 1.125,
      "learning_rate": 0.00046718557326264,
      "loss": 4.9443,
      "mean_token_accuracy": 0.20825288891792298,
      "num_tokens": 41542234.0,
      "step": 20015
    },
    {
      "entropy": 5.545076704025268,
      "epoch": 1.8160377358490565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004671687455036551,
      "loss": 5.0637,
      "mean_token_accuracy": 0.20356175005435945,
      "num_tokens": 41552632.0,
      "step": 20020
    },
    {
      "entropy": 5.518356990814209,
      "epoch": 1.8164912917271407,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004671519137705715,
      "loss": 5.0037,
      "mean_token_accuracy": 0.20862066149711608,
      "num_tokens": 41563723.0,
      "step": 20025
    },
    {
      "entropy": 5.527810955047608,
      "epoch": 1.816944847605225,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046713507806373736,
      "loss": 4.9699,
      "mean_token_accuracy": 0.20548459142446518,
      "num_tokens": 41573680.0,
      "step": 20030
    },
    {
      "entropy": 5.520146417617798,
      "epoch": 1.8173984034833093,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046711823838350086,
      "loss": 5.0348,
      "mean_token_accuracy": 0.20444160848855972,
      "num_tokens": 41584722.0,
      "step": 20035
    },
    {
      "entropy": 5.511092567443848,
      "epoch": 1.8178519593613933,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046710139473021025,
      "loss": 5.0741,
      "mean_token_accuracy": 0.19642874598503113,
      "num_tokens": 41595785.0,
      "step": 20040
    },
    {
      "entropy": 5.545671176910401,
      "epoch": 1.8183055152394774,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046708454710421376,
      "loss": 5.0911,
      "mean_token_accuracy": 0.20113495439291001,
      "num_tokens": 41606023.0,
      "step": 20045
    },
    {
      "entropy": 5.514336347579956,
      "epoch": 1.8187590711175616,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046706769550586,
      "loss": 4.9941,
      "mean_token_accuracy": 0.20624094307422638,
      "num_tokens": 41616010.0,
      "step": 20050
    },
    {
      "entropy": 5.530388689041137,
      "epoch": 1.819212626995646,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046705083993549723,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21033221781253814,
      "num_tokens": 41625105.0,
      "step": 20055
    },
    {
      "entropy": 5.495745611190796,
      "epoch": 1.8196661828737302,
      "grad_norm": 1.125,
      "learning_rate": 0.0004670339803934743,
      "loss": 4.9984,
      "mean_token_accuracy": 0.2080562338232994,
      "num_tokens": 41634989.0,
      "step": 20060
    },
    {
      "entropy": 5.465115642547607,
      "epoch": 1.8201197387518142,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004670171168801396,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2078110843896866,
      "num_tokens": 41645736.0,
      "step": 20065
    },
    {
      "entropy": 5.54468560218811,
      "epoch": 1.8205732946298983,
      "grad_norm": 1.078125,
      "learning_rate": 0.000467000249395842,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20836457312107087,
      "num_tokens": 41655505.0,
      "step": 20070
    },
    {
      "entropy": 5.486360883712768,
      "epoch": 1.8210268505079825,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004669833779409303,
      "loss": 4.9821,
      "mean_token_accuracy": 0.2087858200073242,
      "num_tokens": 41665732.0,
      "step": 20075
    },
    {
      "entropy": 5.493438863754273,
      "epoch": 1.8214804063860668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004669665025157535,
      "loss": 4.9567,
      "mean_token_accuracy": 0.2020236074924469,
      "num_tokens": 41676050.0,
      "step": 20080
    },
    {
      "entropy": 5.591602611541748,
      "epoch": 1.821933962264151,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046694962312066046,
      "loss": 5.1363,
      "mean_token_accuracy": 0.2001710593700409,
      "num_tokens": 41686787.0,
      "step": 20085
    },
    {
      "entropy": 5.499366092681885,
      "epoch": 1.822387518142235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046693273975600036,
      "loss": 4.9289,
      "mean_token_accuracy": 0.20800776779651642,
      "num_tokens": 41697168.0,
      "step": 20090
    },
    {
      "entropy": 5.476192235946655,
      "epoch": 1.8228410740203191,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004669158524221222,
      "loss": 5.0097,
      "mean_token_accuracy": 0.21236103475093843,
      "num_tokens": 41707524.0,
      "step": 20095
    },
    {
      "entropy": 5.556261444091797,
      "epoch": 1.8232946298984034,
      "grad_norm": 1.0,
      "learning_rate": 0.00046689896111937537,
      "loss": 5.0497,
      "mean_token_accuracy": 0.20973631888628005,
      "num_tokens": 41718109.0,
      "step": 20100
    },
    {
      "entropy": 5.610829162597656,
      "epoch": 1.8237481857764877,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004668820658481091,
      "loss": 5.0258,
      "mean_token_accuracy": 0.20574668049812317,
      "num_tokens": 41728862.0,
      "step": 20105
    },
    {
      "entropy": 5.622160243988037,
      "epoch": 1.824201741654572,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046686516660867283,
      "loss": 5.0882,
      "mean_token_accuracy": 0.1992063581943512,
      "num_tokens": 41739024.0,
      "step": 20110
    },
    {
      "entropy": 5.53470139503479,
      "epoch": 1.824655297532656,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004668482634014159,
      "loss": 5.0295,
      "mean_token_accuracy": 0.20258474349975586,
      "num_tokens": 41748979.0,
      "step": 20115
    },
    {
      "entropy": 5.479855966567993,
      "epoch": 1.8251088534107403,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004668313562266882,
      "loss": 5.0228,
      "mean_token_accuracy": 0.20993564575910567,
      "num_tokens": 41759972.0,
      "step": 20120
    },
    {
      "entropy": 5.501933145523071,
      "epoch": 1.8255624092888243,
      "grad_norm": 1.0625,
      "learning_rate": 0.000466814445084839,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21612322181463242,
      "num_tokens": 41770150.0,
      "step": 20125
    },
    {
      "entropy": 5.525883293151855,
      "epoch": 1.8260159651669086,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004667975299762183,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2107188180088997,
      "num_tokens": 41780405.0,
      "step": 20130
    },
    {
      "entropy": 5.475947761535645,
      "epoch": 1.8264695210449928,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046678061090117576,
      "loss": 4.903,
      "mean_token_accuracy": 0.21209864616394042,
      "num_tokens": 41791269.0,
      "step": 20135
    },
    {
      "entropy": 5.428743648529053,
      "epoch": 1.8269230769230769,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004667636878600613,
      "loss": 4.8996,
      "mean_token_accuracy": 0.21087191104888917,
      "num_tokens": 41802513.0,
      "step": 20140
    },
    {
      "entropy": 5.56286849975586,
      "epoch": 1.8273766328011611,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004667467608532249,
      "loss": 5.0565,
      "mean_token_accuracy": 0.19700977504253386,
      "num_tokens": 41814149.0,
      "step": 20145
    },
    {
      "entropy": 5.561482286453247,
      "epoch": 1.8278301886792452,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004667298298810167,
      "loss": 5.024,
      "mean_token_accuracy": 0.20162852555513383,
      "num_tokens": 41825763.0,
      "step": 20150
    },
    {
      "entropy": 5.500921678543091,
      "epoch": 1.8282837445573294,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004667128949437867,
      "loss": 4.9981,
      "mean_token_accuracy": 0.20902491807937623,
      "num_tokens": 41834543.0,
      "step": 20155
    },
    {
      "entropy": 5.523082876205445,
      "epoch": 1.8287373004354137,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004666959560418852,
      "loss": 4.9123,
      "mean_token_accuracy": 0.219485867023468,
      "num_tokens": 41844940.0,
      "step": 20160
    },
    {
      "entropy": 5.4727842807769775,
      "epoch": 1.829190856313498,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046667901317566245,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21303580403327943,
      "num_tokens": 41855532.0,
      "step": 20165
    },
    {
      "entropy": 5.5545471668243405,
      "epoch": 1.829644412191582,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004666620663454689,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20356351137161255,
      "num_tokens": 41865661.0,
      "step": 20170
    },
    {
      "entropy": 5.5023308277130125,
      "epoch": 1.830097968069666,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046664511555165496,
      "loss": 5.0179,
      "mean_token_accuracy": 0.2042405366897583,
      "num_tokens": 41875662.0,
      "step": 20175
    },
    {
      "entropy": 5.495385313034058,
      "epoch": 1.8305515239477503,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046662816079457114,
      "loss": 4.9496,
      "mean_token_accuracy": 0.21413073390722276,
      "num_tokens": 41886211.0,
      "step": 20180
    },
    {
      "entropy": 5.605622100830078,
      "epoch": 1.8310050798258346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004666112020745682,
      "loss": 5.0966,
      "mean_token_accuracy": 0.19857825338840485,
      "num_tokens": 41896737.0,
      "step": 20185
    },
    {
      "entropy": 5.522352790832519,
      "epoch": 1.8314586357039189,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004665942393919967,
      "loss": 4.9692,
      "mean_token_accuracy": 0.2082372471690178,
      "num_tokens": 41907768.0,
      "step": 20190
    },
    {
      "entropy": 5.516893434524536,
      "epoch": 1.831912191582003,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046657727274720755,
      "loss": 4.9448,
      "mean_token_accuracy": 0.20389573127031327,
      "num_tokens": 41917587.0,
      "step": 20195
    },
    {
      "entropy": 5.451837539672852,
      "epoch": 1.832365747460087,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004665603021405516,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21552098989486695,
      "num_tokens": 41927842.0,
      "step": 20200
    },
    {
      "entropy": 5.450876522064209,
      "epoch": 1.8328193033381712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004665433275723798,
      "loss": 4.9512,
      "mean_token_accuracy": 0.21344021558761597,
      "num_tokens": 41939899.0,
      "step": 20205
    },
    {
      "entropy": 5.500430679321289,
      "epoch": 1.8332728592162555,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004665263490430431,
      "loss": 4.9616,
      "mean_token_accuracy": 0.20242294818162918,
      "num_tokens": 41950932.0,
      "step": 20210
    },
    {
      "entropy": 5.546673393249511,
      "epoch": 1.8337264150943398,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004665093665528928,
      "loss": 5.0306,
      "mean_token_accuracy": 0.20848318785429001,
      "num_tokens": 41963512.0,
      "step": 20215
    },
    {
      "entropy": 5.593548536300659,
      "epoch": 1.8341799709724238,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046649238010227983,
      "loss": 5.027,
      "mean_token_accuracy": 0.20841220915317535,
      "num_tokens": 41974677.0,
      "step": 20220
    },
    {
      "entropy": 5.452464294433594,
      "epoch": 1.8346335268505078,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004664753896915558,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21078637689352037,
      "num_tokens": 41984582.0,
      "step": 20225
    },
    {
      "entropy": 5.492018222808838,
      "epoch": 1.8350870827285921,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004664583953210719,
      "loss": 4.9329,
      "mean_token_accuracy": 0.2051735356450081,
      "num_tokens": 41994624.0,
      "step": 20230
    },
    {
      "entropy": 5.518248462677002,
      "epoch": 1.8355406386066764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004664413969911795,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21068610548973082,
      "num_tokens": 42004835.0,
      "step": 20235
    },
    {
      "entropy": 5.446583843231201,
      "epoch": 1.8359941944847606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004664243947022303,
      "loss": 4.95,
      "mean_token_accuracy": 0.2101311355829239,
      "num_tokens": 42014727.0,
      "step": 20240
    },
    {
      "entropy": 5.512728643417359,
      "epoch": 1.8364477503628447,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046640738845457586,
      "loss": 5.0046,
      "mean_token_accuracy": 0.20640296787023543,
      "num_tokens": 42023628.0,
      "step": 20245
    },
    {
      "entropy": 5.55571870803833,
      "epoch": 1.8369013062409287,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004663903782485678,
      "loss": 5.0451,
      "mean_token_accuracy": 0.2015094056725502,
      "num_tokens": 42033587.0,
      "step": 20250
    },
    {
      "entropy": 5.564340829849243,
      "epoch": 1.837354862119013,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046637336408455793,
      "loss": 5.0152,
      "mean_token_accuracy": 0.2039337232708931,
      "num_tokens": 42043765.0,
      "step": 20255
    },
    {
      "entropy": 5.581139183044433,
      "epoch": 1.8378084179970973,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046635634596289816,
      "loss": 5.0727,
      "mean_token_accuracy": 0.2069259002804756,
      "num_tokens": 42054205.0,
      "step": 20260
    },
    {
      "entropy": 5.458800506591797,
      "epoch": 1.8382619738751815,
      "grad_norm": 1.0,
      "learning_rate": 0.0004663393238839404,
      "loss": 4.8914,
      "mean_token_accuracy": 0.21075702756643294,
      "num_tokens": 42064512.0,
      "step": 20265
    },
    {
      "entropy": 5.478949785232544,
      "epoch": 1.8387155297532656,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004663222978480366,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20276922285556792,
      "num_tokens": 42074954.0,
      "step": 20270
    },
    {
      "entropy": 5.5394398212432865,
      "epoch": 1.8391690856313496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000466305267855539,
      "loss": 5.0033,
      "mean_token_accuracy": 0.2032952278852463,
      "num_tokens": 42084914.0,
      "step": 20275
    },
    {
      "entropy": 5.5595615863800045,
      "epoch": 1.8396226415094339,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004662882339067996,
      "loss": 5.0218,
      "mean_token_accuracy": 0.20879424512386321,
      "num_tokens": 42095454.0,
      "step": 20280
    },
    {
      "entropy": 5.524030351638794,
      "epoch": 1.8400761973875182,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004662711960021709,
      "loss": 5.0714,
      "mean_token_accuracy": 0.1993553966283798,
      "num_tokens": 42106960.0,
      "step": 20285
    },
    {
      "entropy": 5.489474582672119,
      "epoch": 1.8405297532656024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046625415414200513,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2108674466609955,
      "num_tokens": 42117009.0,
      "step": 20290
    },
    {
      "entropy": 5.548690843582153,
      "epoch": 1.8409833091436865,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004662371083266546,
      "loss": 4.9778,
      "mean_token_accuracy": 0.20959905236959459,
      "num_tokens": 42126371.0,
      "step": 20295
    },
    {
      "entropy": 5.560553932189942,
      "epoch": 1.8414368650217707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000466220058556472,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20469729453325272,
      "num_tokens": 42135890.0,
      "step": 20300
    },
    {
      "entropy": 5.5765221118927,
      "epoch": 1.8418904208998548,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046620300483180986,
      "loss": 5.1176,
      "mean_token_accuracy": 0.19918676763772963,
      "num_tokens": 42145696.0,
      "step": 20305
    },
    {
      "entropy": 5.524145030975342,
      "epoch": 1.842343976777939,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004661859471530209,
      "loss": 5.0355,
      "mean_token_accuracy": 0.20542675703763963,
      "num_tokens": 42156236.0,
      "step": 20310
    },
    {
      "entropy": 5.587287187576294,
      "epoch": 1.8427975326560233,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004661688855204578,
      "loss": 5.0547,
      "mean_token_accuracy": 0.2009128749370575,
      "num_tokens": 42166179.0,
      "step": 20315
    },
    {
      "entropy": 5.493165826797485,
      "epoch": 1.8432510885341074,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046615181993447344,
      "loss": 4.9283,
      "mean_token_accuracy": 0.21054396480321885,
      "num_tokens": 42176643.0,
      "step": 20320
    },
    {
      "entropy": 5.557625150680542,
      "epoch": 1.8437046444121916,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004661347503954207,
      "loss": 5.1835,
      "mean_token_accuracy": 0.19340878576040268,
      "num_tokens": 42188328.0,
      "step": 20325
    },
    {
      "entropy": 5.5738866329193115,
      "epoch": 1.8441582002902757,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046611767690365265,
      "loss": 5.0129,
      "mean_token_accuracy": 0.2106790453195572,
      "num_tokens": 42198110.0,
      "step": 20330
    },
    {
      "entropy": 5.481729221343994,
      "epoch": 1.84461175616836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004661005994595223,
      "loss": 4.9659,
      "mean_token_accuracy": 0.20133370757102967,
      "num_tokens": 42208438.0,
      "step": 20335
    },
    {
      "entropy": 5.525497341156006,
      "epoch": 1.8450653120464442,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004660835180633829,
      "loss": 4.9576,
      "mean_token_accuracy": 0.20635638386011124,
      "num_tokens": 42220052.0,
      "step": 20340
    },
    {
      "entropy": 5.51782603263855,
      "epoch": 1.8455188679245285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004660664327155876,
      "loss": 4.9629,
      "mean_token_accuracy": 0.20845620036125184,
      "num_tokens": 42229497.0,
      "step": 20345
    },
    {
      "entropy": 5.49897723197937,
      "epoch": 1.8459724238026125,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004660493434164898,
      "loss": 5.0291,
      "mean_token_accuracy": 0.1970988929271698,
      "num_tokens": 42240725.0,
      "step": 20350
    },
    {
      "entropy": 5.582056856155395,
      "epoch": 1.8464259796806965,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004660322501664429,
      "loss": 5.0339,
      "mean_token_accuracy": 0.20600544661283493,
      "num_tokens": 42251114.0,
      "step": 20355
    },
    {
      "entropy": 5.61389651298523,
      "epoch": 1.8468795355587808,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004660151529658004,
      "loss": 5.1813,
      "mean_token_accuracy": 0.1986292600631714,
      "num_tokens": 42262825.0,
      "step": 20360
    },
    {
      "entropy": 5.571655797958374,
      "epoch": 1.847333091436865,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004659980518149158,
      "loss": 5.0061,
      "mean_token_accuracy": 0.2176257774233818,
      "num_tokens": 42272176.0,
      "step": 20365
    },
    {
      "entropy": 5.56455545425415,
      "epoch": 1.8477866473149493,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046598094671414285,
      "loss": 5.0578,
      "mean_token_accuracy": 0.2081517606973648,
      "num_tokens": 42281934.0,
      "step": 20370
    },
    {
      "entropy": 5.537998533248901,
      "epoch": 1.8482402031930334,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046596383766383533,
      "loss": 4.9471,
      "mean_token_accuracy": 0.20919624269008635,
      "num_tokens": 42291927.0,
      "step": 20375
    },
    {
      "entropy": 5.476874446868896,
      "epoch": 1.8486937590711174,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004659467246643469,
      "loss": 4.9406,
      "mean_token_accuracy": 0.20838529020547866,
      "num_tokens": 42301407.0,
      "step": 20380
    },
    {
      "entropy": 5.437132501602173,
      "epoch": 1.8491473149492017,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046592960771603154,
      "loss": 4.8964,
      "mean_token_accuracy": 0.21864065825939177,
      "num_tokens": 42311199.0,
      "step": 20385
    },
    {
      "entropy": 5.498507404327393,
      "epoch": 1.849600870827286,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004659124868192433,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21190747767686843,
      "num_tokens": 42320725.0,
      "step": 20390
    },
    {
      "entropy": 5.598555183410644,
      "epoch": 1.8500544267053702,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004658953619743361,
      "loss": 5.0838,
      "mean_token_accuracy": 0.20678532719612122,
      "num_tokens": 42330857.0,
      "step": 20395
    },
    {
      "entropy": 5.48690185546875,
      "epoch": 1.8505079825834543,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046587823318166427,
      "loss": 4.9036,
      "mean_token_accuracy": 0.20804093033075333,
      "num_tokens": 42340241.0,
      "step": 20400
    },
    {
      "entropy": 5.549601602554321,
      "epoch": 1.8509615384615383,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046586110044158184,
      "loss": 5.1084,
      "mean_token_accuracy": 0.1954648569226265,
      "num_tokens": 42351756.0,
      "step": 20405
    },
    {
      "entropy": 5.530306148529053,
      "epoch": 1.8514150943396226,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046584396375444335,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21770662665367127,
      "num_tokens": 42361649.0,
      "step": 20410
    },
    {
      "entropy": 5.52125186920166,
      "epoch": 1.8518686502177069,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046582682312060295,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21299203336238862,
      "num_tokens": 42371273.0,
      "step": 20415
    },
    {
      "entropy": 5.460595750808716,
      "epoch": 1.8523222060957911,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004658096785404152,
      "loss": 4.9563,
      "mean_token_accuracy": 0.2060428649187088,
      "num_tokens": 42382275.0,
      "step": 20420
    },
    {
      "entropy": 5.58466420173645,
      "epoch": 1.8527757619738752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046579253001423473,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20317630022764205,
      "num_tokens": 42392360.0,
      "step": 20425
    },
    {
      "entropy": 5.54338641166687,
      "epoch": 1.8532293178519592,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046577537754241604,
      "loss": 5.0026,
      "mean_token_accuracy": 0.20023627877235411,
      "num_tokens": 42403695.0,
      "step": 20430
    },
    {
      "entropy": 5.518084383010864,
      "epoch": 1.8536828737300435,
      "grad_norm": 1.0,
      "learning_rate": 0.000465758221125314,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21301815062761306,
      "num_tokens": 42414199.0,
      "step": 20435
    },
    {
      "entropy": 5.533279275894165,
      "epoch": 1.8541364296081277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046574106076328333,
      "loss": 5.0558,
      "mean_token_accuracy": 0.20730967223644256,
      "num_tokens": 42423655.0,
      "step": 20440
    },
    {
      "entropy": 5.498628425598144,
      "epoch": 1.854589985486212,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004657238964566789,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2135024517774582,
      "num_tokens": 42434235.0,
      "step": 20445
    },
    {
      "entropy": 5.472362375259399,
      "epoch": 1.855043541364296,
      "grad_norm": 1.0,
      "learning_rate": 0.0004657067282058557,
      "loss": 4.9913,
      "mean_token_accuracy": 0.20742813795804976,
      "num_tokens": 42444701.0,
      "step": 20450
    },
    {
      "entropy": 5.465218162536621,
      "epoch": 1.85549709724238,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004656895560111687,
      "loss": 4.981,
      "mean_token_accuracy": 0.2064000517129898,
      "num_tokens": 42454499.0,
      "step": 20455
    },
    {
      "entropy": 5.561305713653565,
      "epoch": 1.8559506531204644,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046567237987297323,
      "loss": 4.9639,
      "mean_token_accuracy": 0.19911492466926575,
      "num_tokens": 42465391.0,
      "step": 20460
    },
    {
      "entropy": 5.579448318481445,
      "epoch": 1.8564042089985486,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004656551997916243,
      "loss": 4.9686,
      "mean_token_accuracy": 0.19929346442222595,
      "num_tokens": 42476126.0,
      "step": 20465
    },
    {
      "entropy": 5.5623602867126465,
      "epoch": 1.856857764876633,
      "grad_norm": 1.0,
      "learning_rate": 0.00046563801576747716,
      "loss": 5.044,
      "mean_token_accuracy": 0.20417959094047547,
      "num_tokens": 42486833.0,
      "step": 20470
    },
    {
      "entropy": 5.613506937026978,
      "epoch": 1.857311320754717,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046562082780088733,
      "loss": 5.071,
      "mean_token_accuracy": 0.20036802142858506,
      "num_tokens": 42496986.0,
      "step": 20475
    },
    {
      "entropy": 5.545318651199341,
      "epoch": 1.8577648766328012,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004656036358922103,
      "loss": 4.9858,
      "mean_token_accuracy": 0.206824791431427,
      "num_tokens": 42507352.0,
      "step": 20480
    },
    {
      "entropy": 5.529227924346924,
      "epoch": 1.8582184325108853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004655864400418014,
      "loss": 4.9817,
      "mean_token_accuracy": 0.2064192160964012,
      "num_tokens": 42517655.0,
      "step": 20485
    },
    {
      "entropy": 5.485933399200439,
      "epoch": 1.8586719883889695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004655692402500163,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21793061941862107,
      "num_tokens": 42527502.0,
      "step": 20490
    },
    {
      "entropy": 5.499330377578735,
      "epoch": 1.8591255442670538,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046555203651721085,
      "loss": 4.9677,
      "mean_token_accuracy": 0.20745202898979187,
      "num_tokens": 42538893.0,
      "step": 20495
    },
    {
      "entropy": 5.418507289886475,
      "epoch": 1.8595791001451378,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004655348288437407,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22026781141757965,
      "num_tokens": 42550372.0,
      "step": 20500
    },
    {
      "entropy": 5.5770313262939455,
      "epoch": 1.860032656023222,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004655176172299618,
      "loss": 5.0655,
      "mean_token_accuracy": 0.20463417172431947,
      "num_tokens": 42560814.0,
      "step": 20505
    },
    {
      "entropy": 5.595072937011719,
      "epoch": 1.8604862119013061,
      "grad_norm": 1.515625,
      "learning_rate": 0.00046550040167622986,
      "loss": 5.0803,
      "mean_token_accuracy": 0.1982729837298393,
      "num_tokens": 42570993.0,
      "step": 20510
    },
    {
      "entropy": 5.5297078609466555,
      "epoch": 1.8609397677793904,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046548318218290124,
      "loss": 4.9382,
      "mean_token_accuracy": 0.20696137696504593,
      "num_tokens": 42581357.0,
      "step": 20515
    },
    {
      "entropy": 5.518084716796875,
      "epoch": 1.8613933236574747,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004654659587503318,
      "loss": 5.068,
      "mean_token_accuracy": 0.19770045429468155,
      "num_tokens": 42591653.0,
      "step": 20520
    },
    {
      "entropy": 5.593356609344482,
      "epoch": 1.861846879535559,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004654487313788778,
      "loss": 5.0941,
      "mean_token_accuracy": 0.1982532650232315,
      "num_tokens": 42602232.0,
      "step": 20525
    },
    {
      "entropy": 5.536996698379516,
      "epoch": 1.862300435413643,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046543150006889546,
      "loss": 5.0001,
      "mean_token_accuracy": 0.21202207803726197,
      "num_tokens": 42613283.0,
      "step": 20530
    },
    {
      "entropy": 5.507025051116943,
      "epoch": 1.862753991291727,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004654142648207412,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2139369398355484,
      "num_tokens": 42623594.0,
      "step": 20535
    },
    {
      "entropy": 5.530036115646363,
      "epoch": 1.8632075471698113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004653970256347715,
      "loss": 5.041,
      "mean_token_accuracy": 0.20130124390125276,
      "num_tokens": 42633587.0,
      "step": 20540
    },
    {
      "entropy": 5.548380708694458,
      "epoch": 1.8636611030478956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046537978251134265,
      "loss": 4.9771,
      "mean_token_accuracy": 0.20500110536813737,
      "num_tokens": 42643426.0,
      "step": 20545
    },
    {
      "entropy": 5.490745925903321,
      "epoch": 1.8641146589259798,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004653625354508115,
      "loss": 4.9734,
      "mean_token_accuracy": 0.20890772193670273,
      "num_tokens": 42653521.0,
      "step": 20550
    },
    {
      "entropy": 5.5804942607879635,
      "epoch": 1.8645682148040639,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004653452844535346,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20665431767702103,
      "num_tokens": 42664818.0,
      "step": 20555
    },
    {
      "entropy": 5.613869380950928,
      "epoch": 1.865021770682148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046532802951986863,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20891079157590867,
      "num_tokens": 42674738.0,
      "step": 20560
    },
    {
      "entropy": 5.470247459411621,
      "epoch": 1.8654753265602322,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004653107706501705,
      "loss": 4.9667,
      "mean_token_accuracy": 0.2177759200334549,
      "num_tokens": 42685003.0,
      "step": 20565
    },
    {
      "entropy": 5.454487228393555,
      "epoch": 1.8659288824383164,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004652935078447973,
      "loss": 4.9985,
      "mean_token_accuracy": 0.21224944144487382,
      "num_tokens": 42695428.0,
      "step": 20570
    },
    {
      "entropy": 5.55644907951355,
      "epoch": 1.8663824383164007,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046527624110410574,
      "loss": 4.9615,
      "mean_token_accuracy": 0.21364315897226333,
      "num_tokens": 42704901.0,
      "step": 20575
    },
    {
      "entropy": 5.598369312286377,
      "epoch": 1.8668359941944848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000465258970428453,
      "loss": 5.0522,
      "mean_token_accuracy": 0.2018121674656868,
      "num_tokens": 42714379.0,
      "step": 20580
    },
    {
      "entropy": 5.59432315826416,
      "epoch": 1.8672895500725688,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046524169581819633,
      "loss": 5.153,
      "mean_token_accuracy": 0.19972582757472992,
      "num_tokens": 42724813.0,
      "step": 20585
    },
    {
      "entropy": 5.480282020568848,
      "epoch": 1.867743105950653,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004652244172736929,
      "loss": 4.994,
      "mean_token_accuracy": 0.20874302089214325,
      "num_tokens": 42736590.0,
      "step": 20590
    },
    {
      "entropy": 5.586142301559448,
      "epoch": 1.8681966618287373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004652071347953,
      "loss": 5.0821,
      "mean_token_accuracy": 0.19929955899715424,
      "num_tokens": 42746718.0,
      "step": 20595
    },
    {
      "entropy": 5.540536260604858,
      "epoch": 1.8686502177068216,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046518984838337513,
      "loss": 4.9541,
      "mean_token_accuracy": 0.20269346237182617,
      "num_tokens": 42756484.0,
      "step": 20600
    },
    {
      "entropy": 5.574197816848755,
      "epoch": 1.8691037735849056,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004651725580382757,
      "loss": 5.0577,
      "mean_token_accuracy": 0.2045188307762146,
      "num_tokens": 42768512.0,
      "step": 20605
    },
    {
      "entropy": 5.516998624801635,
      "epoch": 1.8695573294629897,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046515526376035934,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2083277702331543,
      "num_tokens": 42779812.0,
      "step": 20610
    },
    {
      "entropy": 5.427085828781128,
      "epoch": 1.870010885341074,
      "grad_norm": 1.0,
      "learning_rate": 0.0004651379655499837,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21540848016738892,
      "num_tokens": 42791163.0,
      "step": 20615
    },
    {
      "entropy": 5.494724988937378,
      "epoch": 1.8704644412191582,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046512066340750636,
      "loss": 4.9717,
      "mean_token_accuracy": 0.20653896182775497,
      "num_tokens": 42801810.0,
      "step": 20620
    },
    {
      "entropy": 5.5778765201568605,
      "epoch": 1.8709179970972425,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004651033573332853,
      "loss": 4.9995,
      "mean_token_accuracy": 0.2088393285870552,
      "num_tokens": 42813358.0,
      "step": 20625
    },
    {
      "entropy": 5.4334807872772215,
      "epoch": 1.8713715529753265,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004650860473276784,
      "loss": 4.8383,
      "mean_token_accuracy": 0.2156376376748085,
      "num_tokens": 42823519.0,
      "step": 20630
    },
    {
      "entropy": 5.545387315750122,
      "epoch": 1.8718251088534106,
      "grad_norm": 1.125,
      "learning_rate": 0.0004650687333910436,
      "loss": 5.0159,
      "mean_token_accuracy": 0.20554910302162172,
      "num_tokens": 42833257.0,
      "step": 20635
    },
    {
      "entropy": 5.571887588500976,
      "epoch": 1.8722786647314948,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046505141552373897,
      "loss": 5.0211,
      "mean_token_accuracy": 0.21233807355165482,
      "num_tokens": 42843536.0,
      "step": 20640
    },
    {
      "entropy": 5.523766040802002,
      "epoch": 1.8727322206095791,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004650340937261226,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20724039971828462,
      "num_tokens": 42853911.0,
      "step": 20645
    },
    {
      "entropy": 5.6062156677246096,
      "epoch": 1.8731857764876634,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004650167679985528,
      "loss": 5.0954,
      "mean_token_accuracy": 0.19527529627084733,
      "num_tokens": 42864173.0,
      "step": 20650
    },
    {
      "entropy": 5.5203619480133055,
      "epoch": 1.8736393323657474,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004649994383413877,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21267371773719787,
      "num_tokens": 42873470.0,
      "step": 20655
    },
    {
      "entropy": 5.506966209411621,
      "epoch": 1.8740928882438317,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046498210475498595,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21166199296712876,
      "num_tokens": 42882496.0,
      "step": 20660
    },
    {
      "entropy": 5.506556510925293,
      "epoch": 1.8745464441219157,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046496476723970574,
      "loss": 4.9433,
      "mean_token_accuracy": 0.20904543846845627,
      "num_tokens": 42893163.0,
      "step": 20665
    },
    {
      "entropy": 5.50634994506836,
      "epoch": 1.875,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004649474257959058,
      "loss": 4.965,
      "mean_token_accuracy": 0.20617899894714356,
      "num_tokens": 42903406.0,
      "step": 20670
    },
    {
      "entropy": 5.544222402572632,
      "epoch": 1.8754535558780843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004649300804239447,
      "loss": 4.9653,
      "mean_token_accuracy": 0.20702796578407287,
      "num_tokens": 42913702.0,
      "step": 20675
    },
    {
      "entropy": 5.484886598587036,
      "epoch": 1.8759071117561683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046491273112418105,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21404954195022582,
      "num_tokens": 42923877.0,
      "step": 20680
    },
    {
      "entropy": 5.499910020828247,
      "epoch": 1.8763606676342526,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046489537789697386,
      "loss": 4.9555,
      "mean_token_accuracy": 0.20722677558660507,
      "num_tokens": 42934393.0,
      "step": 20685
    },
    {
      "entropy": 5.5295860290527346,
      "epoch": 1.8768142235123366,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046487802074268177,
      "loss": 5.0285,
      "mean_token_accuracy": 0.21113871335983275,
      "num_tokens": 42945803.0,
      "step": 20690
    },
    {
      "entropy": 5.5067907810211185,
      "epoch": 1.8772677793904209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046486065966166377,
      "loss": 4.9527,
      "mean_token_accuracy": 0.20922113656997682,
      "num_tokens": 42955829.0,
      "step": 20695
    },
    {
      "entropy": 5.534265613555908,
      "epoch": 1.8777213352685052,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004648432946542791,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20132580101490022,
      "num_tokens": 42967080.0,
      "step": 20700
    },
    {
      "entropy": 5.487653923034668,
      "epoch": 1.8781748911465894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046482592572088654,
      "loss": 4.9695,
      "mean_token_accuracy": 0.20034733563661575,
      "num_tokens": 42978082.0,
      "step": 20705
    },
    {
      "entropy": 5.573316764831543,
      "epoch": 1.8786284470246735,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046480855286184565,
      "loss": 5.0414,
      "mean_token_accuracy": 0.20434023290872574,
      "num_tokens": 42988982.0,
      "step": 20710
    },
    {
      "entropy": 5.554499721527099,
      "epoch": 1.8790820029027575,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046479117607751535,
      "loss": 4.9838,
      "mean_token_accuracy": 0.19854393154382705,
      "num_tokens": 42998982.0,
      "step": 20715
    },
    {
      "entropy": 5.526091575622559,
      "epoch": 1.8795355587808418,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046477379536825534,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20879895836114884,
      "num_tokens": 43009574.0,
      "step": 20720
    },
    {
      "entropy": 5.548022603988647,
      "epoch": 1.879989114658926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004647564107344247,
      "loss": 4.9842,
      "mean_token_accuracy": 0.20474085807800294,
      "num_tokens": 43019335.0,
      "step": 20725
    },
    {
      "entropy": 5.468513154983521,
      "epoch": 1.8804426705370103,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046473902217638327,
      "loss": 5.0042,
      "mean_token_accuracy": 0.21021214723587037,
      "num_tokens": 43029264.0,
      "step": 20730
    },
    {
      "entropy": 5.4808114528656,
      "epoch": 1.8808962264150944,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046472162969449037,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2108668640255928,
      "num_tokens": 43040519.0,
      "step": 20735
    },
    {
      "entropy": 5.57932825088501,
      "epoch": 1.8813497822931784,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046470423328910594,
      "loss": 4.9967,
      "mean_token_accuracy": 0.21168433129787445,
      "num_tokens": 43051146.0,
      "step": 20740
    },
    {
      "entropy": 5.596247911453247,
      "epoch": 1.8818033381712627,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046468683296058957,
      "loss": 5.0823,
      "mean_token_accuracy": 0.20713951736688613,
      "num_tokens": 43061339.0,
      "step": 20745
    },
    {
      "entropy": 5.497734260559082,
      "epoch": 1.882256894049347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046466942870930126,
      "loss": 4.9941,
      "mean_token_accuracy": 0.2102876588702202,
      "num_tokens": 43072624.0,
      "step": 20750
    },
    {
      "entropy": 5.545460033416748,
      "epoch": 1.8827104499274312,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004646520205356007,
      "loss": 4.9513,
      "mean_token_accuracy": 0.20594087541103362,
      "num_tokens": 43082120.0,
      "step": 20755
    },
    {
      "entropy": 5.609503507614136,
      "epoch": 1.8831640058055152,
      "grad_norm": 1.25,
      "learning_rate": 0.00046463460843984805,
      "loss": 5.0671,
      "mean_token_accuracy": 0.20164793282747268,
      "num_tokens": 43093202.0,
      "step": 20760
    },
    {
      "entropy": 5.674992609024048,
      "epoch": 1.8836175616835993,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004646171924224034,
      "loss": 5.1496,
      "mean_token_accuracy": 0.20784732401371003,
      "num_tokens": 43104801.0,
      "step": 20765
    },
    {
      "entropy": 5.5489527702331545,
      "epoch": 1.8840711175616836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046459977248362687,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20929950177669526,
      "num_tokens": 43114855.0,
      "step": 20770
    },
    {
      "entropy": 5.533009958267212,
      "epoch": 1.8845246734397678,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004645823486238788,
      "loss": 4.974,
      "mean_token_accuracy": 0.20246503204107286,
      "num_tokens": 43124759.0,
      "step": 20775
    },
    {
      "entropy": 5.508164787292481,
      "epoch": 1.884978229317852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004645649208435193,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21345612406730652,
      "num_tokens": 43135526.0,
      "step": 20780
    },
    {
      "entropy": 5.540684652328491,
      "epoch": 1.8854317851959361,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046454748914290906,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20299570262432098,
      "num_tokens": 43145605.0,
      "step": 20785
    },
    {
      "entropy": 5.623768997192383,
      "epoch": 1.8858853410740202,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046453005352240835,
      "loss": 5.1192,
      "mean_token_accuracy": 0.19875621050596237,
      "num_tokens": 43156578.0,
      "step": 20790
    },
    {
      "entropy": 5.578191328048706,
      "epoch": 1.8863388969521044,
      "grad_norm": 1.078125,
      "learning_rate": 0.000464512613982378,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20559273660182953,
      "num_tokens": 43166870.0,
      "step": 20795
    },
    {
      "entropy": 5.5336119651794435,
      "epoch": 1.8867924528301887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004644951705231783,
      "loss": 4.947,
      "mean_token_accuracy": 0.21584009975194932,
      "num_tokens": 43176406.0,
      "step": 20800
    },
    {
      "entropy": 5.625538349151611,
      "epoch": 1.887246008708273,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046447772314517024,
      "loss": 5.1402,
      "mean_token_accuracy": 0.20117899626493455,
      "num_tokens": 43187502.0,
      "step": 20805
    },
    {
      "entropy": 5.471494913101196,
      "epoch": 1.887699564586357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046446027184871464,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21427054107189178,
      "num_tokens": 43197928.0,
      "step": 20810
    },
    {
      "entropy": 5.504427337646485,
      "epoch": 1.888153120464441,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046444281663417236,
      "loss": 4.9733,
      "mean_token_accuracy": 0.21687044948339462,
      "num_tokens": 43209083.0,
      "step": 20815
    },
    {
      "entropy": 5.551455545425415,
      "epoch": 1.8886066763425253,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004644253575019043,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20451149493455886,
      "num_tokens": 43219089.0,
      "step": 20820
    },
    {
      "entropy": 5.584152936935425,
      "epoch": 1.8890602322206096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004644078944522716,
      "loss": 5.0323,
      "mean_token_accuracy": 0.20532690286636351,
      "num_tokens": 43229440.0,
      "step": 20825
    },
    {
      "entropy": 5.519352865219116,
      "epoch": 1.8895137880986939,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004643904274856353,
      "loss": 4.9654,
      "mean_token_accuracy": 0.20806962847709656,
      "num_tokens": 43239995.0,
      "step": 20830
    },
    {
      "entropy": 5.4865211009979244,
      "epoch": 1.889967343976778,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046437295660235674,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21096756607294082,
      "num_tokens": 43249722.0,
      "step": 20835
    },
    {
      "entropy": 5.475346326828003,
      "epoch": 1.8904208998548622,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004643554818027972,
      "loss": 5.011,
      "mean_token_accuracy": 0.2061210662126541,
      "num_tokens": 43260029.0,
      "step": 20840
    },
    {
      "entropy": 5.555378055572509,
      "epoch": 1.8908744557329462,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004643380030873181,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2109837368130684,
      "num_tokens": 43270571.0,
      "step": 20845
    },
    {
      "entropy": 5.538992643356323,
      "epoch": 1.8913280116110305,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004643205204562808,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2108249172568321,
      "num_tokens": 43281310.0,
      "step": 20850
    },
    {
      "entropy": 5.512079238891602,
      "epoch": 1.8917815674891147,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046430303391004687,
      "loss": 4.9582,
      "mean_token_accuracy": 0.2100333169102669,
      "num_tokens": 43292904.0,
      "step": 20855
    },
    {
      "entropy": 5.523993253707886,
      "epoch": 1.8922351233671988,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004642855434489779,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20685950964689254,
      "num_tokens": 43303745.0,
      "step": 20860
    },
    {
      "entropy": 5.570832300186157,
      "epoch": 1.892688679245283,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046426804907343574,
      "loss": 5.0057,
      "mean_token_accuracy": 0.20969969779253006,
      "num_tokens": 43314652.0,
      "step": 20865
    },
    {
      "entropy": 5.496357393264771,
      "epoch": 1.893142235123367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046425055078378206,
      "loss": 4.8764,
      "mean_token_accuracy": 0.2160039648413658,
      "num_tokens": 43324242.0,
      "step": 20870
    },
    {
      "entropy": 5.48613805770874,
      "epoch": 1.8935957910014514,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046423304858037885,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21549898535013198,
      "num_tokens": 43334054.0,
      "step": 20875
    },
    {
      "entropy": 5.474690866470337,
      "epoch": 1.8940493468795356,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004642155424635878,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22431363314390182,
      "num_tokens": 43343389.0,
      "step": 20880
    },
    {
      "entropy": 5.483912801742553,
      "epoch": 1.89450290275762,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046419803243377125,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21698027998209,
      "num_tokens": 43353892.0,
      "step": 20885
    },
    {
      "entropy": 5.466065549850464,
      "epoch": 1.894956458635704,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046418051849129115,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21455236077308654,
      "num_tokens": 43363912.0,
      "step": 20890
    },
    {
      "entropy": 5.513872575759888,
      "epoch": 1.895410014513788,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004641630006365097,
      "loss": 5.0315,
      "mean_token_accuracy": 0.202163402736187,
      "num_tokens": 43374719.0,
      "step": 20895
    },
    {
      "entropy": 5.570595026016235,
      "epoch": 1.8958635703918723,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046414547886978926,
      "loss": 5.0106,
      "mean_token_accuracy": 0.20297456681728362,
      "num_tokens": 43385549.0,
      "step": 20900
    },
    {
      "entropy": 5.568581295013428,
      "epoch": 1.8963171262699565,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046412795319149204,
      "loss": 5.0238,
      "mean_token_accuracy": 0.216174978017807,
      "num_tokens": 43395503.0,
      "step": 20905
    },
    {
      "entropy": 5.515105676651001,
      "epoch": 1.8967706821480408,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046411042360198053,
      "loss": 5.026,
      "mean_token_accuracy": 0.20338638573884965,
      "num_tokens": 43406913.0,
      "step": 20910
    },
    {
      "entropy": 5.531669187545776,
      "epoch": 1.8972242380261248,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004640928901016173,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20720363557338714,
      "num_tokens": 43416459.0,
      "step": 20915
    },
    {
      "entropy": 5.509416341781616,
      "epoch": 1.8976777939042089,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046407535269076493,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20892471969127654,
      "num_tokens": 43426280.0,
      "step": 20920
    },
    {
      "entropy": 5.583643436431885,
      "epoch": 1.8981313497822931,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046405781136978606,
      "loss": 5.0225,
      "mean_token_accuracy": 0.20613077729940416,
      "num_tokens": 43436676.0,
      "step": 20925
    },
    {
      "entropy": 5.63284592628479,
      "epoch": 1.8985849056603774,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046404026613904357,
      "loss": 5.0209,
      "mean_token_accuracy": 0.20198507159948348,
      "num_tokens": 43447019.0,
      "step": 20930
    },
    {
      "entropy": 5.572532033920288,
      "epoch": 1.8990384615384617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004640227169989001,
      "loss": 4.9882,
      "mean_token_accuracy": 0.20607016682624818,
      "num_tokens": 43456609.0,
      "step": 20935
    },
    {
      "entropy": 5.550652933120728,
      "epoch": 1.8994920174165457,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004640051639497187,
      "loss": 5.0357,
      "mean_token_accuracy": 0.20605168044567107,
      "num_tokens": 43467616.0,
      "step": 20940
    },
    {
      "entropy": 5.454766130447387,
      "epoch": 1.8999455732946298,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004639876069918624,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20363384038209914,
      "num_tokens": 43478209.0,
      "step": 20945
    },
    {
      "entropy": 5.518501901626587,
      "epoch": 1.900399129172714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046397004612569416,
      "loss": 4.8878,
      "mean_token_accuracy": 0.21243149936199188,
      "num_tokens": 43487278.0,
      "step": 20950
    },
    {
      "entropy": 5.4941329002380375,
      "epoch": 1.9008526850507983,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046395248135157715,
      "loss": 4.9283,
      "mean_token_accuracy": 0.2183908686041832,
      "num_tokens": 43497457.0,
      "step": 20955
    },
    {
      "entropy": 5.543423986434936,
      "epoch": 1.9013062409288826,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046393491266987484,
      "loss": 5.0449,
      "mean_token_accuracy": 0.20089227706193924,
      "num_tokens": 43507761.0,
      "step": 20960
    },
    {
      "entropy": 5.468902492523194,
      "epoch": 1.9017597968069666,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004639173400809502,
      "loss": 4.9417,
      "mean_token_accuracy": 0.20726902931928634,
      "num_tokens": 43517198.0,
      "step": 20965
    },
    {
      "entropy": 5.527848100662231,
      "epoch": 1.9022133526850507,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004638997635851669,
      "loss": 5.0474,
      "mean_token_accuracy": 0.20085853487253189,
      "num_tokens": 43527429.0,
      "step": 20970
    },
    {
      "entropy": 5.603863906860352,
      "epoch": 1.902666908563135,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046388218318288844,
      "loss": 5.0512,
      "mean_token_accuracy": 0.2058670088648796,
      "num_tokens": 43537867.0,
      "step": 20975
    },
    {
      "entropy": 5.544486427307129,
      "epoch": 1.9031204644412192,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004638645988744782,
      "loss": 4.977,
      "mean_token_accuracy": 0.2110976591706276,
      "num_tokens": 43547284.0,
      "step": 20980
    },
    {
      "entropy": 5.5339678764343265,
      "epoch": 1.9035740203193035,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004638470106603,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20343973636627197,
      "num_tokens": 43557693.0,
      "step": 20985
    },
    {
      "entropy": 5.572199535369873,
      "epoch": 1.9040275761973875,
      "grad_norm": 1.125,
      "learning_rate": 0.0004638294185407174,
      "loss": 4.9785,
      "mean_token_accuracy": 0.2162001460790634,
      "num_tokens": 43566909.0,
      "step": 20990
    },
    {
      "entropy": 5.541551828384399,
      "epoch": 1.9044811320754715,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004638118225160944,
      "loss": 5.0342,
      "mean_token_accuracy": 0.2148605614900589,
      "num_tokens": 43576963.0,
      "step": 20995
    },
    {
      "entropy": 5.45386323928833,
      "epoch": 1.9049346879535558,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046379422258679477,
      "loss": 4.9538,
      "mean_token_accuracy": 0.20698957294225692,
      "num_tokens": 43586649.0,
      "step": 21000
    },
    {
      "epoch": 1.9049346879535558,
      "eval_entropy": 5.257148189674136,
      "eval_loss": 5.037424087524414,
      "eval_mean_token_accuracy": 0.21348288062437357,
      "eval_num_tokens": 43586649.0,
      "eval_runtime": 20.2464,
      "eval_samples_per_second": 1746.479,
      "eval_steps_per_second": 218.31,
      "step": 21000
    },
    {
      "entropy": 5.492565298080445,
      "epoch": 1.90538824383164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004637766187531825,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21100122779607772,
      "num_tokens": 43596506.0,
      "step": 21005
    },
    {
      "entropy": 5.501142501831055,
      "epoch": 1.9058417997097243,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046375901101562163,
      "loss": 5.0252,
      "mean_token_accuracy": 0.2093059837818146,
      "num_tokens": 43607188.0,
      "step": 21010
    },
    {
      "entropy": 5.546844816207885,
      "epoch": 1.9062953555878084,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004637413993744763,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21627327054738998,
      "num_tokens": 43617465.0,
      "step": 21015
    },
    {
      "entropy": 5.531183862686158,
      "epoch": 1.9067489114658926,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046372378383011077,
      "loss": 4.9451,
      "mean_token_accuracy": 0.212823686003685,
      "num_tokens": 43627692.0,
      "step": 21020
    },
    {
      "entropy": 5.572429466247558,
      "epoch": 1.9072024673439767,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046370616438288933,
      "loss": 5.0213,
      "mean_token_accuracy": 0.20401257425546646,
      "num_tokens": 43637350.0,
      "step": 21025
    },
    {
      "entropy": 5.556962871551514,
      "epoch": 1.907656023222061,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046368854103317624,
      "loss": 5.0161,
      "mean_token_accuracy": 0.19954755157232285,
      "num_tokens": 43647092.0,
      "step": 21030
    },
    {
      "entropy": 5.5124814987182615,
      "epoch": 1.9081095791001452,
      "grad_norm": 1.0,
      "learning_rate": 0.00046367091378133615,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21090739369392394,
      "num_tokens": 43657237.0,
      "step": 21035
    },
    {
      "entropy": 5.52500205039978,
      "epoch": 1.9085631349782293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004636532826277333,
      "loss": 4.9693,
      "mean_token_accuracy": 0.20812972038984298,
      "num_tokens": 43667966.0,
      "step": 21040
    },
    {
      "entropy": 5.504333448410034,
      "epoch": 1.9090166908563135,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046363564757273264,
      "loss": 4.9661,
      "mean_token_accuracy": 0.20653597116470337,
      "num_tokens": 43678059.0,
      "step": 21045
    },
    {
      "entropy": 5.491852521896362,
      "epoch": 1.9094702467343976,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046361800861669867,
      "loss": 4.981,
      "mean_token_accuracy": 0.21124444603919984,
      "num_tokens": 43687556.0,
      "step": 21050
    },
    {
      "entropy": 5.58355221748352,
      "epoch": 1.9099238026124818,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046360036575999623,
      "loss": 5.0062,
      "mean_token_accuracy": 0.20604631006717683,
      "num_tokens": 43697791.0,
      "step": 21055
    },
    {
      "entropy": 5.53818097114563,
      "epoch": 1.9103773584905661,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004635827190029902,
      "loss": 4.933,
      "mean_token_accuracy": 0.20206995755434037,
      "num_tokens": 43707390.0,
      "step": 21060
    },
    {
      "entropy": 5.571305894851685,
      "epoch": 1.9108309143686504,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004635650683460454,
      "loss": 5.0588,
      "mean_token_accuracy": 0.20439065843820572,
      "num_tokens": 43716842.0,
      "step": 21065
    },
    {
      "entropy": 5.480151748657226,
      "epoch": 1.9112844702467344,
      "grad_norm": 1.015625,
      "learning_rate": 0.000463547413789527,
      "loss": 4.9678,
      "mean_token_accuracy": 0.20866233855485916,
      "num_tokens": 43726967.0,
      "step": 21070
    },
    {
      "entropy": 5.583720636367798,
      "epoch": 1.9117380261248185,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046352975533379997,
      "loss": 4.987,
      "mean_token_accuracy": 0.21360233575105667,
      "num_tokens": 43737425.0,
      "step": 21075
    },
    {
      "entropy": 5.466370964050293,
      "epoch": 1.9121915820029027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046351209297922964,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21783478409051896,
      "num_tokens": 43747395.0,
      "step": 21080
    },
    {
      "entropy": 5.520073318481446,
      "epoch": 1.912645137880987,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046349442672618114,
      "loss": 5.0256,
      "mean_token_accuracy": 0.2036650225520134,
      "num_tokens": 43756348.0,
      "step": 21085
    },
    {
      "entropy": 5.539807319641113,
      "epoch": 1.9130986937590713,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046347675657501986,
      "loss": 4.9707,
      "mean_token_accuracy": 0.20604749917984008,
      "num_tokens": 43767629.0,
      "step": 21090
    },
    {
      "entropy": 5.500824403762818,
      "epoch": 1.9135522496371553,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004634590825261113,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21267948299646378,
      "num_tokens": 43777398.0,
      "step": 21095
    },
    {
      "entropy": 5.500860261917114,
      "epoch": 1.9140058055152394,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046344140457982075,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20852149575948714,
      "num_tokens": 43787605.0,
      "step": 21100
    },
    {
      "entropy": 5.518041706085205,
      "epoch": 1.9144593613933236,
      "grad_norm": 1.09375,
      "learning_rate": 0.000463423722736514,
      "loss": 5.0053,
      "mean_token_accuracy": 0.20624440014362336,
      "num_tokens": 43796989.0,
      "step": 21105
    },
    {
      "entropy": 5.538582754135132,
      "epoch": 1.9149129172714079,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004634060369965567,
      "loss": 5.0062,
      "mean_token_accuracy": 0.21026088148355485,
      "num_tokens": 43806653.0,
      "step": 21110
    },
    {
      "entropy": 5.4877242088317875,
      "epoch": 1.9153664731494922,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046338834736031446,
      "loss": 4.936,
      "mean_token_accuracy": 0.20859975814819337,
      "num_tokens": 43817152.0,
      "step": 21115
    },
    {
      "entropy": 5.528735399246216,
      "epoch": 1.9158200290275762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046337065382815324,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20312796235084535,
      "num_tokens": 43827341.0,
      "step": 21120
    },
    {
      "entropy": 5.484879827499389,
      "epoch": 1.9162735849056602,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046335295640043887,
      "loss": 4.9304,
      "mean_token_accuracy": 0.22185509651899338,
      "num_tokens": 43837195.0,
      "step": 21125
    },
    {
      "entropy": 5.5073271751403805,
      "epoch": 1.9167271407837445,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046333525507753743,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21319166868925093,
      "num_tokens": 43847473.0,
      "step": 21130
    },
    {
      "entropy": 5.4727256298065186,
      "epoch": 1.9171806966618288,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004633175498598148,
      "loss": 4.9581,
      "mean_token_accuracy": 0.2146495595574379,
      "num_tokens": 43857590.0,
      "step": 21135
    },
    {
      "entropy": 5.512988615036011,
      "epoch": 1.917634252539913,
      "grad_norm": 1.125,
      "learning_rate": 0.00046329984074763733,
      "loss": 5.105,
      "mean_token_accuracy": 0.19720255881547927,
      "num_tokens": 43867085.0,
      "step": 21140
    },
    {
      "entropy": 5.546961879730224,
      "epoch": 1.918087808417997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004632821277413711,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2073230728507042,
      "num_tokens": 43877153.0,
      "step": 21145
    },
    {
      "entropy": 5.630422782897949,
      "epoch": 1.9185413642960811,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004632644108413825,
      "loss": 5.0738,
      "mean_token_accuracy": 0.1899451419711113,
      "num_tokens": 43887977.0,
      "step": 21150
    },
    {
      "entropy": 5.445050382614136,
      "epoch": 1.9189949201741654,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046324669004803803,
      "loss": 4.9749,
      "mean_token_accuracy": 0.20582039207220076,
      "num_tokens": 43898649.0,
      "step": 21155
    },
    {
      "entropy": 5.60186185836792,
      "epoch": 1.9194484760522497,
      "grad_norm": 1.09375,
      "learning_rate": 0.000463228965361704,
      "loss": 5.0774,
      "mean_token_accuracy": 0.2027946949005127,
      "num_tokens": 43909369.0,
      "step": 21160
    },
    {
      "entropy": 5.566307926177979,
      "epoch": 1.919902031930334,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004632112367827469,
      "loss": 4.9758,
      "mean_token_accuracy": 0.2062263160943985,
      "num_tokens": 43919941.0,
      "step": 21165
    },
    {
      "entropy": 5.5621003150939945,
      "epoch": 1.920355587808418,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004631935043115335,
      "loss": 5.0276,
      "mean_token_accuracy": 0.2066472589969635,
      "num_tokens": 43929710.0,
      "step": 21170
    },
    {
      "entropy": 5.524765062332153,
      "epoch": 1.920809143686502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046317576794843056,
      "loss": 4.9662,
      "mean_token_accuracy": 0.20599670261144637,
      "num_tokens": 43939850.0,
      "step": 21175
    },
    {
      "entropy": 5.496295213699341,
      "epoch": 1.9212626995645863,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046315802769380474,
      "loss": 4.8865,
      "mean_token_accuracy": 0.21019480675458907,
      "num_tokens": 43950067.0,
      "step": 21180
    },
    {
      "entropy": 5.523532438278198,
      "epoch": 1.9217162554426706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004631402835480229,
      "loss": 4.9761,
      "mean_token_accuracy": 0.20840566009283065,
      "num_tokens": 43960029.0,
      "step": 21185
    },
    {
      "entropy": 5.515127897262573,
      "epoch": 1.9221698113207548,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046312253551145213,
      "loss": 4.995,
      "mean_token_accuracy": 0.20890448093414307,
      "num_tokens": 43971221.0,
      "step": 21190
    },
    {
      "entropy": 5.546313285827637,
      "epoch": 1.9226233671988389,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046310478358445935,
      "loss": 5.0148,
      "mean_token_accuracy": 0.1989280804991722,
      "num_tokens": 43981579.0,
      "step": 21195
    },
    {
      "entropy": 5.553898859024048,
      "epoch": 1.9230769230769231,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004630870277674117,
      "loss": 5.1133,
      "mean_token_accuracy": 0.1940425992012024,
      "num_tokens": 43991132.0,
      "step": 21200
    },
    {
      "entropy": 5.502590370178223,
      "epoch": 1.9235304789550072,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004630692680606764,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21060607582330704,
      "num_tokens": 44001255.0,
      "step": 21205
    },
    {
      "entropy": 5.478507614135742,
      "epoch": 1.9239840348330914,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004630515044646207,
      "loss": 4.8255,
      "mean_token_accuracy": 0.21846405565738677,
      "num_tokens": 44011275.0,
      "step": 21210
    },
    {
      "entropy": 5.517763280868531,
      "epoch": 1.9244375907111757,
      "grad_norm": 1.046875,
      "learning_rate": 0.000463033736979612,
      "loss": 4.9979,
      "mean_token_accuracy": 0.2088761657476425,
      "num_tokens": 44022524.0,
      "step": 21215
    },
    {
      "entropy": 5.496527481079101,
      "epoch": 1.9248911465892597,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046301596560601767,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2104906052350998,
      "num_tokens": 44033955.0,
      "step": 21220
    },
    {
      "entropy": 5.531476736068726,
      "epoch": 1.925344702467344,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004629981903442052,
      "loss": 5.0161,
      "mean_token_accuracy": 0.2065232053399086,
      "num_tokens": 44045149.0,
      "step": 21225
    },
    {
      "entropy": 5.577050638198853,
      "epoch": 1.925798258345428,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004629804111945424,
      "loss": 5.0143,
      "mean_token_accuracy": 0.2050991639494896,
      "num_tokens": 44055280.0,
      "step": 21230
    },
    {
      "entropy": 5.523096418380737,
      "epoch": 1.9262518142235123,
      "grad_norm": 1.125,
      "learning_rate": 0.0004629626281573966,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21663039922714233,
      "num_tokens": 44065578.0,
      "step": 21235
    },
    {
      "entropy": 5.446359395980835,
      "epoch": 1.9267053701015966,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004629448412331359,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21574582606554032,
      "num_tokens": 44075454.0,
      "step": 21240
    },
    {
      "entropy": 5.484840965270996,
      "epoch": 1.9271589259796809,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004629270504221279,
      "loss": 5.0053,
      "mean_token_accuracy": 0.2080381244421005,
      "num_tokens": 44085680.0,
      "step": 21245
    },
    {
      "entropy": 5.469542503356934,
      "epoch": 1.927612481857765,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004629092557247407,
      "loss": 4.9184,
      "mean_token_accuracy": 0.21238523721694946,
      "num_tokens": 44097177.0,
      "step": 21250
    },
    {
      "entropy": 5.549477672576904,
      "epoch": 1.928066037735849,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004628914571413422,
      "loss": 5.0381,
      "mean_token_accuracy": 0.2058754965662956,
      "num_tokens": 44107251.0,
      "step": 21255
    },
    {
      "entropy": 5.5213793277740475,
      "epoch": 1.9285195936139332,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046287365467230035,
      "loss": 5.0086,
      "mean_token_accuracy": 0.20609732270240783,
      "num_tokens": 44117943.0,
      "step": 21260
    },
    {
      "entropy": 5.595392465591431,
      "epoch": 1.9289731494920175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046285584831798354,
      "loss": 5.0213,
      "mean_token_accuracy": 0.2093285247683525,
      "num_tokens": 44128608.0,
      "step": 21265
    },
    {
      "entropy": 5.554631614685059,
      "epoch": 1.9294267053701017,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046283803807875996,
      "loss": 4.9864,
      "mean_token_accuracy": 0.21226508617401124,
      "num_tokens": 44138809.0,
      "step": 21270
    },
    {
      "entropy": 5.506673860549927,
      "epoch": 1.9298802612481858,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046282022395499785,
      "loss": 5.0116,
      "mean_token_accuracy": 0.21062631011009217,
      "num_tokens": 44148532.0,
      "step": 21275
    },
    {
      "entropy": 5.5527716159820555,
      "epoch": 1.9303338171262698,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004628024059470656,
      "loss": 5.0064,
      "mean_token_accuracy": 0.20672438740730287,
      "num_tokens": 44158628.0,
      "step": 21280
    },
    {
      "entropy": 5.5136350154876705,
      "epoch": 1.930787373004354,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004627845840553318,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2162649229168892,
      "num_tokens": 44168574.0,
      "step": 21285
    },
    {
      "entropy": 5.593709802627563,
      "epoch": 1.9312409288824384,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046276675828016493,
      "loss": 5.0538,
      "mean_token_accuracy": 0.20566692501306533,
      "num_tokens": 44180207.0,
      "step": 21290
    },
    {
      "entropy": 5.582747983932495,
      "epoch": 1.9316944847605226,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004627489286219336,
      "loss": 5.0888,
      "mean_token_accuracy": 0.1981254741549492,
      "num_tokens": 44190473.0,
      "step": 21295
    },
    {
      "entropy": 5.594016456604004,
      "epoch": 1.9321480406386067,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004627310950810066,
      "loss": 5.032,
      "mean_token_accuracy": 0.20988326370716096,
      "num_tokens": 44201150.0,
      "step": 21300
    },
    {
      "entropy": 5.512035655975342,
      "epoch": 1.9326015965166907,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004627132576577528,
      "loss": 4.9863,
      "mean_token_accuracy": 0.20305253118276595,
      "num_tokens": 44211272.0,
      "step": 21305
    },
    {
      "entropy": 5.538138294219971,
      "epoch": 1.933055152394775,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000462695416352541,
      "loss": 5.0528,
      "mean_token_accuracy": 0.2050202250480652,
      "num_tokens": 44223556.0,
      "step": 21310
    },
    {
      "entropy": 5.514124059677124,
      "epoch": 1.9335087082728593,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046267757116574,
      "loss": 4.9526,
      "mean_token_accuracy": 0.2091605097055435,
      "num_tokens": 44233911.0,
      "step": 21315
    },
    {
      "entropy": 5.5861574649810795,
      "epoch": 1.9339622641509435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004626597220977191,
      "loss": 5.0401,
      "mean_token_accuracy": 0.20843394100666046,
      "num_tokens": 44244108.0,
      "step": 21320
    },
    {
      "entropy": 5.4727623462677,
      "epoch": 1.9344158200290276,
      "grad_norm": 1.125,
      "learning_rate": 0.00046264186914884737,
      "loss": 4.918,
      "mean_token_accuracy": 0.2141546592116356,
      "num_tokens": 44253604.0,
      "step": 21325
    },
    {
      "entropy": 5.504957294464111,
      "epoch": 1.9348693759071116,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046262401231949386,
      "loss": 5.0472,
      "mean_token_accuracy": 0.19554894864559175,
      "num_tokens": 44263573.0,
      "step": 21330
    },
    {
      "entropy": 5.489033555984497,
      "epoch": 1.9353229317851959,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046260615161002803,
      "loss": 4.9464,
      "mean_token_accuracy": 0.20706450194120407,
      "num_tokens": 44274071.0,
      "step": 21335
    },
    {
      "entropy": 5.6296285629272464,
      "epoch": 1.9357764876632801,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004625882870208192,
      "loss": 5.0072,
      "mean_token_accuracy": 0.20659762620925903,
      "num_tokens": 44283997.0,
      "step": 21340
    },
    {
      "entropy": 5.57551760673523,
      "epoch": 1.9362300435413644,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046257041855223674,
      "loss": 4.9774,
      "mean_token_accuracy": 0.20467734783887864,
      "num_tokens": 44295024.0,
      "step": 21345
    },
    {
      "entropy": 5.489021444320679,
      "epoch": 1.9366835994194485,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046255254620465033,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21154813915491105,
      "num_tokens": 44305476.0,
      "step": 21350
    },
    {
      "entropy": 5.471819162368774,
      "epoch": 1.9371371552975325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046253466997842936,
      "loss": 4.9043,
      "mean_token_accuracy": 0.21418674886226655,
      "num_tokens": 44316101.0,
      "step": 21355
    },
    {
      "entropy": 5.468229007720947,
      "epoch": 1.9375907111756168,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004625167898739436,
      "loss": 4.8443,
      "mean_token_accuracy": 0.21662822663784026,
      "num_tokens": 44325058.0,
      "step": 21360
    },
    {
      "entropy": 5.664371204376221,
      "epoch": 1.938044267053701,
      "grad_norm": 1.015625,
      "learning_rate": 0.000462498905891563,
      "loss": 5.1664,
      "mean_token_accuracy": 0.19284735172986983,
      "num_tokens": 44336585.0,
      "step": 21365
    },
    {
      "entropy": 5.544726228713989,
      "epoch": 1.9384978229317853,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004624810180316571,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20519959330558776,
      "num_tokens": 44346911.0,
      "step": 21370
    },
    {
      "entropy": 5.486548137664795,
      "epoch": 1.9389513788098693,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046246312629459596,
      "loss": 4.9504,
      "mean_token_accuracy": 0.20622476637363435,
      "num_tokens": 44358040.0,
      "step": 21375
    },
    {
      "entropy": 5.535544443130493,
      "epoch": 1.9394049346879536,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004624452306807497,
      "loss": 4.9669,
      "mean_token_accuracy": 0.2152292862534523,
      "num_tokens": 44368781.0,
      "step": 21380
    },
    {
      "entropy": 5.469103145599365,
      "epoch": 1.9398584905660377,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004624273311904882,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20952262431383134,
      "num_tokens": 44378809.0,
      "step": 21385
    },
    {
      "entropy": 5.582791519165039,
      "epoch": 1.940312046444122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046240942782418184,
      "loss": 5.0741,
      "mean_token_accuracy": 0.20184009224176408,
      "num_tokens": 44388517.0,
      "step": 21390
    },
    {
      "entropy": 5.605941200256348,
      "epoch": 1.9407656023222062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004623915205822007,
      "loss": 5.0311,
      "mean_token_accuracy": 0.2025391176342964,
      "num_tokens": 44399364.0,
      "step": 21395
    },
    {
      "entropy": 5.535068464279175,
      "epoch": 1.9412191582002902,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046237360946491514,
      "loss": 4.9317,
      "mean_token_accuracy": 0.20450536757707596,
      "num_tokens": 44409506.0,
      "step": 21400
    },
    {
      "entropy": 5.570227003097534,
      "epoch": 1.9416727140783745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004623556944726956,
      "loss": 5.0051,
      "mean_token_accuracy": 0.20779819190502166,
      "num_tokens": 44419446.0,
      "step": 21405
    },
    {
      "entropy": 5.5028160572052,
      "epoch": 1.9421262699564585,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046233777560591256,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20704322755336763,
      "num_tokens": 44429560.0,
      "step": 21410
    },
    {
      "entropy": 5.51181058883667,
      "epoch": 1.9425798258345428,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046231985286493654,
      "loss": 4.962,
      "mean_token_accuracy": 0.20402525663375853,
      "num_tokens": 44439640.0,
      "step": 21415
    },
    {
      "entropy": 5.535993194580078,
      "epoch": 1.943033381712627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046230192625013836,
      "loss": 4.976,
      "mean_token_accuracy": 0.20708105117082595,
      "num_tokens": 44450480.0,
      "step": 21420
    },
    {
      "entropy": 5.463518047332764,
      "epoch": 1.9434869375907113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046228399576188844,
      "loss": 4.8783,
      "mean_token_accuracy": 0.20897210538387298,
      "num_tokens": 44460633.0,
      "step": 21425
    },
    {
      "entropy": 5.535352039337158,
      "epoch": 1.9439404934687954,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046226606140055784,
      "loss": 4.946,
      "mean_token_accuracy": 0.21099353432655335,
      "num_tokens": 44470917.0,
      "step": 21430
    },
    {
      "entropy": 5.534382009506226,
      "epoch": 1.9443940493468794,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046224812316651736,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20674648582935334,
      "num_tokens": 44482538.0,
      "step": 21435
    },
    {
      "entropy": 5.571159601211548,
      "epoch": 1.9448476052249637,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000462230181060138,
      "loss": 5.0268,
      "mean_token_accuracy": 0.205121012032032,
      "num_tokens": 44493518.0,
      "step": 21440
    },
    {
      "entropy": 5.5595598220825195,
      "epoch": 1.945301161103048,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046221223508179077,
      "loss": 5.0057,
      "mean_token_accuracy": 0.2067851632833481,
      "num_tokens": 44503685.0,
      "step": 21445
    },
    {
      "entropy": 5.505195236206054,
      "epoch": 1.9457547169811322,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046219428523184685,
      "loss": 4.9515,
      "mean_token_accuracy": 0.21454554796218872,
      "num_tokens": 44514015.0,
      "step": 21450
    },
    {
      "entropy": 5.495475006103516,
      "epoch": 1.9462082728592163,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004621763315106774,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20962912291288377,
      "num_tokens": 44525187.0,
      "step": 21455
    },
    {
      "entropy": 5.443025588989258,
      "epoch": 1.9466618287373003,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046215837391865364,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21432756632566452,
      "num_tokens": 44535964.0,
      "step": 21460
    },
    {
      "entropy": 5.462235355377198,
      "epoch": 1.9471153846153846,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046214041245614713,
      "loss": 4.9347,
      "mean_token_accuracy": 0.20689413845539092,
      "num_tokens": 44546985.0,
      "step": 21465
    },
    {
      "entropy": 5.579189205169678,
      "epoch": 1.9475689404934688,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004621224471235291,
      "loss": 5.0131,
      "mean_token_accuracy": 0.19903770685195923,
      "num_tokens": 44556888.0,
      "step": 21470
    },
    {
      "entropy": 5.552192831039429,
      "epoch": 1.9480224963715531,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046210447792117114,
      "loss": 4.9604,
      "mean_token_accuracy": 0.20719585418701172,
      "num_tokens": 44566857.0,
      "step": 21475
    },
    {
      "entropy": 5.543669605255127,
      "epoch": 1.9484760522496372,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000462086504849445,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20520297884941102,
      "num_tokens": 44578152.0,
      "step": 21480
    },
    {
      "entropy": 5.521569108963012,
      "epoch": 1.9489296081277212,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004620685279087222,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21676838546991348,
      "num_tokens": 44588814.0,
      "step": 21485
    },
    {
      "entropy": 5.5170595169067385,
      "epoch": 1.9493831640058055,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046205054709937474,
      "loss": 4.9627,
      "mean_token_accuracy": 0.2055974692106247,
      "num_tokens": 44599366.0,
      "step": 21490
    },
    {
      "entropy": 5.611104822158813,
      "epoch": 1.9498367198838897,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004620325624217741,
      "loss": 5.0599,
      "mean_token_accuracy": 0.19693043678998948,
      "num_tokens": 44609962.0,
      "step": 21495
    },
    {
      "entropy": 5.5933417797088625,
      "epoch": 1.950290275761974,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004620145738762925,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20933643728494644,
      "num_tokens": 44619917.0,
      "step": 21500
    },
    {
      "entropy": 5.48622498512268,
      "epoch": 1.950743831640058,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004619965814633019,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21082516610622407,
      "num_tokens": 44629723.0,
      "step": 21505
    },
    {
      "entropy": 5.5199370861053465,
      "epoch": 1.951197387518142,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004619785851831742,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20540409535169601,
      "num_tokens": 44641011.0,
      "step": 21510
    },
    {
      "entropy": 5.576232147216797,
      "epoch": 1.9516509433962264,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046196058503628176,
      "loss": 4.9444,
      "mean_token_accuracy": 0.20850757658481597,
      "num_tokens": 44651154.0,
      "step": 21515
    },
    {
      "entropy": 5.541594171524048,
      "epoch": 1.9521044992743106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004619425810229968,
      "loss": 5.0389,
      "mean_token_accuracy": 0.20138053894042968,
      "num_tokens": 44661672.0,
      "step": 21520
    },
    {
      "entropy": 5.451036787033081,
      "epoch": 1.952558055152395,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004619245731436917,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21387535482645034,
      "num_tokens": 44671449.0,
      "step": 21525
    },
    {
      "entropy": 5.533092641830445,
      "epoch": 1.953011611030479,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046190656139873863,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21527331322431564,
      "num_tokens": 44681977.0,
      "step": 21530
    },
    {
      "entropy": 5.525635576248169,
      "epoch": 1.953465166908563,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004618885457885103,
      "loss": 4.9483,
      "mean_token_accuracy": 0.209395132958889,
      "num_tokens": 44691946.0,
      "step": 21535
    },
    {
      "entropy": 5.500061988830566,
      "epoch": 1.9539187227866472,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004618705263133793,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20680051445960998,
      "num_tokens": 44702012.0,
      "step": 21540
    },
    {
      "entropy": 5.501916599273682,
      "epoch": 1.9543722786647315,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004618525029737182,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2127337336540222,
      "num_tokens": 44711605.0,
      "step": 21545
    },
    {
      "entropy": 5.593691682815551,
      "epoch": 1.9548258345428158,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046183447576989966,
      "loss": 5.0426,
      "mean_token_accuracy": 0.2109603613615036,
      "num_tokens": 44722765.0,
      "step": 21550
    },
    {
      "entropy": 5.537740278244018,
      "epoch": 1.9552793904208998,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046181644470229656,
      "loss": 4.9675,
      "mean_token_accuracy": 0.20676883459091186,
      "num_tokens": 44732082.0,
      "step": 21555
    },
    {
      "entropy": 5.447609090805054,
      "epoch": 1.955732946298984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004617984097712817,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21181290000677108,
      "num_tokens": 44742054.0,
      "step": 21560
    },
    {
      "entropy": 5.463639926910401,
      "epoch": 1.9561865021770681,
      "grad_norm": 1.125,
      "learning_rate": 0.0004617803709772282,
      "loss": 4.9973,
      "mean_token_accuracy": 0.2102394863963127,
      "num_tokens": 44752954.0,
      "step": 21565
    },
    {
      "entropy": 5.517415618896484,
      "epoch": 1.9566400580551524,
      "grad_norm": 1.0,
      "learning_rate": 0.0004617623283205091,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20561620146036147,
      "num_tokens": 44764277.0,
      "step": 21570
    },
    {
      "entropy": 5.57895073890686,
      "epoch": 1.9570936139332367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046174428180149733,
      "loss": 5.0503,
      "mean_token_accuracy": 0.2027288094162941,
      "num_tokens": 44775429.0,
      "step": 21575
    },
    {
      "entropy": 5.5902626514434814,
      "epoch": 1.9575471698113207,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004617262314205663,
      "loss": 5.0187,
      "mean_token_accuracy": 0.2027005985379219,
      "num_tokens": 44786390.0,
      "step": 21580
    },
    {
      "entropy": 5.57490029335022,
      "epoch": 1.958000725689405,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004617081771780893,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20770234614610672,
      "num_tokens": 44796473.0,
      "step": 21585
    },
    {
      "entropy": 5.495505332946777,
      "epoch": 1.958454281567489,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046169011907443944,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21066100597381593,
      "num_tokens": 44806379.0,
      "step": 21590
    },
    {
      "entropy": 5.517255973815918,
      "epoch": 1.9589078374455733,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004616720571099904,
      "loss": 5.0099,
      "mean_token_accuracy": 0.20810171961784363,
      "num_tokens": 44817659.0,
      "step": 21595
    },
    {
      "entropy": 5.54626955986023,
      "epoch": 1.9593613933236576,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004616539912851157,
      "loss": 4.9337,
      "mean_token_accuracy": 0.20864105224609375,
      "num_tokens": 44827073.0,
      "step": 21600
    },
    {
      "entropy": 5.513571405410767,
      "epoch": 1.9598149492017418,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046163592160018886,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2087594211101532,
      "num_tokens": 44837019.0,
      "step": 21605
    },
    {
      "entropy": 5.5844926834106445,
      "epoch": 1.9602685050798259,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004616178480555836,
      "loss": 4.955,
      "mean_token_accuracy": 0.2119430497288704,
      "num_tokens": 44846670.0,
      "step": 21610
    },
    {
      "entropy": 5.460532760620117,
      "epoch": 1.96072206095791,
      "grad_norm": 1.125,
      "learning_rate": 0.0004615997706516737,
      "loss": 4.9069,
      "mean_token_accuracy": 0.20875924825668335,
      "num_tokens": 44857016.0,
      "step": 21615
    },
    {
      "entropy": 5.580404996871948,
      "epoch": 1.9611756168359942,
      "grad_norm": 1.03125,
      "learning_rate": 0.000461581689388833,
      "loss": 5.0602,
      "mean_token_accuracy": 0.20059917271137237,
      "num_tokens": 44868180.0,
      "step": 21620
    },
    {
      "entropy": 5.5173357963562015,
      "epoch": 1.9616291727140784,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046156360426743536,
      "loss": 4.9178,
      "mean_token_accuracy": 0.21419098377227783,
      "num_tokens": 44877864.0,
      "step": 21625
    },
    {
      "entropy": 5.570303821563721,
      "epoch": 1.9620827285921627,
      "grad_norm": 1.125,
      "learning_rate": 0.0004615455152878549,
      "loss": 4.9773,
      "mean_token_accuracy": 0.21818986386060715,
      "num_tokens": 44887737.0,
      "step": 21630
    },
    {
      "entropy": 5.449281024932861,
      "epoch": 1.9625362844702468,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004615274224504657,
      "loss": 4.9729,
      "mean_token_accuracy": 0.20659109354019164,
      "num_tokens": 44898669.0,
      "step": 21635
    },
    {
      "entropy": 5.560696220397949,
      "epoch": 1.9629898403483308,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046150932575564173,
      "loss": 5.0438,
      "mean_token_accuracy": 0.20681725144386293,
      "num_tokens": 44909074.0,
      "step": 21640
    },
    {
      "entropy": 5.555341911315918,
      "epoch": 1.963443396226415,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046149122520375746,
      "loss": 4.9362,
      "mean_token_accuracy": 0.20842625498771666,
      "num_tokens": 44919468.0,
      "step": 21645
    },
    {
      "entropy": 5.558181190490723,
      "epoch": 1.9638969521044993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004614731207951871,
      "loss": 5.0424,
      "mean_token_accuracy": 0.19995157122612,
      "num_tokens": 44930470.0,
      "step": 21650
    },
    {
      "entropy": 5.529663133621216,
      "epoch": 1.9643505079825836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004614550125303051,
      "loss": 5.0372,
      "mean_token_accuracy": 0.20677049309015275,
      "num_tokens": 44940611.0,
      "step": 21655
    },
    {
      "entropy": 5.534930992126465,
      "epoch": 1.9648040638606676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004614369004094859,
      "loss": 4.9883,
      "mean_token_accuracy": 0.2098083481192589,
      "num_tokens": 44950985.0,
      "step": 21660
    },
    {
      "entropy": 5.531403207778931,
      "epoch": 1.9652576197387517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046141878443310415,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20378538072109223,
      "num_tokens": 44961858.0,
      "step": 21665
    },
    {
      "entropy": 5.590341997146607,
      "epoch": 1.965711175616836,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046140066460153444,
      "loss": 4.988,
      "mean_token_accuracy": 0.20614514350891114,
      "num_tokens": 44972048.0,
      "step": 21670
    },
    {
      "entropy": 5.538367700576782,
      "epoch": 1.9661647314949202,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046138254091515136,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21528832316398622,
      "num_tokens": 44981835.0,
      "step": 21675
    },
    {
      "entropy": 5.521666669845581,
      "epoch": 1.9666182873730045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046136441337432996,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21960411220788956,
      "num_tokens": 44992108.0,
      "step": 21680
    },
    {
      "entropy": 5.6115960597991945,
      "epoch": 1.9670718432510885,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046134628197944496,
      "loss": 5.0596,
      "mean_token_accuracy": 0.19814335703849792,
      "num_tokens": 45002105.0,
      "step": 21685
    },
    {
      "entropy": 5.499299335479736,
      "epoch": 1.9675253991291726,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046132814673087143,
      "loss": 4.9795,
      "mean_token_accuracy": 0.202993506193161,
      "num_tokens": 45013934.0,
      "step": 21690
    },
    {
      "entropy": 5.5065466403961185,
      "epoch": 1.9679789550072568,
      "grad_norm": 1.0,
      "learning_rate": 0.0004613100076289842,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2106448546051979,
      "num_tokens": 45023851.0,
      "step": 21695
    },
    {
      "entropy": 5.464599180221557,
      "epoch": 1.968432510885341,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004612918646741586,
      "loss": 4.8947,
      "mean_token_accuracy": 0.20776404291391373,
      "num_tokens": 45034140.0,
      "step": 21700
    },
    {
      "entropy": 5.495933675765992,
      "epoch": 1.9688860667634254,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004612737178667698,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2077025592327118,
      "num_tokens": 45044452.0,
      "step": 21705
    },
    {
      "entropy": 5.5003642559051515,
      "epoch": 1.9693396226415094,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046125556720719304,
      "loss": 4.9899,
      "mean_token_accuracy": 0.20421548187732697,
      "num_tokens": 45053845.0,
      "step": 21710
    },
    {
      "entropy": 5.51359977722168,
      "epoch": 1.9697931785195935,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004612374126958037,
      "loss": 4.9186,
      "mean_token_accuracy": 0.21345615088939668,
      "num_tokens": 45063973.0,
      "step": 21715
    },
    {
      "entropy": 5.605167865753174,
      "epoch": 1.9702467343976777,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004612192543329771,
      "loss": 5.0872,
      "mean_token_accuracy": 0.2048759251832962,
      "num_tokens": 45074879.0,
      "step": 21720
    },
    {
      "entropy": 5.502563047409057,
      "epoch": 1.970700290275762,
      "grad_norm": 1.046875,
      "learning_rate": 0.000461201092119089,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21382432878017427,
      "num_tokens": 45085161.0,
      "step": 21725
    },
    {
      "entropy": 5.57512412071228,
      "epoch": 1.9711538461538463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004611829260545148,
      "loss": 5.0563,
      "mean_token_accuracy": 0.20214797109365462,
      "num_tokens": 45095862.0,
      "step": 21730
    },
    {
      "entropy": 5.564107656478882,
      "epoch": 1.9716074020319303,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046116475613963026,
      "loss": 4.971,
      "mean_token_accuracy": 0.20207210183143615,
      "num_tokens": 45106396.0,
      "step": 21735
    },
    {
      "entropy": 5.521020364761353,
      "epoch": 1.9720609579100146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046114658237481114,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21837573498487473,
      "num_tokens": 45117178.0,
      "step": 21740
    },
    {
      "entropy": 5.472550916671753,
      "epoch": 1.9725145137880986,
      "grad_norm": 1.25,
      "learning_rate": 0.0004611284047604332,
      "loss": 4.912,
      "mean_token_accuracy": 0.2144747331738472,
      "num_tokens": 45126571.0,
      "step": 21745
    },
    {
      "entropy": 5.566296529769898,
      "epoch": 1.9729680696661829,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046111022329687246,
      "loss": 5.0264,
      "mean_token_accuracy": 0.20858824998140335,
      "num_tokens": 45137247.0,
      "step": 21750
    },
    {
      "entropy": 5.583434104919434,
      "epoch": 1.9734216255442671,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046109203798450483,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20433346927165985,
      "num_tokens": 45147531.0,
      "step": 21755
    },
    {
      "entropy": 5.517968988418579,
      "epoch": 1.9738751814223512,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004610738488237065,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21058939546346664,
      "num_tokens": 45158312.0,
      "step": 21760
    },
    {
      "entropy": 5.520002317428589,
      "epoch": 1.9743287373004355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004610556558148535,
      "loss": 5.0121,
      "mean_token_accuracy": 0.20637791901826857,
      "num_tokens": 45169754.0,
      "step": 21765
    },
    {
      "entropy": 5.570459270477295,
      "epoch": 1.9747822931785195,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004610374589583222,
      "loss": 5.0215,
      "mean_token_accuracy": 0.2031082585453987,
      "num_tokens": 45180428.0,
      "step": 21770
    },
    {
      "entropy": 5.558872270584106,
      "epoch": 1.9752358490566038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004610192582544888,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21063655018806457,
      "num_tokens": 45191558.0,
      "step": 21775
    },
    {
      "entropy": 5.4693413257598875,
      "epoch": 1.975689404934688,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004610010537037297,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21101723462343216,
      "num_tokens": 45201567.0,
      "step": 21780
    },
    {
      "entropy": 5.566562366485596,
      "epoch": 1.9761429608127723,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004609828453064214,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2021735742688179,
      "num_tokens": 45211359.0,
      "step": 21785
    },
    {
      "entropy": 5.615008401870727,
      "epoch": 1.9765965166908563,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046096463306294057,
      "loss": 4.9639,
      "mean_token_accuracy": 0.20703614503145218,
      "num_tokens": 45222106.0,
      "step": 21790
    },
    {
      "entropy": 5.473619365692139,
      "epoch": 1.9770500725689404,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046094641697366366,
      "loss": 4.9935,
      "mean_token_accuracy": 0.21004395484924315,
      "num_tokens": 45233169.0,
      "step": 21795
    },
    {
      "entropy": 5.594874143600464,
      "epoch": 1.9775036284470247,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004609281970389675,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20054440051317216,
      "num_tokens": 45243627.0,
      "step": 21800
    },
    {
      "entropy": 5.559269714355469,
      "epoch": 1.977957184325109,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046090997325922887,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21414467245340346,
      "num_tokens": 45253730.0,
      "step": 21805
    },
    {
      "entropy": 5.521198081970215,
      "epoch": 1.9784107402031932,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046089174563482456,
      "loss": 5.043,
      "mean_token_accuracy": 0.20758040994405746,
      "num_tokens": 45263443.0,
      "step": 21810
    },
    {
      "entropy": 5.487611389160156,
      "epoch": 1.9788642960812772,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004608735141661316,
      "loss": 4.9378,
      "mean_token_accuracy": 0.21140017360448837,
      "num_tokens": 45275298.0,
      "step": 21815
    },
    {
      "entropy": 5.542535018920899,
      "epoch": 1.9793178519593613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046085527885352694,
      "loss": 5.0238,
      "mean_token_accuracy": 0.2079692006111145,
      "num_tokens": 45285628.0,
      "step": 21820
    },
    {
      "entropy": 5.52398452758789,
      "epoch": 1.9797714078374455,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004608370396973878,
      "loss": 4.8981,
      "mean_token_accuracy": 0.2145746096968651,
      "num_tokens": 45295559.0,
      "step": 21825
    },
    {
      "entropy": 5.490517091751099,
      "epoch": 1.9802249637155298,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046081879669809133,
      "loss": 4.921,
      "mean_token_accuracy": 0.21242493987083436,
      "num_tokens": 45305395.0,
      "step": 21830
    },
    {
      "entropy": 5.490908813476563,
      "epoch": 1.980678519593614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004608005498560148,
      "loss": 4.9368,
      "mean_token_accuracy": 0.21311070770025253,
      "num_tokens": 45316763.0,
      "step": 21835
    },
    {
      "entropy": 5.556914472579956,
      "epoch": 1.9811320754716981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004607822991715356,
      "loss": 5.0763,
      "mean_token_accuracy": 0.2059244155883789,
      "num_tokens": 45326484.0,
      "step": 21840
    },
    {
      "entropy": 5.577116632461548,
      "epoch": 1.9815856313497822,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000460764044645031,
      "loss": 4.9977,
      "mean_token_accuracy": 0.20606459826231002,
      "num_tokens": 45335792.0,
      "step": 21845
    },
    {
      "entropy": 5.533209276199341,
      "epoch": 1.9820391872278664,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004607457862768787,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20594725459814073,
      "num_tokens": 45347247.0,
      "step": 21850
    },
    {
      "entropy": 5.588332223892212,
      "epoch": 1.9824927431059507,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004607275240674562,
      "loss": 5.0576,
      "mean_token_accuracy": 0.20373062789440155,
      "num_tokens": 45357994.0,
      "step": 21855
    },
    {
      "entropy": 5.4819825172424315,
      "epoch": 1.982946298984035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004607092580171411,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21019235402345657,
      "num_tokens": 45368170.0,
      "step": 21860
    },
    {
      "entropy": 5.573336935043335,
      "epoch": 1.983399854862119,
      "grad_norm": 1.125,
      "learning_rate": 0.0004606909881263113,
      "loss": 5.0433,
      "mean_token_accuracy": 0.2052479237318039,
      "num_tokens": 45378696.0,
      "step": 21865
    },
    {
      "entropy": 5.617618989944458,
      "epoch": 1.983853410740203,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046067271439534445,
      "loss": 5.0557,
      "mean_token_accuracy": 0.20690106749534606,
      "num_tokens": 45388578.0,
      "step": 21870
    },
    {
      "entropy": 5.543561553955078,
      "epoch": 1.9843069666182873,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046065443682461864,
      "loss": 5.0371,
      "mean_token_accuracy": 0.20809713900089263,
      "num_tokens": 45398835.0,
      "step": 21875
    },
    {
      "entropy": 5.547297239303589,
      "epoch": 1.9847605224963716,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046063615541451177,
      "loss": 5.0092,
      "mean_token_accuracy": 0.2083718404173851,
      "num_tokens": 45409032.0,
      "step": 21880
    },
    {
      "entropy": 5.592833137512207,
      "epoch": 1.9852140783744558,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046061787016540184,
      "loss": 5.106,
      "mean_token_accuracy": 0.19703632742166519,
      "num_tokens": 45420661.0,
      "step": 21885
    },
    {
      "entropy": 5.501183080673218,
      "epoch": 1.98566763425254,
      "grad_norm": 1.25,
      "learning_rate": 0.00046059958107766706,
      "loss": 4.858,
      "mean_token_accuracy": 0.22111203968524934,
      "num_tokens": 45429751.0,
      "step": 21890
    },
    {
      "entropy": 5.558317708969116,
      "epoch": 1.986121190130624,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004605812881516857,
      "loss": 4.981,
      "mean_token_accuracy": 0.21411139219999314,
      "num_tokens": 45439551.0,
      "step": 21895
    },
    {
      "entropy": 5.444088983535766,
      "epoch": 1.9865747460087082,
      "grad_norm": 1.125,
      "learning_rate": 0.00046056299138783595,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22072216123342514,
      "num_tokens": 45449543.0,
      "step": 21900
    },
    {
      "entropy": 5.53483510017395,
      "epoch": 1.9870283018867925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004605446907864963,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21000481843948365,
      "num_tokens": 45459069.0,
      "step": 21905
    },
    {
      "entropy": 5.5791398048400875,
      "epoch": 1.9874818577648767,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004605263863480452,
      "loss": 5.0618,
      "mean_token_accuracy": 0.19910577982664107,
      "num_tokens": 45468987.0,
      "step": 21910
    },
    {
      "entropy": 5.589298105239868,
      "epoch": 1.9879354136429608,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046050807807286105,
      "loss": 4.9864,
      "mean_token_accuracy": 0.21228085309267045,
      "num_tokens": 45479781.0,
      "step": 21915
    },
    {
      "entropy": 5.62843542098999,
      "epoch": 1.988388969521045,
      "grad_norm": 1.125,
      "learning_rate": 0.00046048976596132265,
      "loss": 5.1198,
      "mean_token_accuracy": 0.19545234590768815,
      "num_tokens": 45490318.0,
      "step": 21920
    },
    {
      "entropy": 5.520717000961303,
      "epoch": 1.988842525399129,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046047145001380856,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2116482824087143,
      "num_tokens": 45501514.0,
      "step": 21925
    },
    {
      "entropy": 5.525570249557495,
      "epoch": 1.9892960812772134,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046045313023069764,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21024584025144577,
      "num_tokens": 45511727.0,
      "step": 21930
    },
    {
      "entropy": 5.571838521957398,
      "epoch": 1.9897496371552976,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004604348066123688,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20526627004146575,
      "num_tokens": 45521807.0,
      "step": 21935
    },
    {
      "entropy": 5.613113021850586,
      "epoch": 1.9902031930333817,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004604164791592008,
      "loss": 5.0881,
      "mean_token_accuracy": 0.20075277239084244,
      "num_tokens": 45532033.0,
      "step": 21940
    },
    {
      "entropy": 5.550717735290528,
      "epoch": 1.990656748911466,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046039814787157283,
      "loss": 5.0,
      "mean_token_accuracy": 0.2103917732834816,
      "num_tokens": 45542806.0,
      "step": 21945
    },
    {
      "entropy": 5.504235935211182,
      "epoch": 1.99111030478955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004603798127498639,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22614379674196244,
      "num_tokens": 45551982.0,
      "step": 21950
    },
    {
      "entropy": 5.535236835479736,
      "epoch": 1.9915638606676342,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004603614737944532,
      "loss": 4.9939,
      "mean_token_accuracy": 0.20743495523929595,
      "num_tokens": 45561998.0,
      "step": 21955
    },
    {
      "entropy": 5.486254262924194,
      "epoch": 1.9920174165457185,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046034313100571994,
      "loss": 4.9879,
      "mean_token_accuracy": 0.2085157796740532,
      "num_tokens": 45572246.0,
      "step": 21960
    },
    {
      "entropy": 5.551699161529541,
      "epoch": 1.9924709724238028,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004603247843840436,
      "loss": 5.0096,
      "mean_token_accuracy": 0.206974858045578,
      "num_tokens": 45582251.0,
      "step": 21965
    },
    {
      "entropy": 5.529483032226563,
      "epoch": 1.9929245283018868,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004603064339298034,
      "loss": 4.9041,
      "mean_token_accuracy": 0.20982683300971985,
      "num_tokens": 45592749.0,
      "step": 21970
    },
    {
      "entropy": 5.510742473602295,
      "epoch": 1.9933780841799709,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004602880796433789,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21203220039606094,
      "num_tokens": 45602412.0,
      "step": 21975
    },
    {
      "entropy": 5.587583255767822,
      "epoch": 1.9938316400580551,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046026972152514976,
      "loss": 5.0042,
      "mean_token_accuracy": 0.2036629617214203,
      "num_tokens": 45612736.0,
      "step": 21980
    },
    {
      "entropy": 5.546084403991699,
      "epoch": 1.9942851959361394,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004602513595754956,
      "loss": 4.9916,
      "mean_token_accuracy": 0.20740335881710054,
      "num_tokens": 45622952.0,
      "step": 21985
    },
    {
      "entropy": 5.587206506729126,
      "epoch": 1.9947387518142237,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046023299379479606,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20512623488903045,
      "num_tokens": 45633569.0,
      "step": 21990
    },
    {
      "entropy": 5.499563360214234,
      "epoch": 1.9951923076923077,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046021462418343104,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21353395879268647,
      "num_tokens": 45644347.0,
      "step": 21995
    },
    {
      "entropy": 5.587524795532227,
      "epoch": 1.9956458635703918,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004601962507417804,
      "loss": 5.0314,
      "mean_token_accuracy": 0.2084423065185547,
      "num_tokens": 45655268.0,
      "step": 22000
    },
    {
      "entropy": 5.518528270721435,
      "epoch": 1.996099419448476,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046017787347022403,
      "loss": 4.943,
      "mean_token_accuracy": 0.2089353621006012,
      "num_tokens": 45664900.0,
      "step": 22005
    },
    {
      "entropy": 5.5573653221130375,
      "epoch": 1.9965529753265603,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046015949236914217,
      "loss": 5.0315,
      "mean_token_accuracy": 0.2094045177102089,
      "num_tokens": 45675252.0,
      "step": 22010
    },
    {
      "entropy": 5.612162399291992,
      "epoch": 1.9970065312046446,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004601411074389147,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21518081575632095,
      "num_tokens": 45686027.0,
      "step": 22015
    },
    {
      "entropy": 5.548440742492676,
      "epoch": 1.9974600870827286,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000460122718679922,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20475952625274657,
      "num_tokens": 45696436.0,
      "step": 22020
    },
    {
      "entropy": 5.586886548995972,
      "epoch": 1.9979136429608126,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004601043260925443,
      "loss": 5.0644,
      "mean_token_accuracy": 0.2107183665037155,
      "num_tokens": 45706509.0,
      "step": 22025
    },
    {
      "entropy": 5.593432807922364,
      "epoch": 1.998367198838897,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046008592967716197,
      "loss": 5.09,
      "mean_token_accuracy": 0.20457664877176285,
      "num_tokens": 45717933.0,
      "step": 22030
    },
    {
      "entropy": 5.543962860107422,
      "epoch": 1.9988207547169812,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004600675294341554,
      "loss": 5.0272,
      "mean_token_accuracy": 0.19962505102157593,
      "num_tokens": 45727966.0,
      "step": 22035
    },
    {
      "entropy": 5.543148803710937,
      "epoch": 1.9992743105950654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046004912536390526,
      "loss": 4.9966,
      "mean_token_accuracy": 0.20704041719436644,
      "num_tokens": 45737649.0,
      "step": 22040
    },
    {
      "entropy": 5.570962762832641,
      "epoch": 1.9997278664731495,
      "grad_norm": 1.140625,
      "learning_rate": 0.000460030717466792,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2104273408651352,
      "num_tokens": 45747561.0,
      "step": 22045
    },
    {
      "entropy": 5.526540374755859,
      "epoch": 2.0001814223512335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004600123057431963,
      "loss": 4.9,
      "mean_token_accuracy": 0.2146461173892021,
      "num_tokens": 45758183.0,
      "step": 22050
    },
    {
      "entropy": 5.5385801792144775,
      "epoch": 2.000634978229318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000459993890193499,
      "loss": 4.865,
      "mean_token_accuracy": 0.22243619263172149,
      "num_tokens": 45768006.0,
      "step": 22055
    },
    {
      "entropy": 5.592305421829224,
      "epoch": 2.001088534107402,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004599754708180809,
      "loss": 5.0066,
      "mean_token_accuracy": 0.20514395833015442,
      "num_tokens": 45777981.0,
      "step": 22060
    },
    {
      "entropy": 5.546550273895264,
      "epoch": 2.0015420899854863,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000459957047617323,
      "loss": 4.9182,
      "mean_token_accuracy": 0.20611172318458557,
      "num_tokens": 45789083.0,
      "step": 22065
    },
    {
      "entropy": 5.561010456085205,
      "epoch": 2.0019956458635706,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045993862059160617,
      "loss": 4.8912,
      "mean_token_accuracy": 0.2149354949593544,
      "num_tokens": 45799750.0,
      "step": 22070
    },
    {
      "entropy": 5.524298477172851,
      "epoch": 2.0024492017416544,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004599201897413116,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2158678650856018,
      "num_tokens": 45810321.0,
      "step": 22075
    },
    {
      "entropy": 5.4613282680511475,
      "epoch": 2.0029027576197387,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045990175506682024,
      "loss": 4.8536,
      "mean_token_accuracy": 0.22158993929624557,
      "num_tokens": 45820581.0,
      "step": 22080
    },
    {
      "entropy": 5.513234090805054,
      "epoch": 2.003356313497823,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045988331656851355,
      "loss": 4.9518,
      "mean_token_accuracy": 0.2053571820259094,
      "num_tokens": 45831504.0,
      "step": 22085
    },
    {
      "entropy": 5.56544303894043,
      "epoch": 2.003809869375907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045986487424677284,
      "loss": 4.9723,
      "mean_token_accuracy": 0.20918092727661133,
      "num_tokens": 45841100.0,
      "step": 22090
    },
    {
      "entropy": 5.561041927337646,
      "epoch": 2.0042634252539915,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004598464281019793,
      "loss": 4.9426,
      "mean_token_accuracy": 0.2117842137813568,
      "num_tokens": 45851647.0,
      "step": 22095
    },
    {
      "entropy": 5.519134044647217,
      "epoch": 2.0047169811320753,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045982797813451463,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2213513061404228,
      "num_tokens": 45862995.0,
      "step": 22100
    },
    {
      "entropy": 5.4666729927062985,
      "epoch": 2.0051705370101596,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045980952434476014,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2112858533859253,
      "num_tokens": 45872360.0,
      "step": 22105
    },
    {
      "entropy": 5.4949592590332035,
      "epoch": 2.005624092888244,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045979106673309765,
      "loss": 4.9584,
      "mean_token_accuracy": 0.20771063715219498,
      "num_tokens": 45883440.0,
      "step": 22110
    },
    {
      "entropy": 5.488120174407959,
      "epoch": 2.006077648766328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045977260529990883,
      "loss": 4.9212,
      "mean_token_accuracy": 0.2086660832166672,
      "num_tokens": 45894323.0,
      "step": 22115
    },
    {
      "entropy": 5.559689617156982,
      "epoch": 2.0065312046444124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004597541400455755,
      "loss": 4.9669,
      "mean_token_accuracy": 0.20842481702566146,
      "num_tokens": 45904544.0,
      "step": 22120
    },
    {
      "entropy": 5.621049499511718,
      "epoch": 2.006984760522496,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045973567097047933,
      "loss": 4.8339,
      "mean_token_accuracy": 0.23325383067131042,
      "num_tokens": 45913225.0,
      "step": 22125
    },
    {
      "entropy": 5.545746946334839,
      "epoch": 2.0074383164005805,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045971719807500253,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21734781414270402,
      "num_tokens": 45923306.0,
      "step": 22130
    },
    {
      "entropy": 5.481171178817749,
      "epoch": 2.0078918722786647,
      "grad_norm": 1.078125,
      "learning_rate": 0.000459698721359527,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21010444760322572,
      "num_tokens": 45933243.0,
      "step": 22135
    },
    {
      "entropy": 5.4783995151519775,
      "epoch": 2.008345428156749,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004596802408244348,
      "loss": 4.9731,
      "mean_token_accuracy": 0.20971969068050383,
      "num_tokens": 45944498.0,
      "step": 22140
    },
    {
      "entropy": 5.482260131835938,
      "epoch": 2.0087989840348333,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004596617564701081,
      "loss": 4.8183,
      "mean_token_accuracy": 0.21418478339910507,
      "num_tokens": 45954550.0,
      "step": 22145
    },
    {
      "entropy": 5.601777172088623,
      "epoch": 2.009252539912917,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045964326829692925,
      "loss": 4.9663,
      "mean_token_accuracy": 0.20779272466897963,
      "num_tokens": 45965425.0,
      "step": 22150
    },
    {
      "entropy": 5.534191751480103,
      "epoch": 2.0097060957910013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045962477630528064,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22023537009954453,
      "num_tokens": 45975637.0,
      "step": 22155
    },
    {
      "entropy": 5.506317615509033,
      "epoch": 2.0101596516690856,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045960628049554453,
      "loss": 4.9474,
      "mean_token_accuracy": 0.2040410190820694,
      "num_tokens": 45985268.0,
      "step": 22160
    },
    {
      "entropy": 5.477055644989013,
      "epoch": 2.01061320754717,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045958778086810354,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21610236614942552,
      "num_tokens": 45995335.0,
      "step": 22165
    },
    {
      "entropy": 5.518622493743896,
      "epoch": 2.011066763425254,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004595692774233402,
      "loss": 4.9384,
      "mean_token_accuracy": 0.20850727409124375,
      "num_tokens": 46006037.0,
      "step": 22170
    },
    {
      "entropy": 5.551631498336792,
      "epoch": 2.011520319303338,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045955077016163713,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21348341554403305,
      "num_tokens": 46017016.0,
      "step": 22175
    },
    {
      "entropy": 5.5876051902771,
      "epoch": 2.0119738751814222,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045953225908337717,
      "loss": 5.0079,
      "mean_token_accuracy": 0.2049964189529419,
      "num_tokens": 46026797.0,
      "step": 22180
    },
    {
      "entropy": 5.468192148208618,
      "epoch": 2.0124274310595065,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045951374418894304,
      "loss": 4.8311,
      "mean_token_accuracy": 0.21861059665679933,
      "num_tokens": 46037799.0,
      "step": 22185
    },
    {
      "entropy": 5.490703010559082,
      "epoch": 2.0128809869375908,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045949522547871765,
      "loss": 4.8544,
      "mean_token_accuracy": 0.21228369772434236,
      "num_tokens": 46048617.0,
      "step": 22190
    },
    {
      "entropy": 5.545925235748291,
      "epoch": 2.013334542815675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000459476702953084,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21240699738264085,
      "num_tokens": 46060122.0,
      "step": 22195
    },
    {
      "entropy": 5.5373071193695065,
      "epoch": 2.013788098693759,
      "grad_norm": 1.0625,
      "learning_rate": 0.000459458176612425,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21469461172819138,
      "num_tokens": 46071488.0,
      "step": 22200
    },
    {
      "entropy": 5.603073406219482,
      "epoch": 2.014241654571843,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00045943964645712407,
      "loss": 4.9731,
      "mean_token_accuracy": 0.2059222310781479,
      "num_tokens": 46081954.0,
      "step": 22205
    },
    {
      "entropy": 5.52503080368042,
      "epoch": 2.0146952104499274,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004594211124875642,
      "loss": 4.889,
      "mean_token_accuracy": 0.21305508017539979,
      "num_tokens": 46092488.0,
      "step": 22210
    },
    {
      "entropy": 5.439353561401367,
      "epoch": 2.0151487663280117,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004594025747041287,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2224961668252945,
      "num_tokens": 46102451.0,
      "step": 22215
    },
    {
      "entropy": 5.4922776222229,
      "epoch": 2.015602322206096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000459384033107201,
      "loss": 4.922,
      "mean_token_accuracy": 0.21498547047376632,
      "num_tokens": 46113862.0,
      "step": 22220
    },
    {
      "entropy": 5.581901884078979,
      "epoch": 2.01605587808418,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004593654876971645,
      "loss": 4.9931,
      "mean_token_accuracy": 0.20773377269506454,
      "num_tokens": 46124517.0,
      "step": 22225
    },
    {
      "entropy": 5.466474819183349,
      "epoch": 2.016509433962264,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004593469384744027,
      "loss": 4.9048,
      "mean_token_accuracy": 0.21205781847238542,
      "num_tokens": 46134221.0,
      "step": 22230
    },
    {
      "entropy": 5.535274600982666,
      "epoch": 2.0169629898403483,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045932838543929924,
      "loss": 4.9908,
      "mean_token_accuracy": 0.21098925024271012,
      "num_tokens": 46144512.0,
      "step": 22235
    },
    {
      "entropy": 5.529980611801148,
      "epoch": 2.0174165457184325,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004593098285922378,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2101367324590683,
      "num_tokens": 46154373.0,
      "step": 22240
    },
    {
      "entropy": 5.564878654479981,
      "epoch": 2.017870101596517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004592912679336021,
      "loss": 5.0098,
      "mean_token_accuracy": 0.20913756787776946,
      "num_tokens": 46164834.0,
      "step": 22245
    },
    {
      "entropy": 5.520668077468872,
      "epoch": 2.018323657474601,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004592727034637761,
      "loss": 4.8545,
      "mean_token_accuracy": 0.21772663593292235,
      "num_tokens": 46174815.0,
      "step": 22250
    },
    {
      "entropy": 5.502844905853271,
      "epoch": 2.018777213352685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045925413518314363,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21282357573509217,
      "num_tokens": 46185155.0,
      "step": 22255
    },
    {
      "entropy": 5.5436310291290285,
      "epoch": 2.019230769230769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004592355630920886,
      "loss": 4.9489,
      "mean_token_accuracy": 0.21195296794176102,
      "num_tokens": 46194537.0,
      "step": 22260
    },
    {
      "entropy": 5.591118860244751,
      "epoch": 2.0196843251088534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045921698719099527,
      "loss": 4.9821,
      "mean_token_accuracy": 0.20837253332138062,
      "num_tokens": 46205494.0,
      "step": 22265
    },
    {
      "entropy": 5.485752916336059,
      "epoch": 2.0201378809869377,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045919840748024764,
      "loss": 4.8794,
      "mean_token_accuracy": 0.21786799281835556,
      "num_tokens": 46216399.0,
      "step": 22270
    },
    {
      "entropy": 5.458440828323364,
      "epoch": 2.020591436865022,
      "grad_norm": 1.125,
      "learning_rate": 0.00045917982396023,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2182154029607773,
      "num_tokens": 46226515.0,
      "step": 22275
    },
    {
      "entropy": 5.5705328464508055,
      "epoch": 2.021044992743106,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045916123663132676,
      "loss": 4.9585,
      "mean_token_accuracy": 0.20380362272262573,
      "num_tokens": 46236688.0,
      "step": 22280
    },
    {
      "entropy": 5.529405927658081,
      "epoch": 2.02149854862119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045914264549392215,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22116508036851884,
      "num_tokens": 46247166.0,
      "step": 22285
    },
    {
      "entropy": 5.535774850845337,
      "epoch": 2.0219521044992743,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004591240505484007,
      "loss": 4.9098,
      "mean_token_accuracy": 0.2172283187508583,
      "num_tokens": 46258902.0,
      "step": 22290
    },
    {
      "entropy": 5.527719402313233,
      "epoch": 2.0224056603773586,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045910545179514694,
      "loss": 5.0005,
      "mean_token_accuracy": 0.21122104078531265,
      "num_tokens": 46269819.0,
      "step": 22295
    },
    {
      "entropy": 5.628784084320069,
      "epoch": 2.022859216255443,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004590868492345455,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21266111582517624,
      "num_tokens": 46280715.0,
      "step": 22300
    },
    {
      "entropy": 5.570830535888672,
      "epoch": 2.0233127721335267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004590682428669813,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20262421369552613,
      "num_tokens": 46291079.0,
      "step": 22305
    },
    {
      "entropy": 5.538163852691651,
      "epoch": 2.023766328011611,
      "grad_norm": 1.125,
      "learning_rate": 0.0004590496326928387,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21355263739824296,
      "num_tokens": 46301362.0,
      "step": 22310
    },
    {
      "entropy": 5.549081039428711,
      "epoch": 2.024219883889695,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045903101871250295,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21949705183506013,
      "num_tokens": 46311820.0,
      "step": 22315
    },
    {
      "entropy": 5.59455623626709,
      "epoch": 2.0246734397677795,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004590124009263587,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21171429008245468,
      "num_tokens": 46321992.0,
      "step": 22320
    },
    {
      "entropy": 5.504718542098999,
      "epoch": 2.0251269956458637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004589937793347912,
      "loss": 4.9583,
      "mean_token_accuracy": 0.20902350842952727,
      "num_tokens": 46331537.0,
      "step": 22325
    },
    {
      "entropy": 5.53143835067749,
      "epoch": 2.0255805515239476,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045897515393818547,
      "loss": 4.9057,
      "mean_token_accuracy": 0.2132235735654831,
      "num_tokens": 46342214.0,
      "step": 22330
    },
    {
      "entropy": 5.525017833709716,
      "epoch": 2.026034107402032,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004589565247369266,
      "loss": 4.9217,
      "mean_token_accuracy": 0.2118954434990883,
      "num_tokens": 46352471.0,
      "step": 22335
    },
    {
      "entropy": 5.514080429077149,
      "epoch": 2.026487663280116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004589378917314,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2153361514210701,
      "num_tokens": 46363221.0,
      "step": 22340
    },
    {
      "entropy": 5.487549591064453,
      "epoch": 2.0269412191582004,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045891925492199087,
      "loss": 4.817,
      "mean_token_accuracy": 0.22507339268922805,
      "num_tokens": 46373454.0,
      "step": 22345
    },
    {
      "entropy": 5.628279495239258,
      "epoch": 2.0273947750362846,
      "grad_norm": 1.125,
      "learning_rate": 0.0004589006143090847,
      "loss": 5.0692,
      "mean_token_accuracy": 0.20835783183574677,
      "num_tokens": 46384403.0,
      "step": 22350
    },
    {
      "entropy": 5.471293783187866,
      "epoch": 2.0278483309143684,
      "grad_norm": 1.25,
      "learning_rate": 0.00045888196989306697,
      "loss": 4.7959,
      "mean_token_accuracy": 0.22343188971281053,
      "num_tokens": 46393362.0,
      "step": 22355
    },
    {
      "entropy": 5.597118139266968,
      "epoch": 2.0283018867924527,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004588633216743232,
      "loss": 5.0084,
      "mean_token_accuracy": 0.21056696474552156,
      "num_tokens": 46405100.0,
      "step": 22360
    },
    {
      "entropy": 5.4636969566345215,
      "epoch": 2.028755442670537,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045884466965323914,
      "loss": 4.8548,
      "mean_token_accuracy": 0.21492289006710052,
      "num_tokens": 46415058.0,
      "step": 22365
    },
    {
      "entropy": 5.518535089492798,
      "epoch": 2.0292089985486212,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045882601383020045,
      "loss": 4.9063,
      "mean_token_accuracy": 0.21413395255804063,
      "num_tokens": 46425863.0,
      "step": 22370
    },
    {
      "entropy": 5.5287059307098385,
      "epoch": 2.0296625544267055,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004588073542055929,
      "loss": 4.9236,
      "mean_token_accuracy": 0.20940161347389222,
      "num_tokens": 46436172.0,
      "step": 22375
    },
    {
      "entropy": 5.582818269729614,
      "epoch": 2.0301161103047893,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045878869077980246,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21076412945985795,
      "num_tokens": 46446717.0,
      "step": 22380
    },
    {
      "entropy": 5.561581325531006,
      "epoch": 2.0305696661828736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045877002355321504,
      "loss": 4.9544,
      "mean_token_accuracy": 0.20341219156980514,
      "num_tokens": 46456944.0,
      "step": 22385
    },
    {
      "entropy": 5.550802707672119,
      "epoch": 2.031023222060958,
      "grad_norm": 1.125,
      "learning_rate": 0.0004587513525262168,
      "loss": 5.0148,
      "mean_token_accuracy": 0.21019573956727983,
      "num_tokens": 46467231.0,
      "step": 22390
    },
    {
      "entropy": 5.504215240478516,
      "epoch": 2.031476777939042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045873267769919364,
      "loss": 4.8692,
      "mean_token_accuracy": 0.21217937171459197,
      "num_tokens": 46477107.0,
      "step": 22395
    },
    {
      "entropy": 5.487004375457763,
      "epoch": 2.0319303338171264,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004587139990725319,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21539823561906815,
      "num_tokens": 46487351.0,
      "step": 22400
    },
    {
      "entropy": 5.543479490280151,
      "epoch": 2.0323838896952107,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004586953166466179,
      "loss": 4.9385,
      "mean_token_accuracy": 0.20999382436275482,
      "num_tokens": 46497685.0,
      "step": 22405
    },
    {
      "entropy": 5.548045253753662,
      "epoch": 2.0328374455732945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004586766304218379,
      "loss": 4.9381,
      "mean_token_accuracy": 0.2079530820250511,
      "num_tokens": 46507866.0,
      "step": 22410
    },
    {
      "entropy": 5.4384204864501955,
      "epoch": 2.0332910014513788,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045865794039857846,
      "loss": 4.8343,
      "mean_token_accuracy": 0.21661165207624436,
      "num_tokens": 46518797.0,
      "step": 22415
    },
    {
      "entropy": 5.527453994750976,
      "epoch": 2.033744557329463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004586392465772259,
      "loss": 4.8926,
      "mean_token_accuracy": 0.21332688331604005,
      "num_tokens": 46528759.0,
      "step": 22420
    },
    {
      "entropy": 5.588012456893921,
      "epoch": 2.0341981132075473,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045862054895816703,
      "loss": 5.049,
      "mean_token_accuracy": 0.20451307594776152,
      "num_tokens": 46539014.0,
      "step": 22425
    },
    {
      "entropy": 5.523033475875854,
      "epoch": 2.0346516690856316,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004586018475417884,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21793646663427352,
      "num_tokens": 46548856.0,
      "step": 22430
    },
    {
      "entropy": 5.4947190284729,
      "epoch": 2.0351052249637154,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004585831423284766,
      "loss": 4.8722,
      "mean_token_accuracy": 0.21460603028535843,
      "num_tokens": 46559164.0,
      "step": 22435
    },
    {
      "entropy": 5.535038471221924,
      "epoch": 2.0355587808417996,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004585644333186188,
      "loss": 4.9717,
      "mean_token_accuracy": 0.2081926241517067,
      "num_tokens": 46569793.0,
      "step": 22440
    },
    {
      "entropy": 5.537733650207519,
      "epoch": 2.036012336719884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045854572051260166,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21144626289606094,
      "num_tokens": 46580257.0,
      "step": 22445
    },
    {
      "entropy": 5.545389032363891,
      "epoch": 2.036465892597968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045852700391081234,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2072218418121338,
      "num_tokens": 46590297.0,
      "step": 22450
    },
    {
      "entropy": 5.494785642623901,
      "epoch": 2.0369194484760524,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004585082835136378,
      "loss": 4.8855,
      "mean_token_accuracy": 0.21596694141626357,
      "num_tokens": 46600647.0,
      "step": 22455
    },
    {
      "entropy": 5.540784883499145,
      "epoch": 2.0373730043541363,
      "grad_norm": 1.125,
      "learning_rate": 0.0004584895593214652,
      "loss": 4.8961,
      "mean_token_accuracy": 0.21155596673488616,
      "num_tokens": 46610478.0,
      "step": 22460
    },
    {
      "entropy": 5.536260795593262,
      "epoch": 2.0378265602322205,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045847083133468164,
      "loss": 4.963,
      "mean_token_accuracy": 0.20509119778871537,
      "num_tokens": 46620159.0,
      "step": 22465
    },
    {
      "entropy": 5.560859537124633,
      "epoch": 2.038280116110305,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045845209955367464,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21908921897411346,
      "num_tokens": 46630555.0,
      "step": 22470
    },
    {
      "entropy": 5.528761196136474,
      "epoch": 2.038733671988389,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045843336397883146,
      "loss": 4.9868,
      "mean_token_accuracy": 0.20852479040622712,
      "num_tokens": 46640188.0,
      "step": 22475
    },
    {
      "entropy": 5.511876916885376,
      "epoch": 2.0391872278664733,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004584146246105396,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21286656558513642,
      "num_tokens": 46650284.0,
      "step": 22480
    },
    {
      "entropy": 5.5852978229522705,
      "epoch": 2.039640783744557,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045839588144918643,
      "loss": 4.913,
      "mean_token_accuracy": 0.21420916318893432,
      "num_tokens": 46661466.0,
      "step": 22485
    },
    {
      "entropy": 5.563222122192383,
      "epoch": 2.0400943396226414,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004583771344951598,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21046785414218902,
      "num_tokens": 46671683.0,
      "step": 22490
    },
    {
      "entropy": 5.528917455673218,
      "epoch": 2.0405478955007257,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045835838374884733,
      "loss": 4.9602,
      "mean_token_accuracy": 0.20760292559862137,
      "num_tokens": 46681787.0,
      "step": 22495
    },
    {
      "entropy": 5.5417286396026615,
      "epoch": 2.04100145137881,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045833962921063665,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2139039784669876,
      "num_tokens": 46692752.0,
      "step": 22500
    },
    {
      "entropy": 5.573590040206909,
      "epoch": 2.041455007256894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045832087088091574,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2105260133743286,
      "num_tokens": 46702980.0,
      "step": 22505
    },
    {
      "entropy": 5.515853214263916,
      "epoch": 2.041908563134978,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045830210876007257,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2156961053609848,
      "num_tokens": 46712281.0,
      "step": 22510
    },
    {
      "entropy": 5.536145210266113,
      "epoch": 2.0423621190130623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000458283342848495,
      "loss": 4.8937,
      "mean_token_accuracy": 0.21441562920808793,
      "num_tokens": 46722097.0,
      "step": 22515
    },
    {
      "entropy": 5.469173765182495,
      "epoch": 2.0428156748911466,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004582645731465713,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22220447808504104,
      "num_tokens": 46733056.0,
      "step": 22520
    },
    {
      "entropy": 5.441329145431519,
      "epoch": 2.043269230769231,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045824579965468944,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2145723357796669,
      "num_tokens": 46742608.0,
      "step": 22525
    },
    {
      "entropy": 5.453447437286377,
      "epoch": 2.043722786647315,
      "grad_norm": 1.125,
      "learning_rate": 0.00045822702237323777,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22509369850158692,
      "num_tokens": 46751762.0,
      "step": 22530
    },
    {
      "entropy": 5.495344400405884,
      "epoch": 2.044176342525399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045820824130260457,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2206840693950653,
      "num_tokens": 46762252.0,
      "step": 22535
    },
    {
      "entropy": 5.508134508132935,
      "epoch": 2.044629898403483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004581894564431783,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20868685096502304,
      "num_tokens": 46772172.0,
      "step": 22540
    },
    {
      "entropy": 5.500317096710205,
      "epoch": 2.0450834542815675,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004581706677953473,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21302879750728607,
      "num_tokens": 46782894.0,
      "step": 22545
    },
    {
      "entropy": 5.534683704376221,
      "epoch": 2.0455370101596517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045815187535950034,
      "loss": 4.9319,
      "mean_token_accuracy": 0.2115266814827919,
      "num_tokens": 46793097.0,
      "step": 22550
    },
    {
      "entropy": 5.5787821292877195,
      "epoch": 2.045990566037736,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045813307913602586,
      "loss": 5.0236,
      "mean_token_accuracy": 0.20176143050193787,
      "num_tokens": 46804196.0,
      "step": 22555
    },
    {
      "entropy": 5.563619995117188,
      "epoch": 2.04644412191582,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004581142791253126,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21111859381198883,
      "num_tokens": 46813985.0,
      "step": 22560
    },
    {
      "entropy": 5.469507551193237,
      "epoch": 2.046897677793904,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045809547532774943,
      "loss": 4.8258,
      "mean_token_accuracy": 0.21234235614538194,
      "num_tokens": 46824125.0,
      "step": 22565
    },
    {
      "entropy": 5.4895566463470455,
      "epoch": 2.0473512336719883,
      "grad_norm": 1.125,
      "learning_rate": 0.00045807666774372514,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2142532303929329,
      "num_tokens": 46834982.0,
      "step": 22570
    },
    {
      "entropy": 5.520470333099365,
      "epoch": 2.0478047895500726,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045805785637362874,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22265964448451997,
      "num_tokens": 46844669.0,
      "step": 22575
    },
    {
      "entropy": 5.494301128387451,
      "epoch": 2.048258345428157,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004580390412178492,
      "loss": 4.9412,
      "mean_token_accuracy": 0.2141905978322029,
      "num_tokens": 46855108.0,
      "step": 22580
    },
    {
      "entropy": 5.508283233642578,
      "epoch": 2.048711901306241,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045802022227677566,
      "loss": 4.8543,
      "mean_token_accuracy": 0.21525073945522308,
      "num_tokens": 46864878.0,
      "step": 22585
    },
    {
      "entropy": 5.518833255767822,
      "epoch": 2.049165457184325,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045800139955079727,
      "loss": 4.9398,
      "mean_token_accuracy": 0.2127300888299942,
      "num_tokens": 46874916.0,
      "step": 22590
    },
    {
      "entropy": 5.585209655761719,
      "epoch": 2.0496190130624092,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004579825730403033,
      "loss": 4.9875,
      "mean_token_accuracy": 0.2057396098971367,
      "num_tokens": 46885167.0,
      "step": 22595
    },
    {
      "entropy": 5.602046298980713,
      "epoch": 2.0500725689404935,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004579637427456831,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2065080761909485,
      "num_tokens": 46895526.0,
      "step": 22600
    },
    {
      "entropy": 5.50861611366272,
      "epoch": 2.0505261248185778,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045794490866732607,
      "loss": 4.9041,
      "mean_token_accuracy": 0.2166951358318329,
      "num_tokens": 46905426.0,
      "step": 22605
    },
    {
      "entropy": 5.5101197242736815,
      "epoch": 2.050979680696662,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004579260708056217,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21042592376470565,
      "num_tokens": 46915728.0,
      "step": 22610
    },
    {
      "entropy": 5.484219360351562,
      "epoch": 2.051433236574746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004579072291609596,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21287375688552856,
      "num_tokens": 46925972.0,
      "step": 22615
    },
    {
      "entropy": 5.504348516464233,
      "epoch": 2.05188679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045788838373372935,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21596241146326065,
      "num_tokens": 46937108.0,
      "step": 22620
    },
    {
      "entropy": 5.484693193435669,
      "epoch": 2.0523403483309144,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045786953452432076,
      "loss": 4.8774,
      "mean_token_accuracy": 0.21151575148105622,
      "num_tokens": 46947684.0,
      "step": 22625
    },
    {
      "entropy": 5.594632816314697,
      "epoch": 2.0527939042089987,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004578506815331236,
      "loss": 4.9956,
      "mean_token_accuracy": 0.20563028007745743,
      "num_tokens": 46958089.0,
      "step": 22630
    },
    {
      "entropy": 5.529718637466431,
      "epoch": 2.053247460087083,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004578318247605277,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22268491387367248,
      "num_tokens": 46968618.0,
      "step": 22635
    },
    {
      "entropy": 5.55033860206604,
      "epoch": 2.0537010159651667,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004578129642069231,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2134891152381897,
      "num_tokens": 46978791.0,
      "step": 22640
    },
    {
      "entropy": 5.5299859046936035,
      "epoch": 2.054154571843251,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004577940998726998,
      "loss": 4.8603,
      "mean_token_accuracy": 0.21561755388975143,
      "num_tokens": 46988706.0,
      "step": 22645
    },
    {
      "entropy": 5.604277801513672,
      "epoch": 2.0546081277213353,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045777523175824796,
      "loss": 5.0524,
      "mean_token_accuracy": 0.19636777341365813,
      "num_tokens": 47000140.0,
      "step": 22650
    },
    {
      "entropy": 5.5475939273834225,
      "epoch": 2.0550616835994195,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004577563598639577,
      "loss": 4.871,
      "mean_token_accuracy": 0.2143001601099968,
      "num_tokens": 47009773.0,
      "step": 22655
    },
    {
      "entropy": 5.599683809280395,
      "epoch": 2.055515239477504,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045773748419021944,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2070373550057411,
      "num_tokens": 47020326.0,
      "step": 22660
    },
    {
      "entropy": 5.502377033233643,
      "epoch": 2.0559687953555876,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045771860473742347,
      "loss": 4.8921,
      "mean_token_accuracy": 0.20433692783117294,
      "num_tokens": 47030410.0,
      "step": 22665
    },
    {
      "entropy": 5.55394434928894,
      "epoch": 2.056422351233672,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045769972150596006,
      "loss": 4.9272,
      "mean_token_accuracy": 0.20738234966993332,
      "num_tokens": 47040546.0,
      "step": 22670
    },
    {
      "entropy": 5.545335149765014,
      "epoch": 2.056875907111756,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045768083449621987,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2232957437634468,
      "num_tokens": 47050210.0,
      "step": 22675
    },
    {
      "entropy": 5.638255786895752,
      "epoch": 2.0573294629898404,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045766194370859347,
      "loss": 5.0452,
      "mean_token_accuracy": 0.20277032554149627,
      "num_tokens": 47060641.0,
      "step": 22680
    },
    {
      "entropy": 5.529693841934204,
      "epoch": 2.0577830188679247,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004576430491434716,
      "loss": 4.963,
      "mean_token_accuracy": 0.21088709086179733,
      "num_tokens": 47070737.0,
      "step": 22685
    },
    {
      "entropy": 5.582369947433472,
      "epoch": 2.0582365747460085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045762415080124487,
      "loss": 4.918,
      "mean_token_accuracy": 0.2172081097960472,
      "num_tokens": 47080884.0,
      "step": 22690
    },
    {
      "entropy": 5.532611751556397,
      "epoch": 2.058690130624093,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045760524868230417,
      "loss": 4.9722,
      "mean_token_accuracy": 0.20996871143579482,
      "num_tokens": 47090865.0,
      "step": 22695
    },
    {
      "entropy": 5.541356229782105,
      "epoch": 2.059143686502177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004575863427870404,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21559229493141174,
      "num_tokens": 47100570.0,
      "step": 22700
    },
    {
      "entropy": 5.520215177536011,
      "epoch": 2.0595972423802613,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004575674331158446,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21572486758232118,
      "num_tokens": 47111199.0,
      "step": 22705
    },
    {
      "entropy": 5.5142409324646,
      "epoch": 2.0600507982583456,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575485196691076,
      "loss": 4.9105,
      "mean_token_accuracy": 0.2087768018245697,
      "num_tokens": 47121191.0,
      "step": 22710
    },
    {
      "entropy": 5.574739074707031,
      "epoch": 2.0605043541364294,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004575296024472209,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21077124625444413,
      "num_tokens": 47131608.0,
      "step": 22715
    },
    {
      "entropy": 5.599815845489502,
      "epoch": 2.0609579100145137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004575106814505753,
      "loss": 4.939,
      "mean_token_accuracy": 0.20619721263647078,
      "num_tokens": 47141492.0,
      "step": 22720
    },
    {
      "entropy": 5.494761610031128,
      "epoch": 2.061411465892598,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004574917566795624,
      "loss": 4.815,
      "mean_token_accuracy": 0.22166116982698442,
      "num_tokens": 47152576.0,
      "step": 22725
    },
    {
      "entropy": 5.531712627410888,
      "epoch": 2.061865021770682,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045747282813457345,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21075544357299805,
      "num_tokens": 47163863.0,
      "step": 22730
    },
    {
      "entropy": 5.5498847484588625,
      "epoch": 2.0623185776487665,
      "grad_norm": 1.171875,
      "learning_rate": 0.000457453895816,
      "loss": 4.8902,
      "mean_token_accuracy": 0.212752041220665,
      "num_tokens": 47174140.0,
      "step": 22735
    },
    {
      "entropy": 5.5516307830810545,
      "epoch": 2.0627721335268503,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045743495972423346,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2083651453256607,
      "num_tokens": 47184708.0,
      "step": 22740
    },
    {
      "entropy": 5.50225715637207,
      "epoch": 2.0632256894049346,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004574160198596653,
      "loss": 4.9248,
      "mean_token_accuracy": 0.20785647481679917,
      "num_tokens": 47196286.0,
      "step": 22745
    },
    {
      "entropy": 5.437078523635864,
      "epoch": 2.063679245283019,
      "grad_norm": 1.125,
      "learning_rate": 0.00045739707622268757,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22240602523088454,
      "num_tokens": 47206774.0,
      "step": 22750
    },
    {
      "entropy": 5.459208011627197,
      "epoch": 2.064132801161103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004573781288136917,
      "loss": 4.7976,
      "mean_token_accuracy": 0.22010296136140822,
      "num_tokens": 47217009.0,
      "step": 22755
    },
    {
      "entropy": 5.516549205780029,
      "epoch": 2.0645863570391874,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045735917763306967,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2119544565677643,
      "num_tokens": 47226718.0,
      "step": 22760
    },
    {
      "entropy": 5.475437021255493,
      "epoch": 2.065039912917271,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004573402226812134,
      "loss": 4.8144,
      "mean_token_accuracy": 0.21321513056755065,
      "num_tokens": 47237666.0,
      "step": 22765
    },
    {
      "entropy": 5.551860761642456,
      "epoch": 2.0654934687953554,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045732126395851483,
      "loss": 4.9481,
      "mean_token_accuracy": 0.2054678812623024,
      "num_tokens": 47249230.0,
      "step": 22770
    },
    {
      "entropy": 5.509599494934082,
      "epoch": 2.0659470246734397,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004573023014653661,
      "loss": 4.879,
      "mean_token_accuracy": 0.21760220378637313,
      "num_tokens": 47259232.0,
      "step": 22775
    },
    {
      "entropy": 5.521780395507813,
      "epoch": 2.066400580551524,
      "grad_norm": 1.125,
      "learning_rate": 0.00045728333520215933,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21306896358728408,
      "num_tokens": 47269006.0,
      "step": 22780
    },
    {
      "entropy": 5.588029479980468,
      "epoch": 2.0668541364296082,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004572643651692867,
      "loss": 5.0109,
      "mean_token_accuracy": 0.20081331729888915,
      "num_tokens": 47280048.0,
      "step": 22785
    },
    {
      "entropy": 5.508160257339478,
      "epoch": 2.0673076923076925,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045724539136714063,
      "loss": 4.85,
      "mean_token_accuracy": 0.21479624062776564,
      "num_tokens": 47290019.0,
      "step": 22790
    },
    {
      "entropy": 5.482350778579712,
      "epoch": 2.0677612481857763,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004572264137961133,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21641735583543778,
      "num_tokens": 47300573.0,
      "step": 22795
    },
    {
      "entropy": 5.472970771789551,
      "epoch": 2.0682148040638606,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045720743245659734,
      "loss": 4.894,
      "mean_token_accuracy": 0.21207104623317719,
      "num_tokens": 47311795.0,
      "step": 22800
    },
    {
      "entropy": 5.479173707962036,
      "epoch": 2.068668359941945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004571884473489853,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22470548003911972,
      "num_tokens": 47321768.0,
      "step": 22805
    },
    {
      "entropy": 5.548162031173706,
      "epoch": 2.069121915820029,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004571694584736696,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21237631440162658,
      "num_tokens": 47332662.0,
      "step": 22810
    },
    {
      "entropy": 5.497155094146729,
      "epoch": 2.0695754716981134,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004571504658310432,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2143186002969742,
      "num_tokens": 47344355.0,
      "step": 22815
    },
    {
      "entropy": 5.504010534286499,
      "epoch": 2.0700290275761972,
      "grad_norm": 1.125,
      "learning_rate": 0.0004571314694214987,
      "loss": 4.8566,
      "mean_token_accuracy": 0.21746133416891097,
      "num_tokens": 47354228.0,
      "step": 22820
    },
    {
      "entropy": 5.539687204360962,
      "epoch": 2.0704825834542815,
      "grad_norm": 1.125,
      "learning_rate": 0.0004571124692454291,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22003219425678253,
      "num_tokens": 47364477.0,
      "step": 22825
    },
    {
      "entropy": 5.5435566902160645,
      "epoch": 2.0709361393323658,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045709346530322715,
      "loss": 4.9695,
      "mean_token_accuracy": 0.21325616538524628,
      "num_tokens": 47375745.0,
      "step": 22830
    },
    {
      "entropy": 5.540888690948487,
      "epoch": 2.07138969521045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004570744575952859,
      "loss": 4.996,
      "mean_token_accuracy": 0.21818966269493104,
      "num_tokens": 47385341.0,
      "step": 22835
    },
    {
      "entropy": 5.525907325744629,
      "epoch": 2.0718432510885343,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004570554461219985,
      "loss": 4.9118,
      "mean_token_accuracy": 0.2187253162264824,
      "num_tokens": 47396266.0,
      "step": 22840
    },
    {
      "entropy": 5.556705474853516,
      "epoch": 2.072296806966618,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045703643088375806,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21159299612045288,
      "num_tokens": 47406772.0,
      "step": 22845
    },
    {
      "entropy": 5.5891398906707765,
      "epoch": 2.0727503628447024,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045701741188095793,
      "loss": 5.0162,
      "mean_token_accuracy": 0.20717564225196838,
      "num_tokens": 47416833.0,
      "step": 22850
    },
    {
      "entropy": 5.524143171310425,
      "epoch": 2.0732039187227866,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004569983891139912,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21510811299085617,
      "num_tokens": 47426954.0,
      "step": 22855
    },
    {
      "entropy": 5.4667887687683105,
      "epoch": 2.073657474600871,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045697936258325155,
      "loss": 4.856,
      "mean_token_accuracy": 0.22599812150001525,
      "num_tokens": 47437520.0,
      "step": 22860
    },
    {
      "entropy": 5.506705570220947,
      "epoch": 2.074111030478955,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045696033228913226,
      "loss": 4.8753,
      "mean_token_accuracy": 0.2109182059764862,
      "num_tokens": 47447977.0,
      "step": 22865
    },
    {
      "entropy": 5.556794309616089,
      "epoch": 2.074564586357039,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045694129823202684,
      "loss": 4.9571,
      "mean_token_accuracy": 0.2082996279001236,
      "num_tokens": 47458458.0,
      "step": 22870
    },
    {
      "entropy": 5.556359338760376,
      "epoch": 2.0750181422351233,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004569222604123291,
      "loss": 4.936,
      "mean_token_accuracy": 0.2108566403388977,
      "num_tokens": 47469048.0,
      "step": 22875
    },
    {
      "entropy": 5.56695237159729,
      "epoch": 2.0754716981132075,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004569032188304326,
      "loss": 4.9747,
      "mean_token_accuracy": 0.20354607701301575,
      "num_tokens": 47480353.0,
      "step": 22880
    },
    {
      "entropy": 5.4839620113372805,
      "epoch": 2.075925253991292,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004568841734867313,
      "loss": 4.8067,
      "mean_token_accuracy": 0.21461614966392517,
      "num_tokens": 47489951.0,
      "step": 22885
    },
    {
      "entropy": 5.475849628448486,
      "epoch": 2.076378809869376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004568651243816188,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21569481045007705,
      "num_tokens": 47499015.0,
      "step": 22890
    },
    {
      "entropy": 5.505986928939819,
      "epoch": 2.07683236574746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045684607151548923,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2143254652619362,
      "num_tokens": 47508694.0,
      "step": 22895
    },
    {
      "entropy": 5.640796804428101,
      "epoch": 2.077285921625544,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045682701488873655,
      "loss": 5.0588,
      "mean_token_accuracy": 0.20158324092626573,
      "num_tokens": 47521336.0,
      "step": 22900
    },
    {
      "entropy": 5.583995056152344,
      "epoch": 2.0777394775036284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004568079545017549,
      "loss": 5.0432,
      "mean_token_accuracy": 0.2028840884566307,
      "num_tokens": 47531857.0,
      "step": 22905
    },
    {
      "entropy": 5.566130113601685,
      "epoch": 2.0781930333817127,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004567888903549385,
      "loss": 4.955,
      "mean_token_accuracy": 0.20707421153783798,
      "num_tokens": 47542224.0,
      "step": 22910
    },
    {
      "entropy": 5.5337542533874515,
      "epoch": 2.078646589259797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004567698224486813,
      "loss": 4.9566,
      "mean_token_accuracy": 0.21091546416282653,
      "num_tokens": 47552370.0,
      "step": 22915
    },
    {
      "entropy": 5.646003675460816,
      "epoch": 2.0791001451378808,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045675075078337803,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21830417364835739,
      "num_tokens": 47561997.0,
      "step": 22920
    },
    {
      "entropy": 5.573488521575928,
      "epoch": 2.079553701015965,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004567316753594229,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21337729543447495,
      "num_tokens": 47571968.0,
      "step": 22925
    },
    {
      "entropy": 5.564603328704834,
      "epoch": 2.0800072568940493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045671259617721037,
      "loss": 4.9454,
      "mean_token_accuracy": 0.20923822224140168,
      "num_tokens": 47583459.0,
      "step": 22930
    },
    {
      "entropy": 5.518136310577392,
      "epoch": 2.0804608127721336,
      "grad_norm": 1.09375,
      "learning_rate": 0.000456693513237135,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22079051584005355,
      "num_tokens": 47594026.0,
      "step": 22935
    },
    {
      "entropy": 5.5820245265960695,
      "epoch": 2.080914368650218,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004566744265395916,
      "loss": 4.9392,
      "mean_token_accuracy": 0.20952713191509248,
      "num_tokens": 47605246.0,
      "step": 22940
    },
    {
      "entropy": 5.599461698532105,
      "epoch": 2.081367924528302,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045665533608497467,
      "loss": 5.0168,
      "mean_token_accuracy": 0.20179500430822372,
      "num_tokens": 47616702.0,
      "step": 22945
    },
    {
      "entropy": 5.577863550186157,
      "epoch": 2.081821480406386,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045663624187367914,
      "loss": 4.9654,
      "mean_token_accuracy": 0.2085647851228714,
      "num_tokens": 47627616.0,
      "step": 22950
    },
    {
      "entropy": 5.512732648849488,
      "epoch": 2.08227503628447,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004566171439060998,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21078101247549058,
      "num_tokens": 47637287.0,
      "step": 22955
    },
    {
      "entropy": 5.480682992935181,
      "epoch": 2.0827285921625545,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004565980421826317,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21255559623241424,
      "num_tokens": 47647474.0,
      "step": 22960
    },
    {
      "entropy": 5.584753847122192,
      "epoch": 2.0831821480406387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004565789367036697,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21943684965372084,
      "num_tokens": 47657036.0,
      "step": 22965
    },
    {
      "entropy": 5.533649063110351,
      "epoch": 2.083635703918723,
      "grad_norm": 1.125,
      "learning_rate": 0.0004565598274696091,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2158496081829071,
      "num_tokens": 47668026.0,
      "step": 22970
    },
    {
      "entropy": 5.598115539550781,
      "epoch": 2.084089259796807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045654071448084506,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20769884884357454,
      "num_tokens": 47679508.0,
      "step": 22975
    },
    {
      "entropy": 5.53518500328064,
      "epoch": 2.084542815674891,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045652159773777273,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21275254786014558,
      "num_tokens": 47688474.0,
      "step": 22980
    },
    {
      "entropy": 5.661990642547607,
      "epoch": 2.0849963715529753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004565024772407875,
      "loss": 5.0858,
      "mean_token_accuracy": 0.19668970108032227,
      "num_tokens": 47698675.0,
      "step": 22985
    },
    {
      "entropy": 5.50940899848938,
      "epoch": 2.0854499274310596,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004564833529902848,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22405688911676408,
      "num_tokens": 47709214.0,
      "step": 22990
    },
    {
      "entropy": 5.5617471694946286,
      "epoch": 2.085903483309144,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045646422498666,
      "loss": 5.0087,
      "mean_token_accuracy": 0.21112808883190154,
      "num_tokens": 47721000.0,
      "step": 22995
    },
    {
      "entropy": 5.492447853088379,
      "epoch": 2.0863570391872277,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045644509323030893,
      "loss": 4.901,
      "mean_token_accuracy": 0.2128939688205719,
      "num_tokens": 47731218.0,
      "step": 23000
    },
    {
      "entropy": 5.569713973999024,
      "epoch": 2.086810595065312,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045642595772162704,
      "loss": 4.9015,
      "mean_token_accuracy": 0.21092294603586198,
      "num_tokens": 47742339.0,
      "step": 23005
    },
    {
      "entropy": 5.592831325531006,
      "epoch": 2.0872641509433962,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045640681846101014,
      "loss": 4.9883,
      "mean_token_accuracy": 0.2099862962961197,
      "num_tokens": 47752858.0,
      "step": 23010
    },
    {
      "entropy": 5.494131135940552,
      "epoch": 2.0877177068214805,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004563876754488539,
      "loss": 4.8562,
      "mean_token_accuracy": 0.21762501895427705,
      "num_tokens": 47763303.0,
      "step": 23015
    },
    {
      "entropy": 5.551550579071045,
      "epoch": 2.0881712626995648,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004563685286855544,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21246798634529113,
      "num_tokens": 47773212.0,
      "step": 23020
    },
    {
      "entropy": 5.590135622024536,
      "epoch": 2.0886248185776486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004563493781715075,
      "loss": 5.0244,
      "mean_token_accuracy": 0.2087484046816826,
      "num_tokens": 47783376.0,
      "step": 23025
    },
    {
      "entropy": 5.500467014312744,
      "epoch": 2.089078374455733,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004563302239071093,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21255954951047898,
      "num_tokens": 47794859.0,
      "step": 23030
    },
    {
      "entropy": 5.581085920333862,
      "epoch": 2.089531930333817,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045631106589275576,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21208027005195618,
      "num_tokens": 47804721.0,
      "step": 23035
    },
    {
      "entropy": 5.517194938659668,
      "epoch": 2.0899854862119014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045629190412884324,
      "loss": 4.8169,
      "mean_token_accuracy": 0.21927345395088196,
      "num_tokens": 47814701.0,
      "step": 23040
    },
    {
      "entropy": 5.545892810821533,
      "epoch": 2.0904390420899857,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045627273861576786,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21184692680835723,
      "num_tokens": 47823941.0,
      "step": 23045
    },
    {
      "entropy": 5.5316606044769285,
      "epoch": 2.0908925979680695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045625356935392614,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21564259082078935,
      "num_tokens": 47835793.0,
      "step": 23050
    },
    {
      "entropy": 5.585255241394043,
      "epoch": 2.0913461538461537,
      "grad_norm": 1.125,
      "learning_rate": 0.0004562343963437144,
      "loss": 4.9802,
      "mean_token_accuracy": 0.2114524558186531,
      "num_tokens": 47845677.0,
      "step": 23055
    },
    {
      "entropy": 5.564328336715699,
      "epoch": 2.091799709724238,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004562152195855291,
      "loss": 4.8786,
      "mean_token_accuracy": 0.22363824993371964,
      "num_tokens": 47855611.0,
      "step": 23060
    },
    {
      "entropy": 5.499240446090698,
      "epoch": 2.0922532656023223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004561960390797669,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2147802948951721,
      "num_tokens": 47865115.0,
      "step": 23065
    },
    {
      "entropy": 5.630834913253784,
      "epoch": 2.0927068214804065,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045617685482682454,
      "loss": 4.997,
      "mean_token_accuracy": 0.21477146595716476,
      "num_tokens": 47876690.0,
      "step": 23070
    },
    {
      "entropy": 5.528329467773437,
      "epoch": 2.0931603773584904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004561576668270986,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2105965033173561,
      "num_tokens": 47887709.0,
      "step": 23075
    },
    {
      "entropy": 5.52317180633545,
      "epoch": 2.0936139332365746,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045613847508098587,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2128834217786789,
      "num_tokens": 47898209.0,
      "step": 23080
    },
    {
      "entropy": 5.534544134140015,
      "epoch": 2.094067489114659,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045611927958888335,
      "loss": 4.813,
      "mean_token_accuracy": 0.2201571837067604,
      "num_tokens": 47907494.0,
      "step": 23085
    },
    {
      "entropy": 5.557372570037842,
      "epoch": 2.094521044992743,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045610008035118797,
      "loss": 4.9606,
      "mean_token_accuracy": 0.21362939476966858,
      "num_tokens": 47916923.0,
      "step": 23090
    },
    {
      "entropy": 5.471154069900512,
      "epoch": 2.0949746008708274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045608087736829687,
      "loss": 4.849,
      "mean_token_accuracy": 0.21858763098716735,
      "num_tokens": 47927043.0,
      "step": 23095
    },
    {
      "entropy": 5.546939325332642,
      "epoch": 2.0954281567489113,
      "grad_norm": 1.0625,
      "learning_rate": 0.000456061670640607,
      "loss": 4.9062,
      "mean_token_accuracy": 0.2108116030693054,
      "num_tokens": 47937296.0,
      "step": 23100
    },
    {
      "entropy": 5.560381174087524,
      "epoch": 2.0958817126269955,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004560424601685157,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21296298801898955,
      "num_tokens": 47948268.0,
      "step": 23105
    },
    {
      "entropy": 5.545624303817749,
      "epoch": 2.09633526850508,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004560232459524201,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2043468490242958,
      "num_tokens": 47958375.0,
      "step": 23110
    },
    {
      "entropy": 5.53116626739502,
      "epoch": 2.096788824383164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045600402799271766,
      "loss": 4.968,
      "mean_token_accuracy": 0.21385799944400788,
      "num_tokens": 47968469.0,
      "step": 23115
    },
    {
      "entropy": 5.519662380218506,
      "epoch": 2.0972423802612483,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004559848062898058,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22248961180448532,
      "num_tokens": 47979630.0,
      "step": 23120
    },
    {
      "entropy": 5.5345861434936525,
      "epoch": 2.097695936139332,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004559655808440821,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21655319035053253,
      "num_tokens": 47989828.0,
      "step": 23125
    },
    {
      "entropy": 5.60052638053894,
      "epoch": 2.0981494920174164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045594635165594404,
      "loss": 5.0252,
      "mean_token_accuracy": 0.20086491256952285,
      "num_tokens": 47999219.0,
      "step": 23130
    },
    {
      "entropy": 5.553136396408081,
      "epoch": 2.0986030478955007,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045592711872578925,
      "loss": 4.913,
      "mean_token_accuracy": 0.22073778212070466,
      "num_tokens": 48009244.0,
      "step": 23135
    },
    {
      "entropy": 5.551522827148437,
      "epoch": 2.099056603773585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045590788205401557,
      "loss": 4.8757,
      "mean_token_accuracy": 0.21680557876825332,
      "num_tokens": 48019066.0,
      "step": 23140
    },
    {
      "entropy": 5.468857049942017,
      "epoch": 2.099510159651669,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045588864164102086,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2165501222014427,
      "num_tokens": 48028704.0,
      "step": 23145
    },
    {
      "entropy": 5.494235324859619,
      "epoch": 2.0999637155297535,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045586939748720284,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2161787763237953,
      "num_tokens": 48039210.0,
      "step": 23150
    },
    {
      "entropy": 5.603004693984985,
      "epoch": 2.1004172714078373,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045585014959295965,
      "loss": 5.0233,
      "mean_token_accuracy": 0.206943379342556,
      "num_tokens": 48048446.0,
      "step": 23155
    },
    {
      "entropy": 5.55800986289978,
      "epoch": 2.1008708272859216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045583089795868934,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21900135576725005,
      "num_tokens": 48058880.0,
      "step": 23160
    },
    {
      "entropy": 5.572850227355957,
      "epoch": 2.101324383164006,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004558116425847899,
      "loss": 4.9633,
      "mean_token_accuracy": 0.21828603297472,
      "num_tokens": 48070224.0,
      "step": 23165
    },
    {
      "entropy": 5.543377351760864,
      "epoch": 2.10177793904209,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004557923834716596,
      "loss": 4.9687,
      "mean_token_accuracy": 0.20585934370756148,
      "num_tokens": 48080914.0,
      "step": 23170
    },
    {
      "entropy": 5.484085178375244,
      "epoch": 2.1022314949201744,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045577312061969677,
      "loss": 4.9461,
      "mean_token_accuracy": 0.20721143782138823,
      "num_tokens": 48090873.0,
      "step": 23175
    },
    {
      "entropy": 5.62333574295044,
      "epoch": 2.102685050798258,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004557538540292997,
      "loss": 4.9605,
      "mean_token_accuracy": 0.20777802020311356,
      "num_tokens": 48101885.0,
      "step": 23180
    },
    {
      "entropy": 5.587438583374023,
      "epoch": 2.1031386066763424,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004557345837008669,
      "loss": 4.978,
      "mean_token_accuracy": 0.21354096680879592,
      "num_tokens": 48111472.0,
      "step": 23185
    },
    {
      "entropy": 5.455753326416016,
      "epoch": 2.1035921625544267,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045571530963479684,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22404519766569136,
      "num_tokens": 48121351.0,
      "step": 23190
    },
    {
      "entropy": 5.572947311401367,
      "epoch": 2.104045718432511,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045569603183148814,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2221685528755188,
      "num_tokens": 48131104.0,
      "step": 23195
    },
    {
      "entropy": 5.599318218231201,
      "epoch": 2.1044992743105952,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004556767502913395,
      "loss": 4.9944,
      "mean_token_accuracy": 0.21002317667007447,
      "num_tokens": 48142895.0,
      "step": 23200
    },
    {
      "entropy": 5.474131917953491,
      "epoch": 2.104952830188679,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045565746501474946,
      "loss": 4.7994,
      "mean_token_accuracy": 0.2177934542298317,
      "num_tokens": 48153294.0,
      "step": 23205
    },
    {
      "entropy": 5.55059494972229,
      "epoch": 2.1054063860667633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045563817600211714,
      "loss": 4.982,
      "mean_token_accuracy": 0.20749528110027313,
      "num_tokens": 48164319.0,
      "step": 23210
    },
    {
      "entropy": 5.560405015945435,
      "epoch": 2.1058599419448476,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045561888325384124,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21026235818862915,
      "num_tokens": 48173918.0,
      "step": 23215
    },
    {
      "entropy": 5.598680114746093,
      "epoch": 2.106313497822932,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004555995867703209,
      "loss": 4.9403,
      "mean_token_accuracy": 0.2063882529735565,
      "num_tokens": 48184127.0,
      "step": 23220
    },
    {
      "entropy": 5.5045835971832275,
      "epoch": 2.106767053701016,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045558028655195504,
      "loss": 4.883,
      "mean_token_accuracy": 0.21133557558059693,
      "num_tokens": 48194198.0,
      "step": 23225
    },
    {
      "entropy": 5.459793138504028,
      "epoch": 2.1072206095791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045556098259914293,
      "loss": 4.893,
      "mean_token_accuracy": 0.210208959877491,
      "num_tokens": 48204654.0,
      "step": 23230
    },
    {
      "entropy": 5.56398286819458,
      "epoch": 2.1076741654571842,
      "grad_norm": 1.125,
      "learning_rate": 0.00045554167491228357,
      "loss": 5.0368,
      "mean_token_accuracy": 0.19844679087400435,
      "num_tokens": 48215322.0,
      "step": 23235
    },
    {
      "entropy": 5.525118637084961,
      "epoch": 2.1081277213352685,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045552236349177637,
      "loss": 4.8134,
      "mean_token_accuracy": 0.22347820699214935,
      "num_tokens": 48226142.0,
      "step": 23240
    },
    {
      "entropy": 5.488425874710083,
      "epoch": 2.1085812772133528,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045550304833802066,
      "loss": 4.8754,
      "mean_token_accuracy": 0.21439291387796403,
      "num_tokens": 48236896.0,
      "step": 23245
    },
    {
      "entropy": 5.463250827789307,
      "epoch": 2.109034833091437,
      "grad_norm": 1.0625,
      "learning_rate": 0.000455483729451416,
      "loss": 4.89,
      "mean_token_accuracy": 0.21024914532899858,
      "num_tokens": 48247425.0,
      "step": 23250
    },
    {
      "entropy": 5.550429773330689,
      "epoch": 2.109488388969521,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045546440683236173,
      "loss": 4.9799,
      "mean_token_accuracy": 0.2096166729927063,
      "num_tokens": 48258560.0,
      "step": 23255
    },
    {
      "entropy": 5.610762453079223,
      "epoch": 2.109941944847605,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045544508048125763,
      "loss": 5.016,
      "mean_token_accuracy": 0.208012393116951,
      "num_tokens": 48269469.0,
      "step": 23260
    },
    {
      "entropy": 5.569597482681274,
      "epoch": 2.1103955007256894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004554257503985032,
      "loss": 4.92,
      "mean_token_accuracy": 0.2107689216732979,
      "num_tokens": 48278847.0,
      "step": 23265
    },
    {
      "entropy": 5.490497255325318,
      "epoch": 2.1108490566037736,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045540641658449834,
      "loss": 4.8601,
      "mean_token_accuracy": 0.21419641524553298,
      "num_tokens": 48288091.0,
      "step": 23270
    },
    {
      "entropy": 5.472599220275879,
      "epoch": 2.111302612481858,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045538707903964275,
      "loss": 4.871,
      "mean_token_accuracy": 0.2120255082845688,
      "num_tokens": 48297946.0,
      "step": 23275
    },
    {
      "entropy": 5.504520463943481,
      "epoch": 2.1117561683599417,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045536773776433647,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21649318933486938,
      "num_tokens": 48307650.0,
      "step": 23280
    },
    {
      "entropy": 5.571960163116455,
      "epoch": 2.112209724238026,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045534839275897937,
      "loss": 5.0161,
      "mean_token_accuracy": 0.1976493254303932,
      "num_tokens": 48318824.0,
      "step": 23285
    },
    {
      "entropy": 5.528266716003418,
      "epoch": 2.1126632801161103,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045532904402397157,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2127891704440117,
      "num_tokens": 48329909.0,
      "step": 23290
    },
    {
      "entropy": 5.564465475082398,
      "epoch": 2.1131168359941945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004553096915597132,
      "loss": 5.0432,
      "mean_token_accuracy": 0.20372509360313415,
      "num_tokens": 48340142.0,
      "step": 23295
    },
    {
      "entropy": 5.521903038024902,
      "epoch": 2.113570391872279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004552903353666044,
      "loss": 4.879,
      "mean_token_accuracy": 0.21618664562702178,
      "num_tokens": 48350540.0,
      "step": 23300
    },
    {
      "entropy": 5.539557456970215,
      "epoch": 2.114023947750363,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045527097544504556,
      "loss": 4.8604,
      "mean_token_accuracy": 0.218427437543869,
      "num_tokens": 48360153.0,
      "step": 23305
    },
    {
      "entropy": 5.525097608566284,
      "epoch": 2.114477503628447,
      "grad_norm": 1.1875,
      "learning_rate": 0.000455251611795437,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21017168164253236,
      "num_tokens": 48370088.0,
      "step": 23310
    },
    {
      "entropy": 5.571975326538086,
      "epoch": 2.114931059506531,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004552322444181792,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21103547513484955,
      "num_tokens": 48381325.0,
      "step": 23315
    },
    {
      "entropy": 5.571200561523438,
      "epoch": 2.1153846153846154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004552128733136726,
      "loss": 4.923,
      "mean_token_accuracy": 0.2082800880074501,
      "num_tokens": 48393077.0,
      "step": 23320
    },
    {
      "entropy": 5.517536258697509,
      "epoch": 2.1158381712626997,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004551934984823179,
      "loss": 4.9132,
      "mean_token_accuracy": 0.211286461353302,
      "num_tokens": 48403260.0,
      "step": 23325
    },
    {
      "entropy": 5.493818759918213,
      "epoch": 2.116291727140784,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004551741199245157,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2162444844841957,
      "num_tokens": 48413616.0,
      "step": 23330
    },
    {
      "entropy": 5.5517590045928955,
      "epoch": 2.1167452830188678,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045515473764066685,
      "loss": 4.9109,
      "mean_token_accuracy": 0.21525130569934844,
      "num_tokens": 48424189.0,
      "step": 23335
    },
    {
      "entropy": 5.544838714599609,
      "epoch": 2.117198838896952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045513535163117205,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2130915865302086,
      "num_tokens": 48433603.0,
      "step": 23340
    },
    {
      "entropy": 5.593479919433594,
      "epoch": 2.1176523947750363,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004551159618964323,
      "loss": 4.9369,
      "mean_token_accuracy": 0.20968596190214156,
      "num_tokens": 48443043.0,
      "step": 23345
    },
    {
      "entropy": 5.507231283187866,
      "epoch": 2.1181059506531206,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004550965684368486,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22793943136930467,
      "num_tokens": 48452765.0,
      "step": 23350
    },
    {
      "entropy": 5.530251407623291,
      "epoch": 2.118559506531205,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004550771712528219,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22040238827466965,
      "num_tokens": 48463191.0,
      "step": 23355
    },
    {
      "entropy": 5.530681943893432,
      "epoch": 2.1190130624092887,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004550577703447534,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2171202376484871,
      "num_tokens": 48473280.0,
      "step": 23360
    },
    {
      "entropy": 5.5225505352020265,
      "epoch": 2.119466618287373,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045503836571304443,
      "loss": 4.8183,
      "mean_token_accuracy": 0.2192261502146721,
      "num_tokens": 48483465.0,
      "step": 23365
    },
    {
      "entropy": 5.509082460403443,
      "epoch": 2.119920174165457,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004550189573580961,
      "loss": 4.8461,
      "mean_token_accuracy": 0.2147024765610695,
      "num_tokens": 48494497.0,
      "step": 23370
    },
    {
      "entropy": 5.524727010726929,
      "epoch": 2.1203737300435415,
      "grad_norm": 1.125,
      "learning_rate": 0.0004549995452803098,
      "loss": 4.9547,
      "mean_token_accuracy": 0.2161712169647217,
      "num_tokens": 48504106.0,
      "step": 23375
    },
    {
      "entropy": 5.599390602111816,
      "epoch": 2.1208272859216257,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549801294800872,
      "loss": 4.9585,
      "mean_token_accuracy": 0.2059381365776062,
      "num_tokens": 48515729.0,
      "step": 23380
    },
    {
      "entropy": 5.59515528678894,
      "epoch": 2.1212808417997095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004549607099578295,
      "loss": 4.9823,
      "mean_token_accuracy": 0.20741215199232102,
      "num_tokens": 48526611.0,
      "step": 23385
    },
    {
      "entropy": 5.5273181915283205,
      "epoch": 2.121734397677794,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004549412867139386,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22016615122556688,
      "num_tokens": 48537016.0,
      "step": 23390
    },
    {
      "entropy": 5.561395692825317,
      "epoch": 2.122187953555878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045492185974881587,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21230339556932448,
      "num_tokens": 48546786.0,
      "step": 23395
    },
    {
      "entropy": 5.564849615097046,
      "epoch": 2.1226415094339623,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004549024290628634,
      "loss": 4.9459,
      "mean_token_accuracy": 0.20879080444574355,
      "num_tokens": 48556306.0,
      "step": 23400
    },
    {
      "entropy": 5.54790415763855,
      "epoch": 2.1230950653120466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004548829946564827,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21276190578937532,
      "num_tokens": 48566303.0,
      "step": 23405
    },
    {
      "entropy": 5.5776018619537355,
      "epoch": 2.1235486211901304,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004548635565300759,
      "loss": 4.933,
      "mean_token_accuracy": 0.21630136370658876,
      "num_tokens": 48576313.0,
      "step": 23410
    },
    {
      "entropy": 5.546379947662354,
      "epoch": 2.1240021770682147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004548441146840449,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21462699621915818,
      "num_tokens": 48587133.0,
      "step": 23415
    },
    {
      "entropy": 5.581494522094727,
      "epoch": 2.124455732946299,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045482466911879184,
      "loss": 5.0445,
      "mean_token_accuracy": 0.20705566704273223,
      "num_tokens": 48598524.0,
      "step": 23420
    },
    {
      "entropy": 5.5020064353942875,
      "epoch": 2.1249092888243832,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045480521983471875,
      "loss": 4.89,
      "mean_token_accuracy": 0.21692655980587006,
      "num_tokens": 48609870.0,
      "step": 23425
    },
    {
      "entropy": 5.562878274917603,
      "epoch": 2.1253628447024675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004547857668322279,
      "loss": 4.935,
      "mean_token_accuracy": 0.2160693511366844,
      "num_tokens": 48619828.0,
      "step": 23430
    },
    {
      "entropy": 5.533704710006714,
      "epoch": 2.1258164005805513,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004547663101117215,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2212892398238182,
      "num_tokens": 48630361.0,
      "step": 23435
    },
    {
      "entropy": 5.539975214004516,
      "epoch": 2.1262699564586356,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045474684967360216,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20398104786872864,
      "num_tokens": 48641113.0,
      "step": 23440
    },
    {
      "entropy": 5.557209253311157,
      "epoch": 2.12672351233672,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000454727385518272,
      "loss": 5.0097,
      "mean_token_accuracy": 0.20273007303476334,
      "num_tokens": 48651777.0,
      "step": 23445
    },
    {
      "entropy": 5.613733005523682,
      "epoch": 2.127177068214804,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004547079176461338,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20714200735092164,
      "num_tokens": 48663237.0,
      "step": 23450
    },
    {
      "entropy": 5.5418140411376955,
      "epoch": 2.1276306240928884,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004546884460575901,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22064516693353653,
      "num_tokens": 48673932.0,
      "step": 23455
    },
    {
      "entropy": 5.5021709442138675,
      "epoch": 2.1280841799709727,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004546689707530434,
      "loss": 4.8518,
      "mean_token_accuracy": 0.21676686853170396,
      "num_tokens": 48684304.0,
      "step": 23460
    },
    {
      "entropy": 5.501614141464233,
      "epoch": 2.1285377358490565,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045464949173289676,
      "loss": 4.8159,
      "mean_token_accuracy": 0.21509646326303483,
      "num_tokens": 48695569.0,
      "step": 23465
    },
    {
      "entropy": 5.470986652374267,
      "epoch": 2.1289912917271407,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004546300089975528,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21531161218881606,
      "num_tokens": 48706550.0,
      "step": 23470
    },
    {
      "entropy": 5.563193845748901,
      "epoch": 2.129444847605225,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045461052254741445,
      "loss": 4.9819,
      "mean_token_accuracy": 0.20409582555294037,
      "num_tokens": 48716852.0,
      "step": 23475
    },
    {
      "entropy": 5.512483167648315,
      "epoch": 2.1298984034833093,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045459103238288474,
      "loss": 4.8502,
      "mean_token_accuracy": 0.21848638653755187,
      "num_tokens": 48726607.0,
      "step": 23480
    },
    {
      "entropy": 5.574798583984375,
      "epoch": 2.130351959361393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004545715385043667,
      "loss": 5.0667,
      "mean_token_accuracy": 0.20632678717374803,
      "num_tokens": 48738011.0,
      "step": 23485
    },
    {
      "entropy": 5.562725830078125,
      "epoch": 2.1308055152394774,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004545520409122635,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20607829391956328,
      "num_tokens": 48747899.0,
      "step": 23490
    },
    {
      "entropy": 5.574234056472778,
      "epoch": 2.1312590711175616,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004545325396069783,
      "loss": 4.957,
      "mean_token_accuracy": 0.21174486726522446,
      "num_tokens": 48758719.0,
      "step": 23495
    },
    {
      "entropy": 5.576707458496093,
      "epoch": 2.131712626995646,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004545130345889145,
      "loss": 4.9838,
      "mean_token_accuracy": 0.21331266462802886,
      "num_tokens": 48769677.0,
      "step": 23500
    },
    {
      "entropy": 5.5739398956298825,
      "epoch": 2.13216618287373,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004544935258584754,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2096804603934288,
      "num_tokens": 48778440.0,
      "step": 23505
    },
    {
      "entropy": 5.597056150436401,
      "epoch": 2.1326197387518144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004544740134160644,
      "loss": 5.0094,
      "mean_token_accuracy": 0.2075168401002884,
      "num_tokens": 48788328.0,
      "step": 23510
    },
    {
      "entropy": 5.6008179664611815,
      "epoch": 2.1330732946298983,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004544544972620851,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20667157769203187,
      "num_tokens": 48799218.0,
      "step": 23515
    },
    {
      "entropy": 5.64255952835083,
      "epoch": 2.1335268505079825,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004544349773969411,
      "loss": 4.9504,
      "mean_token_accuracy": 0.2058921456336975,
      "num_tokens": 48809648.0,
      "step": 23520
    },
    {
      "entropy": 5.579212379455567,
      "epoch": 2.133980406386067,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544154538210359,
      "loss": 4.9427,
      "mean_token_accuracy": 0.21040252447128296,
      "num_tokens": 48819965.0,
      "step": 23525
    },
    {
      "entropy": 5.620339679718017,
      "epoch": 2.134433962264151,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004543959265347735,
      "loss": 5.0054,
      "mean_token_accuracy": 0.21229623407125472,
      "num_tokens": 48829862.0,
      "step": 23530
    },
    {
      "entropy": 5.588329124450683,
      "epoch": 2.1348875181422353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045437639553855757,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2108191192150116,
      "num_tokens": 48840551.0,
      "step": 23535
    },
    {
      "entropy": 5.492155456542969,
      "epoch": 2.135341074020319,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004543568608327921,
      "loss": 4.9349,
      "mean_token_accuracy": 0.2222515746951103,
      "num_tokens": 48851032.0,
      "step": 23540
    },
    {
      "entropy": 5.534734439849854,
      "epoch": 2.1357946298984034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045433732241788105,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2073029398918152,
      "num_tokens": 48861773.0,
      "step": 23545
    },
    {
      "entropy": 5.6082254409790036,
      "epoch": 2.1362481857764877,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045431778029422846,
      "loss": 5.0361,
      "mean_token_accuracy": 0.20967386066913604,
      "num_tokens": 48872430.0,
      "step": 23550
    },
    {
      "entropy": 5.572673034667969,
      "epoch": 2.136701741654572,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004542982344622385,
      "loss": 4.8385,
      "mean_token_accuracy": 0.21917962431907653,
      "num_tokens": 48882575.0,
      "step": 23555
    },
    {
      "entropy": 5.515121555328369,
      "epoch": 2.137155297532656,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045427868492231533,
      "loss": 4.9262,
      "mean_token_accuracy": 0.20800466686487198,
      "num_tokens": 48893337.0,
      "step": 23560
    },
    {
      "entropy": 5.635800504684449,
      "epoch": 2.13760885341074,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045425913167486333,
      "loss": 5.0649,
      "mean_token_accuracy": 0.2086473137140274,
      "num_tokens": 48904675.0,
      "step": 23565
    },
    {
      "entropy": 5.536910200119019,
      "epoch": 2.1380624092888243,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045423957472028677,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22630843371152878,
      "num_tokens": 48914831.0,
      "step": 23570
    },
    {
      "entropy": 5.472995090484619,
      "epoch": 2.1385159651669086,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004542200140589901,
      "loss": 4.8248,
      "mean_token_accuracy": 0.21463076174259185,
      "num_tokens": 48924574.0,
      "step": 23575
    },
    {
      "entropy": 5.558716535568237,
      "epoch": 2.138969521044993,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045420044969137787,
      "loss": 4.9919,
      "mean_token_accuracy": 0.20739877671003343,
      "num_tokens": 48933921.0,
      "step": 23580
    },
    {
      "entropy": 5.544616413116455,
      "epoch": 2.139423076923077,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004541808816178547,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2133629947900772,
      "num_tokens": 48944571.0,
      "step": 23585
    },
    {
      "entropy": 5.554919815063476,
      "epoch": 2.139876632801161,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045416130983882524,
      "loss": 4.9733,
      "mean_token_accuracy": 0.20785541087388992,
      "num_tokens": 48954569.0,
      "step": 23590
    },
    {
      "entropy": 5.536481523513794,
      "epoch": 2.140330188679245,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045414173435469427,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21094706803560256,
      "num_tokens": 48965886.0,
      "step": 23595
    },
    {
      "entropy": 5.584898138046265,
      "epoch": 2.1407837445573294,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045412215516586654,
      "loss": 4.936,
      "mean_token_accuracy": 0.21533112227916718,
      "num_tokens": 48976219.0,
      "step": 23600
    },
    {
      "entropy": 5.504986381530761,
      "epoch": 2.1412373004354137,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000454102572272747,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21037635952234268,
      "num_tokens": 48986322.0,
      "step": 23605
    },
    {
      "entropy": 5.520770120620727,
      "epoch": 2.141690856313498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004540829856757406,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22111576050519943,
      "num_tokens": 48996297.0,
      "step": 23610
    },
    {
      "entropy": 5.533485460281372,
      "epoch": 2.142144412191582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004540633953752525,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21451962292194365,
      "num_tokens": 49006165.0,
      "step": 23615
    },
    {
      "entropy": 5.531984043121338,
      "epoch": 2.142597968069666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004540438013716876,
      "loss": 4.8937,
      "mean_token_accuracy": 0.20562394261360167,
      "num_tokens": 49017278.0,
      "step": 23620
    },
    {
      "entropy": 5.490485906600952,
      "epoch": 2.1430515239477503,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004540242036654514,
      "loss": 4.8653,
      "mean_token_accuracy": 0.21416873782873153,
      "num_tokens": 49027088.0,
      "step": 23625
    },
    {
      "entropy": 5.531488084793091,
      "epoch": 2.1435050798258346,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000454004602256949,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21203094124794006,
      "num_tokens": 49038156.0,
      "step": 23630
    },
    {
      "entropy": 5.589496374130249,
      "epoch": 2.143958635703919,
      "grad_norm": 1.125,
      "learning_rate": 0.0004539849971465858,
      "loss": 4.9478,
      "mean_token_accuracy": 0.20562424808740615,
      "num_tokens": 49047480.0,
      "step": 23635
    },
    {
      "entropy": 5.598181104660034,
      "epoch": 2.1444121915820027,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045396538833476725,
      "loss": 4.9783,
      "mean_token_accuracy": 0.21298412680625917,
      "num_tokens": 49057373.0,
      "step": 23640
    },
    {
      "entropy": 5.517035818099975,
      "epoch": 2.144865747460087,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004539457758218989,
      "loss": 4.974,
      "mean_token_accuracy": 0.2113524854183197,
      "num_tokens": 49068380.0,
      "step": 23645
    },
    {
      "entropy": 5.485847425460816,
      "epoch": 2.1453193033381712,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004539261596083863,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21413416862487794,
      "num_tokens": 49078646.0,
      "step": 23650
    },
    {
      "entropy": 5.571518468856811,
      "epoch": 2.1457728592162555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004539065396946351,
      "loss": 4.8679,
      "mean_token_accuracy": 0.221435546875,
      "num_tokens": 49089281.0,
      "step": 23655
    },
    {
      "entropy": 5.58362226486206,
      "epoch": 2.1462264150943398,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004538869160810511,
      "loss": 4.9563,
      "mean_token_accuracy": 0.20561626553535461,
      "num_tokens": 49099514.0,
      "step": 23660
    },
    {
      "entropy": 5.5541663646698,
      "epoch": 2.146679970972424,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045386728876804,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20973126888275145,
      "num_tokens": 49110247.0,
      "step": 23665
    },
    {
      "entropy": 5.58193736076355,
      "epoch": 2.147133526850508,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045384765775600787,
      "loss": 5.0224,
      "mean_token_accuracy": 0.20902297496795655,
      "num_tokens": 49122492.0,
      "step": 23670
    },
    {
      "entropy": 5.585496091842652,
      "epoch": 2.147587082728592,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004538280230453606,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21070511639118195,
      "num_tokens": 49132395.0,
      "step": 23675
    },
    {
      "entropy": 5.633884716033935,
      "epoch": 2.1480406386066764,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004538083846365042,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21412622928619385,
      "num_tokens": 49142909.0,
      "step": 23680
    },
    {
      "entropy": 5.550940990447998,
      "epoch": 2.1484941944847606,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004537887425298448,
      "loss": 4.9233,
      "mean_token_accuracy": 0.20719007700681685,
      "num_tokens": 49152933.0,
      "step": 23685
    },
    {
      "entropy": 5.514552593231201,
      "epoch": 2.1489477503628445,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045376909672578875,
      "loss": 4.9143,
      "mean_token_accuracy": 0.20896622985601426,
      "num_tokens": 49163595.0,
      "step": 23690
    },
    {
      "entropy": 5.495805311203003,
      "epoch": 2.1494013062409287,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004537494472247422,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22202681452035905,
      "num_tokens": 49174664.0,
      "step": 23695
    },
    {
      "entropy": 5.597122097015381,
      "epoch": 2.149854862119013,
      "grad_norm": 1.125,
      "learning_rate": 0.0004537297940271115,
      "loss": 4.9704,
      "mean_token_accuracy": 0.21686883568763732,
      "num_tokens": 49185105.0,
      "step": 23700
    },
    {
      "entropy": 5.52719554901123,
      "epoch": 2.1503084179970973,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004537101371333031,
      "loss": 4.8652,
      "mean_token_accuracy": 0.21621908098459244,
      "num_tokens": 49194679.0,
      "step": 23705
    },
    {
      "entropy": 5.53243203163147,
      "epoch": 2.1507619738751815,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045369047654372356,
      "loss": 4.8295,
      "mean_token_accuracy": 0.21875529438257219,
      "num_tokens": 49204563.0,
      "step": 23710
    },
    {
      "entropy": 5.564297389984131,
      "epoch": 2.151215529753266,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004536708122587795,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21561508923768996,
      "num_tokens": 49213759.0,
      "step": 23715
    },
    {
      "entropy": 5.5077611923217775,
      "epoch": 2.1516690856313496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004536511442788774,
      "loss": 4.8183,
      "mean_token_accuracy": 0.22056595981121063,
      "num_tokens": 49223604.0,
      "step": 23720
    },
    {
      "entropy": 5.546893548965454,
      "epoch": 2.152122641509434,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004536314726044241,
      "loss": 4.939,
      "mean_token_accuracy": 0.2145179107785225,
      "num_tokens": 49234000.0,
      "step": 23725
    },
    {
      "entropy": 5.572303056716919,
      "epoch": 2.152576197387518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045361179723582654,
      "loss": 4.9297,
      "mean_token_accuracy": 0.21028758436441422,
      "num_tokens": 49244874.0,
      "step": 23730
    },
    {
      "entropy": 5.606242752075195,
      "epoch": 2.1530297532656024,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004535921181734914,
      "loss": 5.0872,
      "mean_token_accuracy": 0.20579334944486619,
      "num_tokens": 49255495.0,
      "step": 23735
    },
    {
      "entropy": 5.544103479385376,
      "epoch": 2.1534833091436867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004535724354178257,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2137600302696228,
      "num_tokens": 49265357.0,
      "step": 23740
    },
    {
      "entropy": 5.6134320259094235,
      "epoch": 2.1539368650217705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045355274896923664,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20627753883600236,
      "num_tokens": 49276396.0,
      "step": 23745
    },
    {
      "entropy": 5.547289276123047,
      "epoch": 2.1543904208998548,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004535330588281312,
      "loss": 4.8503,
      "mean_token_accuracy": 0.21515513211488724,
      "num_tokens": 49285925.0,
      "step": 23750
    },
    {
      "entropy": 5.519663667678833,
      "epoch": 2.154843976777939,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004535133649949166,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21690916121006013,
      "num_tokens": 49296493.0,
      "step": 23755
    },
    {
      "entropy": 5.486098051071167,
      "epoch": 2.1552975326560233,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004534936674700001,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21768162101507188,
      "num_tokens": 49307273.0,
      "step": 23760
    },
    {
      "entropy": 5.522302675247192,
      "epoch": 2.1557510885341076,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045347396625378914,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2208684727549553,
      "num_tokens": 49317055.0,
      "step": 23765
    },
    {
      "entropy": 5.549699831008911,
      "epoch": 2.1562046444121914,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045345426134669103,
      "loss": 4.932,
      "mean_token_accuracy": 0.2084305077791214,
      "num_tokens": 49327420.0,
      "step": 23770
    },
    {
      "entropy": 5.551327896118164,
      "epoch": 2.1566582002902757,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045343455274911325,
      "loss": 5.0076,
      "mean_token_accuracy": 0.2088395893573761,
      "num_tokens": 49338323.0,
      "step": 23775
    },
    {
      "entropy": 5.5568821907043455,
      "epoch": 2.15711175616836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004534148404614635,
      "loss": 4.965,
      "mean_token_accuracy": 0.21425500959157945,
      "num_tokens": 49348364.0,
      "step": 23780
    },
    {
      "entropy": 5.606203365325928,
      "epoch": 2.157565312046444,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045339512448414943,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21414417326450347,
      "num_tokens": 49358274.0,
      "step": 23785
    },
    {
      "entropy": 5.526130723953247,
      "epoch": 2.1580188679245285,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004533754048175787,
      "loss": 4.8551,
      "mean_token_accuracy": 0.2114555448293686,
      "num_tokens": 49368651.0,
      "step": 23790
    },
    {
      "entropy": 5.507954454421997,
      "epoch": 2.1584724238026123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004533556814621591,
      "loss": 4.8761,
      "mean_token_accuracy": 0.20849419385194778,
      "num_tokens": 49378363.0,
      "step": 23795
    },
    {
      "entropy": 5.538680458068848,
      "epoch": 2.1589259796806965,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045333595441829865,
      "loss": 4.9046,
      "mean_token_accuracy": 0.21472134590148925,
      "num_tokens": 49388241.0,
      "step": 23800
    },
    {
      "entropy": 5.557838821411133,
      "epoch": 2.159379535558781,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004533162236864051,
      "loss": 4.9122,
      "mean_token_accuracy": 0.20833361595869065,
      "num_tokens": 49397678.0,
      "step": 23805
    },
    {
      "entropy": 5.59941463470459,
      "epoch": 2.159833091436865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004532964892668867,
      "loss": 4.9869,
      "mean_token_accuracy": 0.2066510796546936,
      "num_tokens": 49407368.0,
      "step": 23810
    },
    {
      "entropy": 5.516260433197021,
      "epoch": 2.1602866473149493,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045327675116015146,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2173643484711647,
      "num_tokens": 49417926.0,
      "step": 23815
    },
    {
      "entropy": 5.629906749725341,
      "epoch": 2.1607402031930336,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004532570093666075,
      "loss": 5.0485,
      "mean_token_accuracy": 0.20045541673898698,
      "num_tokens": 49427899.0,
      "step": 23820
    },
    {
      "entropy": 5.541424989700317,
      "epoch": 2.1611937590711174,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004532372638866633,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22204057574272157,
      "num_tokens": 49439102.0,
      "step": 23825
    },
    {
      "entropy": 5.530374050140381,
      "epoch": 2.1616473149492017,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045321751472072693,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22254859060049056,
      "num_tokens": 49448013.0,
      "step": 23830
    },
    {
      "entropy": 5.476184129714966,
      "epoch": 2.162100870827286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045319776186920695,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2190641477704048,
      "num_tokens": 49457620.0,
      "step": 23835
    },
    {
      "entropy": 5.469401884078979,
      "epoch": 2.1625544267053702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045317800533251193,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2102116495370865,
      "num_tokens": 49468038.0,
      "step": 23840
    },
    {
      "entropy": 5.462647485733032,
      "epoch": 2.163007982583454,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045315824511105027,
      "loss": 4.8686,
      "mean_token_accuracy": 0.211471126973629,
      "num_tokens": 49477714.0,
      "step": 23845
    },
    {
      "entropy": 5.5680131912231445,
      "epoch": 2.1634615384615383,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045313848120523073,
      "loss": 4.8565,
      "mean_token_accuracy": 0.21658321470022202,
      "num_tokens": 49488214.0,
      "step": 23850
    },
    {
      "entropy": 5.588200521469116,
      "epoch": 2.1639150943396226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045311871361546205,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21260038763284683,
      "num_tokens": 49498444.0,
      "step": 23855
    },
    {
      "entropy": 5.477870607376099,
      "epoch": 2.164368650217707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004530989423421529,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22175231277942659,
      "num_tokens": 49508759.0,
      "step": 23860
    },
    {
      "entropy": 5.431105327606201,
      "epoch": 2.164822206095791,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045307916738571223,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22357546389102936,
      "num_tokens": 49518787.0,
      "step": 23865
    },
    {
      "entropy": 5.545242547988892,
      "epoch": 2.1652757619738754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004530593887465491,
      "loss": 4.8909,
      "mean_token_accuracy": 0.2197202131152153,
      "num_tokens": 49528130.0,
      "step": 23870
    },
    {
      "entropy": 5.621055936813354,
      "epoch": 2.165729317851959,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045303960642507235,
      "loss": 5.0597,
      "mean_token_accuracy": 0.20446344017982482,
      "num_tokens": 49539128.0,
      "step": 23875
    },
    {
      "entropy": 5.471271991729736,
      "epoch": 2.1661828737300435,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045301982042169123,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21687667071819305,
      "num_tokens": 49550571.0,
      "step": 23880
    },
    {
      "entropy": 5.535809421539307,
      "epoch": 2.1666364296081277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045300003073681485,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21757669746875763,
      "num_tokens": 49561106.0,
      "step": 23885
    },
    {
      "entropy": 5.490610790252686,
      "epoch": 2.167089985486212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045298023737085243,
      "loss": 4.8275,
      "mean_token_accuracy": 0.2108614921569824,
      "num_tokens": 49571389.0,
      "step": 23890
    },
    {
      "entropy": 5.507013750076294,
      "epoch": 2.1675435413642963,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004529604403242134,
      "loss": 4.944,
      "mean_token_accuracy": 0.21304160356521606,
      "num_tokens": 49582224.0,
      "step": 23895
    },
    {
      "entropy": 5.564214754104614,
      "epoch": 2.16799709724238,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045294063959730707,
      "loss": 4.942,
      "mean_token_accuracy": 0.2131488725543022,
      "num_tokens": 49592413.0,
      "step": 23900
    },
    {
      "entropy": 5.5788802146911625,
      "epoch": 2.1684506531204644,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045292083519054297,
      "loss": 5.0026,
      "mean_token_accuracy": 0.19674703776836394,
      "num_tokens": 49602796.0,
      "step": 23905
    },
    {
      "entropy": 5.494076347351074,
      "epoch": 2.1689042089985486,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004529010271043307,
      "loss": 4.8902,
      "mean_token_accuracy": 0.21899964809417724,
      "num_tokens": 49613625.0,
      "step": 23910
    },
    {
      "entropy": 5.648594236373901,
      "epoch": 2.169357764876633,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045288121533907977,
      "loss": 5.0131,
      "mean_token_accuracy": 0.20691847652196885,
      "num_tokens": 49624719.0,
      "step": 23915
    },
    {
      "entropy": 5.605312490463257,
      "epoch": 2.169811320754717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045286139989520006,
      "loss": 4.9826,
      "mean_token_accuracy": 0.20615526288747787,
      "num_tokens": 49634916.0,
      "step": 23920
    },
    {
      "entropy": 5.566227388381958,
      "epoch": 2.170264876632801,
      "grad_norm": 1.125,
      "learning_rate": 0.0004528415807731012,
      "loss": 4.8755,
      "mean_token_accuracy": 0.21158760339021682,
      "num_tokens": 49645215.0,
      "step": 23925
    },
    {
      "entropy": 5.514484977722168,
      "epoch": 2.1707184325108853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045282175797319317,
      "loss": 4.935,
      "mean_token_accuracy": 0.21427508145570756,
      "num_tokens": 49655805.0,
      "step": 23930
    },
    {
      "entropy": 5.559678316116333,
      "epoch": 2.1711719883889695,
      "grad_norm": 1.125,
      "learning_rate": 0.0004528019314958859,
      "loss": 4.9179,
      "mean_token_accuracy": 0.21378210186958313,
      "num_tokens": 49665264.0,
      "step": 23935
    },
    {
      "entropy": 5.574174642562866,
      "epoch": 2.171625544267054,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045278210134158935,
      "loss": 4.9764,
      "mean_token_accuracy": 0.2088568314909935,
      "num_tokens": 49676152.0,
      "step": 23940
    },
    {
      "entropy": 5.602774238586425,
      "epoch": 2.172079100145138,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004527622675107137,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2090972140431404,
      "num_tokens": 49686762.0,
      "step": 23945
    },
    {
      "entropy": 5.581876993179321,
      "epoch": 2.172532656023222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045274243000366897,
      "loss": 4.9581,
      "mean_token_accuracy": 0.21147668212652207,
      "num_tokens": 49697504.0,
      "step": 23950
    },
    {
      "entropy": 5.548239135742188,
      "epoch": 2.172986211901306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004527225888208655,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20620023906230928,
      "num_tokens": 49709197.0,
      "step": 23955
    },
    {
      "entropy": 5.564700937271118,
      "epoch": 2.1734397677793904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004527027439627137,
      "loss": 4.9424,
      "mean_token_accuracy": 0.20987117439508438,
      "num_tokens": 49719764.0,
      "step": 23960
    },
    {
      "entropy": 5.558755540847779,
      "epoch": 2.1738933236574747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045268289542962377,
      "loss": 4.9224,
      "mean_token_accuracy": 0.2100379914045334,
      "num_tokens": 49730363.0,
      "step": 23965
    },
    {
      "entropy": 5.529247093200683,
      "epoch": 2.174346879535559,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004526630432220063,
      "loss": 4.9636,
      "mean_token_accuracy": 0.20264688432216643,
      "num_tokens": 49741037.0,
      "step": 23970
    },
    {
      "entropy": 5.5099798202514645,
      "epoch": 2.1748004354136428,
      "grad_norm": 1.125,
      "learning_rate": 0.0004526431873402719,
      "loss": 4.8569,
      "mean_token_accuracy": 0.21590733230113984,
      "num_tokens": 49750127.0,
      "step": 23975
    },
    {
      "entropy": 5.621721458435059,
      "epoch": 2.175253991291727,
      "grad_norm": 1.078125,
      "learning_rate": 0.000452623327784831,
      "loss": 5.0349,
      "mean_token_accuracy": 0.2077339082956314,
      "num_tokens": 49760298.0,
      "step": 23980
    },
    {
      "entropy": 5.552127647399902,
      "epoch": 2.1757075471698113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004526034645560945,
      "loss": 4.9452,
      "mean_token_accuracy": 0.20892651677131652,
      "num_tokens": 49770841.0,
      "step": 23985
    },
    {
      "entropy": 5.49432373046875,
      "epoch": 2.1761611030478956,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045258359765447306,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21433293372392653,
      "num_tokens": 49781000.0,
      "step": 23990
    },
    {
      "entropy": 5.535477828979492,
      "epoch": 2.17661465892598,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004525637270803776,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21459790468215942,
      "num_tokens": 49790886.0,
      "step": 23995
    },
    {
      "entropy": 5.53254656791687,
      "epoch": 2.1770682148040637,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045254385283421896,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22086577415466307,
      "num_tokens": 49801477.0,
      "step": 24000
    },
    {
      "epoch": 2.1770682148040637,
      "eval_entropy": 5.286641340449925,
      "eval_loss": 5.0150980949401855,
      "eval_mean_token_accuracy": 0.21701856827978636,
      "eval_num_tokens": 49801477.0,
      "eval_runtime": 20.5247,
      "eval_samples_per_second": 1722.803,
      "eval_steps_per_second": 215.35,
      "step": 24000
    },
    {
      "entropy": 5.543504762649536,
      "epoch": 2.177521770682148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004525239749164082,
      "loss": 4.9308,
      "mean_token_accuracy": 0.21494587808847426,
      "num_tokens": 49812044.0,
      "step": 24005
    },
    {
      "entropy": 5.473853349685669,
      "epoch": 2.177975326560232,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004525040933273564,
      "loss": 4.8456,
      "mean_token_accuracy": 0.21704766750335694,
      "num_tokens": 49821559.0,
      "step": 24010
    },
    {
      "entropy": 5.6032861232757565,
      "epoch": 2.1784288824383164,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045248420806747483,
      "loss": 5.027,
      "mean_token_accuracy": 0.1996487483382225,
      "num_tokens": 49832937.0,
      "step": 24015
    },
    {
      "entropy": 5.620812034606933,
      "epoch": 2.1788824383164007,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004524643191371745,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21164904683828353,
      "num_tokens": 49843176.0,
      "step": 24020
    },
    {
      "entropy": 5.522511529922485,
      "epoch": 2.179335994194485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045244442653686684,
      "loss": 4.8516,
      "mean_token_accuracy": 0.222689488530159,
      "num_tokens": 49852522.0,
      "step": 24025
    },
    {
      "entropy": 5.456665086746216,
      "epoch": 2.179789550072569,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045242453026696325,
      "loss": 4.8593,
      "mean_token_accuracy": 0.21921806633472443,
      "num_tokens": 49863143.0,
      "step": 24030
    },
    {
      "entropy": 5.611735391616821,
      "epoch": 2.180243105950653,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004524046303278752,
      "loss": 4.9663,
      "mean_token_accuracy": 0.2109787121415138,
      "num_tokens": 49873465.0,
      "step": 24035
    },
    {
      "entropy": 5.661686706542969,
      "epoch": 2.1806966618287373,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004523847267200141,
      "loss": 5.0456,
      "mean_token_accuracy": 0.20169859230518342,
      "num_tokens": 49883690.0,
      "step": 24040
    },
    {
      "entropy": 5.590389680862427,
      "epoch": 2.1811502177068216,
      "grad_norm": 1.125,
      "learning_rate": 0.00045236481944379165,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2091896951198578,
      "num_tokens": 49894418.0,
      "step": 24045
    },
    {
      "entropy": 5.5485937118530275,
      "epoch": 2.1816037735849054,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004523449084996196,
      "loss": 4.87,
      "mean_token_accuracy": 0.21958404779434204,
      "num_tokens": 49903927.0,
      "step": 24050
    },
    {
      "entropy": 5.554563283920288,
      "epoch": 2.1820573294629897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004523249938879096,
      "loss": 4.9257,
      "mean_token_accuracy": 0.21105577945709228,
      "num_tokens": 49914203.0,
      "step": 24055
    },
    {
      "entropy": 5.560449600219727,
      "epoch": 2.182510885341074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045230507560907354,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21184269487857818,
      "num_tokens": 49924320.0,
      "step": 24060
    },
    {
      "entropy": 5.499133539199829,
      "epoch": 2.1829644412191582,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045228515366352334,
      "loss": 4.8853,
      "mean_token_accuracy": 0.21574951559305192,
      "num_tokens": 49934634.0,
      "step": 24065
    },
    {
      "entropy": 5.6132559299469,
      "epoch": 2.1834179970972425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045226522805167095,
      "loss": 4.9531,
      "mean_token_accuracy": 0.2121679052710533,
      "num_tokens": 49944586.0,
      "step": 24070
    },
    {
      "entropy": 5.566622686386109,
      "epoch": 2.1838715529753268,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004522452987739286,
      "loss": 4.9205,
      "mean_token_accuracy": 0.21605656445026397,
      "num_tokens": 49955441.0,
      "step": 24075
    },
    {
      "entropy": 5.6409642696380615,
      "epoch": 2.1843251088534106,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045222536583070823,
      "loss": 5.0409,
      "mean_token_accuracy": 0.21066685914993286,
      "num_tokens": 49966718.0,
      "step": 24080
    },
    {
      "entropy": 5.627300453186035,
      "epoch": 2.184778664731495,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045220542922242205,
      "loss": 5.0181,
      "mean_token_accuracy": 0.20259948670864106,
      "num_tokens": 49979483.0,
      "step": 24085
    },
    {
      "entropy": 5.532460403442383,
      "epoch": 2.185232220609579,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045218548894948254,
      "loss": 4.919,
      "mean_token_accuracy": 0.21612850278615953,
      "num_tokens": 49989976.0,
      "step": 24090
    },
    {
      "entropy": 5.500060558319092,
      "epoch": 2.1856857764876634,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045216554501230187,
      "loss": 4.7963,
      "mean_token_accuracy": 0.21847291886806489,
      "num_tokens": 50000069.0,
      "step": 24095
    },
    {
      "entropy": 5.5431397438049315,
      "epoch": 2.1861393323657476,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004521455974112927,
      "loss": 4.9331,
      "mean_token_accuracy": 0.21194951236248016,
      "num_tokens": 50009970.0,
      "step": 24100
    },
    {
      "entropy": 5.5811930179595945,
      "epoch": 2.1865928882438315,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045212564614686735,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21043117791414262,
      "num_tokens": 50019590.0,
      "step": 24105
    },
    {
      "entropy": 5.54469108581543,
      "epoch": 2.1870464441219157,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004521056912194386,
      "loss": 4.9297,
      "mean_token_accuracy": 0.20442597866058348,
      "num_tokens": 50030092.0,
      "step": 24110
    },
    {
      "entropy": 5.494137859344482,
      "epoch": 2.1875,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004520857326294189,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21274950504302978,
      "num_tokens": 50040318.0,
      "step": 24115
    },
    {
      "entropy": 5.588677835464478,
      "epoch": 2.1879535558780843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045206577037722114,
      "loss": 5.0663,
      "mean_token_accuracy": 0.19531594067811966,
      "num_tokens": 50050507.0,
      "step": 24120
    },
    {
      "entropy": 5.684941339492798,
      "epoch": 2.1884071117561685,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004520458044632582,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20891194343566893,
      "num_tokens": 50059727.0,
      "step": 24125
    },
    {
      "entropy": 5.597309207916259,
      "epoch": 2.1888606676342524,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045202583488794285,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21198929697275162,
      "num_tokens": 50070812.0,
      "step": 24130
    },
    {
      "entropy": 5.613303089141846,
      "epoch": 2.1893142235123366,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004520058616516881,
      "loss": 5.0229,
      "mean_token_accuracy": 0.20498594641685486,
      "num_tokens": 50081175.0,
      "step": 24135
    },
    {
      "entropy": 5.565915489196778,
      "epoch": 2.189767779390421,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045198588475490716,
      "loss": 4.9274,
      "mean_token_accuracy": 0.20840646028518678,
      "num_tokens": 50092246.0,
      "step": 24140
    },
    {
      "entropy": 5.618529415130615,
      "epoch": 2.190221335268505,
      "grad_norm": 1.125,
      "learning_rate": 0.0004519659041980129,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21359175592660903,
      "num_tokens": 50101934.0,
      "step": 24145
    },
    {
      "entropy": 5.5796647548675535,
      "epoch": 2.1906748911465894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004519459199814187,
      "loss": 4.9766,
      "mean_token_accuracy": 0.2122833549976349,
      "num_tokens": 50112159.0,
      "step": 24150
    },
    {
      "entropy": 5.549475145339966,
      "epoch": 2.1911284470246732,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004519259321055377,
      "loss": 4.9651,
      "mean_token_accuracy": 0.20783157646656036,
      "num_tokens": 50123079.0,
      "step": 24155
    },
    {
      "entropy": 5.566382884979248,
      "epoch": 2.1915820029027575,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004519059405707834,
      "loss": 4.9325,
      "mean_token_accuracy": 0.20795058459043503,
      "num_tokens": 50133097.0,
      "step": 24160
    },
    {
      "entropy": 5.6384388446807865,
      "epoch": 2.1920355587808418,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045188594537756916,
      "loss": 5.0151,
      "mean_token_accuracy": 0.2086271807551384,
      "num_tokens": 50143062.0,
      "step": 24165
    },
    {
      "entropy": 5.610979318618774,
      "epoch": 2.192489114658926,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045186594652630853,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2154499277472496,
      "num_tokens": 50154482.0,
      "step": 24170
    },
    {
      "entropy": 5.568870210647583,
      "epoch": 2.1929426705370103,
      "grad_norm": 1.125,
      "learning_rate": 0.000451845944017415,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21843838840723037,
      "num_tokens": 50164880.0,
      "step": 24175
    },
    {
      "entropy": 5.528281164169312,
      "epoch": 2.1933962264150946,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045182593785130225,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21812840104103087,
      "num_tokens": 50173896.0,
      "step": 24180
    },
    {
      "entropy": 5.58480806350708,
      "epoch": 2.1938497822931784,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004518059280283842,
      "loss": 4.9737,
      "mean_token_accuracy": 0.20620279312133788,
      "num_tokens": 50185078.0,
      "step": 24185
    },
    {
      "entropy": 5.603467893600464,
      "epoch": 2.1943033381712627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045178591454907434,
      "loss": 4.945,
      "mean_token_accuracy": 0.21329548805952073,
      "num_tokens": 50195500.0,
      "step": 24190
    },
    {
      "entropy": 5.5206348419189455,
      "epoch": 2.194756894049347,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045176589741378675,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21442530006170274,
      "num_tokens": 50206103.0,
      "step": 24195
    },
    {
      "entropy": 5.500463151931763,
      "epoch": 2.195210449927431,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004517458766229354,
      "loss": 4.8715,
      "mean_token_accuracy": 0.21397484242916107,
      "num_tokens": 50216522.0,
      "step": 24200
    },
    {
      "entropy": 5.577598619461059,
      "epoch": 2.195664005805515,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045172585217693426,
      "loss": 4.9375,
      "mean_token_accuracy": 0.21485844850540162,
      "num_tokens": 50226721.0,
      "step": 24205
    },
    {
      "entropy": 5.615157270431519,
      "epoch": 2.1961175616835993,
      "grad_norm": 1.125,
      "learning_rate": 0.00045170582407619747,
      "loss": 4.9915,
      "mean_token_accuracy": 0.20684700608253478,
      "num_tokens": 50236327.0,
      "step": 24210
    },
    {
      "entropy": 5.543987607955932,
      "epoch": 2.1965711175616836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004516857923211392,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21946696937084198,
      "num_tokens": 50245729.0,
      "step": 24215
    },
    {
      "entropy": 5.577301168441773,
      "epoch": 2.197024673439768,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045166575691217375,
      "loss": 5.0294,
      "mean_token_accuracy": 0.20256706774234773,
      "num_tokens": 50257024.0,
      "step": 24220
    },
    {
      "entropy": 5.575317668914795,
      "epoch": 2.197478229317852,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045164571784971536,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21785360276699067,
      "num_tokens": 50266910.0,
      "step": 24225
    },
    {
      "entropy": 5.54524598121643,
      "epoch": 2.1979317851959363,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045162567513417864,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2148076593875885,
      "num_tokens": 50278233.0,
      "step": 24230
    },
    {
      "entropy": 5.537736988067627,
      "epoch": 2.19838534107402,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045160562876597776,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2106078028678894,
      "num_tokens": 50289276.0,
      "step": 24235
    },
    {
      "entropy": 5.536694002151489,
      "epoch": 2.1988388969521044,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004515855787455276,
      "loss": 4.9032,
      "mean_token_accuracy": 0.21403492242097855,
      "num_tokens": 50299336.0,
      "step": 24240
    },
    {
      "entropy": 5.575514221191407,
      "epoch": 2.1992924528301887,
      "grad_norm": 1.125,
      "learning_rate": 0.0004515655250732426,
      "loss": 4.8718,
      "mean_token_accuracy": 0.215108023583889,
      "num_tokens": 50308308.0,
      "step": 24245
    },
    {
      "entropy": 5.533282661437989,
      "epoch": 2.199746008708273,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045154546774953764,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21498030722141265,
      "num_tokens": 50318295.0,
      "step": 24250
    },
    {
      "entropy": 5.488432884216309,
      "epoch": 2.2001995645863572,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004515254067748274,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2206475928425789,
      "num_tokens": 50329296.0,
      "step": 24255
    },
    {
      "entropy": 5.51523003578186,
      "epoch": 2.200653120464441,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004515053421495267,
      "loss": 4.9983,
      "mean_token_accuracy": 0.20527873039245606,
      "num_tokens": 50339389.0,
      "step": 24260
    },
    {
      "entropy": 5.625369882583618,
      "epoch": 2.2011066763425253,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045148527387405056,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20094504952430725,
      "num_tokens": 50349647.0,
      "step": 24265
    },
    {
      "entropy": 5.651530504226685,
      "epoch": 2.2015602322206096,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004514652019488141,
      "loss": 4.9867,
      "mean_token_accuracy": 0.20662293136119841,
      "num_tokens": 50361292.0,
      "step": 24270
    },
    {
      "entropy": 5.562852478027343,
      "epoch": 2.202013788098694,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004514451263742321,
      "loss": 4.917,
      "mean_token_accuracy": 0.21196745187044144,
      "num_tokens": 50372293.0,
      "step": 24275
    },
    {
      "entropy": 5.507154893875122,
      "epoch": 2.202467343976778,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004514250471507201,
      "loss": 4.8919,
      "mean_token_accuracy": 0.2164333716034889,
      "num_tokens": 50382462.0,
      "step": 24280
    },
    {
      "entropy": 5.591971206665039,
      "epoch": 2.202920899854862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045140496427869306,
      "loss": 4.9548,
      "mean_token_accuracy": 0.22349608838558196,
      "num_tokens": 50393157.0,
      "step": 24285
    },
    {
      "entropy": 5.500439548492432,
      "epoch": 2.203374455732946,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004513848777585664,
      "loss": 4.8429,
      "mean_token_accuracy": 0.21860722601413726,
      "num_tokens": 50403474.0,
      "step": 24290
    },
    {
      "entropy": 5.538592195510864,
      "epoch": 2.2038280116110305,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004513647875907556,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2103622630238533,
      "num_tokens": 50414337.0,
      "step": 24295
    },
    {
      "entropy": 5.604422950744629,
      "epoch": 2.2042815674891147,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045134469377567597,
      "loss": 5.0544,
      "mean_token_accuracy": 0.2028390049934387,
      "num_tokens": 50424863.0,
      "step": 24300
    },
    {
      "entropy": 5.580130100250244,
      "epoch": 2.204735123367199,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004513245963137432,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2095629557967186,
      "num_tokens": 50435982.0,
      "step": 24305
    },
    {
      "entropy": 5.55920991897583,
      "epoch": 2.205188679245283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004513044952053728,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21282147318124772,
      "num_tokens": 50445112.0,
      "step": 24310
    },
    {
      "entropy": 5.496017169952393,
      "epoch": 2.205642235123367,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045128439045098045,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22416318953037262,
      "num_tokens": 50455266.0,
      "step": 24315
    },
    {
      "entropy": 5.587290096282959,
      "epoch": 2.2060957910014514,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045126428205098207,
      "loss": 5.0046,
      "mean_token_accuracy": 0.2023736909031868,
      "num_tokens": 50466930.0,
      "step": 24320
    },
    {
      "entropy": 5.57593183517456,
      "epoch": 2.2065493468795356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004512441700057934,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2138916403055191,
      "num_tokens": 50477534.0,
      "step": 24325
    },
    {
      "entropy": 5.549651432037353,
      "epoch": 2.20700290275762,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004512240543158305,
      "loss": 4.9165,
      "mean_token_accuracy": 0.21114435642957688,
      "num_tokens": 50487352.0,
      "step": 24330
    },
    {
      "entropy": 5.511846876144409,
      "epoch": 2.2074564586357037,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045120393498150905,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2231524720788002,
      "num_tokens": 50496227.0,
      "step": 24335
    },
    {
      "entropy": 5.565906810760498,
      "epoch": 2.207910014513788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045118381200324537,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2089566245675087,
      "num_tokens": 50507536.0,
      "step": 24340
    },
    {
      "entropy": 5.619041395187378,
      "epoch": 2.2083635703918723,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004511636853814556,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21169669181108475,
      "num_tokens": 50517659.0,
      "step": 24345
    },
    {
      "entropy": 5.570213842391968,
      "epoch": 2.2088171262699565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045114355511655595,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2047789826989174,
      "num_tokens": 50528717.0,
      "step": 24350
    },
    {
      "entropy": 5.502689170837402,
      "epoch": 2.209270682148041,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045112342120896267,
      "loss": 4.8427,
      "mean_token_accuracy": 0.21502464711666108,
      "num_tokens": 50538449.0,
      "step": 24355
    },
    {
      "entropy": 5.5902646541595455,
      "epoch": 2.2097242380261246,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045110328365909215,
      "loss": 5.0138,
      "mean_token_accuracy": 0.2043323427438736,
      "num_tokens": 50549098.0,
      "step": 24360
    },
    {
      "entropy": 5.59246072769165,
      "epoch": 2.210177793904209,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004510831424673609,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21242857426404954,
      "num_tokens": 50560047.0,
      "step": 24365
    },
    {
      "entropy": 5.5918073654174805,
      "epoch": 2.210631349782293,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004510629976341853,
      "loss": 4.9057,
      "mean_token_accuracy": 0.21342041194438935,
      "num_tokens": 50569704.0,
      "step": 24370
    },
    {
      "entropy": 5.5751079559326175,
      "epoch": 2.2110849056603774,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045104284915998215,
      "loss": 4.967,
      "mean_token_accuracy": 0.21115521490573883,
      "num_tokens": 50580006.0,
      "step": 24375
    },
    {
      "entropy": 5.518049335479736,
      "epoch": 2.2115384615384617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045102269704516793,
      "loss": 4.909,
      "mean_token_accuracy": 0.21886965334415437,
      "num_tokens": 50589593.0,
      "step": 24380
    },
    {
      "entropy": 5.541171455383301,
      "epoch": 2.211992017416546,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004510025412901595,
      "loss": 4.8935,
      "mean_token_accuracy": 0.21193874031305313,
      "num_tokens": 50598774.0,
      "step": 24385
    },
    {
      "entropy": 5.472470045089722,
      "epoch": 2.2124455732946298,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045098238189537367,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22120751887559892,
      "num_tokens": 50608815.0,
      "step": 24390
    },
    {
      "entropy": 5.467582750320434,
      "epoch": 2.212899129172714,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045096221886122733,
      "loss": 5.015,
      "mean_token_accuracy": 0.21808652728796005,
      "num_tokens": 50619255.0,
      "step": 24395
    },
    {
      "entropy": 5.559749412536621,
      "epoch": 2.2133526850507983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004509420521881375,
      "loss": 4.9128,
      "mean_token_accuracy": 0.21610805988311768,
      "num_tokens": 50629483.0,
      "step": 24400
    },
    {
      "entropy": 5.547110080718994,
      "epoch": 2.2138062409288826,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004509218818765211,
      "loss": 4.8649,
      "mean_token_accuracy": 0.21698973774909974,
      "num_tokens": 50638552.0,
      "step": 24405
    },
    {
      "entropy": 5.592579364776611,
      "epoch": 2.2142597968069664,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045090170792679544,
      "loss": 4.8788,
      "mean_token_accuracy": 0.21044542789459228,
      "num_tokens": 50648752.0,
      "step": 24410
    },
    {
      "entropy": 5.581642818450928,
      "epoch": 2.2147133526850507,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045088153033937765,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21678992211818696,
      "num_tokens": 50658373.0,
      "step": 24415
    },
    {
      "entropy": 5.555846929550171,
      "epoch": 2.215166908563135,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045086134911468496,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21655402183532715,
      "num_tokens": 50668529.0,
      "step": 24420
    },
    {
      "entropy": 5.583046579360962,
      "epoch": 2.215620464441219,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004508411642531347,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21519817858934404,
      "num_tokens": 50678038.0,
      "step": 24425
    },
    {
      "entropy": 5.552519655227661,
      "epoch": 2.2160740203193035,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004508209757551445,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22207371443510054,
      "num_tokens": 50687788.0,
      "step": 24430
    },
    {
      "entropy": 5.556279611587525,
      "epoch": 2.2165275761973877,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045080078362113153,
      "loss": 4.9401,
      "mean_token_accuracy": 0.20835802853107452,
      "num_tokens": 50697748.0,
      "step": 24435
    },
    {
      "entropy": 5.498396587371826,
      "epoch": 2.2169811320754715,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045078058785151363,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2229604870080948,
      "num_tokens": 50707776.0,
      "step": 24440
    },
    {
      "entropy": 5.541026210784912,
      "epoch": 2.217434687953556,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004507603884467085,
      "loss": 4.9792,
      "mean_token_accuracy": 0.20944196730852127,
      "num_tokens": 50717547.0,
      "step": 24445
    },
    {
      "entropy": 5.497593784332276,
      "epoch": 2.21788824383164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004507401854071336,
      "loss": 4.8774,
      "mean_token_accuracy": 0.21882927119731904,
      "num_tokens": 50728446.0,
      "step": 24450
    },
    {
      "entropy": 5.526361465454102,
      "epoch": 2.2183417997097243,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045071997873320687,
      "loss": 4.8458,
      "mean_token_accuracy": 0.21045358031988143,
      "num_tokens": 50737909.0,
      "step": 24455
    },
    {
      "entropy": 5.522407722473145,
      "epoch": 2.2187953555878086,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045069976842534635,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21859708428382874,
      "num_tokens": 50749745.0,
      "step": 24460
    },
    {
      "entropy": 5.588571310043335,
      "epoch": 2.2192489114658924,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004506795544839697,
      "loss": 4.9042,
      "mean_token_accuracy": 0.21195634603500366,
      "num_tokens": 50760073.0,
      "step": 24465
    },
    {
      "entropy": 5.601437377929687,
      "epoch": 2.2197024673439767,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004506593369094951,
      "loss": 4.9033,
      "mean_token_accuracy": 0.20859922468662262,
      "num_tokens": 50769261.0,
      "step": 24470
    },
    {
      "entropy": 5.610005664825439,
      "epoch": 2.220156023222061,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045063911570234074,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20316256880760192,
      "num_tokens": 50779572.0,
      "step": 24475
    },
    {
      "entropy": 5.5070610523223875,
      "epoch": 2.2206095791001452,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004506188908629247,
      "loss": 4.8688,
      "mean_token_accuracy": 0.21925585418939592,
      "num_tokens": 50789893.0,
      "step": 24480
    },
    {
      "entropy": 5.476873111724854,
      "epoch": 2.2210631349782295,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004505986623916652,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2140084221959114,
      "num_tokens": 50799177.0,
      "step": 24485
    },
    {
      "entropy": 5.513564729690552,
      "epoch": 2.2215166908563133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004505784302889806,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22047033309936523,
      "num_tokens": 50810093.0,
      "step": 24490
    },
    {
      "entropy": 5.623832654953003,
      "epoch": 2.2219702467343976,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045055819455528935,
      "loss": 5.0515,
      "mean_token_accuracy": 0.20419042110443114,
      "num_tokens": 50820272.0,
      "step": 24495
    },
    {
      "entropy": 5.577584600448608,
      "epoch": 2.222423802612482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004505379551910099,
      "loss": 4.9935,
      "mean_token_accuracy": 0.2101212188601494,
      "num_tokens": 50830301.0,
      "step": 24500
    },
    {
      "entropy": 5.627776765823365,
      "epoch": 2.222877358490566,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004505177121965608,
      "loss": 4.9856,
      "mean_token_accuracy": 0.21157042533159257,
      "num_tokens": 50839937.0,
      "step": 24505
    },
    {
      "entropy": 5.536531209945679,
      "epoch": 2.2233309143686504,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045049746557236067,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2204652175307274,
      "num_tokens": 50850295.0,
      "step": 24510
    },
    {
      "entropy": 5.562899589538574,
      "epoch": 2.223784470246734,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045047721531882827,
      "loss": 4.9682,
      "mean_token_accuracy": 0.21106036007404327,
      "num_tokens": 50861889.0,
      "step": 24515
    },
    {
      "entropy": 5.569835805892945,
      "epoch": 2.2242380261248185,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004504569614363823,
      "loss": 4.9118,
      "mean_token_accuracy": 0.20758492797613143,
      "num_tokens": 50872011.0,
      "step": 24520
    },
    {
      "entropy": 5.627222299575806,
      "epoch": 2.2246915820029027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004504367039254416,
      "loss": 4.959,
      "mean_token_accuracy": 0.21358663737773895,
      "num_tokens": 50882559.0,
      "step": 24525
    },
    {
      "entropy": 5.555338525772095,
      "epoch": 2.225145137880987,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045041644278642523,
      "loss": 5.0096,
      "mean_token_accuracy": 0.20741062015295028,
      "num_tokens": 50894132.0,
      "step": 24530
    },
    {
      "entropy": 5.605420446395874,
      "epoch": 2.2255986937590713,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045039617801975214,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2162576884031296,
      "num_tokens": 50904337.0,
      "step": 24535
    },
    {
      "entropy": 5.619713401794433,
      "epoch": 2.2260522496371555,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045037590962584143,
      "loss": 4.9741,
      "mean_token_accuracy": 0.20595652610063553,
      "num_tokens": 50914961.0,
      "step": 24540
    },
    {
      "entropy": 5.605706739425659,
      "epoch": 2.2265058055152394,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004503556376051121,
      "loss": 4.9378,
      "mean_token_accuracy": 0.2151291236281395,
      "num_tokens": 50925682.0,
      "step": 24545
    },
    {
      "entropy": 5.531191778182984,
      "epoch": 2.2269593613933236,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045033536195798355,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22426579594612123,
      "num_tokens": 50935809.0,
      "step": 24550
    },
    {
      "entropy": 5.6418431282043455,
      "epoch": 2.227412917271408,
      "grad_norm": 1.140625,
      "learning_rate": 0.000450315082684875,
      "loss": 5.0412,
      "mean_token_accuracy": 0.20077943205833435,
      "num_tokens": 50946098.0,
      "step": 24555
    },
    {
      "entropy": 5.605287170410156,
      "epoch": 2.227866473149492,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004502947997862059,
      "loss": 4.9862,
      "mean_token_accuracy": 0.20992807894945145,
      "num_tokens": 50956240.0,
      "step": 24560
    },
    {
      "entropy": 5.55462646484375,
      "epoch": 2.228320029027576,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004502745132623957,
      "loss": 4.904,
      "mean_token_accuracy": 0.2209315612912178,
      "num_tokens": 50966870.0,
      "step": 24565
    },
    {
      "entropy": 5.5376228332519535,
      "epoch": 2.2287735849056602,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004502542231138639,
      "loss": 4.9088,
      "mean_token_accuracy": 0.20991381406784057,
      "num_tokens": 50977528.0,
      "step": 24570
    },
    {
      "entropy": 5.548512029647827,
      "epoch": 2.2292271407837445,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045023392934103005,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2012219101190567,
      "num_tokens": 50988565.0,
      "step": 24575
    },
    {
      "entropy": 5.566080474853516,
      "epoch": 2.2296806966618288,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045021363194431397,
      "loss": 4.9812,
      "mean_token_accuracy": 0.21750404238700866,
      "num_tokens": 50998410.0,
      "step": 24580
    },
    {
      "entropy": 5.601751327514648,
      "epoch": 2.230134252539913,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004501933309241352,
      "loss": 4.9617,
      "mean_token_accuracy": 0.2097803086042404,
      "num_tokens": 51008890.0,
      "step": 24585
    },
    {
      "entropy": 5.530166912078857,
      "epoch": 2.2305878084179973,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004501730262809138,
      "loss": 4.8927,
      "mean_token_accuracy": 0.21603165864944457,
      "num_tokens": 51020264.0,
      "step": 24590
    },
    {
      "entropy": 5.545501804351806,
      "epoch": 2.231041364296081,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501527180150696,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22150825411081315,
      "num_tokens": 51029892.0,
      "step": 24595
    },
    {
      "entropy": 5.632744598388672,
      "epoch": 2.2314949201741654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004501324061270224,
      "loss": 5.0545,
      "mean_token_accuracy": 0.20776804238557817,
      "num_tokens": 51040499.0,
      "step": 24600
    },
    {
      "entropy": 5.457366609573365,
      "epoch": 2.2319484760522497,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004501120906171926,
      "loss": 4.8432,
      "mean_token_accuracy": 0.21575955152511597,
      "num_tokens": 51050395.0,
      "step": 24605
    },
    {
      "entropy": 5.535169076919556,
      "epoch": 2.232402031930334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045009177148600004,
      "loss": 4.8493,
      "mean_token_accuracy": 0.22044509500265122,
      "num_tokens": 51060316.0,
      "step": 24610
    },
    {
      "entropy": 5.4981481552124025,
      "epoch": 2.232855587808418,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045007144873386514,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21886556297540666,
      "num_tokens": 51070032.0,
      "step": 24615
    },
    {
      "entropy": 5.629982995986938,
      "epoch": 2.233309143686502,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045005112236120795,
      "loss": 5.0524,
      "mean_token_accuracy": 0.20584442913532258,
      "num_tokens": 51080738.0,
      "step": 24620
    },
    {
      "entropy": 5.600133991241455,
      "epoch": 2.2337626995645863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000450030792368449,
      "loss": 4.8803,
      "mean_token_accuracy": 0.21692751944065095,
      "num_tokens": 51090287.0,
      "step": 24625
    },
    {
      "entropy": 5.544170236587524,
      "epoch": 2.2342162554426706,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004500104587560086,
      "loss": 4.9496,
      "mean_token_accuracy": 0.20941774398088456,
      "num_tokens": 51101188.0,
      "step": 24630
    },
    {
      "entropy": 5.518419361114502,
      "epoch": 2.234669811320755,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004499901215243074,
      "loss": 4.836,
      "mean_token_accuracy": 0.21789028644561767,
      "num_tokens": 51112310.0,
      "step": 24635
    },
    {
      "entropy": 5.574601936340332,
      "epoch": 2.235123367198839,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044996978067376575,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21599217802286147,
      "num_tokens": 51122481.0,
      "step": 24640
    },
    {
      "entropy": 5.517053031921387,
      "epoch": 2.235576923076923,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044994943620480453,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21055559664964676,
      "num_tokens": 51131918.0,
      "step": 24645
    },
    {
      "entropy": 5.556273317337036,
      "epoch": 2.236030478955007,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044992908811784446,
      "loss": 4.9937,
      "mean_token_accuracy": 0.19958586990833282,
      "num_tokens": 51142861.0,
      "step": 24650
    },
    {
      "entropy": 5.604096174240112,
      "epoch": 2.2364840348330914,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044990873641330615,
      "loss": 4.9757,
      "mean_token_accuracy": 0.2109047293663025,
      "num_tokens": 51153956.0,
      "step": 24655
    },
    {
      "entropy": 5.589402198791504,
      "epoch": 2.2369375907111757,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004498883810916107,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22332633137702942,
      "num_tokens": 51164499.0,
      "step": 24660
    },
    {
      "entropy": 5.587441253662109,
      "epoch": 2.23739114658926,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004498680221531789,
      "loss": 4.9967,
      "mean_token_accuracy": 0.21051541119813919,
      "num_tokens": 51175722.0,
      "step": 24665
    },
    {
      "entropy": 5.54037561416626,
      "epoch": 2.237844702467344,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004498476595984319,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21334589272737503,
      "num_tokens": 51184597.0,
      "step": 24670
    },
    {
      "entropy": 5.506781387329101,
      "epoch": 2.238298258345428,
      "grad_norm": 1.125,
      "learning_rate": 0.0004498272934277906,
      "loss": 4.9484,
      "mean_token_accuracy": 0.21041173338890076,
      "num_tokens": 51194751.0,
      "step": 24675
    },
    {
      "entropy": 5.559561920166016,
      "epoch": 2.2387518142235123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044980692364167646,
      "loss": 4.9167,
      "mean_token_accuracy": 0.21523239463567734,
      "num_tokens": 51204184.0,
      "step": 24680
    },
    {
      "entropy": 5.557795667648316,
      "epoch": 2.2392053701015966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004497865502405105,
      "loss": 4.9643,
      "mean_token_accuracy": 0.21766977161169052,
      "num_tokens": 51213761.0,
      "step": 24685
    },
    {
      "entropy": 5.557820653915405,
      "epoch": 2.239658925979681,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044976617322471424,
      "loss": 4.905,
      "mean_token_accuracy": 0.21327862292528152,
      "num_tokens": 51224150.0,
      "step": 24690
    },
    {
      "entropy": 5.612838315963745,
      "epoch": 2.2401124818577647,
      "grad_norm": 1.109375,
      "learning_rate": 0.000449745792594709,
      "loss": 5.0084,
      "mean_token_accuracy": 0.20827849209308624,
      "num_tokens": 51235133.0,
      "step": 24695
    },
    {
      "entropy": 5.604679155349731,
      "epoch": 2.240566037735849,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044972540835091617,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21470061540603638,
      "num_tokens": 51245632.0,
      "step": 24700
    },
    {
      "entropy": 5.6119279861450195,
      "epoch": 2.241019593613933,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044970502049375736,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21497476994991302,
      "num_tokens": 51255747.0,
      "step": 24705
    },
    {
      "entropy": 5.5471607685089115,
      "epoch": 2.2414731494920175,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004496846290236542,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21820943355560302,
      "num_tokens": 51265531.0,
      "step": 24710
    },
    {
      "entropy": 5.501578712463379,
      "epoch": 2.2419267053701017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044966423394102845,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21295949518680574,
      "num_tokens": 51275319.0,
      "step": 24715
    },
    {
      "entropy": 5.543272304534912,
      "epoch": 2.2423802612481856,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044964383524630176,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21648724377155304,
      "num_tokens": 51285655.0,
      "step": 24720
    },
    {
      "entropy": 5.514241027832031,
      "epoch": 2.24283381712627,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004496234329398961,
      "loss": 4.8808,
      "mean_token_accuracy": 0.20576780140399933,
      "num_tokens": 51295938.0,
      "step": 24725
    },
    {
      "entropy": 5.559138441085816,
      "epoch": 2.243287373004354,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004496030270222333,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2219208985567093,
      "num_tokens": 51306483.0,
      "step": 24730
    },
    {
      "entropy": 5.567774248123169,
      "epoch": 2.2437409288824384,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044958261749373547,
      "loss": 4.9319,
      "mean_token_accuracy": 0.21395543664693834,
      "num_tokens": 51316446.0,
      "step": 24735
    },
    {
      "entropy": 5.529565382003784,
      "epoch": 2.2441944847605226,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00044956220435482465,
      "loss": 4.946,
      "mean_token_accuracy": 0.21085806041955948,
      "num_tokens": 51326840.0,
      "step": 24740
    },
    {
      "entropy": 5.566893863677978,
      "epoch": 2.244648040638607,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044954178760592283,
      "loss": 4.9929,
      "mean_token_accuracy": 0.20667845010757446,
      "num_tokens": 51337573.0,
      "step": 24745
    },
    {
      "entropy": 5.595806646347046,
      "epoch": 2.2451015965166907,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044952136724745254,
      "loss": 4.9496,
      "mean_token_accuracy": 0.2157149940729141,
      "num_tokens": 51347827.0,
      "step": 24750
    },
    {
      "entropy": 5.525757932662964,
      "epoch": 2.245555152394775,
      "grad_norm": 1.125,
      "learning_rate": 0.0004495009432798358,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21181443631649016,
      "num_tokens": 51358396.0,
      "step": 24755
    },
    {
      "entropy": 5.586498546600342,
      "epoch": 2.2460087082728593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044948051570349503,
      "loss": 4.9406,
      "mean_token_accuracy": 0.2058605894446373,
      "num_tokens": 51368882.0,
      "step": 24760
    },
    {
      "entropy": 5.606159830093384,
      "epoch": 2.2464622641509435,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004494600845188528,
      "loss": 4.9873,
      "mean_token_accuracy": 0.21260903924703597,
      "num_tokens": 51378717.0,
      "step": 24765
    },
    {
      "entropy": 5.585380268096924,
      "epoch": 2.2469158200290273,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044943964972633154,
      "loss": 4.9284,
      "mean_token_accuracy": 0.20769332647323607,
      "num_tokens": 51389270.0,
      "step": 24770
    },
    {
      "entropy": 5.5634852886199955,
      "epoch": 2.2473693759071116,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004494192113263539,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21435392796993255,
      "num_tokens": 51400025.0,
      "step": 24775
    },
    {
      "entropy": 5.56493649482727,
      "epoch": 2.247822931785196,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044939876931934255,
      "loss": 4.9476,
      "mean_token_accuracy": 0.22143661975860596,
      "num_tokens": 51410903.0,
      "step": 24780
    },
    {
      "entropy": 5.512462711334228,
      "epoch": 2.24827648766328,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004493783237057201,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21482669711112976,
      "num_tokens": 51420246.0,
      "step": 24785
    },
    {
      "entropy": 5.625479125976563,
      "epoch": 2.2487300435413644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004493578744859095,
      "loss": 5.047,
      "mean_token_accuracy": 0.20918410867452622,
      "num_tokens": 51431309.0,
      "step": 24790
    },
    {
      "entropy": 5.617939710617065,
      "epoch": 2.2491835994194487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004493374216603338,
      "loss": 4.996,
      "mean_token_accuracy": 0.20713568925857545,
      "num_tokens": 51443142.0,
      "step": 24795
    },
    {
      "entropy": 5.6893837451934814,
      "epoch": 2.2496371552975325,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004493169652294156,
      "loss": 5.1357,
      "mean_token_accuracy": 0.20327261090278625,
      "num_tokens": 51454499.0,
      "step": 24800
    },
    {
      "entropy": 5.541071605682373,
      "epoch": 2.2500907111756168,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044929650519357823,
      "loss": 4.9307,
      "mean_token_accuracy": 0.21504440158605576,
      "num_tokens": 51464667.0,
      "step": 24805
    },
    {
      "entropy": 5.596696424484253,
      "epoch": 2.250544267053701,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004492760415532447,
      "loss": 4.9565,
      "mean_token_accuracy": 0.20902094841003419,
      "num_tokens": 51473627.0,
      "step": 24810
    },
    {
      "entropy": 5.543713760375977,
      "epoch": 2.2509978229317853,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004492555743088382,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2210956558585167,
      "num_tokens": 51483052.0,
      "step": 24815
    },
    {
      "entropy": 5.528712749481201,
      "epoch": 2.2514513788098696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000449235103460782,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21018511950969695,
      "num_tokens": 51492422.0,
      "step": 24820
    },
    {
      "entropy": 5.491948366165161,
      "epoch": 2.2519049346879534,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004492146290094995,
      "loss": 4.8847,
      "mean_token_accuracy": 0.2147200211882591,
      "num_tokens": 51503566.0,
      "step": 24825
    },
    {
      "entropy": 5.5462616920471195,
      "epoch": 2.2523584905660377,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044919415095541406,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21568074971437454,
      "num_tokens": 51513699.0,
      "step": 24830
    },
    {
      "entropy": 5.554300117492676,
      "epoch": 2.252812046444122,
      "grad_norm": 1.125,
      "learning_rate": 0.00044917366929894914,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21047624796628953,
      "num_tokens": 51524745.0,
      "step": 24835
    },
    {
      "entropy": 5.532713842391968,
      "epoch": 2.253265602322206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044915318404052846,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21105870753526687,
      "num_tokens": 51536516.0,
      "step": 24840
    },
    {
      "entropy": 5.602594089508057,
      "epoch": 2.2537191582002905,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044913269518057547,
      "loss": 4.9822,
      "mean_token_accuracy": 0.20331063121557236,
      "num_tokens": 51547063.0,
      "step": 24845
    },
    {
      "entropy": 5.586305618286133,
      "epoch": 2.2541727140783743,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000449112202719514,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21130322068929672,
      "num_tokens": 51557244.0,
      "step": 24850
    },
    {
      "entropy": 5.584157228469849,
      "epoch": 2.2546262699564585,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004490917066577678,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2129945710301399,
      "num_tokens": 51567630.0,
      "step": 24855
    },
    {
      "entropy": 5.530771207809448,
      "epoch": 2.255079825834543,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044907120699576075,
      "loss": 4.9695,
      "mean_token_accuracy": 0.215284264087677,
      "num_tokens": 51577951.0,
      "step": 24860
    },
    {
      "entropy": 5.574575757980346,
      "epoch": 2.255533381712627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044905070373391684,
      "loss": 4.9179,
      "mean_token_accuracy": 0.21349529325962066,
      "num_tokens": 51587533.0,
      "step": 24865
    },
    {
      "entropy": 5.65045428276062,
      "epoch": 2.2559869375907113,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004490301968726599,
      "loss": 4.9159,
      "mean_token_accuracy": 0.2208865687251091,
      "num_tokens": 51597212.0,
      "step": 24870
    },
    {
      "entropy": 5.577109241485596,
      "epoch": 2.256440493468795,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044900968641241426,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21696819365024567,
      "num_tokens": 51608520.0,
      "step": 24875
    },
    {
      "entropy": 5.5734734535217285,
      "epoch": 2.2568940493468794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004489891723536039,
      "loss": 5.0197,
      "mean_token_accuracy": 0.2043403685092926,
      "num_tokens": 51619651.0,
      "step": 24880
    },
    {
      "entropy": 5.536041688919068,
      "epoch": 2.2573476052249637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004489686546966531,
      "loss": 4.8418,
      "mean_token_accuracy": 0.21351330280303954,
      "num_tokens": 51630184.0,
      "step": 24885
    },
    {
      "entropy": 5.47769684791565,
      "epoch": 2.257801161103048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044894813344198626,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22362605184316636,
      "num_tokens": 51641293.0,
      "step": 24890
    },
    {
      "entropy": 5.528341245651245,
      "epoch": 2.2582547169811322,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004489276085900276,
      "loss": 4.938,
      "mean_token_accuracy": 0.2173669993877411,
      "num_tokens": 51652002.0,
      "step": 24895
    },
    {
      "entropy": 5.480300807952881,
      "epoch": 2.2587082728592165,
      "grad_norm": 1.125,
      "learning_rate": 0.00044890708014120174,
      "loss": 4.8129,
      "mean_token_accuracy": 0.2264448881149292,
      "num_tokens": 51661668.0,
      "step": 24900
    },
    {
      "entropy": 5.631719541549683,
      "epoch": 2.2591618287373003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004488865480959331,
      "loss": 4.9814,
      "mean_token_accuracy": 0.20587112456560136,
      "num_tokens": 51672879.0,
      "step": 24905
    },
    {
      "entropy": 5.564335060119629,
      "epoch": 2.2596153846153846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004488660124546464,
      "loss": 4.9163,
      "mean_token_accuracy": 0.21133174747228622,
      "num_tokens": 51683878.0,
      "step": 24910
    },
    {
      "entropy": 5.508681201934815,
      "epoch": 2.260068940493469,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004488454732177662,
      "loss": 4.9375,
      "mean_token_accuracy": 0.20718490779399873,
      "num_tokens": 51693971.0,
      "step": 24915
    },
    {
      "entropy": 5.544399070739746,
      "epoch": 2.260522496371553,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044882493038571737,
      "loss": 4.8465,
      "mean_token_accuracy": 0.22426974326372145,
      "num_tokens": 51703685.0,
      "step": 24920
    },
    {
      "entropy": 5.633752059936524,
      "epoch": 2.260976052249637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044880438395892457,
      "loss": 4.9492,
      "mean_token_accuracy": 0.2102421909570694,
      "num_tokens": 51714469.0,
      "step": 24925
    },
    {
      "entropy": 5.59210615158081,
      "epoch": 2.261429608127721,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004487838339378129,
      "loss": 5.0184,
      "mean_token_accuracy": 0.20027500838041307,
      "num_tokens": 51725283.0,
      "step": 24930
    },
    {
      "entropy": 5.5712477684021,
      "epoch": 2.2618831640058055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004487632803228072,
      "loss": 4.931,
      "mean_token_accuracy": 0.2125330924987793,
      "num_tokens": 51735628.0,
      "step": 24935
    },
    {
      "entropy": 5.488474035263062,
      "epoch": 2.2623367198838897,
      "grad_norm": 1.125,
      "learning_rate": 0.00044874272311433265,
      "loss": 4.8051,
      "mean_token_accuracy": 0.22529734969139098,
      "num_tokens": 51745991.0,
      "step": 24940
    },
    {
      "entropy": 5.501957178115845,
      "epoch": 2.262790275761974,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004487221623128143,
      "loss": 4.857,
      "mean_token_accuracy": 0.22106154412031173,
      "num_tokens": 51756306.0,
      "step": 24945
    },
    {
      "entropy": 5.574098920822143,
      "epoch": 2.2632438316400583,
      "grad_norm": 1.125,
      "learning_rate": 0.00044870159791867743,
      "loss": 4.96,
      "mean_token_accuracy": 0.21319975554943085,
      "num_tokens": 51765466.0,
      "step": 24950
    },
    {
      "entropy": 5.573635721206665,
      "epoch": 2.263697387518142,
      "grad_norm": 1.125,
      "learning_rate": 0.00044868102993234716,
      "loss": 4.939,
      "mean_token_accuracy": 0.20836655646562577,
      "num_tokens": 51775526.0,
      "step": 24955
    },
    {
      "entropy": 5.572900772094727,
      "epoch": 2.2641509433962264,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044866045835424896,
      "loss": 4.9349,
      "mean_token_accuracy": 0.21200913339853286,
      "num_tokens": 51785434.0,
      "step": 24960
    },
    {
      "entropy": 5.601145887374878,
      "epoch": 2.2646044992743106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004486398831848082,
      "loss": 4.9999,
      "mean_token_accuracy": 0.2036530628800392,
      "num_tokens": 51796047.0,
      "step": 24965
    },
    {
      "entropy": 5.6254688739776615,
      "epoch": 2.265058055152395,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044861930442445037,
      "loss": 4.9476,
      "mean_token_accuracy": 0.20650483071804046,
      "num_tokens": 51806173.0,
      "step": 24970
    },
    {
      "entropy": 5.667303800582886,
      "epoch": 2.2655116110304787,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004485987220736012,
      "loss": 5.0944,
      "mean_token_accuracy": 0.20252127647399903,
      "num_tokens": 51817644.0,
      "step": 24975
    },
    {
      "entropy": 5.620352458953858,
      "epoch": 2.265965166908563,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004485781361326861,
      "loss": 5.0598,
      "mean_token_accuracy": 0.20552604049444198,
      "num_tokens": 51829746.0,
      "step": 24980
    },
    {
      "entropy": 5.589034223556519,
      "epoch": 2.2664187227866472,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000448557546602131,
      "loss": 5.0035,
      "mean_token_accuracy": 0.20356702655553818,
      "num_tokens": 51840868.0,
      "step": 24985
    },
    {
      "entropy": 5.599922800064087,
      "epoch": 2.2668722786647315,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044853695348236156,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2149260401725769,
      "num_tokens": 51851385.0,
      "step": 24990
    },
    {
      "entropy": 5.584407377243042,
      "epoch": 2.2673258345428158,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044851635677380374,
      "loss": 4.9005,
      "mean_token_accuracy": 0.2099836364388466,
      "num_tokens": 51861021.0,
      "step": 24995
    },
    {
      "entropy": 5.577030849456787,
      "epoch": 2.2677793904209,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004484957564768834,
      "loss": 5.0014,
      "mean_token_accuracy": 0.2061031013727188,
      "num_tokens": 51870688.0,
      "step": 25000
    },
    {
      "entropy": 5.558215999603272,
      "epoch": 2.268232946298984,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044847515259202657,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21066987812519072,
      "num_tokens": 51880439.0,
      "step": 25005
    },
    {
      "entropy": 5.508165693283081,
      "epoch": 2.268686502177068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044845454511965947,
      "loss": 4.9196,
      "mean_token_accuracy": 0.21572455167770385,
      "num_tokens": 51890861.0,
      "step": 25010
    },
    {
      "entropy": 5.621864318847656,
      "epoch": 2.2691400580551524,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044843393406020823,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21414869725704194,
      "num_tokens": 51900687.0,
      "step": 25015
    },
    {
      "entropy": 5.621375179290771,
      "epoch": 2.2695936139332367,
      "grad_norm": 1.0,
      "learning_rate": 0.00044841331941409884,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20031085163354873,
      "num_tokens": 51911805.0,
      "step": 25020
    },
    {
      "entropy": 5.576607608795166,
      "epoch": 2.270047169811321,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044839270118175793,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22532748878002168,
      "num_tokens": 51922520.0,
      "step": 25025
    },
    {
      "entropy": 5.6344075202941895,
      "epoch": 2.2705007256894048,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004483720793636118,
      "loss": 5.0217,
      "mean_token_accuracy": 0.20068564414978027,
      "num_tokens": 51934056.0,
      "step": 25030
    },
    {
      "entropy": 5.586591911315918,
      "epoch": 2.270954281567489,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044835145396008685,
      "loss": 4.9926,
      "mean_token_accuracy": 0.21061157286167145,
      "num_tokens": 51944696.0,
      "step": 25035
    },
    {
      "entropy": 5.604296636581421,
      "epoch": 2.2714078374455733,
      "grad_norm": 1.125,
      "learning_rate": 0.0004483308249716097,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20802507996559144,
      "num_tokens": 51955111.0,
      "step": 25040
    },
    {
      "entropy": 5.547308683395386,
      "epoch": 2.2718613933236576,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004483101923986068,
      "loss": 4.933,
      "mean_token_accuracy": 0.21139310896396638,
      "num_tokens": 51966036.0,
      "step": 25045
    },
    {
      "entropy": 5.517368030548096,
      "epoch": 2.272314949201742,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044828955624150505,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21404066532850266,
      "num_tokens": 51976060.0,
      "step": 25050
    },
    {
      "entropy": 5.567092990875244,
      "epoch": 2.2727685050798256,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004482689165007311,
      "loss": 4.9173,
      "mean_token_accuracy": 0.20896288901567459,
      "num_tokens": 51986016.0,
      "step": 25055
    },
    {
      "entropy": 5.550427341461182,
      "epoch": 2.27322206095791,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044824827317671177,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22529981285333633,
      "num_tokens": 51994794.0,
      "step": 25060
    },
    {
      "entropy": 5.486112499237061,
      "epoch": 2.273675616835994,
      "grad_norm": 1.109375,
      "learning_rate": 0.000448227626269874,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2184893622994423,
      "num_tokens": 52004976.0,
      "step": 25065
    },
    {
      "entropy": 5.596336221694946,
      "epoch": 2.2741291727140784,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044820697578064485,
      "loss": 4.9622,
      "mean_token_accuracy": 0.20845344811677932,
      "num_tokens": 52016380.0,
      "step": 25070
    },
    {
      "entropy": 5.6203375339508055,
      "epoch": 2.2745827285921627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004481863217094513,
      "loss": 4.8984,
      "mean_token_accuracy": 0.21370139569044114,
      "num_tokens": 52026439.0,
      "step": 25075
    },
    {
      "entropy": 5.558390951156616,
      "epoch": 2.2750362844702465,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004481656640567204,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21745213121175766,
      "num_tokens": 52037111.0,
      "step": 25080
    },
    {
      "entropy": 5.516178846359253,
      "epoch": 2.275489840348331,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004481450028228795,
      "loss": 4.9217,
      "mean_token_accuracy": 0.21593354493379593,
      "num_tokens": 52047463.0,
      "step": 25085
    },
    {
      "entropy": 5.544280910491944,
      "epoch": 2.275943396226415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004481243380083558,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2106416940689087,
      "num_tokens": 52058081.0,
      "step": 25090
    },
    {
      "entropy": 5.556483793258667,
      "epoch": 2.2763969521044993,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004481036696135767,
      "loss": 4.9559,
      "mean_token_accuracy": 0.214629465341568,
      "num_tokens": 52069367.0,
      "step": 25095
    },
    {
      "entropy": 5.602753829956055,
      "epoch": 2.2768505079825836,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004480829976389695,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2128605216741562,
      "num_tokens": 52080023.0,
      "step": 25100
    },
    {
      "entropy": 5.51301679611206,
      "epoch": 2.277304063860668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044806232208496194,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21707749515771865,
      "num_tokens": 52089184.0,
      "step": 25105
    },
    {
      "entropy": 5.581623125076294,
      "epoch": 2.2777576197387517,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004480416429519814,
      "loss": 4.9907,
      "mean_token_accuracy": 0.2084593206644058,
      "num_tokens": 52098711.0,
      "step": 25110
    },
    {
      "entropy": 5.612613296508789,
      "epoch": 2.278211175616836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004480209602404556,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2174218028783798,
      "num_tokens": 52108826.0,
      "step": 25115
    },
    {
      "entropy": 5.566018104553223,
      "epoch": 2.27866473149492,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004480002739508122,
      "loss": 4.8844,
      "mean_token_accuracy": 0.21918694525957108,
      "num_tokens": 52118934.0,
      "step": 25120
    },
    {
      "entropy": 5.513492393493652,
      "epoch": 2.2791182873730045,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004479795840834791,
      "loss": 4.8763,
      "mean_token_accuracy": 0.2190500468015671,
      "num_tokens": 52129803.0,
      "step": 25125
    },
    {
      "entropy": 5.481580638885498,
      "epoch": 2.2795718432510883,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044795889063888413,
      "loss": 4.8137,
      "mean_token_accuracy": 0.22220141142606736,
      "num_tokens": 52140070.0,
      "step": 25130
    },
    {
      "entropy": 5.70681586265564,
      "epoch": 2.2800253991291726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479381936174552,
      "loss": 5.0745,
      "mean_token_accuracy": 0.20509280562400817,
      "num_tokens": 52150462.0,
      "step": 25135
    },
    {
      "entropy": 5.606572008132934,
      "epoch": 2.280478955007257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479174930196204,
      "loss": 4.9423,
      "mean_token_accuracy": 0.212855526804924,
      "num_tokens": 52160552.0,
      "step": 25140
    },
    {
      "entropy": 5.462222623825073,
      "epoch": 2.280932510885341,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044789678884580767,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22200330793857576,
      "num_tokens": 52169780.0,
      "step": 25145
    },
    {
      "entropy": 5.556251764297485,
      "epoch": 2.2813860667634254,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044787608109644536,
      "loss": 4.9596,
      "mean_token_accuracy": 0.2206101283431053,
      "num_tokens": 52180055.0,
      "step": 25150
    },
    {
      "entropy": 5.554968166351318,
      "epoch": 2.2818396226415096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004478553697719616,
      "loss": 4.8975,
      "mean_token_accuracy": 0.21497833132743835,
      "num_tokens": 52189772.0,
      "step": 25155
    },
    {
      "entropy": 5.56885142326355,
      "epoch": 2.2822931785195935,
      "grad_norm": 1.125,
      "learning_rate": 0.0004478346548727848,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22618289589881896,
      "num_tokens": 52199608.0,
      "step": 25160
    },
    {
      "entropy": 5.525118494033814,
      "epoch": 2.2827467343976777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004478139363993433,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21198580116033555,
      "num_tokens": 52208236.0,
      "step": 25165
    },
    {
      "entropy": 5.565851926803589,
      "epoch": 2.283200290275762,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004477932143520655,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21849876195192336,
      "num_tokens": 52218403.0,
      "step": 25170
    },
    {
      "entropy": 5.592417573928833,
      "epoch": 2.2836538461538463,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044777248873138005,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21791344732046128,
      "num_tokens": 52228255.0,
      "step": 25175
    },
    {
      "entropy": 5.5939384460449215,
      "epoch": 2.2841074020319305,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044775175953771547,
      "loss": 4.9609,
      "mean_token_accuracy": 0.20872234255075456,
      "num_tokens": 52237732.0,
      "step": 25180
    },
    {
      "entropy": 5.674498128890991,
      "epoch": 2.2845609579100143,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004477310267715005,
      "loss": 5.0925,
      "mean_token_accuracy": 0.2007480666041374,
      "num_tokens": 52249226.0,
      "step": 25185
    },
    {
      "entropy": 5.60660138130188,
      "epoch": 2.2850145137880986,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004477102904331639,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21450406461954116,
      "num_tokens": 52259550.0,
      "step": 25190
    },
    {
      "entropy": 5.597080850601197,
      "epoch": 2.285468069666183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044768955052313443,
      "loss": 4.959,
      "mean_token_accuracy": 0.21191644221544265,
      "num_tokens": 52269670.0,
      "step": 25195
    },
    {
      "entropy": 5.582578229904175,
      "epoch": 2.285921625544267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044766880704184104,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21737164855003357,
      "num_tokens": 52280135.0,
      "step": 25200
    },
    {
      "entropy": 5.536432695388794,
      "epoch": 2.2863751814223514,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004476480599897128,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22213241308927537,
      "num_tokens": 52291806.0,
      "step": 25205
    },
    {
      "entropy": 5.476989459991455,
      "epoch": 2.2868287373004352,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004476273093671786,
      "loss": 4.8952,
      "mean_token_accuracy": 0.21518257707357408,
      "num_tokens": 52302118.0,
      "step": 25210
    },
    {
      "entropy": 5.582285928726196,
      "epoch": 2.2872822931785195,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044760655517466766,
      "loss": 4.9907,
      "mean_token_accuracy": 0.20804875046014787,
      "num_tokens": 52312762.0,
      "step": 25215
    },
    {
      "entropy": 5.598785352706909,
      "epoch": 2.2877358490566038,
      "grad_norm": 1.125,
      "learning_rate": 0.00044758579741260916,
      "loss": 4.9681,
      "mean_token_accuracy": 0.20767925232648848,
      "num_tokens": 52323367.0,
      "step": 25220
    },
    {
      "entropy": 5.641805982589721,
      "epoch": 2.288189404934688,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044756503608143247,
      "loss": 5.0077,
      "mean_token_accuracy": 0.20431312620639802,
      "num_tokens": 52335231.0,
      "step": 25225
    },
    {
      "entropy": 5.6252013683319095,
      "epoch": 2.2886429608127723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044754427118156675,
      "loss": 4.892,
      "mean_token_accuracy": 0.2135847494006157,
      "num_tokens": 52345066.0,
      "step": 25230
    },
    {
      "entropy": 5.587812328338623,
      "epoch": 2.289096516690856,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004475235027134416,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21514323800802232,
      "num_tokens": 52356383.0,
      "step": 25235
    },
    {
      "entropy": 5.597359371185303,
      "epoch": 2.2895500725689404,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004475027306774865,
      "loss": 5.0579,
      "mean_token_accuracy": 0.21374959498643875,
      "num_tokens": 52366013.0,
      "step": 25240
    },
    {
      "entropy": 5.571437549591065,
      "epoch": 2.2900036284470247,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044748195507413095,
      "loss": 4.9785,
      "mean_token_accuracy": 0.21053512394428253,
      "num_tokens": 52376298.0,
      "step": 25245
    },
    {
      "entropy": 5.464027690887451,
      "epoch": 2.290457184325109,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044746117590380456,
      "loss": 4.8398,
      "mean_token_accuracy": 0.21738963425159455,
      "num_tokens": 52387134.0,
      "step": 25250
    },
    {
      "entropy": 5.527815198898315,
      "epoch": 2.290910740203193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004474403931669371,
      "loss": 4.8453,
      "mean_token_accuracy": 0.21763660460710527,
      "num_tokens": 52397389.0,
      "step": 25255
    },
    {
      "entropy": 5.538775634765625,
      "epoch": 2.2913642960812775,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004474196068639584,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22069861441850663,
      "num_tokens": 52407456.0,
      "step": 25260
    },
    {
      "entropy": 5.51659197807312,
      "epoch": 2.2918178519593613,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004473988169952983,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21715328991413116,
      "num_tokens": 52417967.0,
      "step": 25265
    },
    {
      "entropy": 5.618671989440918,
      "epoch": 2.2922714078374455,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044737802356138674,
      "loss": 5.0031,
      "mean_token_accuracy": 0.211630679666996,
      "num_tokens": 52428757.0,
      "step": 25270
    },
    {
      "entropy": 5.5691547870635985,
      "epoch": 2.29272496371553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004473572265626537,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21781871914863588,
      "num_tokens": 52438973.0,
      "step": 25275
    },
    {
      "entropy": 5.514788389205933,
      "epoch": 2.293178519593614,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004473364259995293,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21047080010175706,
      "num_tokens": 52449323.0,
      "step": 25280
    },
    {
      "entropy": 5.545239353179932,
      "epoch": 2.293632075471698,
      "grad_norm": 1.25,
      "learning_rate": 0.0004473156218724436,
      "loss": 4.9207,
      "mean_token_accuracy": 0.21533533781766892,
      "num_tokens": 52459299.0,
      "step": 25285
    },
    {
      "entropy": 5.515853452682495,
      "epoch": 2.294085631349782,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044729481418182704,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21182161271572114,
      "num_tokens": 52470026.0,
      "step": 25290
    },
    {
      "entropy": 5.577751684188843,
      "epoch": 2.2945391872278664,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044727400292810985,
      "loss": 4.881,
      "mean_token_accuracy": 0.2183116689324379,
      "num_tokens": 52479975.0,
      "step": 25295
    },
    {
      "entropy": 5.651311159133911,
      "epoch": 2.2949927431059507,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044725318811172227,
      "loss": 5.0195,
      "mean_token_accuracy": 0.2090300902724266,
      "num_tokens": 52491067.0,
      "step": 25300
    },
    {
      "entropy": 5.538263368606567,
      "epoch": 2.295446298984035,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447232369733095,
      "loss": 4.9176,
      "mean_token_accuracy": 0.21957393139600753,
      "num_tokens": 52500671.0,
      "step": 25305
    },
    {
      "entropy": 5.5429870128631595,
      "epoch": 2.2958998548621192,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004472115477926583,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2129712849855423,
      "num_tokens": 52509910.0,
      "step": 25310
    },
    {
      "entropy": 5.569476890563965,
      "epoch": 2.296353410740203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044719072229084295,
      "loss": 4.904,
      "mean_token_accuracy": 0.2201918125152588,
      "num_tokens": 52520517.0,
      "step": 25315
    },
    {
      "entropy": 5.5857010841369625,
      "epoch": 2.2968069666182873,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004471698932280796,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21291290670633317,
      "num_tokens": 52530760.0,
      "step": 25320
    },
    {
      "entropy": 5.632915258407593,
      "epoch": 2.2972605224963716,
      "grad_norm": 1.15625,
      "learning_rate": 0.000447149060604799,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20496514737606047,
      "num_tokens": 52541791.0,
      "step": 25325
    },
    {
      "entropy": 5.672983455657959,
      "epoch": 2.297714078374456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044712822442143194,
      "loss": 4.9781,
      "mean_token_accuracy": 0.2142412021756172,
      "num_tokens": 52551885.0,
      "step": 25330
    },
    {
      "entropy": 5.577205610275269,
      "epoch": 2.2981676342525397,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044710738467840927,
      "loss": 4.8845,
      "mean_token_accuracy": 0.21626444458961486,
      "num_tokens": 52563080.0,
      "step": 25335
    },
    {
      "entropy": 5.543568563461304,
      "epoch": 2.298621190130624,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004470865413761621,
      "loss": 4.8648,
      "mean_token_accuracy": 0.2250712513923645,
      "num_tokens": 52573564.0,
      "step": 25340
    },
    {
      "entropy": 5.541332769393921,
      "epoch": 2.299074746008708,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044706569451512143,
      "loss": 4.9147,
      "mean_token_accuracy": 0.21609037667512893,
      "num_tokens": 52583801.0,
      "step": 25345
    },
    {
      "entropy": 5.583554410934449,
      "epoch": 2.2995283018867925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044704484409571825,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20890864431858064,
      "num_tokens": 52594908.0,
      "step": 25350
    },
    {
      "entropy": 5.56727294921875,
      "epoch": 2.2999818577648767,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044702399011838393,
      "loss": 4.8755,
      "mean_token_accuracy": 0.20669530481100082,
      "num_tokens": 52604861.0,
      "step": 25355
    },
    {
      "entropy": 5.571467590332031,
      "epoch": 2.300435413642961,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004470031325835497,
      "loss": 4.9099,
      "mean_token_accuracy": 0.215513151884079,
      "num_tokens": 52616203.0,
      "step": 25360
    },
    {
      "entropy": 5.5409613132476805,
      "epoch": 2.300888969521045,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469822714916468,
      "loss": 4.8639,
      "mean_token_accuracy": 0.21549715995788574,
      "num_tokens": 52625838.0,
      "step": 25365
    },
    {
      "entropy": 5.497993755340576,
      "epoch": 2.301342525399129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004469614068431067,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21902282536029816,
      "num_tokens": 52635741.0,
      "step": 25370
    },
    {
      "entropy": 5.523423385620117,
      "epoch": 2.3017960812772134,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004469405386383608,
      "loss": 4.8465,
      "mean_token_accuracy": 0.21964609920978545,
      "num_tokens": 52645948.0,
      "step": 25375
    },
    {
      "entropy": 5.583684587478638,
      "epoch": 2.3022496371552976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044691966687784087,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21478646248579025,
      "num_tokens": 52655073.0,
      "step": 25380
    },
    {
      "entropy": 5.5510749340057375,
      "epoch": 2.302703193033382,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044689879156197833,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22576086223125458,
      "num_tokens": 52664429.0,
      "step": 25385
    },
    {
      "entropy": 5.61441559791565,
      "epoch": 2.3031567489114657,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044687791269120504,
      "loss": 5.0129,
      "mean_token_accuracy": 0.210846945643425,
      "num_tokens": 52674890.0,
      "step": 25390
    },
    {
      "entropy": 5.665025424957276,
      "epoch": 2.30361030478955,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044685703026595266,
      "loss": 5.0149,
      "mean_token_accuracy": 0.2166868880391121,
      "num_tokens": 52685673.0,
      "step": 25395
    },
    {
      "entropy": 5.668634223937988,
      "epoch": 2.3040638606676342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044683614428665314,
      "loss": 5.0265,
      "mean_token_accuracy": 0.20779452323913575,
      "num_tokens": 52696696.0,
      "step": 25400
    },
    {
      "entropy": 5.577009057998657,
      "epoch": 2.3045174165457185,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044681525475373843,
      "loss": 4.9657,
      "mean_token_accuracy": 0.2119971767067909,
      "num_tokens": 52707662.0,
      "step": 25405
    },
    {
      "entropy": 5.550281524658203,
      "epoch": 2.3049709724238028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004467943616676403,
      "loss": 4.9129,
      "mean_token_accuracy": 0.2126191660761833,
      "num_tokens": 52718438.0,
      "step": 25410
    },
    {
      "entropy": 5.530935335159302,
      "epoch": 2.3054245283018866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004467734650287911,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21752914190292358,
      "num_tokens": 52729267.0,
      "step": 25415
    },
    {
      "entropy": 5.522969532012939,
      "epoch": 2.305878084179971,
      "grad_norm": 1.125,
      "learning_rate": 0.0004467525648376229,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22307204753160476,
      "num_tokens": 52738754.0,
      "step": 25420
    },
    {
      "entropy": 5.505483818054199,
      "epoch": 2.306331640058055,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004467316610945679,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22157405763864518,
      "num_tokens": 52749086.0,
      "step": 25425
    },
    {
      "entropy": 5.645213985443116,
      "epoch": 2.3067851959361394,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00044671075380005825,
      "loss": 4.9963,
      "mean_token_accuracy": 0.20951149761676788,
      "num_tokens": 52758961.0,
      "step": 25430
    },
    {
      "entropy": 5.650061225891113,
      "epoch": 2.3072387518142237,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004466898429545266,
      "loss": 4.9521,
      "mean_token_accuracy": 0.21575318723917009,
      "num_tokens": 52769447.0,
      "step": 25435
    },
    {
      "entropy": 5.625081634521484,
      "epoch": 2.3076923076923075,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044666892855840513,
      "loss": 5.0129,
      "mean_token_accuracy": 0.20519701093435289,
      "num_tokens": 52780306.0,
      "step": 25440
    },
    {
      "entropy": 5.524453973770141,
      "epoch": 2.3081458635703918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004466480106121266,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21147909313440322,
      "num_tokens": 52791263.0,
      "step": 25445
    },
    {
      "entropy": 5.574639177322387,
      "epoch": 2.308599419448476,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004466270891161234,
      "loss": 4.9353,
      "mean_token_accuracy": 0.20846904218196868,
      "num_tokens": 52802194.0,
      "step": 25450
    },
    {
      "entropy": 5.5787606716156,
      "epoch": 2.3090529753265603,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044660616407082827,
      "loss": 4.907,
      "mean_token_accuracy": 0.2227462962269783,
      "num_tokens": 52812391.0,
      "step": 25455
    },
    {
      "entropy": 5.578348207473755,
      "epoch": 2.3095065312046446,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044658523547667385,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2089507147669792,
      "num_tokens": 52823540.0,
      "step": 25460
    },
    {
      "entropy": 5.5237115859985355,
      "epoch": 2.309960087082729,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004465643033340931,
      "loss": 4.8716,
      "mean_token_accuracy": 0.21838271766901016,
      "num_tokens": 52833238.0,
      "step": 25465
    },
    {
      "entropy": 5.581785345077515,
      "epoch": 2.3104136429608126,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044654336764351886,
      "loss": 4.9934,
      "mean_token_accuracy": 0.20452653169631957,
      "num_tokens": 52844510.0,
      "step": 25470
    },
    {
      "entropy": 5.571903562545776,
      "epoch": 2.310867198838897,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044652242840538406,
      "loss": 4.9707,
      "mean_token_accuracy": 0.2163538858294487,
      "num_tokens": 52856311.0,
      "step": 25475
    },
    {
      "entropy": 5.513763427734375,
      "epoch": 2.311320754716981,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004465014856201218,
      "loss": 4.8734,
      "mean_token_accuracy": 0.21733807027339935,
      "num_tokens": 52866295.0,
      "step": 25480
    },
    {
      "entropy": 5.568298292160034,
      "epoch": 2.3117743105950654,
      "grad_norm": 1.140625,
      "learning_rate": 0.000446480539288165,
      "loss": 4.967,
      "mean_token_accuracy": 0.21369968354701996,
      "num_tokens": 52876818.0,
      "step": 25485
    },
    {
      "entropy": 5.589290142059326,
      "epoch": 2.3122278664731493,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004464595894099469,
      "loss": 4.8925,
      "mean_token_accuracy": 0.21179439425468444,
      "num_tokens": 52886173.0,
      "step": 25490
    },
    {
      "entropy": 5.5394899368286135,
      "epoch": 2.3126814223512335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044643863598590094,
      "loss": 4.9107,
      "mean_token_accuracy": 0.20862531960010527,
      "num_tokens": 52895875.0,
      "step": 25495
    },
    {
      "entropy": 5.581629228591919,
      "epoch": 2.313134978229318,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044641767901646025,
      "loss": 5.0125,
      "mean_token_accuracy": 0.20567458868026733,
      "num_tokens": 52906465.0,
      "step": 25500
    },
    {
      "entropy": 5.610698986053467,
      "epoch": 2.313588534107402,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004463967185020583,
      "loss": 4.9123,
      "mean_token_accuracy": 0.21516390144824982,
      "num_tokens": 52916378.0,
      "step": 25505
    },
    {
      "entropy": 5.580998086929322,
      "epoch": 2.3140420899854863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004463757544431285,
      "loss": 4.9922,
      "mean_token_accuracy": 0.21916391402482988,
      "num_tokens": 52926698.0,
      "step": 25510
    },
    {
      "entropy": 5.63509840965271,
      "epoch": 2.3144956458635706,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004463547868401044,
      "loss": 4.9287,
      "mean_token_accuracy": 0.21260821223258972,
      "num_tokens": 52936469.0,
      "step": 25515
    },
    {
      "entropy": 5.586462688446045,
      "epoch": 2.3149492017416544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044633381569341964,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21342955082654952,
      "num_tokens": 52947006.0,
      "step": 25520
    },
    {
      "entropy": 5.454700040817261,
      "epoch": 2.3154027576197387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004463128410035079,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22044927030801773,
      "num_tokens": 52956663.0,
      "step": 25525
    },
    {
      "entropy": 5.531123924255371,
      "epoch": 2.315856313497823,
      "grad_norm": 1.125,
      "learning_rate": 0.00044629186277080307,
      "loss": 4.867,
      "mean_token_accuracy": 0.21318848580121993,
      "num_tokens": 52967383.0,
      "step": 25530
    },
    {
      "entropy": 5.5405303001403805,
      "epoch": 2.316309869375907,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044627088099573884,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22379258126020432,
      "num_tokens": 52976567.0,
      "step": 25535
    },
    {
      "entropy": 5.531987237930298,
      "epoch": 2.3167634252539915,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044624989567874907,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21205026507377625,
      "num_tokens": 52986159.0,
      "step": 25540
    },
    {
      "entropy": 5.55957145690918,
      "epoch": 2.3172169811320753,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044622890682026787,
      "loss": 4.9794,
      "mean_token_accuracy": 0.2064168006181717,
      "num_tokens": 52996028.0,
      "step": 25545
    },
    {
      "entropy": 5.5923529148101805,
      "epoch": 2.3176705370101596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004462079144207292,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2132789224386215,
      "num_tokens": 53006446.0,
      "step": 25550
    },
    {
      "entropy": 5.602336072921753,
      "epoch": 2.318124092888244,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044618691848056726,
      "loss": 4.8853,
      "mean_token_accuracy": 0.2120554581284523,
      "num_tokens": 53016497.0,
      "step": 25555
    },
    {
      "entropy": 5.601254940032959,
      "epoch": 2.318577648766328,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004461659190002162,
      "loss": 4.9663,
      "mean_token_accuracy": 0.22072144597768784,
      "num_tokens": 53027811.0,
      "step": 25560
    },
    {
      "entropy": 5.559877109527588,
      "epoch": 2.3190312046444124,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044614491598011027,
      "loss": 4.9366,
      "mean_token_accuracy": 0.20997716784477233,
      "num_tokens": 53038206.0,
      "step": 25565
    },
    {
      "entropy": 5.6215955257415775,
      "epoch": 2.319484760522496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004461239094206839,
      "loss": 4.9746,
      "mean_token_accuracy": 0.2092909336090088,
      "num_tokens": 53047974.0,
      "step": 25570
    },
    {
      "entropy": 5.607485675811768,
      "epoch": 2.3199383164005805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044610289932237154,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21629616469144822,
      "num_tokens": 53057904.0,
      "step": 25575
    },
    {
      "entropy": 5.53427791595459,
      "epoch": 2.3203918722786647,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044608188568560756,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21096992045640944,
      "num_tokens": 53068447.0,
      "step": 25580
    },
    {
      "entropy": 5.529596662521362,
      "epoch": 2.320845428156749,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004460608685108266,
      "loss": 4.9769,
      "mean_token_accuracy": 0.20401107966899873,
      "num_tokens": 53079369.0,
      "step": 25585
    },
    {
      "entropy": 5.656600046157837,
      "epoch": 2.3212989840348333,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044603984779846334,
      "loss": 5.0023,
      "mean_token_accuracy": 0.20463142395019532,
      "num_tokens": 53090794.0,
      "step": 25590
    },
    {
      "entropy": 5.60491714477539,
      "epoch": 2.321752539912917,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004460188235489524,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21317342817783355,
      "num_tokens": 53100802.0,
      "step": 25595
    },
    {
      "entropy": 5.513609790802002,
      "epoch": 2.3222060957910013,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004459977957627285,
      "loss": 4.9601,
      "mean_token_accuracy": 0.21295360177755357,
      "num_tokens": 53111950.0,
      "step": 25600
    },
    {
      "entropy": 5.440905141830444,
      "epoch": 2.3226596516690856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044597676444022683,
      "loss": 4.7955,
      "mean_token_accuracy": 0.219007308781147,
      "num_tokens": 53121908.0,
      "step": 25605
    },
    {
      "entropy": 5.572998332977295,
      "epoch": 2.32311320754717,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044595572958188194,
      "loss": 4.932,
      "mean_token_accuracy": 0.2076426401734352,
      "num_tokens": 53133065.0,
      "step": 25610
    },
    {
      "entropy": 5.490257263183594,
      "epoch": 2.323566763425254,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004459346911881291,
      "loss": 4.8781,
      "mean_token_accuracy": 0.21220331639051437,
      "num_tokens": 53144135.0,
      "step": 25615
    },
    {
      "entropy": 5.536397409439087,
      "epoch": 2.3240203193033384,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004459136492594032,
      "loss": 4.854,
      "mean_token_accuracy": 0.22584492713212967,
      "num_tokens": 53153916.0,
      "step": 25620
    },
    {
      "entropy": 5.483391284942627,
      "epoch": 2.3244738751814222,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004458926037961394,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22567257136106492,
      "num_tokens": 53163937.0,
      "step": 25625
    },
    {
      "entropy": 5.606769371032715,
      "epoch": 2.3249274310595065,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004458715547987732,
      "loss": 5.0044,
      "mean_token_accuracy": 0.20156700760126114,
      "num_tokens": 53174327.0,
      "step": 25630
    },
    {
      "entropy": 5.593902730941773,
      "epoch": 2.3253809869375908,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004458505022677396,
      "loss": 4.9319,
      "mean_token_accuracy": 0.2109874963760376,
      "num_tokens": 53185946.0,
      "step": 25635
    },
    {
      "entropy": 5.570191860198975,
      "epoch": 2.325834542815675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458294462034741,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2161573588848114,
      "num_tokens": 53195944.0,
      "step": 25640
    },
    {
      "entropy": 5.511809539794922,
      "epoch": 2.326288098693759,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004458083866064121,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21828660666942595,
      "num_tokens": 53206054.0,
      "step": 25645
    },
    {
      "entropy": 5.565154647827148,
      "epoch": 2.326741654571843,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004457873234769891,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21072538793087006,
      "num_tokens": 53216923.0,
      "step": 25650
    },
    {
      "entropy": 5.5570258617401125,
      "epoch": 2.3271952104499274,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044576625681564086,
      "loss": 4.9217,
      "mean_token_accuracy": 0.2092050075531006,
      "num_tokens": 53226788.0,
      "step": 25655
    },
    {
      "entropy": 5.587557744979859,
      "epoch": 2.3276487663280117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044574518662280276,
      "loss": 4.978,
      "mean_token_accuracy": 0.20888438075780869,
      "num_tokens": 53237573.0,
      "step": 25660
    },
    {
      "entropy": 5.580534887313843,
      "epoch": 2.328102322206096,
      "grad_norm": 1.125,
      "learning_rate": 0.0004457241128989108,
      "loss": 4.9835,
      "mean_token_accuracy": 0.20926669090986252,
      "num_tokens": 53247724.0,
      "step": 25665
    },
    {
      "entropy": 5.597061061859131,
      "epoch": 2.32855587808418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004457030356444006,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21281372904777526,
      "num_tokens": 53258774.0,
      "step": 25670
    },
    {
      "entropy": 5.55437126159668,
      "epoch": 2.329009433962264,
      "grad_norm": 1.25,
      "learning_rate": 0.0004456819548597082,
      "loss": 4.9018,
      "mean_token_accuracy": 0.217744442820549,
      "num_tokens": 53268550.0,
      "step": 25675
    },
    {
      "entropy": 5.462180709838867,
      "epoch": 2.3294629898403483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044566087054526946,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22017797976732253,
      "num_tokens": 53277975.0,
      "step": 25680
    },
    {
      "entropy": 5.600870895385742,
      "epoch": 2.3299165457184325,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004456397827015204,
      "loss": 5.0166,
      "mean_token_accuracy": 0.20910515934228896,
      "num_tokens": 53287939.0,
      "step": 25685
    },
    {
      "entropy": 5.59006552696228,
      "epoch": 2.330370101596517,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004456186913288971,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2152164563536644,
      "num_tokens": 53298355.0,
      "step": 25690
    },
    {
      "entropy": 5.636488723754883,
      "epoch": 2.3308236574746006,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004455975964278359,
      "loss": 5.0295,
      "mean_token_accuracy": 0.20766967087984084,
      "num_tokens": 53308513.0,
      "step": 25695
    },
    {
      "entropy": 5.609822177886963,
      "epoch": 2.331277213352685,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044557649799877285,
      "loss": 4.9666,
      "mean_token_accuracy": 0.21177448630332946,
      "num_tokens": 53318249.0,
      "step": 25700
    },
    {
      "entropy": 5.512476062774658,
      "epoch": 2.331730769230769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004455553960421443,
      "loss": 4.8657,
      "mean_token_accuracy": 0.2197270154953003,
      "num_tokens": 53328460.0,
      "step": 25705
    },
    {
      "entropy": 5.54220986366272,
      "epoch": 2.3321843251088534,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044553429055838685,
      "loss": 4.9167,
      "mean_token_accuracy": 0.2191251263022423,
      "num_tokens": 53338984.0,
      "step": 25710
    },
    {
      "entropy": 5.60816764831543,
      "epoch": 2.3326378809869377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044551318154793676,
      "loss": 4.9828,
      "mean_token_accuracy": 0.20916682183742524,
      "num_tokens": 53348092.0,
      "step": 25715
    },
    {
      "entropy": 5.531657934188843,
      "epoch": 2.333091436865022,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044549206901123056,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2128957986831665,
      "num_tokens": 53357343.0,
      "step": 25720
    },
    {
      "entropy": 5.536874151229858,
      "epoch": 2.333544992743106,
      "grad_norm": 1.0625,
      "learning_rate": 0.000445470952948705,
      "loss": 4.8287,
      "mean_token_accuracy": 0.21804427355527878,
      "num_tokens": 53367343.0,
      "step": 25725
    },
    {
      "entropy": 5.6577342510223385,
      "epoch": 2.33399854862119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044544983336079665,
      "loss": 5.0246,
      "mean_token_accuracy": 0.21225915402173995,
      "num_tokens": 53379093.0,
      "step": 25730
    },
    {
      "entropy": 5.524349689483643,
      "epoch": 2.3344521044992743,
      "grad_norm": 1.125,
      "learning_rate": 0.0004454287102479423,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21798638701438905,
      "num_tokens": 53388883.0,
      "step": 25735
    },
    {
      "entropy": 5.543749475479126,
      "epoch": 2.3349056603773586,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004454075836105788,
      "loss": 4.902,
      "mean_token_accuracy": 0.22149984836578368,
      "num_tokens": 53399393.0,
      "step": 25740
    },
    {
      "entropy": 5.520349931716919,
      "epoch": 2.335359216255443,
      "grad_norm": 1.125,
      "learning_rate": 0.00044538645344914307,
      "loss": 4.863,
      "mean_token_accuracy": 0.22062693536281586,
      "num_tokens": 53408545.0,
      "step": 25745
    },
    {
      "entropy": 5.547723007202149,
      "epoch": 2.3358127721335267,
      "grad_norm": 1.125,
      "learning_rate": 0.0004453653197640721,
      "loss": 4.9615,
      "mean_token_accuracy": 0.2091108575463295,
      "num_tokens": 53418325.0,
      "step": 25750
    },
    {
      "entropy": 5.58025975227356,
      "epoch": 2.336266328011611,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044534418255580286,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20408245474100112,
      "num_tokens": 53427984.0,
      "step": 25755
    },
    {
      "entropy": 5.641548728942871,
      "epoch": 2.336719883889695,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044532304182477245,
      "loss": 5.0336,
      "mean_token_accuracy": 0.2045176386833191,
      "num_tokens": 53438765.0,
      "step": 25760
    },
    {
      "entropy": 5.623081302642822,
      "epoch": 2.3371734397677795,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004453018975714182,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21022448539733887,
      "num_tokens": 53448792.0,
      "step": 25765
    },
    {
      "entropy": 5.606881952285766,
      "epoch": 2.3376269956458637,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004452807497961773,
      "loss": 4.9799,
      "mean_token_accuracy": 0.20908829420804978,
      "num_tokens": 53459334.0,
      "step": 25770
    },
    {
      "entropy": 5.565575027465821,
      "epoch": 2.3380805515239476,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044525959849948704,
      "loss": 5.0009,
      "mean_token_accuracy": 0.2106160268187523,
      "num_tokens": 53469841.0,
      "step": 25775
    },
    {
      "entropy": 5.547070503234863,
      "epoch": 2.338534107402032,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044523844368178503,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21561982184648515,
      "num_tokens": 53480660.0,
      "step": 25780
    },
    {
      "entropy": 5.549220895767212,
      "epoch": 2.338987663280116,
      "grad_norm": 1.125,
      "learning_rate": 0.0004452172853435085,
      "loss": 4.8693,
      "mean_token_accuracy": 0.21835670620203018,
      "num_tokens": 53490276.0,
      "step": 25785
    },
    {
      "entropy": 5.463447093963623,
      "epoch": 2.3394412191582004,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004451961234850953,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2247002512216568,
      "num_tokens": 53501100.0,
      "step": 25790
    },
    {
      "entropy": 5.540241622924805,
      "epoch": 2.3398947750362846,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004451749581069827,
      "loss": 4.9202,
      "mean_token_accuracy": 0.22009168714284896,
      "num_tokens": 53511680.0,
      "step": 25795
    },
    {
      "entropy": 5.603099584579468,
      "epoch": 2.3403483309143684,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004451537892096087,
      "loss": 5.0091,
      "mean_token_accuracy": 0.21360567212104797,
      "num_tokens": 53521298.0,
      "step": 25800
    },
    {
      "entropy": 5.537745809555053,
      "epoch": 2.3408018867924527,
      "grad_norm": 1.25,
      "learning_rate": 0.00044513261679341094,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2255215972661972,
      "num_tokens": 53530413.0,
      "step": 25805
    },
    {
      "entropy": 5.5274241924285885,
      "epoch": 2.341255442670537,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004451114408588273,
      "loss": 4.9381,
      "mean_token_accuracy": 0.21218885481357574,
      "num_tokens": 53540959.0,
      "step": 25810
    },
    {
      "entropy": 5.555226993560791,
      "epoch": 2.3417089985486212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044509026140629584,
      "loss": 4.923,
      "mean_token_accuracy": 0.21374281197786332,
      "num_tokens": 53551265.0,
      "step": 25815
    },
    {
      "entropy": 5.60271372795105,
      "epoch": 2.3421625544267055,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044506907843625427,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20834640115499498,
      "num_tokens": 53562666.0,
      "step": 25820
    },
    {
      "entropy": 5.5669739723205565,
      "epoch": 2.34261611030479,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044504789194914086,
      "loss": 5.0123,
      "mean_token_accuracy": 0.2065434217453003,
      "num_tokens": 53573688.0,
      "step": 25825
    },
    {
      "entropy": 5.622451686859131,
      "epoch": 2.3430696661828736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044502670194539374,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2181478723883629,
      "num_tokens": 53584055.0,
      "step": 25830
    },
    {
      "entropy": 5.497566986083984,
      "epoch": 2.343523222060958,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044500550842545104,
      "loss": 4.7986,
      "mean_token_accuracy": 0.22547466605901717,
      "num_tokens": 53593765.0,
      "step": 25835
    },
    {
      "entropy": 5.479328250885009,
      "epoch": 2.343976777939042,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004449843113897512,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22054440379142762,
      "num_tokens": 53603480.0,
      "step": 25840
    },
    {
      "entropy": 5.546900081634521,
      "epoch": 2.3444303338171264,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004449631108387325,
      "loss": 4.9094,
      "mean_token_accuracy": 0.21567704677581787,
      "num_tokens": 53613797.0,
      "step": 25845
    },
    {
      "entropy": 5.545313835144043,
      "epoch": 2.34488388969521,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004449419067728333,
      "loss": 4.8708,
      "mean_token_accuracy": 0.21695394963026046,
      "num_tokens": 53623646.0,
      "step": 25850
    },
    {
      "entropy": 5.52366452217102,
      "epoch": 2.3453374455732945,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044492069919249206,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21340417563915254,
      "num_tokens": 53634118.0,
      "step": 25855
    },
    {
      "entropy": 5.562177038192749,
      "epoch": 2.3457910014513788,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004448994880981476,
      "loss": 4.9214,
      "mean_token_accuracy": 0.21526856273412703,
      "num_tokens": 53644370.0,
      "step": 25860
    },
    {
      "entropy": 5.581945323944092,
      "epoch": 2.346244557329463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044487827349023833,
      "loss": 4.9782,
      "mean_token_accuracy": 0.20599272847175598,
      "num_tokens": 53655560.0,
      "step": 25865
    },
    {
      "entropy": 5.579787492752075,
      "epoch": 2.3466981132075473,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044485705536920315,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21228385120630264,
      "num_tokens": 53666401.0,
      "step": 25870
    },
    {
      "entropy": 5.641883850097656,
      "epoch": 2.3471516690856316,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044483583373548075,
      "loss": 4.9829,
      "mean_token_accuracy": 0.2216608226299286,
      "num_tokens": 53676133.0,
      "step": 25875
    },
    {
      "entropy": 5.543575668334961,
      "epoch": 2.3476052249637154,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004448146085895101,
      "loss": 4.924,
      "mean_token_accuracy": 0.20757105499505996,
      "num_tokens": 53687683.0,
      "step": 25880
    },
    {
      "entropy": 5.5466375827789305,
      "epoch": 2.3480587808417996,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004447933799317299,
      "loss": 5.0061,
      "mean_token_accuracy": 0.20865627825260163,
      "num_tokens": 53698343.0,
      "step": 25885
    },
    {
      "entropy": 5.542923545837402,
      "epoch": 2.348512336719884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004447721477625795,
      "loss": 4.9051,
      "mean_token_accuracy": 0.213993664085865,
      "num_tokens": 53709014.0,
      "step": 25890
    },
    {
      "entropy": 5.51533784866333,
      "epoch": 2.348965892597968,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044475091208249774,
      "loss": 4.929,
      "mean_token_accuracy": 0.21108341068029404,
      "num_tokens": 53718838.0,
      "step": 25895
    },
    {
      "entropy": 5.512377691268921,
      "epoch": 2.3494194484760524,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004447296728919239,
      "loss": 4.8814,
      "mean_token_accuracy": 0.21453598737716675,
      "num_tokens": 53728526.0,
      "step": 25900
    },
    {
      "entropy": 5.618234300613404,
      "epoch": 2.3498730043541363,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004447084301912972,
      "loss": 5.0088,
      "mean_token_accuracy": 0.21235333681106566,
      "num_tokens": 53739261.0,
      "step": 25905
    },
    {
      "entropy": 5.603725099563599,
      "epoch": 2.3503265602322205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004446871839810568,
      "loss": 4.9735,
      "mean_token_accuracy": 0.21099812388420106,
      "num_tokens": 53749599.0,
      "step": 25910
    },
    {
      "entropy": 5.550251817703247,
      "epoch": 2.350780116110305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044466593426164224,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21090869903564452,
      "num_tokens": 53759898.0,
      "step": 25915
    },
    {
      "entropy": 5.5395654201507565,
      "epoch": 2.351233671988389,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000444644681033493,
      "loss": 4.8928,
      "mean_token_accuracy": 0.2077125996351242,
      "num_tokens": 53770561.0,
      "step": 25920
    },
    {
      "entropy": 5.575889825820923,
      "epoch": 2.3516872278664733,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044462342429704845,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21541345864534378,
      "num_tokens": 53780408.0,
      "step": 25925
    },
    {
      "entropy": 5.466110706329346,
      "epoch": 2.352140783744557,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044460216405274827,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22825224399566652,
      "num_tokens": 53789125.0,
      "step": 25930
    },
    {
      "entropy": 5.498211622238159,
      "epoch": 2.3525943396226414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044458090030103216,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2235909730195999,
      "num_tokens": 53799164.0,
      "step": 25935
    },
    {
      "entropy": 5.571869707107544,
      "epoch": 2.3530478955007257,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044455963304233976,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22502944022417068,
      "num_tokens": 53810710.0,
      "step": 25940
    },
    {
      "entropy": 5.525624465942383,
      "epoch": 2.35350145137881,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000444538362277111,
      "loss": 4.9047,
      "mean_token_accuracy": 0.20516117960214614,
      "num_tokens": 53820692.0,
      "step": 25945
    },
    {
      "entropy": 5.603377342224121,
      "epoch": 2.353955007256894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004445170880057857,
      "loss": 4.9691,
      "mean_token_accuracy": 0.2112611025571823,
      "num_tokens": 53831082.0,
      "step": 25950
    },
    {
      "entropy": 5.626092195510864,
      "epoch": 2.354408563134978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044449581022880375,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20191119909286498,
      "num_tokens": 53842602.0,
      "step": 25955
    },
    {
      "entropy": 5.542799186706543,
      "epoch": 2.3548621190130623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004444745289466053,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21708871573209762,
      "num_tokens": 53853784.0,
      "step": 25960
    },
    {
      "entropy": 5.502876472473145,
      "epoch": 2.3553156748911466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044445324415963043,
      "loss": 4.818,
      "mean_token_accuracy": 0.22221260517835617,
      "num_tokens": 53863526.0,
      "step": 25965
    },
    {
      "entropy": 5.510083913803101,
      "epoch": 2.355769230769231,
      "grad_norm": 1.125,
      "learning_rate": 0.0004444319558683192,
      "loss": 4.9415,
      "mean_token_accuracy": 0.2122773364186287,
      "num_tokens": 53874594.0,
      "step": 25970
    },
    {
      "entropy": 5.554979515075684,
      "epoch": 2.356222786647315,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044441066407311205,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21383181363344192,
      "num_tokens": 53884771.0,
      "step": 25975
    },
    {
      "entropy": 5.601384544372559,
      "epoch": 2.3566763425253994,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004443893687744492,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21509104669094087,
      "num_tokens": 53895181.0,
      "step": 25980
    },
    {
      "entropy": 5.522740364074707,
      "epoch": 2.357129898403483,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044436806997277096,
      "loss": 4.8068,
      "mean_token_accuracy": 0.230086250603199,
      "num_tokens": 53904153.0,
      "step": 25985
    },
    {
      "entropy": 5.509780073165894,
      "epoch": 2.3575834542815675,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044434676766851787,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21705929040908814,
      "num_tokens": 53914064.0,
      "step": 25990
    },
    {
      "entropy": 5.554558753967285,
      "epoch": 2.3580370101596517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044432546186213054,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21097599416971208,
      "num_tokens": 53924240.0,
      "step": 25995
    },
    {
      "entropy": 5.601577186584473,
      "epoch": 2.358490566037736,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044430415255404944,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21583618372678756,
      "num_tokens": 53935395.0,
      "step": 26000
    },
    {
      "entropy": 5.573638534545898,
      "epoch": 2.35894412191582,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044428283974471533,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21309202611446382,
      "num_tokens": 53945768.0,
      "step": 26005
    },
    {
      "entropy": 5.475731945037841,
      "epoch": 2.359397677793904,
      "grad_norm": 1.140625,
      "learning_rate": 0.000444261523434569,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21839204132556916,
      "num_tokens": 53956467.0,
      "step": 26010
    },
    {
      "entropy": 5.523874568939209,
      "epoch": 2.3598512336719883,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044424020362405126,
      "loss": 4.8475,
      "mean_token_accuracy": 0.21928653717041016,
      "num_tokens": 53967587.0,
      "step": 26015
    },
    {
      "entropy": 5.512750720977783,
      "epoch": 2.3603047895500726,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004442188803136029,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2134658306837082,
      "num_tokens": 53977691.0,
      "step": 26020
    },
    {
      "entropy": 5.51469316482544,
      "epoch": 2.360758345428157,
      "grad_norm": 1.203125,
      "learning_rate": 0.000444197553503665,
      "loss": 4.9677,
      "mean_token_accuracy": 0.21052328795194625,
      "num_tokens": 53987608.0,
      "step": 26025
    },
    {
      "entropy": 5.5814004898071286,
      "epoch": 2.361211901306241,
      "grad_norm": 1.125,
      "learning_rate": 0.00044417622319467857,
      "loss": 5.0379,
      "mean_token_accuracy": 0.20708208084106444,
      "num_tokens": 53998010.0,
      "step": 26030
    },
    {
      "entropy": 5.543911790847778,
      "epoch": 2.361665457184325,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044415488938708475,
      "loss": 4.9201,
      "mean_token_accuracy": 0.2219969943165779,
      "num_tokens": 54009852.0,
      "step": 26035
    },
    {
      "entropy": 5.559852886199951,
      "epoch": 2.3621190130624092,
      "grad_norm": 1.125,
      "learning_rate": 0.0004441335520813247,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2171434849500656,
      "num_tokens": 54019635.0,
      "step": 26040
    },
    {
      "entropy": 5.567590808868408,
      "epoch": 2.3625725689404935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044411221127783966,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2123323366045952,
      "num_tokens": 54030046.0,
      "step": 26045
    },
    {
      "entropy": 5.5861293315887455,
      "epoch": 2.3630261248185778,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004440908669770711,
      "loss": 4.9115,
      "mean_token_accuracy": 0.2105046182870865,
      "num_tokens": 54041197.0,
      "step": 26050
    },
    {
      "entropy": 5.574221849441528,
      "epoch": 2.3634796806966616,
      "grad_norm": 1.125,
      "learning_rate": 0.0004440695191794602,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22281113117933274,
      "num_tokens": 54051592.0,
      "step": 26055
    },
    {
      "entropy": 5.575102996826172,
      "epoch": 2.363933236574746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004440481678854487,
      "loss": 4.9742,
      "mean_token_accuracy": 0.2099858731031418,
      "num_tokens": 54062893.0,
      "step": 26060
    },
    {
      "entropy": 5.610869121551514,
      "epoch": 2.36438679245283,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044402681309547786,
      "loss": 4.9993,
      "mean_token_accuracy": 0.20897722840309144,
      "num_tokens": 54073651.0,
      "step": 26065
    },
    {
      "entropy": 5.523925399780273,
      "epoch": 2.3648403483309144,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044400545480998956,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21935094743967057,
      "num_tokens": 54084889.0,
      "step": 26070
    },
    {
      "entropy": 5.52763147354126,
      "epoch": 2.3652939042089987,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044398409302942533,
      "loss": 4.927,
      "mean_token_accuracy": 0.2120141163468361,
      "num_tokens": 54095724.0,
      "step": 26075
    },
    {
      "entropy": 5.552655458450317,
      "epoch": 2.365747460087083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000443962727754227,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2145445927977562,
      "num_tokens": 54106581.0,
      "step": 26080
    },
    {
      "entropy": 5.612347888946533,
      "epoch": 2.3662010159651667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004439413589848364,
      "loss": 4.9524,
      "mean_token_accuracy": 0.20964030176401138,
      "num_tokens": 54116074.0,
      "step": 26085
    },
    {
      "entropy": 5.559743595123291,
      "epoch": 2.366654571843251,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004439199867216955,
      "loss": 4.9502,
      "mean_token_accuracy": 0.20922233015298844,
      "num_tokens": 54126902.0,
      "step": 26090
    },
    {
      "entropy": 5.540619421005249,
      "epoch": 2.3671081277213353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004438986109652462,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22057385444641114,
      "num_tokens": 54137554.0,
      "step": 26095
    },
    {
      "entropy": 5.57423677444458,
      "epoch": 2.3675616835994195,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004438772317159306,
      "loss": 4.8839,
      "mean_token_accuracy": 0.21637672781944275,
      "num_tokens": 54148035.0,
      "step": 26100
    },
    {
      "entropy": 5.544586420059204,
      "epoch": 2.368015239477504,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044385584897419075,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22016146630048752,
      "num_tokens": 54157417.0,
      "step": 26105
    },
    {
      "entropy": 5.588238620758057,
      "epoch": 2.3684687953555876,
      "grad_norm": 1.125,
      "learning_rate": 0.000443834462740469,
      "loss": 4.9607,
      "mean_token_accuracy": 0.20928663909435272,
      "num_tokens": 54168353.0,
      "step": 26110
    },
    {
      "entropy": 5.541150522232056,
      "epoch": 2.368922351233672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004438130730152074,
      "loss": 4.9995,
      "mean_token_accuracy": 0.2050572693347931,
      "num_tokens": 54180182.0,
      "step": 26115
    },
    {
      "entropy": 5.597230339050293,
      "epoch": 2.369375907111756,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044379167979884847,
      "loss": 4.953,
      "mean_token_accuracy": 0.2137366935610771,
      "num_tokens": 54190973.0,
      "step": 26120
    },
    {
      "entropy": 5.62618317604065,
      "epoch": 2.3698294629898404,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004437702830918347,
      "loss": 5.0257,
      "mean_token_accuracy": 0.20662141144275664,
      "num_tokens": 54202077.0,
      "step": 26125
    },
    {
      "entropy": 5.5631670475006105,
      "epoch": 2.3702830188679247,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004437488828946084,
      "loss": 4.935,
      "mean_token_accuracy": 0.2112831950187683,
      "num_tokens": 54211778.0,
      "step": 26130
    },
    {
      "entropy": 5.549208068847657,
      "epoch": 2.3707365747460085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004437274792076122,
      "loss": 4.8006,
      "mean_token_accuracy": 0.2305554285645485,
      "num_tokens": 54220941.0,
      "step": 26135
    },
    {
      "entropy": 5.578075695037842,
      "epoch": 2.371190130624093,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044370607203128874,
      "loss": 5.0191,
      "mean_token_accuracy": 0.20727423280477525,
      "num_tokens": 54231602.0,
      "step": 26140
    },
    {
      "entropy": 5.563685321807862,
      "epoch": 2.371643686502177,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004436846613660807,
      "loss": 4.9141,
      "mean_token_accuracy": 0.21764083951711655,
      "num_tokens": 54241797.0,
      "step": 26145
    },
    {
      "entropy": 5.6177104949951175,
      "epoch": 2.3720972423802613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004436632472124308,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21078586876392363,
      "num_tokens": 54253085.0,
      "step": 26150
    },
    {
      "entropy": 5.635371589660645,
      "epoch": 2.3725507982583456,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004436418295707821,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21754845827817917,
      "num_tokens": 54262164.0,
      "step": 26155
    },
    {
      "entropy": 5.568545722961426,
      "epoch": 2.3730043541364294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004436204084415774,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2107449308037758,
      "num_tokens": 54271922.0,
      "step": 26160
    },
    {
      "entropy": 5.5746949195861815,
      "epoch": 2.3734579100145137,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435989838252596,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22012820839881897,
      "num_tokens": 54281738.0,
      "step": 26165
    },
    {
      "entropy": 5.546205186843872,
      "epoch": 2.373911465892598,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004435775557222719,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21573433727025987,
      "num_tokens": 54292252.0,
      "step": 26170
    },
    {
      "entropy": 5.557355642318726,
      "epoch": 2.374365021770682,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044355612413305736,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21355361491441727,
      "num_tokens": 54302518.0,
      "step": 26175
    },
    {
      "entropy": 5.578246688842773,
      "epoch": 2.3748185776487665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435346890580593,
      "loss": 4.9379,
      "mean_token_accuracy": 0.22081943601369858,
      "num_tokens": 54312917.0,
      "step": 26180
    },
    {
      "entropy": 5.600694990158081,
      "epoch": 2.3752721335268507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044351325049772085,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22004953026771545,
      "num_tokens": 54322552.0,
      "step": 26185
    },
    {
      "entropy": 5.5750819683074955,
      "epoch": 2.3757256894049346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044349180845248536,
      "loss": 4.9362,
      "mean_token_accuracy": 0.2173297792673111,
      "num_tokens": 54332917.0,
      "step": 26190
    },
    {
      "entropy": 5.504472875595093,
      "epoch": 2.376179245283019,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004434703629227964,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2206265300512314,
      "num_tokens": 54342791.0,
      "step": 26195
    },
    {
      "entropy": 5.504059219360352,
      "epoch": 2.376632801161103,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004434489139090975,
      "loss": 4.7965,
      "mean_token_accuracy": 0.22644730061292648,
      "num_tokens": 54352821.0,
      "step": 26200
    },
    {
      "entropy": 5.650099897384644,
      "epoch": 2.3770863570391874,
      "grad_norm": 1.125,
      "learning_rate": 0.00044342746141183206,
      "loss": 5.0003,
      "mean_token_accuracy": 0.2065485179424286,
      "num_tokens": 54363095.0,
      "step": 26205
    },
    {
      "entropy": 5.55801420211792,
      "epoch": 2.377539912917271,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004434060054314438,
      "loss": 4.8983,
      "mean_token_accuracy": 0.20976397395133972,
      "num_tokens": 54373841.0,
      "step": 26210
    },
    {
      "entropy": 5.496986341476441,
      "epoch": 2.3779934687953554,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044338454596837637,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21487009525299072,
      "num_tokens": 54385012.0,
      "step": 26215
    },
    {
      "entropy": 5.545676565170288,
      "epoch": 2.3784470246734397,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004433630830230737,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21470629274845124,
      "num_tokens": 54395298.0,
      "step": 26220
    },
    {
      "entropy": 5.532968187332154,
      "epoch": 2.378900580551524,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004433416165959795,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2103761076927185,
      "num_tokens": 54405752.0,
      "step": 26225
    },
    {
      "entropy": 5.585411787033081,
      "epoch": 2.3793541364296082,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004433201466875378,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21393432915210725,
      "num_tokens": 54416607.0,
      "step": 26230
    },
    {
      "entropy": 5.589664602279663,
      "epoch": 2.3798076923076925,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004432986732981926,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21721350252628327,
      "num_tokens": 54426873.0,
      "step": 26235
    },
    {
      "entropy": 5.527328586578369,
      "epoch": 2.3802612481857763,
      "grad_norm": 1.0625,
      "learning_rate": 0.000443277196428388,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22448039352893828,
      "num_tokens": 54437611.0,
      "step": 26240
    },
    {
      "entropy": 5.617331075668335,
      "epoch": 2.3807148040638606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000443255716078568,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21499357223510743,
      "num_tokens": 54448992.0,
      "step": 26245
    },
    {
      "entropy": 5.553133487701416,
      "epoch": 2.381168359941945,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004432342322491769,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2211564749479294,
      "num_tokens": 54459172.0,
      "step": 26250
    },
    {
      "entropy": 5.494709539413452,
      "epoch": 2.381621915820029,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044321274494065903,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2159615322947502,
      "num_tokens": 54469660.0,
      "step": 26255
    },
    {
      "entropy": 5.593491315841675,
      "epoch": 2.3820754716981134,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044319125415345886,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20835172683000563,
      "num_tokens": 54479567.0,
      "step": 26260
    },
    {
      "entropy": 5.565223503112793,
      "epoch": 2.3825290275761972,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004431697598880205,
      "loss": 4.9727,
      "mean_token_accuracy": 0.21185938268899918,
      "num_tokens": 54489690.0,
      "step": 26265
    },
    {
      "entropy": 5.580861616134643,
      "epoch": 2.3829825834542815,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044314826214478877,
      "loss": 4.936,
      "mean_token_accuracy": 0.21563651412725449,
      "num_tokens": 54500876.0,
      "step": 26270
    },
    {
      "entropy": 5.582981395721435,
      "epoch": 2.3834361393323658,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044312676092420804,
      "loss": 4.976,
      "mean_token_accuracy": 0.20720279216766357,
      "num_tokens": 54510988.0,
      "step": 26275
    },
    {
      "entropy": 5.542959880828858,
      "epoch": 2.38388969521045,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004431052562267231,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22552803605794908,
      "num_tokens": 54522149.0,
      "step": 26280
    },
    {
      "entropy": 5.624943256378174,
      "epoch": 2.3843432510885343,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044308374805277867,
      "loss": 5.0531,
      "mean_token_accuracy": 0.20275964438915253,
      "num_tokens": 54533482.0,
      "step": 26285
    },
    {
      "entropy": 5.594573068618774,
      "epoch": 2.384796806966618,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004430622364028194,
      "loss": 4.9624,
      "mean_token_accuracy": 0.2087162896990776,
      "num_tokens": 54543058.0,
      "step": 26290
    },
    {
      "entropy": 5.600107145309448,
      "epoch": 2.3852503628447024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044304072127729035,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21142887473106384,
      "num_tokens": 54552655.0,
      "step": 26295
    },
    {
      "entropy": 5.535739231109619,
      "epoch": 2.3857039187227866,
      "grad_norm": 1.125,
      "learning_rate": 0.00044301920267663633,
      "loss": 4.8832,
      "mean_token_accuracy": 0.20965133607387543,
      "num_tokens": 54562696.0,
      "step": 26300
    },
    {
      "entropy": 5.563492202758789,
      "epoch": 2.386157474600871,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004429976806013024,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22042635530233384,
      "num_tokens": 54572803.0,
      "step": 26305
    },
    {
      "entropy": 5.51749997138977,
      "epoch": 2.386611030478955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004429761550517336,
      "loss": 4.9619,
      "mean_token_accuracy": 0.21359134763479232,
      "num_tokens": 54582450.0,
      "step": 26310
    },
    {
      "entropy": 5.457658576965332,
      "epoch": 2.387064586357039,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044295462602837504,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21416723728179932,
      "num_tokens": 54591868.0,
      "step": 26315
    },
    {
      "entropy": 5.571581697463989,
      "epoch": 2.3875181422351233,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004429330935316721,
      "loss": 4.893,
      "mean_token_accuracy": 0.2109416604042053,
      "num_tokens": 54601858.0,
      "step": 26320
    },
    {
      "entropy": 5.565107631683349,
      "epoch": 2.3879716981132075,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044291155756206996,
      "loss": 4.9541,
      "mean_token_accuracy": 0.21277913600206375,
      "num_tokens": 54612243.0,
      "step": 26325
    },
    {
      "entropy": 5.568923759460449,
      "epoch": 2.388425253991292,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044289001812001404,
      "loss": 4.8371,
      "mean_token_accuracy": 0.21800880283117294,
      "num_tokens": 54622365.0,
      "step": 26330
    },
    {
      "entropy": 5.552705001831055,
      "epoch": 2.388878809869376,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004428684752059497,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22188138961791992,
      "num_tokens": 54632356.0,
      "step": 26335
    },
    {
      "entropy": 5.614539003372192,
      "epoch": 2.3893323657474603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004428469288203226,
      "loss": 4.9541,
      "mean_token_accuracy": 0.20630268901586532,
      "num_tokens": 54642976.0,
      "step": 26340
    },
    {
      "entropy": 5.6432267189025875,
      "epoch": 2.389785921625544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004428253789635783,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21350667774677276,
      "num_tokens": 54652597.0,
      "step": 26345
    },
    {
      "entropy": 5.547775650024414,
      "epoch": 2.3902394775036284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004428038256361622,
      "loss": 5.063,
      "mean_token_accuracy": 0.20882029980421066,
      "num_tokens": 54662930.0,
      "step": 26350
    },
    {
      "entropy": 5.686775493621826,
      "epoch": 2.3906930333817127,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004427822688385203,
      "loss": 5.1626,
      "mean_token_accuracy": 0.20565766394138335,
      "num_tokens": 54674562.0,
      "step": 26355
    },
    {
      "entropy": 5.636586713790893,
      "epoch": 2.391146589259797,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004427607085710984,
      "loss": 4.9455,
      "mean_token_accuracy": 0.21146572828292848,
      "num_tokens": 54684432.0,
      "step": 26360
    },
    {
      "entropy": 5.644398021697998,
      "epoch": 2.3916001451378808,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044273914483434226,
      "loss": 4.9649,
      "mean_token_accuracy": 0.21229048073291779,
      "num_tokens": 54694060.0,
      "step": 26365
    },
    {
      "entropy": 5.552495098114013,
      "epoch": 2.392053701015965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044271757762869777,
      "loss": 4.957,
      "mean_token_accuracy": 0.2057665228843689,
      "num_tokens": 54704411.0,
      "step": 26370
    },
    {
      "entropy": 5.556673908233643,
      "epoch": 2.3925072568940493,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004426960069546112,
      "loss": 4.9555,
      "mean_token_accuracy": 0.21307219415903092,
      "num_tokens": 54715185.0,
      "step": 26375
    },
    {
      "entropy": 5.578215169906616,
      "epoch": 2.3929608127721336,
      "grad_norm": 1.125,
      "learning_rate": 0.0004426744328125284,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22331098318099976,
      "num_tokens": 54725529.0,
      "step": 26380
    },
    {
      "entropy": 5.670542526245117,
      "epoch": 2.393414368650218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004426528552028955,
      "loss": 4.9955,
      "mean_token_accuracy": 0.21452448666095733,
      "num_tokens": 54736992.0,
      "step": 26385
    },
    {
      "entropy": 5.561185073852539,
      "epoch": 2.393867924528302,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044263127412615897,
      "loss": 4.8931,
      "mean_token_accuracy": 0.2126631960272789,
      "num_tokens": 54746951.0,
      "step": 26390
    },
    {
      "entropy": 5.510678958892822,
      "epoch": 2.394321480406386,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004426096895827649,
      "loss": 4.8384,
      "mean_token_accuracy": 0.23029089570045472,
      "num_tokens": 54757272.0,
      "step": 26395
    },
    {
      "entropy": 5.561772012710572,
      "epoch": 2.39477503628447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044258810157315977,
      "loss": 4.9999,
      "mean_token_accuracy": 0.21058303415775298,
      "num_tokens": 54767835.0,
      "step": 26400
    },
    {
      "entropy": 5.563774251937867,
      "epoch": 2.3952285921625545,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044256651009779,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21124887615442275,
      "num_tokens": 54778432.0,
      "step": 26405
    },
    {
      "entropy": 5.624471235275268,
      "epoch": 2.3956821480406387,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044254491515710213,
      "loss": 4.9873,
      "mean_token_accuracy": 0.208717180788517,
      "num_tokens": 54788582.0,
      "step": 26410
    },
    {
      "entropy": 5.622468137741089,
      "epoch": 2.3961357039187225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004425233167515426,
      "loss": 4.9538,
      "mean_token_accuracy": 0.2182428553700447,
      "num_tokens": 54798848.0,
      "step": 26415
    },
    {
      "entropy": 5.541126251220703,
      "epoch": 2.396589259796807,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004425017148815583,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2144728347659111,
      "num_tokens": 54809867.0,
      "step": 26420
    },
    {
      "entropy": 5.556239938735962,
      "epoch": 2.397042815674891,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044248010954759575,
      "loss": 4.916,
      "mean_token_accuracy": 0.2128989115357399,
      "num_tokens": 54819905.0,
      "step": 26425
    },
    {
      "entropy": 5.593782711029053,
      "epoch": 2.3974963715529753,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044245850075010193,
      "loss": 4.935,
      "mean_token_accuracy": 0.2162592440843582,
      "num_tokens": 54829723.0,
      "step": 26430
    },
    {
      "entropy": 5.5636224269866945,
      "epoch": 2.3979499274310596,
      "grad_norm": 1.125,
      "learning_rate": 0.00044243688848952364,
      "loss": 4.864,
      "mean_token_accuracy": 0.22072971910238265,
      "num_tokens": 54840008.0,
      "step": 26435
    },
    {
      "entropy": 5.5568469047546385,
      "epoch": 2.398403483309144,
      "grad_norm": 1.125,
      "learning_rate": 0.0004424152727663078,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21008261740207673,
      "num_tokens": 54851693.0,
      "step": 26440
    },
    {
      "entropy": 5.591697311401367,
      "epoch": 2.3988570391872277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004423936535809015,
      "loss": 4.8933,
      "mean_token_accuracy": 0.21347566246986388,
      "num_tokens": 54862086.0,
      "step": 26445
    },
    {
      "entropy": 5.68751277923584,
      "epoch": 2.399310595065312,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004423720309337517,
      "loss": 5.0254,
      "mean_token_accuracy": 0.21116573065519334,
      "num_tokens": 54872883.0,
      "step": 26450
    },
    {
      "entropy": 5.627489137649536,
      "epoch": 2.3997641509433962,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044235040482530576,
      "loss": 4.9478,
      "mean_token_accuracy": 0.206145341694355,
      "num_tokens": 54882815.0,
      "step": 26455
    },
    {
      "entropy": 5.43229341506958,
      "epoch": 2.4002177068214805,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004423287752560108,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23027214258909226,
      "num_tokens": 54892480.0,
      "step": 26460
    },
    {
      "entropy": 5.458580017089844,
      "epoch": 2.4006712626995648,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044230714222631397,
      "loss": 4.8182,
      "mean_token_accuracy": 0.22742388248443604,
      "num_tokens": 54901607.0,
      "step": 26465
    },
    {
      "entropy": 5.636425542831421,
      "epoch": 2.4011248185776486,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044228550573666294,
      "loss": 5.0555,
      "mean_token_accuracy": 0.21088265925645827,
      "num_tokens": 54913545.0,
      "step": 26470
    },
    {
      "entropy": 5.614617252349854,
      "epoch": 2.401578374455733,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000442263865787505,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21196524500846864,
      "num_tokens": 54924392.0,
      "step": 26475
    },
    {
      "entropy": 5.59484543800354,
      "epoch": 2.402031930333817,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004422422223792877,
      "loss": 4.919,
      "mean_token_accuracy": 0.2162139818072319,
      "num_tokens": 54934114.0,
      "step": 26480
    },
    {
      "entropy": 5.662018775939941,
      "epoch": 2.4024854862119014,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004422205755124586,
      "loss": 4.943,
      "mean_token_accuracy": 0.21348754465579986,
      "num_tokens": 54944268.0,
      "step": 26485
    },
    {
      "entropy": 5.6119029998779295,
      "epoch": 2.4029390420899857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044219892518746536,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21220731884241104,
      "num_tokens": 54954877.0,
      "step": 26490
    },
    {
      "entropy": 5.561251306533814,
      "epoch": 2.4033925979680695,
      "grad_norm": 1.125,
      "learning_rate": 0.00044217727140475585,
      "loss": 4.898,
      "mean_token_accuracy": 0.2196974501013756,
      "num_tokens": 54964745.0,
      "step": 26495
    },
    {
      "entropy": 5.557149219512939,
      "epoch": 2.4038461538461537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044215561416477764,
      "loss": 4.8162,
      "mean_token_accuracy": 0.21659809201955796,
      "num_tokens": 54975256.0,
      "step": 26500
    },
    {
      "entropy": 5.623452854156494,
      "epoch": 2.404299709724238,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004421339534679788,
      "loss": 4.9743,
      "mean_token_accuracy": 0.20540614873170854,
      "num_tokens": 54986226.0,
      "step": 26505
    },
    {
      "entropy": 5.6022529125213625,
      "epoch": 2.4047532656023223,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004421122893148072,
      "loss": 4.9536,
      "mean_token_accuracy": 0.20172326266765594,
      "num_tokens": 54997350.0,
      "step": 26510
    },
    {
      "entropy": 5.595852947235107,
      "epoch": 2.4052068214804065,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044209062170571086,
      "loss": 4.9144,
      "mean_token_accuracy": 0.21679168492555617,
      "num_tokens": 55007947.0,
      "step": 26515
    },
    {
      "entropy": 5.590509462356567,
      "epoch": 2.4056603773584904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004420689506411379,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2100113198161125,
      "num_tokens": 55018789.0,
      "step": 26520
    },
    {
      "entropy": 5.575251626968384,
      "epoch": 2.4061139332365746,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044204727612153643,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21316459625959397,
      "num_tokens": 55028934.0,
      "step": 26525
    },
    {
      "entropy": 5.554839658737182,
      "epoch": 2.406567489114659,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044202559814735475,
      "loss": 4.9058,
      "mean_token_accuracy": 0.21916611790657042,
      "num_tokens": 55039395.0,
      "step": 26530
    },
    {
      "entropy": 5.517670392990112,
      "epoch": 2.407021044992743,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004420039167190411,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2145033672451973,
      "num_tokens": 55049932.0,
      "step": 26535
    },
    {
      "entropy": 5.600569534301758,
      "epoch": 2.4074746008708274,
      "grad_norm": 1.296875,
      "learning_rate": 0.000441982231837044,
      "loss": 4.968,
      "mean_token_accuracy": 0.20802204608917235,
      "num_tokens": 55061532.0,
      "step": 26540
    },
    {
      "entropy": 5.632411813735962,
      "epoch": 2.4079281567489117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004419605435018117,
      "loss": 4.9791,
      "mean_token_accuracy": 0.21000634878873825,
      "num_tokens": 55072368.0,
      "step": 26545
    },
    {
      "entropy": 5.572691774368286,
      "epoch": 2.4083817126269955,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004419388517137928,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21373072117567063,
      "num_tokens": 55083128.0,
      "step": 26550
    },
    {
      "entropy": 5.588914299011231,
      "epoch": 2.40883526850508,
      "grad_norm": 1.125,
      "learning_rate": 0.00044191715647343594,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21337571293115615,
      "num_tokens": 55094071.0,
      "step": 26555
    },
    {
      "entropy": 5.577487754821777,
      "epoch": 2.409288824383164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044189545778118973,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21225529313087463,
      "num_tokens": 55104504.0,
      "step": 26560
    },
    {
      "entropy": 5.57329421043396,
      "epoch": 2.4097423802612483,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000441873755637503,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21232262700796128,
      "num_tokens": 55115722.0,
      "step": 26565
    },
    {
      "entropy": 5.54393162727356,
      "epoch": 2.410195936139332,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004418520500428244,
      "loss": 4.889,
      "mean_token_accuracy": 0.2180961474776268,
      "num_tokens": 55126628.0,
      "step": 26570
    },
    {
      "entropy": 5.531755256652832,
      "epoch": 2.4106494920174164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044183034099760287,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2147435113787651,
      "num_tokens": 55137352.0,
      "step": 26575
    },
    {
      "entropy": 5.604616975784301,
      "epoch": 2.4111030478955007,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004418086285022874,
      "loss": 4.92,
      "mean_token_accuracy": 0.22164202779531478,
      "num_tokens": 55147145.0,
      "step": 26580
    },
    {
      "entropy": 5.615935277938843,
      "epoch": 2.411556603773585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044178691255732707,
      "loss": 4.9338,
      "mean_token_accuracy": 0.20993286967277527,
      "num_tokens": 55157386.0,
      "step": 26585
    },
    {
      "entropy": 5.605233764648437,
      "epoch": 2.412010159651669,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004417651931631708,
      "loss": 5.0123,
      "mean_token_accuracy": 0.20733411312103273,
      "num_tokens": 55168969.0,
      "step": 26590
    },
    {
      "entropy": 5.529330682754517,
      "epoch": 2.4124637155297535,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004417434703202678,
      "loss": 4.9292,
      "mean_token_accuracy": 0.20711630284786225,
      "num_tokens": 55180123.0,
      "step": 26595
    },
    {
      "entropy": 5.556660127639771,
      "epoch": 2.4129172714078373,
      "grad_norm": 1.125,
      "learning_rate": 0.0004417217440290673,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21950370371341704,
      "num_tokens": 55190668.0,
      "step": 26600
    },
    {
      "entropy": 5.53235445022583,
      "epoch": 2.4133708272859216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004417000142900188,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22001198530197144,
      "num_tokens": 55200852.0,
      "step": 26605
    },
    {
      "entropy": 5.670111274719238,
      "epoch": 2.413824383164006,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004416782811035715,
      "loss": 5.0529,
      "mean_token_accuracy": 0.20413116067647935,
      "num_tokens": 55211174.0,
      "step": 26610
    },
    {
      "entropy": 5.521115350723266,
      "epoch": 2.41427793904209,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044165654447017485,
      "loss": 4.8161,
      "mean_token_accuracy": 0.22323009967803956,
      "num_tokens": 55222042.0,
      "step": 26615
    },
    {
      "entropy": 5.574773693084717,
      "epoch": 2.4147314949201744,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044163480439027845,
      "loss": 4.9621,
      "mean_token_accuracy": 0.20403494834899902,
      "num_tokens": 55231688.0,
      "step": 26620
    },
    {
      "entropy": 5.57382287979126,
      "epoch": 2.415185050798258,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004416130608643317,
      "loss": 4.9644,
      "mean_token_accuracy": 0.2135877266526222,
      "num_tokens": 55241089.0,
      "step": 26625
    },
    {
      "entropy": 5.587157773971557,
      "epoch": 2.4156386066763424,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004415913138927844,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22084160894155502,
      "num_tokens": 55251511.0,
      "step": 26630
    },
    {
      "entropy": 5.528352546691894,
      "epoch": 2.4160921625544267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004415695634760864,
      "loss": 4.9023,
      "mean_token_accuracy": 0.20916433185338973,
      "num_tokens": 55262855.0,
      "step": 26635
    },
    {
      "entropy": 5.504735469818115,
      "epoch": 2.416545718432511,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004415478096146873,
      "loss": 4.8203,
      "mean_token_accuracy": 0.21695016771554948,
      "num_tokens": 55273035.0,
      "step": 26640
    },
    {
      "entropy": 5.550880479812622,
      "epoch": 2.4169992743105952,
      "grad_norm": 1.0625,
      "learning_rate": 0.000441526052309037,
      "loss": 4.9102,
      "mean_token_accuracy": 0.2157744586467743,
      "num_tokens": 55284033.0,
      "step": 26645
    },
    {
      "entropy": 5.551938724517822,
      "epoch": 2.417452830188679,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004415042915595856,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21973585337400436,
      "num_tokens": 55294290.0,
      "step": 26650
    },
    {
      "entropy": 5.609615612030029,
      "epoch": 2.4179063860667633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000441482527366783,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21230510920286177,
      "num_tokens": 55305681.0,
      "step": 26655
    },
    {
      "entropy": 5.552863883972168,
      "epoch": 2.4183599419448476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004414607597310793,
      "loss": 4.8873,
      "mean_token_accuracy": 0.2162024825811386,
      "num_tokens": 55316647.0,
      "step": 26660
    },
    {
      "entropy": 5.609606456756592,
      "epoch": 2.418813497822932,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004414389886529247,
      "loss": 5.038,
      "mean_token_accuracy": 0.21029458492994307,
      "num_tokens": 55328318.0,
      "step": 26665
    },
    {
      "entropy": 5.617603969573975,
      "epoch": 2.419267053701016,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004414172141327693,
      "loss": 4.9202,
      "mean_token_accuracy": 0.21053712368011473,
      "num_tokens": 55338483.0,
      "step": 26670
    },
    {
      "entropy": 5.59386944770813,
      "epoch": 2.4197206095791,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004413954361710636,
      "loss": 5.0069,
      "mean_token_accuracy": 0.210391066968441,
      "num_tokens": 55348648.0,
      "step": 26675
    },
    {
      "entropy": 5.559199810028076,
      "epoch": 2.4201741654571842,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044137365476825785,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21521667242050171,
      "num_tokens": 55358599.0,
      "step": 26680
    },
    {
      "entropy": 5.642380428314209,
      "epoch": 2.4206277213352685,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044135186992480253,
      "loss": 4.9475,
      "mean_token_accuracy": 0.21077101826667785,
      "num_tokens": 55369847.0,
      "step": 26685
    },
    {
      "entropy": 5.574283504486084,
      "epoch": 2.4210812772133528,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044133008164114814,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21662528067827225,
      "num_tokens": 55381235.0,
      "step": 26690
    },
    {
      "entropy": 5.569499158859253,
      "epoch": 2.421534833091437,
      "grad_norm": 1.125,
      "learning_rate": 0.00044130828991774526,
      "loss": 4.9692,
      "mean_token_accuracy": 0.20827213823795318,
      "num_tokens": 55391527.0,
      "step": 26695
    },
    {
      "entropy": 5.611489915847779,
      "epoch": 2.4219883889695213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044128649475504454,
      "loss": 4.9956,
      "mean_token_accuracy": 0.21032423824071883,
      "num_tokens": 55401528.0,
      "step": 26700
    },
    {
      "entropy": 5.6436731815338135,
      "epoch": 2.422441944847605,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004412646961534968,
      "loss": 4.9913,
      "mean_token_accuracy": 0.21021522879600524,
      "num_tokens": 55411411.0,
      "step": 26705
    },
    {
      "entropy": 5.514085721969605,
      "epoch": 2.4228955007256894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004412428941135528,
      "loss": 4.789,
      "mean_token_accuracy": 0.2239726349711418,
      "num_tokens": 55423164.0,
      "step": 26710
    },
    {
      "entropy": 5.533010196685791,
      "epoch": 2.4233490566037736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004412210886356633,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22164860367774963,
      "num_tokens": 55433397.0,
      "step": 26715
    },
    {
      "entropy": 5.569516324996949,
      "epoch": 2.423802612481858,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044119927972027936,
      "loss": 4.9263,
      "mean_token_accuracy": 0.20557917952537536,
      "num_tokens": 55442898.0,
      "step": 26720
    },
    {
      "entropy": 5.5873212814331055,
      "epoch": 2.4242561683599417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044117746736785195,
      "loss": 4.9274,
      "mean_token_accuracy": 0.21936164647340775,
      "num_tokens": 55452226.0,
      "step": 26725
    },
    {
      "entropy": 5.574273347854614,
      "epoch": 2.424709724238026,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004411556515788322,
      "loss": 4.9235,
      "mean_token_accuracy": 0.20912425816059113,
      "num_tokens": 55462503.0,
      "step": 26730
    },
    {
      "entropy": 5.621296453475952,
      "epoch": 2.4251632801161103,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004411338323536712,
      "loss": 5.0566,
      "mean_token_accuracy": 0.2071949914097786,
      "num_tokens": 55472789.0,
      "step": 26735
    },
    {
      "entropy": 5.5308976650238035,
      "epoch": 2.4256168359941945,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004411120096928203,
      "loss": 4.8572,
      "mean_token_accuracy": 0.21623512357473373,
      "num_tokens": 55482572.0,
      "step": 26740
    },
    {
      "entropy": 5.583556270599365,
      "epoch": 2.426070391872279,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004410901835967306,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22074851393699646,
      "num_tokens": 55493033.0,
      "step": 26745
    },
    {
      "entropy": 5.59099440574646,
      "epoch": 2.426523947750363,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044106835406585354,
      "loss": 4.8675,
      "mean_token_accuracy": 0.21514177620410918,
      "num_tokens": 55503574.0,
      "step": 26750
    },
    {
      "entropy": 5.59430160522461,
      "epoch": 2.426977503628447,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044104652110064067,
      "loss": 4.9698,
      "mean_token_accuracy": 0.2123052790760994,
      "num_tokens": 55514119.0,
      "step": 26755
    },
    {
      "entropy": 5.550724220275879,
      "epoch": 2.427431059506531,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004410246847015435,
      "loss": 4.889,
      "mean_token_accuracy": 0.21459876149892806,
      "num_tokens": 55524973.0,
      "step": 26760
    },
    {
      "entropy": 5.574378585815429,
      "epoch": 2.4278846153846154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004410028448690134,
      "loss": 4.9764,
      "mean_token_accuracy": 0.21177350878715515,
      "num_tokens": 55534847.0,
      "step": 26765
    },
    {
      "entropy": 5.622271871566772,
      "epoch": 2.4283381712626997,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004409810016035023,
      "loss": 4.9542,
      "mean_token_accuracy": 0.20416250973939895,
      "num_tokens": 55544886.0,
      "step": 26770
    },
    {
      "entropy": 5.604365062713623,
      "epoch": 2.4287917271407835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004409591549054617,
      "loss": 5.0031,
      "mean_token_accuracy": 0.2096606522798538,
      "num_tokens": 55554534.0,
      "step": 26775
    },
    {
      "entropy": 5.612155151367188,
      "epoch": 2.4292452830188678,
      "grad_norm": 1.125,
      "learning_rate": 0.00044093730477534355,
      "loss": 4.9791,
      "mean_token_accuracy": 0.20514402389526368,
      "num_tokens": 55564302.0,
      "step": 26780
    },
    {
      "entropy": 5.526300573348999,
      "epoch": 2.429698838896952,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004409154512135996,
      "loss": 4.888,
      "mean_token_accuracy": 0.21692268401384354,
      "num_tokens": 55574543.0,
      "step": 26785
    },
    {
      "entropy": 5.502074384689331,
      "epoch": 2.4301523947750363,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004408935942206818,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21633322387933732,
      "num_tokens": 55583849.0,
      "step": 26790
    },
    {
      "entropy": 5.558918476104736,
      "epoch": 2.4306059506531206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004408717337970423,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21295633018016816,
      "num_tokens": 55594294.0,
      "step": 26795
    },
    {
      "entropy": 5.591272354125977,
      "epoch": 2.431059506531205,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044084986994313304,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21233857423067093,
      "num_tokens": 55604185.0,
      "step": 26800
    },
    {
      "entropy": 5.554242134094238,
      "epoch": 2.4315130624092887,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004408280026594062,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21258280277252198,
      "num_tokens": 55615203.0,
      "step": 26805
    },
    {
      "entropy": 5.549846744537353,
      "epoch": 2.431966618287373,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000440806131946314,
      "loss": 4.9752,
      "mean_token_accuracy": 0.21261704415082933,
      "num_tokens": 55626157.0,
      "step": 26810
    },
    {
      "entropy": 5.629889440536499,
      "epoch": 2.432420174165457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044078425780430877,
      "loss": 4.938,
      "mean_token_accuracy": 0.20552310198545456,
      "num_tokens": 55636862.0,
      "step": 26815
    },
    {
      "entropy": 5.554861354827881,
      "epoch": 2.4328737300435415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004407623802338428,
      "loss": 4.8499,
      "mean_token_accuracy": 0.21850804090499878,
      "num_tokens": 55647646.0,
      "step": 26820
    },
    {
      "entropy": 5.560325670242309,
      "epoch": 2.4333272859216257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044074049923536855,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21901639997959138,
      "num_tokens": 55659847.0,
      "step": 26825
    },
    {
      "entropy": 5.506578254699707,
      "epoch": 2.4337808417997095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004407186148093386,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21475048065185548,
      "num_tokens": 55670626.0,
      "step": 26830
    },
    {
      "entropy": 5.575541162490845,
      "epoch": 2.434234397677794,
      "grad_norm": 1.125,
      "learning_rate": 0.00044069672695620543,
      "loss": 4.991,
      "mean_token_accuracy": 0.21187302321195603,
      "num_tokens": 55681368.0,
      "step": 26835
    },
    {
      "entropy": 5.546980142593384,
      "epoch": 2.434687953555878,
      "grad_norm": 1.125,
      "learning_rate": 0.0004406748356764217,
      "loss": 4.8672,
      "mean_token_accuracy": 0.21301209777593613,
      "num_tokens": 55691606.0,
      "step": 26840
    },
    {
      "entropy": 5.584707355499267,
      "epoch": 2.4351415094339623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044065294097044017,
      "loss": 5.0394,
      "mean_token_accuracy": 0.2070117712020874,
      "num_tokens": 55702223.0,
      "step": 26845
    },
    {
      "entropy": 5.522512340545655,
      "epoch": 2.4355950653120466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004406310428387136,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2171640768647194,
      "num_tokens": 55712729.0,
      "step": 26850
    },
    {
      "entropy": 5.616864824295044,
      "epoch": 2.4360486211901304,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004406091412816949,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2157175973057747,
      "num_tokens": 55722863.0,
      "step": 26855
    },
    {
      "entropy": 5.630240869522095,
      "epoch": 2.4365021770682147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044058723629983694,
      "loss": 5.0105,
      "mean_token_accuracy": 0.214661405980587,
      "num_tokens": 55733124.0,
      "step": 26860
    },
    {
      "entropy": 5.607588434219361,
      "epoch": 2.436955732946299,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004405653278935927,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21800507754087448,
      "num_tokens": 55743438.0,
      "step": 26865
    },
    {
      "entropy": 5.4981067180633545,
      "epoch": 2.4374092888243832,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044054341606341533,
      "loss": 4.8243,
      "mean_token_accuracy": 0.21509604901075363,
      "num_tokens": 55753602.0,
      "step": 26870
    },
    {
      "entropy": 5.501210069656372,
      "epoch": 2.4378628447024675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004405215008097579,
      "loss": 4.8758,
      "mean_token_accuracy": 0.2248426780104637,
      "num_tokens": 55763473.0,
      "step": 26875
    },
    {
      "entropy": 5.564803123474121,
      "epoch": 2.4383164005805513,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044049958213307364,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21540975868701934,
      "num_tokens": 55772848.0,
      "step": 26880
    },
    {
      "entropy": 5.517916631698609,
      "epoch": 2.4387699564586356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044047766003381584,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22073815315961837,
      "num_tokens": 55783468.0,
      "step": 26885
    },
    {
      "entropy": 5.547765111923217,
      "epoch": 2.43922351233672,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044045573451243793,
      "loss": 4.9617,
      "mean_token_accuracy": 0.21874792426824569,
      "num_tokens": 55793619.0,
      "step": 26890
    },
    {
      "entropy": 5.568424558639526,
      "epoch": 2.439677068214804,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044043380556939326,
      "loss": 4.9314,
      "mean_token_accuracy": 0.21112943142652513,
      "num_tokens": 55803869.0,
      "step": 26895
    },
    {
      "entropy": 5.606073093414307,
      "epoch": 2.4401306240928884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004404118732051353,
      "loss": 4.9389,
      "mean_token_accuracy": 0.2127373695373535,
      "num_tokens": 55814754.0,
      "step": 26900
    },
    {
      "entropy": 5.544326639175415,
      "epoch": 2.4405841799709727,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004403899374201176,
      "loss": 4.8973,
      "mean_token_accuracy": 0.21217919439077376,
      "num_tokens": 55824862.0,
      "step": 26905
    },
    {
      "entropy": 5.497791385650634,
      "epoch": 2.4410377358490565,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044036799821479397,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21363878399133682,
      "num_tokens": 55835506.0,
      "step": 26910
    },
    {
      "entropy": 5.630941486358642,
      "epoch": 2.4414912917271407,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044034605558961786,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2071603938937187,
      "num_tokens": 55846244.0,
      "step": 26915
    },
    {
      "entropy": 5.54923152923584,
      "epoch": 2.441944847605225,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044032410954504325,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22099509090185165,
      "num_tokens": 55856192.0,
      "step": 26920
    },
    {
      "entropy": 5.554870080947876,
      "epoch": 2.4423984034833093,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044030216008152394,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2092841923236847,
      "num_tokens": 55866509.0,
      "step": 26925
    },
    {
      "entropy": 5.566070508956909,
      "epoch": 2.442851959361393,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044028020719951376,
      "loss": 4.9508,
      "mean_token_accuracy": 0.2158271536231041,
      "num_tokens": 55877321.0,
      "step": 26930
    },
    {
      "entropy": 5.582284832000733,
      "epoch": 2.4433055152394774,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004402582508994669,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21382608711719514,
      "num_tokens": 55887933.0,
      "step": 26935
    },
    {
      "entropy": 5.5485701084136965,
      "epoch": 2.4437590711175616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044023629118183726,
      "loss": 4.8666,
      "mean_token_accuracy": 0.21526138633489608,
      "num_tokens": 55897274.0,
      "step": 26940
    },
    {
      "entropy": 5.602759599685669,
      "epoch": 2.444212626995646,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044021432804707895,
      "loss": 5.0143,
      "mean_token_accuracy": 0.2082632750272751,
      "num_tokens": 55907393.0,
      "step": 26945
    },
    {
      "entropy": 5.559384059906006,
      "epoch": 2.44466618287373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004401923614956463,
      "loss": 4.9076,
      "mean_token_accuracy": 0.21300158500671387,
      "num_tokens": 55918429.0,
      "step": 26950
    },
    {
      "entropy": 5.570344924926758,
      "epoch": 2.4451197387518144,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004401703915279935,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2174890011548996,
      "num_tokens": 55928733.0,
      "step": 26955
    },
    {
      "entropy": 5.599796772003174,
      "epoch": 2.4455732946298983,
      "grad_norm": 1.125,
      "learning_rate": 0.00044014841814457486,
      "loss": 4.8779,
      "mean_token_accuracy": 0.21433673202991485,
      "num_tokens": 55939128.0,
      "step": 26960
    },
    {
      "entropy": 5.59528694152832,
      "epoch": 2.4460268505079825,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004401264413458449,
      "loss": 5.0381,
      "mean_token_accuracy": 0.20667206943035127,
      "num_tokens": 55950237.0,
      "step": 26965
    },
    {
      "entropy": 5.5890590190887455,
      "epoch": 2.446480406386067,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004401044611322581,
      "loss": 4.8559,
      "mean_token_accuracy": 0.21903896182775498,
      "num_tokens": 55960436.0,
      "step": 26970
    },
    {
      "entropy": 5.528964996337891,
      "epoch": 2.446933962264151,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004400824775042689,
      "loss": 4.8225,
      "mean_token_accuracy": 0.21980970054864885,
      "num_tokens": 55970446.0,
      "step": 26975
    },
    {
      "entropy": 5.5585630416870115,
      "epoch": 2.447387518142235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000440060490462332,
      "loss": 4.9502,
      "mean_token_accuracy": 0.21068621873855592,
      "num_tokens": 55980706.0,
      "step": 26980
    },
    {
      "entropy": 5.56493034362793,
      "epoch": 2.447841074020319,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004400385000069021,
      "loss": 4.9972,
      "mean_token_accuracy": 0.20877644121646882,
      "num_tokens": 55990669.0,
      "step": 26985
    },
    {
      "entropy": 5.672916507720947,
      "epoch": 2.4482946298984034,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044001650613843395,
      "loss": 5.0034,
      "mean_token_accuracy": 0.20895944237709047,
      "num_tokens": 56000709.0,
      "step": 26990
    },
    {
      "entropy": 5.607293367385864,
      "epoch": 2.4487481857764877,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004399945088573824,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22434985190629958,
      "num_tokens": 56011435.0,
      "step": 26995
    },
    {
      "entropy": 5.561026859283447,
      "epoch": 2.449201741654572,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043997250816420246,
      "loss": 4.9143,
      "mean_token_accuracy": 0.22065363824367523,
      "num_tokens": 56021264.0,
      "step": 27000
    },
    {
      "epoch": 2.449201741654572,
      "eval_entropy": 5.289766723537877,
      "eval_loss": 5.00201416015625,
      "eval_mean_token_accuracy": 0.21932787828956524,
      "eval_num_tokens": 56021264.0,
      "eval_runtime": 20.4923,
      "eval_samples_per_second": 1725.527,
      "eval_steps_per_second": 215.691,
      "step": 27000
    },
    {
      "entropy": 5.54692530632019,
      "epoch": 2.449655297532656,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004399505040593489,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2112852618098259,
      "num_tokens": 56031463.0,
      "step": 27005
    },
    {
      "entropy": 5.536106204986572,
      "epoch": 2.45010885341074,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043992849654327695,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22149303704500198,
      "num_tokens": 56041195.0,
      "step": 27010
    },
    {
      "entropy": 5.572748851776123,
      "epoch": 2.4505624092888243,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004399064856164416,
      "loss": 4.9596,
      "mean_token_accuracy": 0.213093401491642,
      "num_tokens": 56051840.0,
      "step": 27015
    },
    {
      "entropy": 5.56742491722107,
      "epoch": 2.4510159651669086,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043988447127929817,
      "loss": 4.9862,
      "mean_token_accuracy": 0.21002749651670455,
      "num_tokens": 56062535.0,
      "step": 27020
    },
    {
      "entropy": 5.581688594818115,
      "epoch": 2.451469521044993,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043986245353230173,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22294532358646393,
      "num_tokens": 56072177.0,
      "step": 27025
    },
    {
      "entropy": 5.552148294448853,
      "epoch": 2.451923076923077,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043984043237590785,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21574793756008148,
      "num_tokens": 56081607.0,
      "step": 27030
    },
    {
      "entropy": 5.577052927017212,
      "epoch": 2.452376632801161,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043981840781057185,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21317142099142075,
      "num_tokens": 56093371.0,
      "step": 27035
    },
    {
      "entropy": 5.57371940612793,
      "epoch": 2.452830188679245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043979637983674913,
      "loss": 4.9295,
      "mean_token_accuracy": 0.21297799348831176,
      "num_tokens": 56103973.0,
      "step": 27040
    },
    {
      "entropy": 5.545422887802124,
      "epoch": 2.4532837445573294,
      "grad_norm": 1.0,
      "learning_rate": 0.0004397743484548953,
      "loss": 4.9261,
      "mean_token_accuracy": 0.2118041381239891,
      "num_tokens": 56113906.0,
      "step": 27045
    },
    {
      "entropy": 5.561880779266358,
      "epoch": 2.4537373004354137,
      "grad_norm": 1.125,
      "learning_rate": 0.0004397523136654659,
      "loss": 4.9131,
      "mean_token_accuracy": 0.21048663556575775,
      "num_tokens": 56124172.0,
      "step": 27050
    },
    {
      "entropy": 5.566390943527222,
      "epoch": 2.454190856313498,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043973027546891677,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2262653574347496,
      "num_tokens": 56133088.0,
      "step": 27055
    },
    {
      "entropy": 5.57589111328125,
      "epoch": 2.4546444121915822,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004397082338657034,
      "loss": 4.893,
      "mean_token_accuracy": 0.2158320352435112,
      "num_tokens": 56143098.0,
      "step": 27060
    },
    {
      "entropy": 5.537955331802368,
      "epoch": 2.455097968069666,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043968618885628185,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22055192589759826,
      "num_tokens": 56153889.0,
      "step": 27065
    },
    {
      "entropy": 5.523283624649048,
      "epoch": 2.4555515239477503,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043966414044110805,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2203761950135231,
      "num_tokens": 56164164.0,
      "step": 27070
    },
    {
      "entropy": 5.609129858016968,
      "epoch": 2.4560050798258346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043964208862063775,
      "loss": 4.9666,
      "mean_token_accuracy": 0.20902420133352279,
      "num_tokens": 56173172.0,
      "step": 27075
    },
    {
      "entropy": 5.5944620132446286,
      "epoch": 2.456458635703919,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043962003339532714,
      "loss": 4.9185,
      "mean_token_accuracy": 0.21544283479452134,
      "num_tokens": 56183127.0,
      "step": 27080
    },
    {
      "entropy": 5.589930009841919,
      "epoch": 2.4569121915820027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043959797476563226,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21085061132907867,
      "num_tokens": 56192528.0,
      "step": 27085
    },
    {
      "entropy": 5.5305253028869625,
      "epoch": 2.457365747460087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043957591273200943,
      "loss": 4.904,
      "mean_token_accuracy": 0.21747169494628907,
      "num_tokens": 56203239.0,
      "step": 27090
    },
    {
      "entropy": 5.595845031738281,
      "epoch": 2.4578193033381712,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004395538472949146,
      "loss": 4.9568,
      "mean_token_accuracy": 0.20946492105722428,
      "num_tokens": 56214068.0,
      "step": 27095
    },
    {
      "entropy": 5.637240266799926,
      "epoch": 2.4582728592162555,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004395317784548045,
      "loss": 4.9673,
      "mean_token_accuracy": 0.20381802022457124,
      "num_tokens": 56224829.0,
      "step": 27100
    },
    {
      "entropy": 5.686540317535401,
      "epoch": 2.4587264150943398,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004395097062121351,
      "loss": 4.9694,
      "mean_token_accuracy": 0.20938710272312164,
      "num_tokens": 56235412.0,
      "step": 27105
    },
    {
      "entropy": 5.634702491760254,
      "epoch": 2.459179970972424,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043948763056736315,
      "loss": 5.0101,
      "mean_token_accuracy": 0.20617360025644302,
      "num_tokens": 56246785.0,
      "step": 27110
    },
    {
      "entropy": 5.575610113143921,
      "epoch": 2.459633526850508,
      "grad_norm": 1.09375,
      "learning_rate": 0.000439465551520945,
      "loss": 4.98,
      "mean_token_accuracy": 0.20847520381212234,
      "num_tokens": 56256701.0,
      "step": 27115
    },
    {
      "entropy": 5.551024341583252,
      "epoch": 2.460087082728592,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043944346907333744,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21718012243509294,
      "num_tokens": 56267418.0,
      "step": 27120
    },
    {
      "entropy": 5.529728555679322,
      "epoch": 2.4605406386066764,
      "grad_norm": 1.0625,
      "learning_rate": 0.000439421383224997,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21338353157043458,
      "num_tokens": 56278232.0,
      "step": 27125
    },
    {
      "entropy": 5.6256965637207035,
      "epoch": 2.4609941944847606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004393992939763804,
      "loss": 4.9442,
      "mean_token_accuracy": 0.2154671236872673,
      "num_tokens": 56289866.0,
      "step": 27130
    },
    {
      "entropy": 5.578684329986572,
      "epoch": 2.4614477503628445,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043937720132794457,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21840518116950988,
      "num_tokens": 56299595.0,
      "step": 27135
    },
    {
      "entropy": 5.611116170883179,
      "epoch": 2.4619013062409287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043935510528014625,
      "loss": 4.9751,
      "mean_token_accuracy": 0.21603075712919234,
      "num_tokens": 56310152.0,
      "step": 27140
    },
    {
      "entropy": 5.590248346328735,
      "epoch": 2.462354862119013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004393330058334426,
      "loss": 5.029,
      "mean_token_accuracy": 0.2056109055876732,
      "num_tokens": 56320288.0,
      "step": 27145
    },
    {
      "entropy": 5.621517086029053,
      "epoch": 2.4628084179970973,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043931090298829047,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21012260913848876,
      "num_tokens": 56330345.0,
      "step": 27150
    },
    {
      "entropy": 5.669356870651245,
      "epoch": 2.4632619738751815,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043928879674514694,
      "loss": 5.0769,
      "mean_token_accuracy": 0.2166757345199585,
      "num_tokens": 56340996.0,
      "step": 27155
    },
    {
      "entropy": 5.477344942092896,
      "epoch": 2.463715529753266,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004392666871044692,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22312122732400894,
      "num_tokens": 56350326.0,
      "step": 27160
    },
    {
      "entropy": 5.546573209762573,
      "epoch": 2.4641690856313496,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043924457406671456,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22116460502147675,
      "num_tokens": 56360556.0,
      "step": 27165
    },
    {
      "entropy": 5.630509376525879,
      "epoch": 2.464622641509434,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004392224576323403,
      "loss": 4.887,
      "mean_token_accuracy": 0.21700661778450012,
      "num_tokens": 56369980.0,
      "step": 27170
    },
    {
      "entropy": 5.607940340042115,
      "epoch": 2.465076197387518,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043920033780180367,
      "loss": 4.942,
      "mean_token_accuracy": 0.20928603708744048,
      "num_tokens": 56379974.0,
      "step": 27175
    },
    {
      "entropy": 5.587023639678955,
      "epoch": 2.4655297532656024,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004391782145755623,
      "loss": 4.88,
      "mean_token_accuracy": 0.2152649074792862,
      "num_tokens": 56390542.0,
      "step": 27180
    },
    {
      "entropy": 5.671121883392334,
      "epoch": 2.4659833091436867,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004391560879540736,
      "loss": 5.0283,
      "mean_token_accuracy": 0.20608559101819993,
      "num_tokens": 56401755.0,
      "step": 27185
    },
    {
      "entropy": 5.542839527130127,
      "epoch": 2.4664368650217705,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004391339579377952,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2187672436237335,
      "num_tokens": 56412223.0,
      "step": 27190
    },
    {
      "entropy": 5.5960150241851805,
      "epoch": 2.4668904208998548,
      "grad_norm": 1.125,
      "learning_rate": 0.00043911182452718476,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2109716072678566,
      "num_tokens": 56422947.0,
      "step": 27195
    },
    {
      "entropy": 5.5772764682769775,
      "epoch": 2.467343976777939,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043908968772269984,
      "loss": 4.9325,
      "mean_token_accuracy": 0.21519991308450698,
      "num_tokens": 56433189.0,
      "step": 27200
    },
    {
      "entropy": 5.616091299057007,
      "epoch": 2.4677975326560233,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043906754752479836,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2168402835726738,
      "num_tokens": 56444460.0,
      "step": 27205
    },
    {
      "entropy": 5.63706750869751,
      "epoch": 2.4682510885341076,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004390454039339383,
      "loss": 4.9816,
      "mean_token_accuracy": 0.2053310304880142,
      "num_tokens": 56454368.0,
      "step": 27210
    },
    {
      "entropy": 5.5530811786651615,
      "epoch": 2.4687046444121914,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043902325695057743,
      "loss": 4.8778,
      "mean_token_accuracy": 0.21912561357021332,
      "num_tokens": 56464153.0,
      "step": 27215
    },
    {
      "entropy": 5.565906381607055,
      "epoch": 2.4691582002902757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043900110657517377,
      "loss": 4.8961,
      "mean_token_accuracy": 0.21710194349288942,
      "num_tokens": 56473627.0,
      "step": 27220
    },
    {
      "entropy": 5.532009553909302,
      "epoch": 2.46961175616836,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004389789528081855,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2186640113592148,
      "num_tokens": 56483660.0,
      "step": 27225
    },
    {
      "entropy": 5.600687694549561,
      "epoch": 2.470065312046444,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004389567956500707,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2213448166847229,
      "num_tokens": 56492653.0,
      "step": 27230
    },
    {
      "entropy": 5.681814813613892,
      "epoch": 2.4705188679245285,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004389346351012876,
      "loss": 5.0429,
      "mean_token_accuracy": 0.20562908053398132,
      "num_tokens": 56503289.0,
      "step": 27235
    },
    {
      "entropy": 5.618892860412598,
      "epoch": 2.4709724238026123,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043891247116229436,
      "loss": 4.9953,
      "mean_token_accuracy": 0.20237219780683519,
      "num_tokens": 56513843.0,
      "step": 27240
    },
    {
      "entropy": 5.61939811706543,
      "epoch": 2.4714259796806965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043889030383354957,
      "loss": 4.9532,
      "mean_token_accuracy": 0.20816065669059752,
      "num_tokens": 56524175.0,
      "step": 27245
    },
    {
      "entropy": 5.552542304992675,
      "epoch": 2.471879535558781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004388681331155114,
      "loss": 4.8773,
      "mean_token_accuracy": 0.222781939804554,
      "num_tokens": 56534976.0,
      "step": 27250
    },
    {
      "entropy": 5.591837978363037,
      "epoch": 2.472333091436865,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004388459590086386,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21086273491382598,
      "num_tokens": 56545967.0,
      "step": 27255
    },
    {
      "entropy": 5.615090417861938,
      "epoch": 2.4727866473149493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043882378151338956,
      "loss": 4.9801,
      "mean_token_accuracy": 0.20738523900508882,
      "num_tokens": 56555527.0,
      "step": 27260
    },
    {
      "entropy": 5.633236455917358,
      "epoch": 2.4732402031930336,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000438801600630223,
      "loss": 4.951,
      "mean_token_accuracy": 0.2082256481051445,
      "num_tokens": 56565874.0,
      "step": 27265
    },
    {
      "entropy": 5.581714534759522,
      "epoch": 2.4736937590711174,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043877941635959755,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2179457053542137,
      "num_tokens": 56575452.0,
      "step": 27270
    },
    {
      "entropy": 5.464635944366455,
      "epoch": 2.4741473149492017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043875722870197214,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2233095422387123,
      "num_tokens": 56586176.0,
      "step": 27275
    },
    {
      "entropy": 5.5927777767181395,
      "epoch": 2.474600870827286,
      "grad_norm": 1.25,
      "learning_rate": 0.0004387350376578054,
      "loss": 4.9139,
      "mean_token_accuracy": 0.215044829249382,
      "num_tokens": 56595473.0,
      "step": 27280
    },
    {
      "entropy": 5.6027380466461185,
      "epoch": 2.4750544267053702,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004387128432275563,
      "loss": 4.9372,
      "mean_token_accuracy": 0.21109627634286882,
      "num_tokens": 56606016.0,
      "step": 27285
    },
    {
      "entropy": 5.549075984954834,
      "epoch": 2.475507982583454,
      "grad_norm": 1.203125,
      "learning_rate": 0.000438690645411684,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21403739005327224,
      "num_tokens": 56617674.0,
      "step": 27290
    },
    {
      "entropy": 5.604084110260009,
      "epoch": 2.4759615384615383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043866844421064745,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2182503819465637,
      "num_tokens": 56628077.0,
      "step": 27295
    },
    {
      "entropy": 5.588220167160034,
      "epoch": 2.4764150943396226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043864623962490575,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21731435060501098,
      "num_tokens": 56638691.0,
      "step": 27300
    },
    {
      "entropy": 5.5865119934082035,
      "epoch": 2.476868650217707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004386240316549181,
      "loss": 4.9069,
      "mean_token_accuracy": 0.21447901725769042,
      "num_tokens": 56650338.0,
      "step": 27305
    },
    {
      "entropy": 5.5811741828918455,
      "epoch": 2.477322206095791,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043860182030114374,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22308145016431807,
      "num_tokens": 56660328.0,
      "step": 27310
    },
    {
      "entropy": 5.638687705993652,
      "epoch": 2.4777757619738754,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043857960556404214,
      "loss": 4.9975,
      "mean_token_accuracy": 0.213313028216362,
      "num_tokens": 56670674.0,
      "step": 27315
    },
    {
      "entropy": 5.559856414794922,
      "epoch": 2.478229317851959,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004385573874440725,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21704309284687043,
      "num_tokens": 56680656.0,
      "step": 27320
    },
    {
      "entropy": 5.54161057472229,
      "epoch": 2.4786828737300435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043853516594169446,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2209966316819191,
      "num_tokens": 56691039.0,
      "step": 27325
    },
    {
      "entropy": 5.588390779495239,
      "epoch": 2.4791364296081277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004385129410573675,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2088440477848053,
      "num_tokens": 56703101.0,
      "step": 27330
    },
    {
      "entropy": 5.620885372161865,
      "epoch": 2.479589985486212,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043849071279155124,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21611011177301406,
      "num_tokens": 56712351.0,
      "step": 27335
    },
    {
      "entropy": 5.573245096206665,
      "epoch": 2.480043541364296,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043846848114470537,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2197880432009697,
      "num_tokens": 56722836.0,
      "step": 27340
    },
    {
      "entropy": 5.609938049316407,
      "epoch": 2.48049709724238,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043844624611728974,
      "loss": 4.9409,
      "mean_token_accuracy": 0.20814346224069596,
      "num_tokens": 56734032.0,
      "step": 27345
    },
    {
      "entropy": 5.626892518997193,
      "epoch": 2.4809506531204644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043842400770976406,
      "loss": 4.8946,
      "mean_token_accuracy": 0.21539338678121567,
      "num_tokens": 56743536.0,
      "step": 27350
    },
    {
      "entropy": 5.574132966995239,
      "epoch": 2.4814042089985486,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004384017659225882,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2113188087940216,
      "num_tokens": 56753475.0,
      "step": 27355
    },
    {
      "entropy": 5.552309989929199,
      "epoch": 2.481857764876633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043837952075622224,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22213870882987977,
      "num_tokens": 56763602.0,
      "step": 27360
    },
    {
      "entropy": 5.589760065078735,
      "epoch": 2.482311320754717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004383572722111261,
      "loss": 4.878,
      "mean_token_accuracy": 0.21835470646619798,
      "num_tokens": 56773859.0,
      "step": 27365
    },
    {
      "entropy": 5.568104839324951,
      "epoch": 2.482764876632801,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043833502028775995,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21688507199287416,
      "num_tokens": 56783810.0,
      "step": 27370
    },
    {
      "entropy": 5.590130233764649,
      "epoch": 2.4832184325108853,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043831276498658395,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2060217559337616,
      "num_tokens": 56794088.0,
      "step": 27375
    },
    {
      "entropy": 5.520579624176025,
      "epoch": 2.4836719883889695,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004382905063080584,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2182580292224884,
      "num_tokens": 56804623.0,
      "step": 27380
    },
    {
      "entropy": 5.55784387588501,
      "epoch": 2.484125544267054,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043826824425264357,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21142170578241348,
      "num_tokens": 56813842.0,
      "step": 27385
    },
    {
      "entropy": 5.624997568130493,
      "epoch": 2.484579100145138,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004382459788207997,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21277585476636887,
      "num_tokens": 56823501.0,
      "step": 27390
    },
    {
      "entropy": 5.594214582443238,
      "epoch": 2.485032656023222,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043822371001298754,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21373566389083862,
      "num_tokens": 56834601.0,
      "step": 27395
    },
    {
      "entropy": 5.5291581630706785,
      "epoch": 2.485486211901306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004382014378296674,
      "loss": 4.9488,
      "mean_token_accuracy": 0.20987190306186676,
      "num_tokens": 56844701.0,
      "step": 27400
    },
    {
      "entropy": 5.538779449462891,
      "epoch": 2.4859397677793904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004381791622713,
      "loss": 4.9152,
      "mean_token_accuracy": 0.21994872093200685,
      "num_tokens": 56854867.0,
      "step": 27405
    },
    {
      "entropy": 5.617107200622558,
      "epoch": 2.4863933236574747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000438156883338346,
      "loss": 4.9442,
      "mean_token_accuracy": 0.2178375318646431,
      "num_tokens": 56865688.0,
      "step": 27410
    },
    {
      "entropy": 5.642718410491943,
      "epoch": 2.486846879535559,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043813460103126595,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2102121114730835,
      "num_tokens": 56876504.0,
      "step": 27415
    },
    {
      "entropy": 5.597728157043457,
      "epoch": 2.487300435413643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043811231535052076,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21712858229875565,
      "num_tokens": 56886083.0,
      "step": 27420
    },
    {
      "entropy": 5.55987582206726,
      "epoch": 2.487753991291727,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004380900262965714,
      "loss": 4.9666,
      "mean_token_accuracy": 0.2168812245130539,
      "num_tokens": 56897079.0,
      "step": 27425
    },
    {
      "entropy": 5.552517604827881,
      "epoch": 2.4882075471698113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004380677338698787,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21822305619716645,
      "num_tokens": 56906962.0,
      "step": 27430
    },
    {
      "entropy": 5.635354709625244,
      "epoch": 2.4886611030478956,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043804543807090374,
      "loss": 5.0006,
      "mean_token_accuracy": 0.20962221473455428,
      "num_tokens": 56918156.0,
      "step": 27435
    },
    {
      "entropy": 5.592575979232788,
      "epoch": 2.48911465892598,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043802313890010757,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2106891468167305,
      "num_tokens": 56928888.0,
      "step": 27440
    },
    {
      "entropy": 5.620944166183472,
      "epoch": 2.4895682148040637,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004380008363579513,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21380129009485244,
      "num_tokens": 56939929.0,
      "step": 27445
    },
    {
      "entropy": 5.609107446670532,
      "epoch": 2.490021770682148,
      "grad_norm": 1.125,
      "learning_rate": 0.00043797853044489623,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2185220092535019,
      "num_tokens": 56950992.0,
      "step": 27450
    },
    {
      "entropy": 5.640049171447754,
      "epoch": 2.490475326560232,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004379562211614036,
      "loss": 5.0222,
      "mean_token_accuracy": 0.20864610373973846,
      "num_tokens": 56962118.0,
      "step": 27455
    },
    {
      "entropy": 5.584109354019165,
      "epoch": 2.4909288824383164,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004379339085079348,
      "loss": 4.92,
      "mean_token_accuracy": 0.2172509253025055,
      "num_tokens": 56972837.0,
      "step": 27460
    },
    {
      "entropy": 5.618616151809692,
      "epoch": 2.4913824383164007,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004379115924849512,
      "loss": 4.9303,
      "mean_token_accuracy": 0.2137095585465431,
      "num_tokens": 56982794.0,
      "step": 27465
    },
    {
      "entropy": 5.679965829849243,
      "epoch": 2.491835994194485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004378892730929145,
      "loss": 5.0684,
      "mean_token_accuracy": 0.2122359424829483,
      "num_tokens": 56993979.0,
      "step": 27470
    },
    {
      "entropy": 5.615882349014282,
      "epoch": 2.492289550072569,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043786695033228604,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2104012757539749,
      "num_tokens": 57004505.0,
      "step": 27475
    },
    {
      "entropy": 5.638076829910278,
      "epoch": 2.492743105950653,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043784462420352747,
      "loss": 4.9619,
      "mean_token_accuracy": 0.20832054167985917,
      "num_tokens": 57014121.0,
      "step": 27480
    },
    {
      "entropy": 5.652388286590576,
      "epoch": 2.4931966618287373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043782229470710067,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21631906181573868,
      "num_tokens": 57023748.0,
      "step": 27485
    },
    {
      "entropy": 5.5536274909973145,
      "epoch": 2.4936502177068216,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004377999618434673,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21256810575723648,
      "num_tokens": 57034782.0,
      "step": 27490
    },
    {
      "entropy": 5.5397687435150145,
      "epoch": 2.4941037735849054,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004377776256130892,
      "loss": 4.8485,
      "mean_token_accuracy": 0.21600691974163055,
      "num_tokens": 57045136.0,
      "step": 27495
    },
    {
      "entropy": 5.586387968063354,
      "epoch": 2.4945573294629897,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043775528601642833,
      "loss": 5.0745,
      "mean_token_accuracy": 0.20072693675756453,
      "num_tokens": 57055893.0,
      "step": 27500
    },
    {
      "entropy": 5.581463718414307,
      "epoch": 2.495010885341074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043773294305394665,
      "loss": 4.9701,
      "mean_token_accuracy": 0.20962213575839997,
      "num_tokens": 57067189.0,
      "step": 27505
    },
    {
      "entropy": 5.643249082565307,
      "epoch": 2.4954644412191582,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043771059672610636,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2198803037405014,
      "num_tokens": 57077374.0,
      "step": 27510
    },
    {
      "entropy": 5.608972072601318,
      "epoch": 2.4959179970972425,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043768824703336933,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21179517805576326,
      "num_tokens": 57088168.0,
      "step": 27515
    },
    {
      "entropy": 5.581168937683105,
      "epoch": 2.4963715529753268,
      "grad_norm": 1.125,
      "learning_rate": 0.0004376658939761979,
      "loss": 4.8767,
      "mean_token_accuracy": 0.2155425950884819,
      "num_tokens": 57098743.0,
      "step": 27520
    },
    {
      "entropy": 5.555883741378784,
      "epoch": 2.4968251088534106,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004376435375550544,
      "loss": 5.0202,
      "mean_token_accuracy": 0.208633217215538,
      "num_tokens": 57109854.0,
      "step": 27525
    },
    {
      "entropy": 5.543722105026245,
      "epoch": 2.497278664731495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043762117777040106,
      "loss": 4.865,
      "mean_token_accuracy": 0.22181468158960344,
      "num_tokens": 57120296.0,
      "step": 27530
    },
    {
      "entropy": 5.557712888717651,
      "epoch": 2.497732220609579,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004375988146227003,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2137293577194214,
      "num_tokens": 57131295.0,
      "step": 27535
    },
    {
      "entropy": 5.606823825836182,
      "epoch": 2.4981857764876634,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004375764481124147,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21407651156187057,
      "num_tokens": 57141936.0,
      "step": 27540
    },
    {
      "entropy": 5.584395217895508,
      "epoch": 2.4986393323657476,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043755407824000664,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22266368716955184,
      "num_tokens": 57152510.0,
      "step": 27545
    },
    {
      "entropy": 5.59586238861084,
      "epoch": 2.4990928882438315,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043753170500593884,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21178195476531983,
      "num_tokens": 57162546.0,
      "step": 27550
    },
    {
      "entropy": 5.501942348480225,
      "epoch": 2.4995464441219157,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043750932841067397,
      "loss": 4.8337,
      "mean_token_accuracy": 0.22415578961372376,
      "num_tokens": 57174658.0,
      "step": 27555
    },
    {
      "entropy": 5.627531433105469,
      "epoch": 2.5,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043748694845467483,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20090083330869674,
      "num_tokens": 57185157.0,
      "step": 27560
    },
    {
      "entropy": 5.573130941390991,
      "epoch": 2.5004535558780843,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043746456513840405,
      "loss": 4.8534,
      "mean_token_accuracy": 0.217386594414711,
      "num_tokens": 57195277.0,
      "step": 27565
    },
    {
      "entropy": 5.60779185295105,
      "epoch": 2.5009071117561685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004374421784623248,
      "loss": 4.9412,
      "mean_token_accuracy": 0.21299373656511306,
      "num_tokens": 57205374.0,
      "step": 27570
    },
    {
      "entropy": 5.570887422561645,
      "epoch": 2.501360667634253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043741978842689987,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2148380994796753,
      "num_tokens": 57215700.0,
      "step": 27575
    },
    {
      "entropy": 5.604409313201904,
      "epoch": 2.5018142235123366,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004373973950325923,
      "loss": 4.9472,
      "mean_token_accuracy": 0.213739313185215,
      "num_tokens": 57225765.0,
      "step": 27580
    },
    {
      "entropy": 5.6124614715576175,
      "epoch": 2.502267779390421,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043737499827986527,
      "loss": 4.9322,
      "mean_token_accuracy": 0.21606886386871338,
      "num_tokens": 57235913.0,
      "step": 27585
    },
    {
      "entropy": 5.530026865005493,
      "epoch": 2.502721335268505,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043735259816918184,
      "loss": 4.8638,
      "mean_token_accuracy": 0.21802701354026793,
      "num_tokens": 57246204.0,
      "step": 27590
    },
    {
      "entropy": 5.5358093738555905,
      "epoch": 2.5031748911465894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004373301947010053,
      "loss": 4.9588,
      "mean_token_accuracy": 0.20669886022806166,
      "num_tokens": 57256167.0,
      "step": 27595
    },
    {
      "entropy": 5.535846805572509,
      "epoch": 2.5036284470246732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043730778787579895,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2164482817053795,
      "num_tokens": 57266621.0,
      "step": 27600
    },
    {
      "entropy": 5.603772687911987,
      "epoch": 2.5040820029027575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043728537769402627,
      "loss": 4.9213,
      "mean_token_accuracy": 0.2138948306441307,
      "num_tokens": 57277282.0,
      "step": 27605
    },
    {
      "entropy": 5.62356128692627,
      "epoch": 2.5045355587808418,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004372629641561506,
      "loss": 5.003,
      "mean_token_accuracy": 0.2088750794529915,
      "num_tokens": 57287541.0,
      "step": 27610
    },
    {
      "entropy": 5.590396451950073,
      "epoch": 2.504989114658926,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043724054726263547,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21239400953054427,
      "num_tokens": 57298158.0,
      "step": 27615
    },
    {
      "entropy": 5.601081848144531,
      "epoch": 2.5054426705370103,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004372181270139444,
      "loss": 4.956,
      "mean_token_accuracy": 0.2102935031056404,
      "num_tokens": 57308356.0,
      "step": 27620
    },
    {
      "entropy": 5.547576427459717,
      "epoch": 2.5058962264150946,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043719570341054127,
      "loss": 4.8593,
      "mean_token_accuracy": 0.22023206800222397,
      "num_tokens": 57317905.0,
      "step": 27625
    },
    {
      "entropy": 5.5640476703643795,
      "epoch": 2.5063497822931784,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043717327645288956,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21614961624145507,
      "num_tokens": 57328277.0,
      "step": 27630
    },
    {
      "entropy": 5.621697998046875,
      "epoch": 2.5068033381712627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043715084614145316,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21061322391033171,
      "num_tokens": 57338975.0,
      "step": 27635
    },
    {
      "entropy": 5.581604814529419,
      "epoch": 2.507256894049347,
      "grad_norm": 1.078125,
      "learning_rate": 0.000437128412476696,
      "loss": 4.9057,
      "mean_token_accuracy": 0.21776700913906097,
      "num_tokens": 57349439.0,
      "step": 27640
    },
    {
      "entropy": 5.572036504745483,
      "epoch": 2.507710449927431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004371059754590819,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20855688899755478,
      "num_tokens": 57359414.0,
      "step": 27645
    },
    {
      "entropy": 5.640950727462768,
      "epoch": 2.508164005805515,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043708353508907494,
      "loss": 4.9983,
      "mean_token_accuracy": 0.21514788568019866,
      "num_tokens": 57369711.0,
      "step": 27650
    },
    {
      "entropy": 5.567636442184448,
      "epoch": 2.5086175616835993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004370610913671392,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22319878339767457,
      "num_tokens": 57380001.0,
      "step": 27655
    },
    {
      "entropy": 5.616363906860352,
      "epoch": 2.5090711175616836,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043703864429373875,
      "loss": 5.0483,
      "mean_token_accuracy": 0.19964646697044372,
      "num_tokens": 57390794.0,
      "step": 27660
    },
    {
      "entropy": 5.578546762466431,
      "epoch": 2.509524673439768,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043701619386933793,
      "loss": 4.9569,
      "mean_token_accuracy": 0.2049718514084816,
      "num_tokens": 57401975.0,
      "step": 27665
    },
    {
      "entropy": 5.524433374404907,
      "epoch": 2.509978229317852,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043699374009440085,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22720505446195602,
      "num_tokens": 57411249.0,
      "step": 27670
    },
    {
      "entropy": 5.520901203155518,
      "epoch": 2.5104317851959363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004369712829693919,
      "loss": 4.8714,
      "mean_token_accuracy": 0.2197423607110977,
      "num_tokens": 57422679.0,
      "step": 27675
    },
    {
      "entropy": 5.576691913604736,
      "epoch": 2.51088534107402,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004369488224947757,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2181418776512146,
      "num_tokens": 57433265.0,
      "step": 27680
    },
    {
      "entropy": 5.598511695861816,
      "epoch": 2.5113388969521044,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004369263586710164,
      "loss": 4.9558,
      "mean_token_accuracy": 0.21273164451122284,
      "num_tokens": 57444095.0,
      "step": 27685
    },
    {
      "entropy": 5.5759705066680905,
      "epoch": 2.5117924528301887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004369038914985789,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21777071356773375,
      "num_tokens": 57454059.0,
      "step": 27690
    },
    {
      "entropy": 5.567894315719604,
      "epoch": 2.512246008708273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004368814209779276,
      "loss": 4.8755,
      "mean_token_accuracy": 0.21597269624471666,
      "num_tokens": 57463940.0,
      "step": 27695
    },
    {
      "entropy": 5.505574750900268,
      "epoch": 2.512699564586357,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043685894710952724,
      "loss": 4.948,
      "mean_token_accuracy": 0.2123625859618187,
      "num_tokens": 57474731.0,
      "step": 27700
    },
    {
      "entropy": 5.547702932357788,
      "epoch": 2.513153120464441,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004368364698938426,
      "loss": 4.8667,
      "mean_token_accuracy": 0.21568514108657838,
      "num_tokens": 57485096.0,
      "step": 27705
    },
    {
      "entropy": 5.559608888626099,
      "epoch": 2.5136066763425253,
      "grad_norm": 1.125,
      "learning_rate": 0.0004368139893313385,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21700042486190796,
      "num_tokens": 57496371.0,
      "step": 27710
    },
    {
      "entropy": 5.612035655975342,
      "epoch": 2.5140602322206096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043679150542248,
      "loss": 4.9041,
      "mean_token_accuracy": 0.21756649315357207,
      "num_tokens": 57506012.0,
      "step": 27715
    },
    {
      "entropy": 5.5967076301574705,
      "epoch": 2.514513788098694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043676901816773174,
      "loss": 4.968,
      "mean_token_accuracy": 0.20733955204486848,
      "num_tokens": 57516536.0,
      "step": 27720
    },
    {
      "entropy": 5.505822992324829,
      "epoch": 2.514967343976778,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004367465275675591,
      "loss": 4.8578,
      "mean_token_accuracy": 0.21870542615652083,
      "num_tokens": 57527450.0,
      "step": 27725
    },
    {
      "entropy": 5.60909743309021,
      "epoch": 2.5154208998548624,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043672403362242704,
      "loss": 4.9912,
      "mean_token_accuracy": 0.21314203292131423,
      "num_tokens": 57538234.0,
      "step": 27730
    },
    {
      "entropy": 5.6083019256591795,
      "epoch": 2.515874455732946,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043670153633280066,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22190325409173967,
      "num_tokens": 57547938.0,
      "step": 27735
    },
    {
      "entropy": 5.617470788955688,
      "epoch": 2.5163280116110305,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004366790356991454,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21650672554969788,
      "num_tokens": 57560553.0,
      "step": 27740
    },
    {
      "entropy": 5.557185792922974,
      "epoch": 2.5167815674891147,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004366565317219265,
      "loss": 4.896,
      "mean_token_accuracy": 0.214796045422554,
      "num_tokens": 57570670.0,
      "step": 27745
    },
    {
      "entropy": 5.686214399337769,
      "epoch": 2.5172351233671986,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043663402440160924,
      "loss": 5.1186,
      "mean_token_accuracy": 0.20182747393846512,
      "num_tokens": 57582377.0,
      "step": 27750
    },
    {
      "entropy": 5.5826903820037845,
      "epoch": 2.517688679245283,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004366115137386592,
      "loss": 4.8053,
      "mean_token_accuracy": 0.2303880289196968,
      "num_tokens": 57592764.0,
      "step": 27755
    },
    {
      "entropy": 5.56025800704956,
      "epoch": 2.518142235123367,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004365889997335419,
      "loss": 4.9378,
      "mean_token_accuracy": 0.20805972814559937,
      "num_tokens": 57603768.0,
      "step": 27760
    },
    {
      "entropy": 5.5337381839752195,
      "epoch": 2.5185957910014514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043656648238672285,
      "loss": 4.918,
      "mean_token_accuracy": 0.2153510645031929,
      "num_tokens": 57614346.0,
      "step": 27765
    },
    {
      "entropy": 5.672798728942871,
      "epoch": 2.5190493468795356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004365439616986678,
      "loss": 5.0471,
      "mean_token_accuracy": 0.20816917717456818,
      "num_tokens": 57625015.0,
      "step": 27770
    },
    {
      "entropy": 5.632098245620727,
      "epoch": 2.51950290275762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043652143766984247,
      "loss": 4.9872,
      "mean_token_accuracy": 0.21003783345222474,
      "num_tokens": 57635476.0,
      "step": 27775
    },
    {
      "entropy": 5.559882688522339,
      "epoch": 2.519956458635704,
      "grad_norm": 1.125,
      "learning_rate": 0.00043649891030071264,
      "loss": 4.8905,
      "mean_token_accuracy": 0.2153201714158058,
      "num_tokens": 57645512.0,
      "step": 27780
    },
    {
      "entropy": 5.5952517032623295,
      "epoch": 2.520410014513788,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043647637959174425,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21549456119537352,
      "num_tokens": 57655766.0,
      "step": 27785
    },
    {
      "entropy": 5.591435956954956,
      "epoch": 2.5208635703918723,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043645384554340315,
      "loss": 4.9064,
      "mean_token_accuracy": 0.2220040425658226,
      "num_tokens": 57667325.0,
      "step": 27790
    },
    {
      "entropy": 5.539133930206299,
      "epoch": 2.5213171262699565,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043643130815615536,
      "loss": 4.9227,
      "mean_token_accuracy": 0.20776327252388,
      "num_tokens": 57677884.0,
      "step": 27795
    },
    {
      "entropy": 5.642353630065918,
      "epoch": 2.521770682148041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043640876743046694,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21789935380220413,
      "num_tokens": 57687765.0,
      "step": 27800
    },
    {
      "entropy": 5.576065111160278,
      "epoch": 2.5222242380261246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004363862233668041,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21047825068235398,
      "num_tokens": 57697777.0,
      "step": 27805
    },
    {
      "entropy": 5.541267967224121,
      "epoch": 2.522677793904209,
      "grad_norm": 1.078125,
      "learning_rate": 0.000436363675965633,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21198201179504395,
      "num_tokens": 57708765.0,
      "step": 27810
    },
    {
      "entropy": 5.614379167556763,
      "epoch": 2.523131349782293,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004363411252274201,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21668234020471572,
      "num_tokens": 57719658.0,
      "step": 27815
    },
    {
      "entropy": 5.6014598369598385,
      "epoch": 2.5235849056603774,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043631857115263153,
      "loss": 4.8978,
      "mean_token_accuracy": 0.21975661516189576,
      "num_tokens": 57730018.0,
      "step": 27820
    },
    {
      "entropy": 5.510796213150025,
      "epoch": 2.5240384615384617,
      "grad_norm": 1.125,
      "learning_rate": 0.0004362960137417338,
      "loss": 4.8623,
      "mean_token_accuracy": 0.21918608546257018,
      "num_tokens": 57741251.0,
      "step": 27825
    },
    {
      "entropy": 5.582495164871216,
      "epoch": 2.524492017416546,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004362734529951933,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21350988298654555,
      "num_tokens": 57751020.0,
      "step": 27830
    },
    {
      "entropy": 5.593598127365112,
      "epoch": 2.5249455732946298,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043625088891347685,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22202199399471284,
      "num_tokens": 57760989.0,
      "step": 27835
    },
    {
      "entropy": 5.676693344116211,
      "epoch": 2.525399129172714,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043622832149705086,
      "loss": 4.9718,
      "mean_token_accuracy": 0.2155179649591446,
      "num_tokens": 57770882.0,
      "step": 27840
    },
    {
      "entropy": 5.687330150604248,
      "epoch": 2.5258526850507983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043620575074638206,
      "loss": 4.9951,
      "mean_token_accuracy": 0.21127077788114548,
      "num_tokens": 57781812.0,
      "step": 27845
    },
    {
      "entropy": 5.582814693450928,
      "epoch": 2.5263062409288826,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043618317666193734,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22397453784942628,
      "num_tokens": 57792009.0,
      "step": 27850
    },
    {
      "entropy": 5.5197436809539795,
      "epoch": 2.5267597968069664,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043616059924418346,
      "loss": 4.8898,
      "mean_token_accuracy": 0.22000424414873124,
      "num_tokens": 57802109.0,
      "step": 27855
    },
    {
      "entropy": 5.605290079116822,
      "epoch": 2.5272133526850507,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004361380184935872,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21872030943632126,
      "num_tokens": 57812374.0,
      "step": 27860
    },
    {
      "entropy": 5.577760505676269,
      "epoch": 2.527666908563135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004361154344106158,
      "loss": 4.8658,
      "mean_token_accuracy": 0.21653814017772674,
      "num_tokens": 57822936.0,
      "step": 27865
    },
    {
      "entropy": 5.68985276222229,
      "epoch": 2.528120464441219,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043609284699573604,
      "loss": 5.0872,
      "mean_token_accuracy": 0.20586992502212526,
      "num_tokens": 57832693.0,
      "step": 27870
    },
    {
      "entropy": 5.654791831970215,
      "epoch": 2.5285740203193035,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043607025624941523,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21327386498451234,
      "num_tokens": 57843161.0,
      "step": 27875
    },
    {
      "entropy": 5.6050772190094,
      "epoch": 2.5290275761973877,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004360476621721204,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22539540380239487,
      "num_tokens": 57853027.0,
      "step": 27880
    },
    {
      "entropy": 5.605326414108276,
      "epoch": 2.5294811320754715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004360250647643189,
      "loss": 4.9268,
      "mean_token_accuracy": 0.22102978825569153,
      "num_tokens": 57862821.0,
      "step": 27885
    },
    {
      "entropy": 5.557902002334595,
      "epoch": 2.529934687953556,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043600246402647805,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2189985066652298,
      "num_tokens": 57874065.0,
      "step": 27890
    },
    {
      "entropy": 5.576665449142456,
      "epoch": 2.53038824383164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004359798599590652,
      "loss": 4.9452,
      "mean_token_accuracy": 0.21320515722036362,
      "num_tokens": 57885190.0,
      "step": 27895
    },
    {
      "entropy": 5.557775211334229,
      "epoch": 2.5308417997097243,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043595725256254786,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21442791372537612,
      "num_tokens": 57894234.0,
      "step": 27900
    },
    {
      "entropy": 5.622046661376953,
      "epoch": 2.531295355587808,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004359346418373935,
      "loss": 5.0611,
      "mean_token_accuracy": 0.20003791749477387,
      "num_tokens": 57905401.0,
      "step": 27905
    },
    {
      "entropy": 5.567955589294433,
      "epoch": 2.5317489114658924,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043591202778406975,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21410039365291594,
      "num_tokens": 57916097.0,
      "step": 27910
    },
    {
      "entropy": 5.531673717498779,
      "epoch": 2.5322024673439767,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004358894104030442,
      "loss": 4.8047,
      "mean_token_accuracy": 0.2229517713189125,
      "num_tokens": 57926871.0,
      "step": 27915
    },
    {
      "entropy": 5.539941787719727,
      "epoch": 2.532656023222061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043586678969478464,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22349265217781067,
      "num_tokens": 57937344.0,
      "step": 27920
    },
    {
      "entropy": 5.640328741073608,
      "epoch": 2.5331095791001452,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000435844165659759,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21388643831014634,
      "num_tokens": 57947633.0,
      "step": 27925
    },
    {
      "entropy": 5.550778436660766,
      "epoch": 2.5335631349782295,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043582153829843493,
      "loss": 4.8597,
      "mean_token_accuracy": 0.21751099228858947,
      "num_tokens": 57957948.0,
      "step": 27930
    },
    {
      "entropy": 5.599670553207398,
      "epoch": 2.5340166908563138,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004357989076112805,
      "loss": 4.904,
      "mean_token_accuracy": 0.21520741134881974,
      "num_tokens": 57967719.0,
      "step": 27935
    },
    {
      "entropy": 5.615107440948487,
      "epoch": 2.5344702467343976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004357762735987637,
      "loss": 4.9704,
      "mean_token_accuracy": 0.20863263607025145,
      "num_tokens": 57978218.0,
      "step": 27940
    },
    {
      "entropy": 5.605807113647461,
      "epoch": 2.534923802612482,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004357536362613525,
      "loss": 4.8639,
      "mean_token_accuracy": 0.21907026022672654,
      "num_tokens": 57987860.0,
      "step": 27945
    },
    {
      "entropy": 5.567364120483399,
      "epoch": 2.535377358490566,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043573099559951516,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21898820847272873,
      "num_tokens": 57997872.0,
      "step": 27950
    },
    {
      "entropy": 5.636494779586792,
      "epoch": 2.5358309143686504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004357083516137199,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2082849159836769,
      "num_tokens": 58007771.0,
      "step": 27955
    },
    {
      "entropy": 5.6334034442901615,
      "epoch": 2.536284470246734,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000435685704304435,
      "loss": 4.9225,
      "mean_token_accuracy": 0.21224740147590637,
      "num_tokens": 58018972.0,
      "step": 27960
    },
    {
      "entropy": 5.58109564781189,
      "epoch": 2.5367380261248185,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043566305367212884,
      "loss": 4.8098,
      "mean_token_accuracy": 0.218258860707283,
      "num_tokens": 58028581.0,
      "step": 27965
    },
    {
      "entropy": 5.561747598648071,
      "epoch": 2.5371915820029027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043564039971726964,
      "loss": 4.9657,
      "mean_token_accuracy": 0.21264857053756714,
      "num_tokens": 58038877.0,
      "step": 27970
    },
    {
      "entropy": 5.557735967636108,
      "epoch": 2.537645137880987,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043561774244032614,
      "loss": 4.9037,
      "mean_token_accuracy": 0.21876992732286454,
      "num_tokens": 58048777.0,
      "step": 27975
    },
    {
      "entropy": 5.612783288955688,
      "epoch": 2.5380986937590713,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043559508184176675,
      "loss": 4.986,
      "mean_token_accuracy": 0.21423932164907455,
      "num_tokens": 58060270.0,
      "step": 27980
    },
    {
      "entropy": 5.566536617279053,
      "epoch": 2.5385522496371555,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043557241792206014,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22022177428007125,
      "num_tokens": 58070838.0,
      "step": 27985
    },
    {
      "entropy": 5.613703918457031,
      "epoch": 2.5390058055152394,
      "grad_norm": 1.125,
      "learning_rate": 0.00043554975068167505,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2126248747110367,
      "num_tokens": 58079933.0,
      "step": 27990
    },
    {
      "entropy": 5.566472721099854,
      "epoch": 2.5394593613933236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004355270801210802,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21569198220968247,
      "num_tokens": 58090673.0,
      "step": 27995
    },
    {
      "entropy": 5.555805349349976,
      "epoch": 2.539912917271408,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004355044062407444,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2296976089477539,
      "num_tokens": 58100172.0,
      "step": 28000
    },
    {
      "entropy": 5.563894176483155,
      "epoch": 2.540366473149492,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004354817290411366,
      "loss": 4.9408,
      "mean_token_accuracy": 0.21394419074058532,
      "num_tokens": 58108965.0,
      "step": 28005
    },
    {
      "entropy": 5.56345763206482,
      "epoch": 2.540820029027576,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043545904852272573,
      "loss": 4.8364,
      "mean_token_accuracy": 0.22171283960342408,
      "num_tokens": 58120094.0,
      "step": 28010
    },
    {
      "entropy": 5.605643272399902,
      "epoch": 2.5412735849056602,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043543636468598083,
      "loss": 4.9763,
      "mean_token_accuracy": 0.20851194560527803,
      "num_tokens": 58131525.0,
      "step": 28015
    },
    {
      "entropy": 5.622106266021729,
      "epoch": 2.5417271407837445,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043541367753137107,
      "loss": 4.8766,
      "mean_token_accuracy": 0.21641097068786622,
      "num_tokens": 58141986.0,
      "step": 28020
    },
    {
      "entropy": 5.615628290176391,
      "epoch": 2.5421806966618288,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004353909870593656,
      "loss": 4.9318,
      "mean_token_accuracy": 0.21234720051288605,
      "num_tokens": 58154122.0,
      "step": 28025
    },
    {
      "entropy": 5.584965801239013,
      "epoch": 2.542634252539913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004353682932704336,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22646308839321136,
      "num_tokens": 58164148.0,
      "step": 28030
    },
    {
      "entropy": 5.532323169708252,
      "epoch": 2.5430878084179973,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004353455961650444,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2159460723400116,
      "num_tokens": 58173836.0,
      "step": 28035
    },
    {
      "entropy": 5.558132839202881,
      "epoch": 2.543541364296081,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004353228957436675,
      "loss": 4.909,
      "mean_token_accuracy": 0.2180062174797058,
      "num_tokens": 58184598.0,
      "step": 28040
    },
    {
      "entropy": 5.54286298751831,
      "epoch": 2.5439949201741654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043530019200677215,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21547307521104814,
      "num_tokens": 58194820.0,
      "step": 28045
    },
    {
      "entropy": 5.596372079849243,
      "epoch": 2.5444484760522497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043527748495482806,
      "loss": 4.9832,
      "mean_token_accuracy": 0.21225499212741852,
      "num_tokens": 58206589.0,
      "step": 28050
    },
    {
      "entropy": 5.641532850265503,
      "epoch": 2.544902031930334,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043525477458830473,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21534069031476974,
      "num_tokens": 58216750.0,
      "step": 28055
    },
    {
      "entropy": 5.525184869766235,
      "epoch": 2.5453555878084178,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004352320609076718,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2207137256860733,
      "num_tokens": 58228496.0,
      "step": 28060
    },
    {
      "entropy": 5.566481876373291,
      "epoch": 2.545809143686502,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043520934391339894,
      "loss": 4.8886,
      "mean_token_accuracy": 0.21039522737264632,
      "num_tokens": 58237844.0,
      "step": 28065
    },
    {
      "entropy": 5.546635341644287,
      "epoch": 2.5462626995645863,
      "grad_norm": 1.125,
      "learning_rate": 0.0004351866236059561,
      "loss": 4.8617,
      "mean_token_accuracy": 0.21599849313497543,
      "num_tokens": 58248522.0,
      "step": 28070
    },
    {
      "entropy": 5.560593509674073,
      "epoch": 2.5467162554426706,
      "grad_norm": 1.25,
      "learning_rate": 0.00043516389998581305,
      "loss": 4.8814,
      "mean_token_accuracy": 0.21837327480316163,
      "num_tokens": 58258492.0,
      "step": 28075
    },
    {
      "entropy": 5.536153364181518,
      "epoch": 2.547169811320755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004351411730534398,
      "loss": 4.915,
      "mean_token_accuracy": 0.2091811031103134,
      "num_tokens": 58268683.0,
      "step": 28080
    },
    {
      "entropy": 5.596719551086426,
      "epoch": 2.547623367198839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004351184428093062,
      "loss": 4.9482,
      "mean_token_accuracy": 0.20734853595495223,
      "num_tokens": 58279479.0,
      "step": 28085
    },
    {
      "entropy": 5.602116537094116,
      "epoch": 2.5480769230769234,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004350957092538825,
      "loss": 4.9942,
      "mean_token_accuracy": 0.21143606901168824,
      "num_tokens": 58290751.0,
      "step": 28090
    },
    {
      "entropy": 5.633344268798828,
      "epoch": 2.548530478955007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004350729723876386,
      "loss": 5.0151,
      "mean_token_accuracy": 0.2052756980061531,
      "num_tokens": 58302163.0,
      "step": 28095
    },
    {
      "entropy": 5.606661701202393,
      "epoch": 2.5489840348330914,
      "grad_norm": 1.125,
      "learning_rate": 0.00043505023221104497,
      "loss": 4.9334,
      "mean_token_accuracy": 0.20499745458364488,
      "num_tokens": 58311488.0,
      "step": 28100
    },
    {
      "entropy": 5.5649652004241945,
      "epoch": 2.5494375907111757,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043502748872457166,
      "loss": 4.9038,
      "mean_token_accuracy": 0.21362664550542831,
      "num_tokens": 58321599.0,
      "step": 28105
    },
    {
      "entropy": 5.576816129684448,
      "epoch": 2.5498911465892595,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004350047419286892,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21436455845832825,
      "num_tokens": 58330696.0,
      "step": 28110
    },
    {
      "entropy": 5.555698108673096,
      "epoch": 2.550344702467344,
      "grad_norm": 1.125,
      "learning_rate": 0.00043498199182386775,
      "loss": 4.9367,
      "mean_token_accuracy": 0.2198529452085495,
      "num_tokens": 58341793.0,
      "step": 28115
    },
    {
      "entropy": 5.609289979934692,
      "epoch": 2.550798258345428,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043495923841057807,
      "loss": 5.0066,
      "mean_token_accuracy": 0.2080481007695198,
      "num_tokens": 58352260.0,
      "step": 28120
    },
    {
      "entropy": 5.672750091552734,
      "epoch": 2.5512518142235123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043493648168929053,
      "loss": 5.0483,
      "mean_token_accuracy": 0.2021309405565262,
      "num_tokens": 58363052.0,
      "step": 28125
    },
    {
      "entropy": 5.655731821060181,
      "epoch": 2.5517053701015966,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004349137216604758,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21103172600269318,
      "num_tokens": 58374039.0,
      "step": 28130
    },
    {
      "entropy": 5.547771644592285,
      "epoch": 2.552158925979681,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004348909583246045,
      "loss": 4.7928,
      "mean_token_accuracy": 0.22742677330970765,
      "num_tokens": 58383441.0,
      "step": 28135
    },
    {
      "entropy": 5.537250709533692,
      "epoch": 2.552612481857765,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004348681916821475,
      "loss": 4.8999,
      "mean_token_accuracy": 0.21714603304862976,
      "num_tokens": 58394164.0,
      "step": 28140
    },
    {
      "entropy": 5.5491735458374025,
      "epoch": 2.553066037735849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004348454217335755,
      "loss": 4.936,
      "mean_token_accuracy": 0.21317442059516906,
      "num_tokens": 58404610.0,
      "step": 28145
    },
    {
      "entropy": 5.55910439491272,
      "epoch": 2.553519593613933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043482264847935953,
      "loss": 4.8724,
      "mean_token_accuracy": 0.21210339665412903,
      "num_tokens": 58414512.0,
      "step": 28150
    },
    {
      "entropy": 5.63139762878418,
      "epoch": 2.5539731494920175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043479987191997045,
      "loss": 5.0075,
      "mean_token_accuracy": 0.2129155859351158,
      "num_tokens": 58424669.0,
      "step": 28155
    },
    {
      "entropy": 5.528117895126343,
      "epoch": 2.5544267053701017,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004347770920558792,
      "loss": 4.8257,
      "mean_token_accuracy": 0.2197267085313797,
      "num_tokens": 58435348.0,
      "step": 28160
    },
    {
      "entropy": 5.594818925857544,
      "epoch": 2.5548802612481856,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043475430888755706,
      "loss": 4.908,
      "mean_token_accuracy": 0.216103295981884,
      "num_tokens": 58445534.0,
      "step": 28165
    },
    {
      "entropy": 5.5924924373626705,
      "epoch": 2.55533381712627,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004347315224154751,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21385388374328612,
      "num_tokens": 58456718.0,
      "step": 28170
    },
    {
      "entropy": 5.681323575973511,
      "epoch": 2.555787373004354,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004347087326401045,
      "loss": 4.9842,
      "mean_token_accuracy": 0.2102986067533493,
      "num_tokens": 58467176.0,
      "step": 28175
    },
    {
      "entropy": 5.568201303482056,
      "epoch": 2.5562409288824384,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043468593956191663,
      "loss": 4.8586,
      "mean_token_accuracy": 0.21857065111398696,
      "num_tokens": 58477642.0,
      "step": 28180
    },
    {
      "entropy": 5.55831356048584,
      "epoch": 2.5566944847605226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004346631431813829,
      "loss": 4.9214,
      "mean_token_accuracy": 0.20626770704984665,
      "num_tokens": 58489055.0,
      "step": 28185
    },
    {
      "entropy": 5.640730619430542,
      "epoch": 2.557148040638607,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004346403434989746,
      "loss": 4.937,
      "mean_token_accuracy": 0.2086753860116005,
      "num_tokens": 58499062.0,
      "step": 28190
    },
    {
      "entropy": 5.627087163925171,
      "epoch": 2.5576015965166907,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004346175405151633,
      "loss": 4.943,
      "mean_token_accuracy": 0.20462347567081451,
      "num_tokens": 58509154.0,
      "step": 28195
    },
    {
      "entropy": 5.546791887283325,
      "epoch": 2.558055152394775,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004345947342304207,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22026186734437941,
      "num_tokens": 58520056.0,
      "step": 28200
    },
    {
      "entropy": 5.571124839782715,
      "epoch": 2.5585087082728593,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043457192464521827,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2149498790502548,
      "num_tokens": 58529934.0,
      "step": 28205
    },
    {
      "entropy": 5.519578313827514,
      "epoch": 2.5589622641509435,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004345491117600278,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22049045860767363,
      "num_tokens": 58540013.0,
      "step": 28210
    },
    {
      "entropy": 5.643443441390991,
      "epoch": 2.5594158200290273,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043452629557532097,
      "loss": 4.9324,
      "mean_token_accuracy": 0.21330540776252746,
      "num_tokens": 58550282.0,
      "step": 28215
    },
    {
      "entropy": 5.633735847473145,
      "epoch": 2.5598693759071116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043450347609156974,
      "loss": 4.9583,
      "mean_token_accuracy": 0.2072821319103241,
      "num_tokens": 58560490.0,
      "step": 28220
    },
    {
      "entropy": 5.561642169952393,
      "epoch": 2.560322931785196,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043448065330924595,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21342907696962357,
      "num_tokens": 58570371.0,
      "step": 28225
    },
    {
      "entropy": 5.603939247131348,
      "epoch": 2.56077648766328,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004344578272288216,
      "loss": 4.8826,
      "mean_token_accuracy": 0.2158885806798935,
      "num_tokens": 58580701.0,
      "step": 28230
    },
    {
      "entropy": 5.63672366142273,
      "epoch": 2.5612300435413644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043443499785076885,
      "loss": 4.9278,
      "mean_token_accuracy": 0.21494228392839432,
      "num_tokens": 58590903.0,
      "step": 28235
    },
    {
      "entropy": 5.610977602005005,
      "epoch": 2.5616835994194487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004344121651755596,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2165874645113945,
      "num_tokens": 58600924.0,
      "step": 28240
    },
    {
      "entropy": 5.6390541076660154,
      "epoch": 2.5621371552975325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043438932920366615,
      "loss": 4.9994,
      "mean_token_accuracy": 0.21650861501693724,
      "num_tokens": 58611369.0,
      "step": 28245
    },
    {
      "entropy": 5.653045177459717,
      "epoch": 2.5625907111756168,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004343664899355608,
      "loss": 4.9917,
      "mean_token_accuracy": 0.21420637220144273,
      "num_tokens": 58622725.0,
      "step": 28250
    },
    {
      "entropy": 5.5967926502227785,
      "epoch": 2.563044267053701,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043434364737171587,
      "loss": 4.9164,
      "mean_token_accuracy": 0.21355426758527757,
      "num_tokens": 58634157.0,
      "step": 28255
    },
    {
      "entropy": 5.5868336200714115,
      "epoch": 2.5634978229317853,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043432080151260366,
      "loss": 4.8452,
      "mean_token_accuracy": 0.21530336439609526,
      "num_tokens": 58643384.0,
      "step": 28260
    },
    {
      "entropy": 5.599169397354126,
      "epoch": 2.563951378809869,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043429795235869667,
      "loss": 4.8941,
      "mean_token_accuracy": 0.21368952691555024,
      "num_tokens": 58653748.0,
      "step": 28265
    },
    {
      "entropy": 5.5993890285491945,
      "epoch": 2.5644049346879534,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043427509991046746,
      "loss": 4.9259,
      "mean_token_accuracy": 0.20760954022407532,
      "num_tokens": 58664619.0,
      "step": 28270
    },
    {
      "entropy": 5.547529888153076,
      "epoch": 2.5648584905660377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043425224416838857,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22502666413784028,
      "num_tokens": 58674506.0,
      "step": 28275
    },
    {
      "entropy": 5.619511985778809,
      "epoch": 2.565312046444122,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004342293851329327,
      "loss": 4.9788,
      "mean_token_accuracy": 0.20527592599391936,
      "num_tokens": 58686252.0,
      "step": 28280
    },
    {
      "entropy": 5.615843009948731,
      "epoch": 2.565765602322206,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043420652280457255,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2105938822031021,
      "num_tokens": 58696477.0,
      "step": 28285
    },
    {
      "entropy": 5.507663059234619,
      "epoch": 2.5662191582002905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043418365718378096,
      "loss": 4.8027,
      "mean_token_accuracy": 0.22498465329408646,
      "num_tokens": 58706664.0,
      "step": 28290
    },
    {
      "entropy": 5.50329360961914,
      "epoch": 2.5666727140783747,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004341607882710307,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21467096954584122,
      "num_tokens": 58716950.0,
      "step": 28295
    },
    {
      "entropy": 5.565147399902344,
      "epoch": 2.5671262699564585,
      "grad_norm": 1.125,
      "learning_rate": 0.0004341379160667949,
      "loss": 4.9137,
      "mean_token_accuracy": 0.21829004138708114,
      "num_tokens": 58727723.0,
      "step": 28300
    },
    {
      "entropy": 5.5678483009338375,
      "epoch": 2.567579825834543,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004341150405715464,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21011751443147658,
      "num_tokens": 58737956.0,
      "step": 28305
    },
    {
      "entropy": 5.559805011749267,
      "epoch": 2.568033381712627,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043409216178575826,
      "loss": 4.8553,
      "mean_token_accuracy": 0.2246466100215912,
      "num_tokens": 58748476.0,
      "step": 28310
    },
    {
      "entropy": 5.635240745544434,
      "epoch": 2.5684869375907113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004340692797099037,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21887056827545165,
      "num_tokens": 58759604.0,
      "step": 28315
    },
    {
      "entropy": 5.618993616104126,
      "epoch": 2.568940493468795,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004340463943444559,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2222944125533104,
      "num_tokens": 58769781.0,
      "step": 28320
    },
    {
      "entropy": 5.5820365905761715,
      "epoch": 2.5693940493468794,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004340235056898882,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21411850452423095,
      "num_tokens": 58779384.0,
      "step": 28325
    },
    {
      "entropy": 5.65414514541626,
      "epoch": 2.5698476052249637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043400061374667374,
      "loss": 5.0036,
      "mean_token_accuracy": 0.21057495027780532,
      "num_tokens": 58791953.0,
      "step": 28330
    },
    {
      "entropy": 5.560049772262573,
      "epoch": 2.570301161103048,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004339777185152862,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21515665501356124,
      "num_tokens": 58802415.0,
      "step": 28335
    },
    {
      "entropy": 5.652825021743775,
      "epoch": 2.5707547169811322,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004339548199961989,
      "loss": 4.9224,
      "mean_token_accuracy": 0.21943846046924592,
      "num_tokens": 58812582.0,
      "step": 28340
    },
    {
      "entropy": 5.588954877853394,
      "epoch": 2.5712082728592165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043393191818988543,
      "loss": 4.9685,
      "mean_token_accuracy": 0.210238978266716,
      "num_tokens": 58822987.0,
      "step": 28345
    },
    {
      "entropy": 5.553316926956176,
      "epoch": 2.5716618287373003,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004339090130968193,
      "loss": 4.8664,
      "mean_token_accuracy": 0.21653890013694763,
      "num_tokens": 58832651.0,
      "step": 28350
    },
    {
      "entropy": 5.5735753059387205,
      "epoch": 2.5721153846153846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004338861047174743,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21268133521080018,
      "num_tokens": 58842493.0,
      "step": 28355
    },
    {
      "entropy": 5.569461393356323,
      "epoch": 2.572568940493469,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004338631930523242,
      "loss": 4.8436,
      "mean_token_accuracy": 0.2204466074705124,
      "num_tokens": 58852428.0,
      "step": 28360
    },
    {
      "entropy": 5.555241394042969,
      "epoch": 2.573022496371553,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004338402781018429,
      "loss": 4.8462,
      "mean_token_accuracy": 0.22737339735031128,
      "num_tokens": 58863234.0,
      "step": 28365
    },
    {
      "entropy": 5.545900154113769,
      "epoch": 2.573476052249637,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000433817359866504,
      "loss": 4.9134,
      "mean_token_accuracy": 0.21409471929073334,
      "num_tokens": 58872632.0,
      "step": 28370
    },
    {
      "entropy": 5.642688655853272,
      "epoch": 2.573929608127721,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004337944383467817,
      "loss": 5.0817,
      "mean_token_accuracy": 0.20328803956508637,
      "num_tokens": 58882774.0,
      "step": 28375
    },
    {
      "entropy": 5.516797590255737,
      "epoch": 2.5743831640058055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043377151354314996,
      "loss": 4.8623,
      "mean_token_accuracy": 0.22080742418766022,
      "num_tokens": 58892380.0,
      "step": 28380
    },
    {
      "entropy": 5.651259136199951,
      "epoch": 2.5748367198838897,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004337485854560829,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2133295565843582,
      "num_tokens": 58902659.0,
      "step": 28385
    },
    {
      "entropy": 5.5647858619689945,
      "epoch": 2.575290275761974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004337256540860545,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2178839772939682,
      "num_tokens": 58913979.0,
      "step": 28390
    },
    {
      "entropy": 5.5853187084198,
      "epoch": 2.5757438316400583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004337027194335393,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21901330649852752,
      "num_tokens": 58924481.0,
      "step": 28395
    },
    {
      "entropy": 5.56138596534729,
      "epoch": 2.576197387518142,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043367978149901125,
      "loss": 4.9483,
      "mean_token_accuracy": 0.20727741420269014,
      "num_tokens": 58936287.0,
      "step": 28400
    },
    {
      "entropy": 5.610691165924072,
      "epoch": 2.5766509433962264,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043365684028294493,
      "loss": 4.9479,
      "mean_token_accuracy": 0.21400222182273865,
      "num_tokens": 58946070.0,
      "step": 28405
    },
    {
      "entropy": 5.63266544342041,
      "epoch": 2.5771044992743106,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043363389578581483,
      "loss": 5.0065,
      "mean_token_accuracy": 0.2023698091506958,
      "num_tokens": 58956083.0,
      "step": 28410
    },
    {
      "entropy": 5.6129203796386715,
      "epoch": 2.577558055152395,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004336109480080953,
      "loss": 4.9204,
      "mean_token_accuracy": 0.20855825990438462,
      "num_tokens": 58967557.0,
      "step": 28415
    },
    {
      "entropy": 5.5420880794525145,
      "epoch": 2.5780116110304787,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004335879969502608,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2167479306459427,
      "num_tokens": 58977509.0,
      "step": 28420
    },
    {
      "entropy": 5.491224718093872,
      "epoch": 2.578465166908563,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043356504261278625,
      "loss": 4.7702,
      "mean_token_accuracy": 0.232230943441391,
      "num_tokens": 58987498.0,
      "step": 28425
    },
    {
      "entropy": 5.640783309936523,
      "epoch": 2.5789187227866472,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004335420849961462,
      "loss": 5.0403,
      "mean_token_accuracy": 0.20119925886392592,
      "num_tokens": 58998503.0,
      "step": 28430
    },
    {
      "entropy": 5.594738912582398,
      "epoch": 2.5793722786647315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004335191241008154,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2120930328965187,
      "num_tokens": 59007384.0,
      "step": 28435
    },
    {
      "entropy": 5.664136171340942,
      "epoch": 2.5798258345428158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004334961599272687,
      "loss": 5.0121,
      "mean_token_accuracy": 0.21087125837802886,
      "num_tokens": 59017739.0,
      "step": 28440
    },
    {
      "entropy": 5.622853803634643,
      "epoch": 2.5802793904209,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004334731924759811,
      "loss": 4.927,
      "mean_token_accuracy": 0.21730037033557892,
      "num_tokens": 59028439.0,
      "step": 28445
    },
    {
      "entropy": 5.573846054077149,
      "epoch": 2.5807329462989843,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004334502217474274,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21530888825654984,
      "num_tokens": 59039250.0,
      "step": 28450
    },
    {
      "entropy": 5.642838096618652,
      "epoch": 2.581186502177068,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043342724774208277,
      "loss": 4.996,
      "mean_token_accuracy": 0.21597775667905808,
      "num_tokens": 59049375.0,
      "step": 28455
    },
    {
      "entropy": 5.678769969940186,
      "epoch": 2.5816400580551524,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043340427046042237,
      "loss": 5.0192,
      "mean_token_accuracy": 0.2115707889199257,
      "num_tokens": 59059873.0,
      "step": 28460
    },
    {
      "entropy": 5.585235214233398,
      "epoch": 2.5820936139332367,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004333812899029212,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22070954442024232,
      "num_tokens": 59069288.0,
      "step": 28465
    },
    {
      "entropy": 5.567311477661133,
      "epoch": 2.5825471698113205,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043335830607005456,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21043307334184647,
      "num_tokens": 59079339.0,
      "step": 28470
    },
    {
      "entropy": 5.602032566070557,
      "epoch": 2.5830007256894048,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004333353189622979,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21338420063257219,
      "num_tokens": 59089883.0,
      "step": 28475
    },
    {
      "entropy": 5.6487586975097654,
      "epoch": 2.583454281567489,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004333123285801264,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21646983474493026,
      "num_tokens": 59100532.0,
      "step": 28480
    },
    {
      "entropy": 5.586737251281738,
      "epoch": 2.5839078374455733,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004332893349240156,
      "loss": 4.9172,
      "mean_token_accuracy": 0.218587788939476,
      "num_tokens": 59111295.0,
      "step": 28485
    },
    {
      "entropy": 5.625041961669922,
      "epoch": 2.5843613933236576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004332663379944411,
      "loss": 4.9594,
      "mean_token_accuracy": 0.21187435686588288,
      "num_tokens": 59120808.0,
      "step": 28490
    },
    {
      "entropy": 5.677102375030517,
      "epoch": 2.584814949201742,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043324333779187844,
      "loss": 4.9752,
      "mean_token_accuracy": 0.22130003720521926,
      "num_tokens": 59131438.0,
      "step": 28495
    },
    {
      "entropy": 5.556276845932007,
      "epoch": 2.585268505079826,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043322033431680314,
      "loss": 4.9279,
      "mean_token_accuracy": 0.22077663689851762,
      "num_tokens": 59141701.0,
      "step": 28500
    },
    {
      "entropy": 5.528799057006836,
      "epoch": 2.58572206095791,
      "grad_norm": 1.203125,
      "learning_rate": 0.000433197327569691,
      "loss": 4.901,
      "mean_token_accuracy": 0.21541220396757127,
      "num_tokens": 59152184.0,
      "step": 28505
    },
    {
      "entropy": 5.566218519210816,
      "epoch": 2.586175616835994,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043317431755101785,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22377150803804396,
      "num_tokens": 59162177.0,
      "step": 28510
    },
    {
      "entropy": 5.6689772605896,
      "epoch": 2.5866291727140784,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004331513042612594,
      "loss": 5.03,
      "mean_token_accuracy": 0.20977116823196412,
      "num_tokens": 59171964.0,
      "step": 28515
    },
    {
      "entropy": 5.591834878921508,
      "epoch": 2.5870827285921627,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004331282877008918,
      "loss": 4.9081,
      "mean_token_accuracy": 0.21542945206165315,
      "num_tokens": 59182224.0,
      "step": 28520
    },
    {
      "entropy": 5.616686725616455,
      "epoch": 2.5875362844702465,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004331052678703909,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21107533425092698,
      "num_tokens": 59191912.0,
      "step": 28525
    },
    {
      "entropy": 5.595475578308106,
      "epoch": 2.587989840348331,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004330822447702327,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2153389111161232,
      "num_tokens": 59200984.0,
      "step": 28530
    },
    {
      "entropy": 5.60999903678894,
      "epoch": 2.588443396226415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043305921840089347,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21253565698862076,
      "num_tokens": 59210672.0,
      "step": 28535
    },
    {
      "entropy": 5.605475854873657,
      "epoch": 2.5888969521044993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043303618876284935,
      "loss": 4.9613,
      "mean_token_accuracy": 0.2158065229654312,
      "num_tokens": 59220230.0,
      "step": 28540
    },
    {
      "entropy": 5.537205410003662,
      "epoch": 2.5893505079825836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043301315585657645,
      "loss": 4.8657,
      "mean_token_accuracy": 0.21830154359340667,
      "num_tokens": 59230615.0,
      "step": 28545
    },
    {
      "entropy": 5.6161479473114015,
      "epoch": 2.589804063860668,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004329901196825513,
      "loss": 4.9664,
      "mean_token_accuracy": 0.21798596084117888,
      "num_tokens": 59241717.0,
      "step": 28550
    },
    {
      "entropy": 5.660571336746216,
      "epoch": 2.5902576197387517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043296708024125025,
      "loss": 4.988,
      "mean_token_accuracy": 0.21035898923873902,
      "num_tokens": 59251791.0,
      "step": 28555
    },
    {
      "entropy": 5.539387845993042,
      "epoch": 2.590711175616836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043294403753314965,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22088188230991362,
      "num_tokens": 59262434.0,
      "step": 28560
    },
    {
      "entropy": 5.56079592704773,
      "epoch": 2.59116473149492,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004329209915587261,
      "loss": 4.9901,
      "mean_token_accuracy": 0.2113431602716446,
      "num_tokens": 59272138.0,
      "step": 28565
    },
    {
      "entropy": 5.5781495571136475,
      "epoch": 2.5916182873730045,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043289794231845624,
      "loss": 4.9405,
      "mean_token_accuracy": 0.2156928926706314,
      "num_tokens": 59283470.0,
      "step": 28570
    },
    {
      "entropy": 5.572504186630249,
      "epoch": 2.5920718432510883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004328748898128168,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21307137608528137,
      "num_tokens": 59293670.0,
      "step": 28575
    },
    {
      "entropy": 5.608151197433472,
      "epoch": 2.5925253991291726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004328518340422842,
      "loss": 4.9338,
      "mean_token_accuracy": 0.21377547830343246,
      "num_tokens": 59303728.0,
      "step": 28580
    },
    {
      "entropy": 5.5668891906738285,
      "epoch": 2.592978955007257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043282877500733555,
      "loss": 4.8861,
      "mean_token_accuracy": 0.21935559064149857,
      "num_tokens": 59314271.0,
      "step": 28585
    },
    {
      "entropy": 5.563998508453369,
      "epoch": 2.593432510885341,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004328057127084476,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21345902383327484,
      "num_tokens": 59324581.0,
      "step": 28590
    },
    {
      "entropy": 5.599312162399292,
      "epoch": 2.5938860667634254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043278264714609733,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21982351988554,
      "num_tokens": 59333912.0,
      "step": 28595
    },
    {
      "entropy": 5.570941925048828,
      "epoch": 2.5943396226415096,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043275957832076173,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21508774310350418,
      "num_tokens": 59344245.0,
      "step": 28600
    },
    {
      "entropy": 5.580321502685547,
      "epoch": 2.5947931785195935,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043273650623291775,
      "loss": 4.9962,
      "mean_token_accuracy": 0.21122524738311768,
      "num_tokens": 59353825.0,
      "step": 28605
    },
    {
      "entropy": 5.575416994094849,
      "epoch": 2.5952467343976777,
      "grad_norm": 1.125,
      "learning_rate": 0.0004327134308830427,
      "loss": 4.9561,
      "mean_token_accuracy": 0.2110458046197891,
      "num_tokens": 59364901.0,
      "step": 28610
    },
    {
      "entropy": 5.545351457595825,
      "epoch": 2.595700290275762,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004326903522716137,
      "loss": 4.9008,
      "mean_token_accuracy": 0.22079697400331497,
      "num_tokens": 59374772.0,
      "step": 28615
    },
    {
      "entropy": 5.517592048645019,
      "epoch": 2.5961538461538463,
      "grad_norm": 1.25,
      "learning_rate": 0.000432667270399108,
      "loss": 4.8123,
      "mean_token_accuracy": 0.2290320485830307,
      "num_tokens": 59383640.0,
      "step": 28620
    },
    {
      "entropy": 5.606025743484497,
      "epoch": 2.59660740203193,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043264418526600305,
      "loss": 4.8846,
      "mean_token_accuracy": 0.21920408457517623,
      "num_tokens": 59392928.0,
      "step": 28625
    },
    {
      "entropy": 5.659003400802613,
      "epoch": 2.5970609579100143,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004326210968727761,
      "loss": 4.9942,
      "mean_token_accuracy": 0.2083485022187233,
      "num_tokens": 59402969.0,
      "step": 28630
    },
    {
      "entropy": 5.651481866836548,
      "epoch": 2.5975145137880986,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004325980052199047,
      "loss": 4.9963,
      "mean_token_accuracy": 0.2039766788482666,
      "num_tokens": 59413439.0,
      "step": 28635
    },
    {
      "entropy": 5.61940188407898,
      "epoch": 2.597968069666183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043257491030786644,
      "loss": 4.946,
      "mean_token_accuracy": 0.21343785971403123,
      "num_tokens": 59423403.0,
      "step": 28640
    },
    {
      "entropy": 5.5780657768249515,
      "epoch": 2.598421625544267,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043255181213713883,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20909651815891267,
      "num_tokens": 59434519.0,
      "step": 28645
    },
    {
      "entropy": 5.612778282165527,
      "epoch": 2.5988751814223514,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004325287107081996,
      "loss": 4.9495,
      "mean_token_accuracy": 0.20814028531312942,
      "num_tokens": 59445056.0,
      "step": 28650
    },
    {
      "entropy": 5.616443634033203,
      "epoch": 2.5993287373004357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004325056060215265,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21357515305280686,
      "num_tokens": 59455037.0,
      "step": 28655
    },
    {
      "entropy": 5.576290130615234,
      "epoch": 2.5997822931785195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004324824980775973,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21416040807962416,
      "num_tokens": 59465093.0,
      "step": 28660
    },
    {
      "entropy": 5.583121490478516,
      "epoch": 2.6002358490566038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043245938687688995,
      "loss": 4.9958,
      "mean_token_accuracy": 0.21132656633853913,
      "num_tokens": 59475449.0,
      "step": 28665
    },
    {
      "entropy": 5.64788408279419,
      "epoch": 2.600689404934688,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043243627241988235,
      "loss": 4.9336,
      "mean_token_accuracy": 0.21484825909137725,
      "num_tokens": 59484990.0,
      "step": 28670
    },
    {
      "entropy": 5.651201438903809,
      "epoch": 2.6011429608127723,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043241315470705254,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21332077383995057,
      "num_tokens": 59495739.0,
      "step": 28675
    },
    {
      "entropy": 5.594068717956543,
      "epoch": 2.601596516690856,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043239003373887855,
      "loss": 4.9307,
      "mean_token_accuracy": 0.215201672911644,
      "num_tokens": 59506432.0,
      "step": 28680
    },
    {
      "entropy": 5.583340787887574,
      "epoch": 2.6020500725689404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004323669095158385,
      "loss": 4.9233,
      "mean_token_accuracy": 0.21424262821674347,
      "num_tokens": 59516585.0,
      "step": 28685
    },
    {
      "entropy": 5.589278984069824,
      "epoch": 2.6025036284470247,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004323437820384107,
      "loss": 4.9229,
      "mean_token_accuracy": 0.21192097961902617,
      "num_tokens": 59526824.0,
      "step": 28690
    },
    {
      "entropy": 5.669124794006348,
      "epoch": 2.602957184325109,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004323206513070734,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2121256709098816,
      "num_tokens": 59537749.0,
      "step": 28695
    },
    {
      "entropy": 5.631524705886841,
      "epoch": 2.603410740203193,
      "grad_norm": 1.171875,
      "learning_rate": 0.000432297517322305,
      "loss": 4.9166,
      "mean_token_accuracy": 0.20218633413314818,
      "num_tokens": 59548310.0,
      "step": 28700
    },
    {
      "entropy": 5.619440507888794,
      "epoch": 2.6038642960812775,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004322743800845838,
      "loss": 4.9925,
      "mean_token_accuracy": 0.21113066375255585,
      "num_tokens": 59558191.0,
      "step": 28705
    },
    {
      "entropy": 5.618631219863891,
      "epoch": 2.6043178519593613,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043225123959438836,
      "loss": 4.9912,
      "mean_token_accuracy": 0.20114259719848632,
      "num_tokens": 59568961.0,
      "step": 28710
    },
    {
      "entropy": 5.622322368621826,
      "epoch": 2.6047714078374455,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043222809585219716,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2100427493453026,
      "num_tokens": 59579137.0,
      "step": 28715
    },
    {
      "entropy": 5.466851854324341,
      "epoch": 2.60522496371553,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043220494885848896,
      "loss": 4.7428,
      "mean_token_accuracy": 0.22778067290782927,
      "num_tokens": 59589612.0,
      "step": 28720
    },
    {
      "entropy": 5.608293294906616,
      "epoch": 2.605678519593614,
      "grad_norm": 1.125,
      "learning_rate": 0.0004321817986137423,
      "loss": 4.976,
      "mean_token_accuracy": 0.20830547958612441,
      "num_tokens": 59600332.0,
      "step": 28725
    },
    {
      "entropy": 5.523149156570435,
      "epoch": 2.606132075471698,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000432158645118436,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22856512814760208,
      "num_tokens": 59610491.0,
      "step": 28730
    },
    {
      "entropy": 5.718927383422852,
      "epoch": 2.606585631349782,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004321354883730488,
      "loss": 5.0969,
      "mean_token_accuracy": 0.19743888974189758,
      "num_tokens": 59620729.0,
      "step": 28735
    },
    {
      "entropy": 5.558250188827515,
      "epoch": 2.6070391872278664,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043211232837805973,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22068310976028443,
      "num_tokens": 59631243.0,
      "step": 28740
    },
    {
      "entropy": 5.572921371459961,
      "epoch": 2.6074927431059507,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043208916513394767,
      "loss": 4.9054,
      "mean_token_accuracy": 0.2196188360452652,
      "num_tokens": 59642055.0,
      "step": 28745
    },
    {
      "entropy": 5.654205846786499,
      "epoch": 2.607946298984035,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043206599864119164,
      "loss": 5.002,
      "mean_token_accuracy": 0.20765141397714615,
      "num_tokens": 59652144.0,
      "step": 28750
    },
    {
      "entropy": 5.617706203460694,
      "epoch": 2.6083998548621192,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043204282890027074,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22844355851411818,
      "num_tokens": 59662367.0,
      "step": 28755
    },
    {
      "entropy": 5.608437967300415,
      "epoch": 2.608853410740203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043201965591166413,
      "loss": 4.939,
      "mean_token_accuracy": 0.21596370190382003,
      "num_tokens": 59672729.0,
      "step": 28760
    },
    {
      "entropy": 5.647892761230469,
      "epoch": 2.6093069666182873,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000431996479675851,
      "loss": 4.9886,
      "mean_token_accuracy": 0.21753159463405608,
      "num_tokens": 59683282.0,
      "step": 28765
    },
    {
      "entropy": 5.655825138092041,
      "epoch": 2.6097605224963716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043197330019331074,
      "loss": 5.0321,
      "mean_token_accuracy": 0.20757057070732116,
      "num_tokens": 59694297.0,
      "step": 28770
    },
    {
      "entropy": 5.538294172286987,
      "epoch": 2.610214078374456,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043195011746452244,
      "loss": 4.8668,
      "mean_token_accuracy": 0.21508006602525712,
      "num_tokens": 59704149.0,
      "step": 28775
    },
    {
      "entropy": 5.61726484298706,
      "epoch": 2.6106676342525397,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043192693148996585,
      "loss": 4.9731,
      "mean_token_accuracy": 0.21504817754030228,
      "num_tokens": 59714811.0,
      "step": 28780
    },
    {
      "entropy": 5.525910997390747,
      "epoch": 2.611121190130624,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004319037422701203,
      "loss": 4.8922,
      "mean_token_accuracy": 0.2191385045647621,
      "num_tokens": 59724950.0,
      "step": 28785
    },
    {
      "entropy": 5.591194438934326,
      "epoch": 2.611574746008708,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004318805498054654,
      "loss": 5.0376,
      "mean_token_accuracy": 0.2009529396891594,
      "num_tokens": 59735905.0,
      "step": 28790
    },
    {
      "entropy": 5.6811747550964355,
      "epoch": 2.6120283018867925,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043185735409648077,
      "loss": 4.9835,
      "mean_token_accuracy": 0.21211819648742675,
      "num_tokens": 59746743.0,
      "step": 28795
    },
    {
      "entropy": 5.665025806427002,
      "epoch": 2.6124818577648767,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043183415514364606,
      "loss": 4.9251,
      "mean_token_accuracy": 0.21005556285381316,
      "num_tokens": 59757205.0,
      "step": 28800
    },
    {
      "entropy": 5.549183750152588,
      "epoch": 2.612935413642961,
      "grad_norm": 1.125,
      "learning_rate": 0.0004318109529474411,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21935643553733825,
      "num_tokens": 59767691.0,
      "step": 28805
    },
    {
      "entropy": 5.54490180015564,
      "epoch": 2.6133889695210453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004317877475083456,
      "loss": 4.9517,
      "mean_token_accuracy": 0.21445105969905853,
      "num_tokens": 59778618.0,
      "step": 28810
    },
    {
      "entropy": 5.634345531463623,
      "epoch": 2.613842525399129,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043176453882683955,
      "loss": 4.9736,
      "mean_token_accuracy": 0.20804711133241655,
      "num_tokens": 59789423.0,
      "step": 28815
    },
    {
      "entropy": 5.669890403747559,
      "epoch": 2.6142960812772134,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043174132690340287,
      "loss": 5.0148,
      "mean_token_accuracy": 0.20097881108522414,
      "num_tokens": 59800021.0,
      "step": 28820
    },
    {
      "entropy": 5.581205606460571,
      "epoch": 2.6147496371552976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004317181117385156,
      "loss": 4.9036,
      "mean_token_accuracy": 0.2117966428399086,
      "num_tokens": 59810006.0,
      "step": 28825
    },
    {
      "entropy": 5.610345554351807,
      "epoch": 2.6152031930333814,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043169489333265795,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21069630235433578,
      "num_tokens": 59819505.0,
      "step": 28830
    },
    {
      "entropy": 5.612638664245606,
      "epoch": 2.6156567489114657,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043167167168630977,
      "loss": 4.9811,
      "mean_token_accuracy": 0.21464820057153702,
      "num_tokens": 59830439.0,
      "step": 28835
    },
    {
      "entropy": 5.5897603034973145,
      "epoch": 2.61611030478955,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004316484467999516,
      "loss": 4.937,
      "mean_token_accuracy": 0.21509720236063004,
      "num_tokens": 59841281.0,
      "step": 28840
    },
    {
      "entropy": 5.59977502822876,
      "epoch": 2.6165638606676342,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043162521867406376,
      "loss": 4.9781,
      "mean_token_accuracy": 0.20722197592258454,
      "num_tokens": 59853581.0,
      "step": 28845
    },
    {
      "entropy": 5.690955543518067,
      "epoch": 2.6170174165457185,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004316019873091263,
      "loss": 4.9933,
      "mean_token_accuracy": 0.21022857278585433,
      "num_tokens": 59864787.0,
      "step": 28850
    },
    {
      "entropy": 5.661240768432617,
      "epoch": 2.6174709724238028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004315787527056199,
      "loss": 4.975,
      "mean_token_accuracy": 0.21226702630519867,
      "num_tokens": 59875244.0,
      "step": 28855
    },
    {
      "entropy": 5.59158730506897,
      "epoch": 2.617924528301887,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004315555148640251,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21436477154493333,
      "num_tokens": 59886535.0,
      "step": 28860
    },
    {
      "entropy": 5.611928462982178,
      "epoch": 2.618378084179971,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004315322737848222,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2149127334356308,
      "num_tokens": 59896298.0,
      "step": 28865
    },
    {
      "entropy": 5.48563928604126,
      "epoch": 2.618831640058055,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000431509029468492,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21713176965713502,
      "num_tokens": 59907010.0,
      "step": 28870
    },
    {
      "entropy": 5.626279067993164,
      "epoch": 2.6192851959361394,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004314857819155153,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22194833308458328,
      "num_tokens": 59916247.0,
      "step": 28875
    },
    {
      "entropy": 5.540341186523437,
      "epoch": 2.6197387518142237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043146253112637267,
      "loss": 4.8474,
      "mean_token_accuracy": 0.21896859258413315,
      "num_tokens": 59926680.0,
      "step": 28880
    },
    {
      "entropy": 5.528435754776001,
      "epoch": 2.6201923076923075,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043143927710154504,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22238457202911377,
      "num_tokens": 59936346.0,
      "step": 28885
    },
    {
      "entropy": 5.598195266723633,
      "epoch": 2.6206458635703918,
      "grad_norm": 1.125,
      "learning_rate": 0.0004314160198415133,
      "loss": 4.9366,
      "mean_token_accuracy": 0.22239785343408586,
      "num_tokens": 59947007.0,
      "step": 28890
    },
    {
      "entropy": 5.6324211120605465,
      "epoch": 2.621099419448476,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043139275934675844,
      "loss": 4.9737,
      "mean_token_accuracy": 0.21413871496915818,
      "num_tokens": 59957123.0,
      "step": 28895
    },
    {
      "entropy": 5.599764251708985,
      "epoch": 2.6215529753265603,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004313694956177613,
      "loss": 4.9486,
      "mean_token_accuracy": 0.21563960164785384,
      "num_tokens": 59966566.0,
      "step": 28900
    },
    {
      "entropy": 5.5931439876556395,
      "epoch": 2.6220065312046446,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043134622865500334,
      "loss": 4.9161,
      "mean_token_accuracy": 0.21438695192337037,
      "num_tokens": 59976065.0,
      "step": 28905
    },
    {
      "entropy": 5.630001449584961,
      "epoch": 2.622460087082729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004313229584589655,
      "loss": 4.924,
      "mean_token_accuracy": 0.21155037581920624,
      "num_tokens": 59986201.0,
      "step": 28910
    },
    {
      "entropy": 5.574226903915405,
      "epoch": 2.6229136429608126,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000431299685030129,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21302037090063095,
      "num_tokens": 59996574.0,
      "step": 28915
    },
    {
      "entropy": 5.616730880737305,
      "epoch": 2.623367198838897,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043127640836897516,
      "loss": 4.9454,
      "mean_token_accuracy": 0.21935513913631438,
      "num_tokens": 60006082.0,
      "step": 28920
    },
    {
      "entropy": 5.623209714889526,
      "epoch": 2.623820754716981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004312531284759854,
      "loss": 4.9337,
      "mean_token_accuracy": 0.21560416072607042,
      "num_tokens": 60015483.0,
      "step": 28925
    },
    {
      "entropy": 5.582947397232056,
      "epoch": 2.6242743105950654,
      "grad_norm": 1.515625,
      "learning_rate": 0.00043122984535164116,
      "loss": 4.8552,
      "mean_token_accuracy": 0.2278554379940033,
      "num_tokens": 60025984.0,
      "step": 28930
    },
    {
      "entropy": 5.624043226242065,
      "epoch": 2.6247278664731493,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043120655899642386,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21044308692216873,
      "num_tokens": 60036109.0,
      "step": 28935
    },
    {
      "entropy": 5.69779224395752,
      "epoch": 2.6251814223512335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043118326941081513,
      "loss": 5.0141,
      "mean_token_accuracy": 0.20475431233644487,
      "num_tokens": 60046546.0,
      "step": 28940
    },
    {
      "entropy": 5.633314371109009,
      "epoch": 2.625634978229318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004311599765952966,
      "loss": 4.9909,
      "mean_token_accuracy": 0.21793765127658843,
      "num_tokens": 60056768.0,
      "step": 28945
    },
    {
      "entropy": 5.605279207229614,
      "epoch": 2.626088534107402,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043113668055034995,
      "loss": 5.0589,
      "mean_token_accuracy": 0.20609117895364762,
      "num_tokens": 60066962.0,
      "step": 28950
    },
    {
      "entropy": 5.6159998893737795,
      "epoch": 2.6265420899854863,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004311133812764569,
      "loss": 4.9213,
      "mean_token_accuracy": 0.21738533228635787,
      "num_tokens": 60076477.0,
      "step": 28955
    },
    {
      "entropy": 5.566936492919922,
      "epoch": 2.6269956458635706,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043109007877409946,
      "loss": 4.8259,
      "mean_token_accuracy": 0.2241242468357086,
      "num_tokens": 60088060.0,
      "step": 28960
    },
    {
      "entropy": 5.600653743743896,
      "epoch": 2.6274492017416544,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043106677304375935,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21998287737369537,
      "num_tokens": 60098677.0,
      "step": 28965
    },
    {
      "entropy": 5.652439975738526,
      "epoch": 2.6279027576197387,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004310434640859186,
      "loss": 5.0124,
      "mean_token_accuracy": 0.20822567492723465,
      "num_tokens": 60109054.0,
      "step": 28970
    },
    {
      "entropy": 5.5923264503479,
      "epoch": 2.628356313497823,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004310201519010592,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2189892426133156,
      "num_tokens": 60119951.0,
      "step": 28975
    },
    {
      "entropy": 5.50391411781311,
      "epoch": 2.628809869375907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043099683648966335,
      "loss": 4.8338,
      "mean_token_accuracy": 0.21378185600042343,
      "num_tokens": 60130026.0,
      "step": 28980
    },
    {
      "entropy": 5.507849407196045,
      "epoch": 2.629263425253991,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004309735178522132,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2198483407497406,
      "num_tokens": 60140748.0,
      "step": 28985
    },
    {
      "entropy": 5.581177139282227,
      "epoch": 2.6297169811320753,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004309501959891909,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21935796737670898,
      "num_tokens": 60151042.0,
      "step": 28990
    },
    {
      "entropy": 5.652481746673584,
      "epoch": 2.6301705370101596,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043092687090107883,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2078447535634041,
      "num_tokens": 60160758.0,
      "step": 28995
    },
    {
      "entropy": 5.579534435272217,
      "epoch": 2.630624092888244,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043090354258835934,
      "loss": 4.9328,
      "mean_token_accuracy": 0.20992718189954757,
      "num_tokens": 60170249.0,
      "step": 29000
    },
    {
      "entropy": 5.539300203323364,
      "epoch": 2.631077648766328,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043088021105151483,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21028091609477997,
      "num_tokens": 60181216.0,
      "step": 29005
    },
    {
      "entropy": 5.57244462966919,
      "epoch": 2.6315312046444124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004308568762910279,
      "loss": 4.9066,
      "mean_token_accuracy": 0.213537734746933,
      "num_tokens": 60190559.0,
      "step": 29010
    },
    {
      "entropy": 5.654372119903565,
      "epoch": 2.6319847605224966,
      "grad_norm": 1.125,
      "learning_rate": 0.00043083353830738097,
      "loss": 5.0367,
      "mean_token_accuracy": 0.20308063328266143,
      "num_tokens": 60201268.0,
      "step": 29015
    },
    {
      "entropy": 5.532228899002075,
      "epoch": 2.6324383164005805,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043081019710105683,
      "loss": 4.7871,
      "mean_token_accuracy": 0.220168037712574,
      "num_tokens": 60210907.0,
      "step": 29020
    },
    {
      "entropy": 5.524450016021729,
      "epoch": 2.6328918722786647,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004307868526725381,
      "loss": 4.8358,
      "mean_token_accuracy": 0.22366890609264373,
      "num_tokens": 60220298.0,
      "step": 29025
    },
    {
      "entropy": 5.526149320602417,
      "epoch": 2.633345428156749,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004307635050223075,
      "loss": 4.888,
      "mean_token_accuracy": 0.22235353291034698,
      "num_tokens": 60230811.0,
      "step": 29030
    },
    {
      "entropy": 5.617976570129395,
      "epoch": 2.6337989840348333,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000430740154150848,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2154410481452942,
      "num_tokens": 60241211.0,
      "step": 29035
    },
    {
      "entropy": 5.545801973342895,
      "epoch": 2.634252539912917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004307168000586423,
      "loss": 4.8986,
      "mean_token_accuracy": 0.2125350221991539,
      "num_tokens": 60251506.0,
      "step": 29040
    },
    {
      "entropy": 5.545943021774292,
      "epoch": 2.6347060957910013,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004306934427461736,
      "loss": 4.862,
      "mean_token_accuracy": 0.23238665163516997,
      "num_tokens": 60261416.0,
      "step": 29045
    },
    {
      "entropy": 5.611690998077393,
      "epoch": 2.6351596516690856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004306700822139248,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21499481797218323,
      "num_tokens": 60272315.0,
      "step": 29050
    },
    {
      "entropy": 5.599742984771728,
      "epoch": 2.63561320754717,
      "grad_norm": 1.109375,
      "learning_rate": 0.000430646718462379,
      "loss": 4.9996,
      "mean_token_accuracy": 0.2130164995789528,
      "num_tokens": 60282722.0,
      "step": 29055
    },
    {
      "entropy": 5.582980632781982,
      "epoch": 2.636066763425254,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004306233514920194,
      "loss": 4.92,
      "mean_token_accuracy": 0.21766471415758132,
      "num_tokens": 60292636.0,
      "step": 29060
    },
    {
      "entropy": 5.601656532287597,
      "epoch": 2.6365203193033384,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043059998130332933,
      "loss": 4.9127,
      "mean_token_accuracy": 0.21722197085618972,
      "num_tokens": 60302979.0,
      "step": 29065
    },
    {
      "entropy": 5.616178894042969,
      "epoch": 2.6369738751814222,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043057660789679194,
      "loss": 4.9328,
      "mean_token_accuracy": 0.2176376223564148,
      "num_tokens": 60313181.0,
      "step": 29070
    },
    {
      "entropy": 5.628565979003906,
      "epoch": 2.6374274310595065,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004305532312728906,
      "loss": 4.9012,
      "mean_token_accuracy": 0.21035630851984025,
      "num_tokens": 60323401.0,
      "step": 29075
    },
    {
      "entropy": 5.64501986503601,
      "epoch": 2.6378809869375908,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004305298514321088,
      "loss": 5.004,
      "mean_token_accuracy": 0.2065522253513336,
      "num_tokens": 60333478.0,
      "step": 29080
    },
    {
      "entropy": 5.593452310562133,
      "epoch": 2.638334542815675,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043050646837493014,
      "loss": 4.8896,
      "mean_token_accuracy": 0.21438779085874557,
      "num_tokens": 60343980.0,
      "step": 29085
    },
    {
      "entropy": 5.61275691986084,
      "epoch": 2.638788098693759,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004304830821018381,
      "loss": 4.9064,
      "mean_token_accuracy": 0.21305711567401886,
      "num_tokens": 60354262.0,
      "step": 29090
    },
    {
      "entropy": 5.617759943008423,
      "epoch": 2.639241654571843,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004304596926133162,
      "loss": 4.9809,
      "mean_token_accuracy": 0.21523819863796234,
      "num_tokens": 60364137.0,
      "step": 29095
    },
    {
      "entropy": 5.584303331375122,
      "epoch": 2.6396952104499274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043043629990984823,
      "loss": 4.97,
      "mean_token_accuracy": 0.21195448487997054,
      "num_tokens": 60374528.0,
      "step": 29100
    },
    {
      "entropy": 5.585899829864502,
      "epoch": 2.6401487663280117,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043041290399191797,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22169270664453505,
      "num_tokens": 60384880.0,
      "step": 29105
    },
    {
      "entropy": 5.582402563095092,
      "epoch": 2.640602322206096,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004303895048600093,
      "loss": 4.8853,
      "mean_token_accuracy": 0.21463870406150817,
      "num_tokens": 60395543.0,
      "step": 29110
    },
    {
      "entropy": 5.630666351318359,
      "epoch": 2.64105587808418,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004303661025146061,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21815116554498673,
      "num_tokens": 60405432.0,
      "step": 29115
    },
    {
      "entropy": 5.66688985824585,
      "epoch": 2.641509433962264,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303426969561922,
      "loss": 4.9948,
      "mean_token_accuracy": 0.20774481892585756,
      "num_tokens": 60416360.0,
      "step": 29120
    },
    {
      "entropy": 5.5959725856781,
      "epoch": 2.6419629898403483,
      "grad_norm": 1.125,
      "learning_rate": 0.0004303192881852518,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2127983957529068,
      "num_tokens": 60426371.0,
      "step": 29125
    },
    {
      "entropy": 5.6134583950042725,
      "epoch": 2.6424165457184325,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043029587620226897,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21314037293195726,
      "num_tokens": 60437505.0,
      "step": 29130
    },
    {
      "entropy": 5.5950376987457275,
      "epoch": 2.642870101596517,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004302724610077278,
      "loss": 4.92,
      "mean_token_accuracy": 0.2118452623486519,
      "num_tokens": 60448170.0,
      "step": 29135
    },
    {
      "entropy": 5.640395212173462,
      "epoch": 2.6433236574746006,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004302490426021126,
      "loss": 5.007,
      "mean_token_accuracy": 0.2072097435593605,
      "num_tokens": 60458776.0,
      "step": 29140
    },
    {
      "entropy": 5.619085168838501,
      "epoch": 2.643777213352685,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004302256209859076,
      "loss": 4.9009,
      "mean_token_accuracy": 0.2223048537969589,
      "num_tokens": 60468279.0,
      "step": 29145
    },
    {
      "entropy": 5.603021097183228,
      "epoch": 2.644230769230769,
      "grad_norm": 1.125,
      "learning_rate": 0.00043020219615959727,
      "loss": 4.9948,
      "mean_token_accuracy": 0.21203717291355134,
      "num_tokens": 60478507.0,
      "step": 29150
    },
    {
      "entropy": 5.644442701339722,
      "epoch": 2.6446843251088534,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004301787681236659,
      "loss": 5.0433,
      "mean_token_accuracy": 0.20372387617826462,
      "num_tokens": 60489016.0,
      "step": 29155
    },
    {
      "entropy": 5.698155450820923,
      "epoch": 2.6451378809869377,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004301553368785981,
      "loss": 4.992,
      "mean_token_accuracy": 0.21064584255218505,
      "num_tokens": 60500169.0,
      "step": 29160
    },
    {
      "entropy": 5.585400390625,
      "epoch": 2.645591436865022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004301319024248784,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21672481000423433,
      "num_tokens": 60511109.0,
      "step": 29165
    },
    {
      "entropy": 5.574535465240478,
      "epoch": 2.6460449927431062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043010846476299145,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21943282186985016,
      "num_tokens": 60521735.0,
      "step": 29170
    },
    {
      "entropy": 5.654412269592285,
      "epoch": 2.64649854862119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004300850238934219,
      "loss": 4.9755,
      "mean_token_accuracy": 0.20859822481870652,
      "num_tokens": 60532071.0,
      "step": 29175
    },
    {
      "entropy": 5.650846529006958,
      "epoch": 2.6469521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043006157981665446,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22102375328540802,
      "num_tokens": 60541909.0,
      "step": 29180
    },
    {
      "entropy": 5.54217677116394,
      "epoch": 2.6474056603773586,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043003813253317415,
      "loss": 4.861,
      "mean_token_accuracy": 0.22088648676872252,
      "num_tokens": 60551361.0,
      "step": 29185
    },
    {
      "entropy": 5.5638352394104,
      "epoch": 2.6478592162554424,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043001468204346564,
      "loss": 4.9619,
      "mean_token_accuracy": 0.21426273882389069,
      "num_tokens": 60562112.0,
      "step": 29190
    },
    {
      "entropy": 5.6178632259368895,
      "epoch": 2.6483127721335267,
      "grad_norm": 1.25,
      "learning_rate": 0.00042999122834801406,
      "loss": 4.9631,
      "mean_token_accuracy": 0.2092098265886307,
      "num_tokens": 60572827.0,
      "step": 29195
    },
    {
      "entropy": 5.5406323909759525,
      "epoch": 2.648766328011611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004299677714473044,
      "loss": 4.8852,
      "mean_token_accuracy": 0.21783188730478287,
      "num_tokens": 60582222.0,
      "step": 29200
    },
    {
      "entropy": 5.5999448776245115,
      "epoch": 2.649219883889695,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004299443113418217,
      "loss": 4.9574,
      "mean_token_accuracy": 0.22016395926475524,
      "num_tokens": 60592943.0,
      "step": 29205
    },
    {
      "entropy": 5.633694648742676,
      "epoch": 2.6496734397677795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042992084803205124,
      "loss": 4.9397,
      "mean_token_accuracy": 0.2124394193291664,
      "num_tokens": 60603206.0,
      "step": 29210
    },
    {
      "entropy": 5.572782373428344,
      "epoch": 2.6501269956458637,
      "grad_norm": 1.21875,
      "learning_rate": 0.000429897381518478,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22109796106815338,
      "num_tokens": 60614776.0,
      "step": 29215
    },
    {
      "entropy": 5.571521139144897,
      "epoch": 2.650580551523948,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042987391180158757,
      "loss": 4.904,
      "mean_token_accuracy": 0.21229830980300904,
      "num_tokens": 60625914.0,
      "step": 29220
    },
    {
      "entropy": 5.559616374969482,
      "epoch": 2.651034107402032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042985043888186516,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21735753864049911,
      "num_tokens": 60636934.0,
      "step": 29225
    },
    {
      "entropy": 5.591310977935791,
      "epoch": 2.651487663280116,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004298269627597962,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2102430447936058,
      "num_tokens": 60647707.0,
      "step": 29230
    },
    {
      "entropy": 5.52634449005127,
      "epoch": 2.6519412191582004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042980348343586624,
      "loss": 4.8767,
      "mean_token_accuracy": 0.21380297243595123,
      "num_tokens": 60658691.0,
      "step": 29235
    },
    {
      "entropy": 5.611511945724487,
      "epoch": 2.6523947750362846,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042978000091056074,
      "loss": 4.9736,
      "mean_token_accuracy": 0.2099282130599022,
      "num_tokens": 60668479.0,
      "step": 29240
    },
    {
      "entropy": 5.585924673080444,
      "epoch": 2.6528483309143684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004297565151843653,
      "loss": 4.8305,
      "mean_token_accuracy": 0.21798519045114517,
      "num_tokens": 60679158.0,
      "step": 29245
    },
    {
      "entropy": 5.641716337203979,
      "epoch": 2.6533018867924527,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004297330262577659,
      "loss": 5.0657,
      "mean_token_accuracy": 0.20683201551437377,
      "num_tokens": 60689363.0,
      "step": 29250
    },
    {
      "entropy": 5.580539608001709,
      "epoch": 2.653755442670537,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004297095341312479,
      "loss": 4.824,
      "mean_token_accuracy": 0.2192675366997719,
      "num_tokens": 60699650.0,
      "step": 29255
    },
    {
      "entropy": 5.554980659484864,
      "epoch": 2.6542089985486212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004296860388052974,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2142020046710968,
      "num_tokens": 60709921.0,
      "step": 29260
    },
    {
      "entropy": 5.593706369400024,
      "epoch": 2.6546625544267055,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042966254028040015,
      "loss": 4.8799,
      "mean_token_accuracy": 0.2268610790371895,
      "num_tokens": 60720112.0,
      "step": 29265
    },
    {
      "entropy": 5.557190990447998,
      "epoch": 2.65511611030479,
      "grad_norm": 1.25,
      "learning_rate": 0.00042963903855704216,
      "loss": 4.8611,
      "mean_token_accuracy": 0.21162960231304168,
      "num_tokens": 60730604.0,
      "step": 29270
    },
    {
      "entropy": 5.562040853500366,
      "epoch": 2.6555696661828736,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042961553363570946,
      "loss": 4.893,
      "mean_token_accuracy": 0.21914690136909484,
      "num_tokens": 60740896.0,
      "step": 29275
    },
    {
      "entropy": 5.609918928146362,
      "epoch": 2.656023222060958,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004295920255168881,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21599795818328857,
      "num_tokens": 60751771.0,
      "step": 29280
    },
    {
      "entropy": 5.655177879333496,
      "epoch": 2.656476777939042,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042956851420106423,
      "loss": 4.9728,
      "mean_token_accuracy": 0.20803488045930862,
      "num_tokens": 60762802.0,
      "step": 29285
    },
    {
      "entropy": 5.560130453109741,
      "epoch": 2.6569303338171264,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004295449996887242,
      "loss": 4.8557,
      "mean_token_accuracy": 0.21522625982761384,
      "num_tokens": 60774102.0,
      "step": 29290
    },
    {
      "entropy": 5.587153816223145,
      "epoch": 2.65738388969521,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004295214819803541,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21435084491968154,
      "num_tokens": 60785383.0,
      "step": 29295
    },
    {
      "entropy": 5.5969987392425535,
      "epoch": 2.6578374455732945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042949796107644053,
      "loss": 4.919,
      "mean_token_accuracy": 0.2216167777776718,
      "num_tokens": 60795173.0,
      "step": 29300
    },
    {
      "entropy": 5.600811243057251,
      "epoch": 2.6582910014513788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042947443697746956,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22251317799091339,
      "num_tokens": 60805621.0,
      "step": 29305
    },
    {
      "entropy": 5.621386384963989,
      "epoch": 2.658744557329463,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004294509096839279,
      "loss": 5.0362,
      "mean_token_accuracy": 0.20104185342788697,
      "num_tokens": 60816799.0,
      "step": 29310
    },
    {
      "entropy": 5.514614009857178,
      "epoch": 2.6591981132075473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042942737919630215,
      "loss": 4.8677,
      "mean_token_accuracy": 0.2156401827931404,
      "num_tokens": 60826827.0,
      "step": 29315
    },
    {
      "entropy": 5.614333868026733,
      "epoch": 2.6596516690856316,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042940384551507876,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21849855929613113,
      "num_tokens": 60836567.0,
      "step": 29320
    },
    {
      "entropy": 5.6320164680480955,
      "epoch": 2.6601052249637154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004293803086407447,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21504542976617813,
      "num_tokens": 60846723.0,
      "step": 29325
    },
    {
      "entropy": 5.655851411819458,
      "epoch": 2.6605587808417996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004293567685737863,
      "loss": 5.0267,
      "mean_token_accuracy": 0.2074614018201828,
      "num_tokens": 60857434.0,
      "step": 29330
    },
    {
      "entropy": 5.640121030807495,
      "epoch": 2.661012336719884,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042933322531469074,
      "loss": 4.975,
      "mean_token_accuracy": 0.21061351746320725,
      "num_tokens": 60868780.0,
      "step": 29335
    },
    {
      "entropy": 5.679599332809448,
      "epoch": 2.661465892597968,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042930967886394456,
      "loss": 4.973,
      "mean_token_accuracy": 0.2145544707775116,
      "num_tokens": 60879037.0,
      "step": 29340
    },
    {
      "entropy": 5.543292665481568,
      "epoch": 2.661919448476052,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000429286129222035,
      "loss": 4.92,
      "mean_token_accuracy": 0.21262886375188828,
      "num_tokens": 60889543.0,
      "step": 29345
    },
    {
      "entropy": 5.549352979660034,
      "epoch": 2.6623730043541363,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000429262576389449,
      "loss": 4.9031,
      "mean_token_accuracy": 0.208954057097435,
      "num_tokens": 60900141.0,
      "step": 29350
    },
    {
      "entropy": 5.6620992660522464,
      "epoch": 2.6628265602322205,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004292390203666736,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21727977842092513,
      "num_tokens": 60910244.0,
      "step": 29355
    },
    {
      "entropy": 5.567185878753662,
      "epoch": 2.663280116110305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000429215461154196,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22440990954637527,
      "num_tokens": 60919908.0,
      "step": 29360
    },
    {
      "entropy": 5.607943105697632,
      "epoch": 2.663733671988389,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042919189875250324,
      "loss": 4.9362,
      "mean_token_accuracy": 0.21307821124792098,
      "num_tokens": 60929096.0,
      "step": 29365
    },
    {
      "entropy": 5.600187015533447,
      "epoch": 2.6641872278664733,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042916833316208284,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21086373329162597,
      "num_tokens": 60939346.0,
      "step": 29370
    },
    {
      "entropy": 5.592050838470459,
      "epoch": 2.6646407837445576,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004291447643834219,
      "loss": 4.932,
      "mean_token_accuracy": 0.2121726617217064,
      "num_tokens": 60950364.0,
      "step": 29375
    },
    {
      "entropy": 5.581553411483765,
      "epoch": 2.6650943396226414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000429121192417008,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22185025066137315,
      "num_tokens": 60960905.0,
      "step": 29380
    },
    {
      "entropy": 5.639850807189942,
      "epoch": 2.6655478955007257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042909761726332864,
      "loss": 4.952,
      "mean_token_accuracy": 0.21890058368444443,
      "num_tokens": 60971281.0,
      "step": 29385
    },
    {
      "entropy": 5.59586091041565,
      "epoch": 2.66600145137881,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004290740389228712,
      "loss": 4.9028,
      "mean_token_accuracy": 0.21439144909381866,
      "num_tokens": 60982522.0,
      "step": 29390
    },
    {
      "entropy": 5.600847148895264,
      "epoch": 2.6664550072568938,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042905045739612347,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21046410202980043,
      "num_tokens": 60992738.0,
      "step": 29395
    },
    {
      "entropy": 5.672179460525513,
      "epoch": 2.666908563134978,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042902687268357297,
      "loss": 5.047,
      "mean_token_accuracy": 0.21110555678606033,
      "num_tokens": 61003665.0,
      "step": 29400
    },
    {
      "entropy": 5.721938371658325,
      "epoch": 2.6673621190130623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042900328478570745,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20804743319749833,
      "num_tokens": 61014663.0,
      "step": 29405
    },
    {
      "entropy": 5.63833589553833,
      "epoch": 2.6678156748911466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004289796937030148,
      "loss": 4.852,
      "mean_token_accuracy": 0.21949941515922547,
      "num_tokens": 61024916.0,
      "step": 29410
    },
    {
      "entropy": 5.508666467666626,
      "epoch": 2.668269230769231,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042895609943598286,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2158627212047577,
      "num_tokens": 61035043.0,
      "step": 29415
    },
    {
      "entropy": 5.574836874008179,
      "epoch": 2.668722786647315,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004289325019850995,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21113785803318025,
      "num_tokens": 61046153.0,
      "step": 29420
    },
    {
      "entropy": 5.640136671066284,
      "epoch": 2.6691763425253994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004289089013508528,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2173328846693039,
      "num_tokens": 61057009.0,
      "step": 29425
    },
    {
      "entropy": 5.659178018569946,
      "epoch": 2.669629898403483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004288852975337308,
      "loss": 5.0459,
      "mean_token_accuracy": 0.2037871852517128,
      "num_tokens": 61068514.0,
      "step": 29430
    },
    {
      "entropy": 5.646041584014893,
      "epoch": 2.6700834542815675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042886169053422165,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21725002974271773,
      "num_tokens": 61078902.0,
      "step": 29435
    },
    {
      "entropy": 5.615032815933228,
      "epoch": 2.6705370101596517,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004288380803528135,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21771776527166367,
      "num_tokens": 61089216.0,
      "step": 29440
    },
    {
      "entropy": 5.569168138504028,
      "epoch": 2.670990566037736,
      "grad_norm": 1.125,
      "learning_rate": 0.00042881446698999465,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22087444961071015,
      "num_tokens": 61098470.0,
      "step": 29445
    },
    {
      "entropy": 5.702025556564331,
      "epoch": 2.67144412191582,
      "grad_norm": 1.25,
      "learning_rate": 0.0004287908504462534,
      "loss": 5.0628,
      "mean_token_accuracy": 0.21940363049507142,
      "num_tokens": 61108694.0,
      "step": 29450
    },
    {
      "entropy": 5.577374601364136,
      "epoch": 2.671897677793904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042876723072207825,
      "loss": 4.9223,
      "mean_token_accuracy": 0.21640316098928453,
      "num_tokens": 61119389.0,
      "step": 29455
    },
    {
      "entropy": 5.662784910202026,
      "epoch": 2.6723512336719883,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042874360781795757,
      "loss": 4.9927,
      "mean_token_accuracy": 0.20761950463056564,
      "num_tokens": 61130491.0,
      "step": 29460
    },
    {
      "entropy": 5.6134308815002445,
      "epoch": 2.6728047895500726,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042871998173437987,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21356953233480452,
      "num_tokens": 61140404.0,
      "step": 29465
    },
    {
      "entropy": 5.659073734283448,
      "epoch": 2.673258345428157,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042869635247183373,
      "loss": 5.0109,
      "mean_token_accuracy": 0.21197839230298995,
      "num_tokens": 61150777.0,
      "step": 29470
    },
    {
      "entropy": 5.616831302642822,
      "epoch": 2.673711901306241,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042867272003080793,
      "loss": 4.9278,
      "mean_token_accuracy": 0.2129394754767418,
      "num_tokens": 61160685.0,
      "step": 29475
    },
    {
      "entropy": 5.623648309707642,
      "epoch": 2.674165457184325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004286490844117911,
      "loss": 4.9498,
      "mean_token_accuracy": 0.21064317375421523,
      "num_tokens": 61170553.0,
      "step": 29480
    },
    {
      "entropy": 5.66174168586731,
      "epoch": 2.6746190130624092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042862544561527203,
      "loss": 4.97,
      "mean_token_accuracy": 0.21203428953886033,
      "num_tokens": 61181396.0,
      "step": 29485
    },
    {
      "entropy": 5.600545644760132,
      "epoch": 2.6750725689404935,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286018036417396,
      "loss": 4.9606,
      "mean_token_accuracy": 0.2145301043987274,
      "num_tokens": 61192315.0,
      "step": 29490
    },
    {
      "entropy": 5.619437599182129,
      "epoch": 2.6755261248185778,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042857815849168277,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21751265227794647,
      "num_tokens": 61202719.0,
      "step": 29495
    },
    {
      "entropy": 5.528173351287842,
      "epoch": 2.6759796806966616,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004285545101655904,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22230553925037383,
      "num_tokens": 61212703.0,
      "step": 29500
    },
    {
      "entropy": 5.511766719818115,
      "epoch": 2.676433236574746,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004285308586639517,
      "loss": 4.7835,
      "mean_token_accuracy": 0.22890749722719192,
      "num_tokens": 61222406.0,
      "step": 29505
    },
    {
      "entropy": 5.589002370834351,
      "epoch": 2.67688679245283,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042850720398725575,
      "loss": 5.0071,
      "mean_token_accuracy": 0.20766305178403854,
      "num_tokens": 61232758.0,
      "step": 29510
    },
    {
      "entropy": 5.550859069824218,
      "epoch": 2.6773403483309144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042848354613599164,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21874978691339492,
      "num_tokens": 61243715.0,
      "step": 29515
    },
    {
      "entropy": 5.648174571990967,
      "epoch": 2.6777939042089987,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042845988511064863,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21885040700435637,
      "num_tokens": 61254669.0,
      "step": 29520
    },
    {
      "entropy": 5.6080406665802,
      "epoch": 2.678247460087083,
      "grad_norm": 1.125,
      "learning_rate": 0.00042843622091171625,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21830051988363267,
      "num_tokens": 61264377.0,
      "step": 29525
    },
    {
      "entropy": 5.561658477783203,
      "epoch": 2.6787010159651667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004284125535396836,
      "loss": 4.9179,
      "mean_token_accuracy": 0.21624565422534942,
      "num_tokens": 61275348.0,
      "step": 29530
    },
    {
      "entropy": 5.548919153213501,
      "epoch": 2.679154571843251,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042838888299504024,
      "loss": 4.8155,
      "mean_token_accuracy": 0.225259667634964,
      "num_tokens": 61285579.0,
      "step": 29535
    },
    {
      "entropy": 5.566143321990967,
      "epoch": 2.6796081277213353,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042836520927827575,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2171211928129196,
      "num_tokens": 61296037.0,
      "step": 29540
    },
    {
      "entropy": 5.585876083374023,
      "epoch": 2.6800616835994195,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004283415323898797,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2230330526828766,
      "num_tokens": 61305882.0,
      "step": 29545
    },
    {
      "entropy": 5.564358282089233,
      "epoch": 2.6805152394775034,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042831785233034156,
      "loss": 4.8477,
      "mean_token_accuracy": 0.2230583757162094,
      "num_tokens": 61317117.0,
      "step": 29550
    },
    {
      "entropy": 5.636760807037353,
      "epoch": 2.6809687953555876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042829416910015116,
      "loss": 4.9382,
      "mean_token_accuracy": 0.2105455443263054,
      "num_tokens": 61328190.0,
      "step": 29555
    },
    {
      "entropy": 5.582792806625366,
      "epoch": 2.681422351233672,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042827048269979834,
      "loss": 4.9435,
      "mean_token_accuracy": 0.21554917246103286,
      "num_tokens": 61337978.0,
      "step": 29560
    },
    {
      "entropy": 5.662825441360473,
      "epoch": 2.681875907111756,
      "grad_norm": 1.125,
      "learning_rate": 0.0004282467931297729,
      "loss": 4.9894,
      "mean_token_accuracy": 0.20879105627536773,
      "num_tokens": 61348523.0,
      "step": 29565
    },
    {
      "entropy": 5.552019214630127,
      "epoch": 2.6823294629898404,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004282231003905647,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2270594909787178,
      "num_tokens": 61359250.0,
      "step": 29570
    },
    {
      "entropy": 5.542756128311157,
      "epoch": 2.6827830188679247,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042819940448266376,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2201579749584198,
      "num_tokens": 61369999.0,
      "step": 29575
    },
    {
      "entropy": 5.665325546264649,
      "epoch": 2.683236574746009,
      "grad_norm": 1.125,
      "learning_rate": 0.00042817570540656003,
      "loss": 4.9713,
      "mean_token_accuracy": 0.20968573838472365,
      "num_tokens": 61380279.0,
      "step": 29580
    },
    {
      "entropy": 5.556784152984619,
      "epoch": 2.683690130624093,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004281520031627438,
      "loss": 4.919,
      "mean_token_accuracy": 0.2241036683320999,
      "num_tokens": 61389794.0,
      "step": 29585
    },
    {
      "entropy": 5.640665197372437,
      "epoch": 2.684143686502177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042812829775170497,
      "loss": 5.0329,
      "mean_token_accuracy": 0.20681362301111222,
      "num_tokens": 61401136.0,
      "step": 29590
    },
    {
      "entropy": 5.610365676879883,
      "epoch": 2.6845972423802613,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042810458917393404,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21376110315322877,
      "num_tokens": 61412830.0,
      "step": 29595
    },
    {
      "entropy": 5.608079862594605,
      "epoch": 2.6850507982583456,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004280808774299211,
      "loss": 4.8713,
      "mean_token_accuracy": 0.22550671994686128,
      "num_tokens": 61422168.0,
      "step": 29600
    },
    {
      "entropy": 5.556282091140747,
      "epoch": 2.6855043541364294,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042805716252015665,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21926275342702867,
      "num_tokens": 61432471.0,
      "step": 29605
    },
    {
      "entropy": 5.595130729675293,
      "epoch": 2.6859579100145137,
      "grad_norm": 1.25,
      "learning_rate": 0.00042803344444513106,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22124001532793044,
      "num_tokens": 61443402.0,
      "step": 29610
    },
    {
      "entropy": 5.647167491912842,
      "epoch": 2.686411465892598,
      "grad_norm": 1.125,
      "learning_rate": 0.0004280097232053349,
      "loss": 4.8981,
      "mean_token_accuracy": 0.21352253556251527,
      "num_tokens": 61454465.0,
      "step": 29615
    },
    {
      "entropy": 5.64017162322998,
      "epoch": 2.686865021770682,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004279859988012586,
      "loss": 4.9366,
      "mean_token_accuracy": 0.2160960078239441,
      "num_tokens": 61464671.0,
      "step": 29620
    },
    {
      "entropy": 5.583374547958374,
      "epoch": 2.6873185776487665,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004279622712333929,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21660369038581848,
      "num_tokens": 61474940.0,
      "step": 29625
    },
    {
      "entropy": 5.622679710388184,
      "epoch": 2.6877721335268507,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004279385405022284,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21151655912399292,
      "num_tokens": 61485240.0,
      "step": 29630
    },
    {
      "entropy": 5.595038795471192,
      "epoch": 2.6882256894049346,
      "grad_norm": 1.140625,
      "learning_rate": 0.000427914806608256,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22014024257659912,
      "num_tokens": 61495382.0,
      "step": 29635
    },
    {
      "entropy": 5.573564624786377,
      "epoch": 2.688679245283019,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004278910695519663,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22232517302036287,
      "num_tokens": 61505372.0,
      "step": 29640
    },
    {
      "entropy": 5.610299301147461,
      "epoch": 2.689132801161103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042786732933385046,
      "loss": 4.912,
      "mean_token_accuracy": 0.21934127658605576,
      "num_tokens": 61516039.0,
      "step": 29645
    },
    {
      "entropy": 5.617102432250976,
      "epoch": 2.6895863570391874,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004278435859543992,
      "loss": 4.9845,
      "mean_token_accuracy": 0.21144675463438034,
      "num_tokens": 61525435.0,
      "step": 29650
    },
    {
      "entropy": 5.669988012313842,
      "epoch": 2.690039912917271,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004278198394141037,
      "loss": 4.9093,
      "mean_token_accuracy": 0.20882124602794647,
      "num_tokens": 61534714.0,
      "step": 29655
    },
    {
      "entropy": 5.690169620513916,
      "epoch": 2.6904934687953554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042779608971345496,
      "loss": 5.073,
      "mean_token_accuracy": 0.20781083256006241,
      "num_tokens": 61545841.0,
      "step": 29660
    },
    {
      "entropy": 5.57533745765686,
      "epoch": 2.6909470246734397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042777233685294407,
      "loss": 4.8455,
      "mean_token_accuracy": 0.22460610121488572,
      "num_tokens": 61555355.0,
      "step": 29665
    },
    {
      "entropy": 5.602438116073609,
      "epoch": 2.691400580551524,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042774858083306246,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2140805557370186,
      "num_tokens": 61565474.0,
      "step": 29670
    },
    {
      "entropy": 5.6092884063720705,
      "epoch": 2.6918541364296082,
      "grad_norm": 1.125,
      "learning_rate": 0.00042772482165430117,
      "loss": 4.863,
      "mean_token_accuracy": 0.223926542699337,
      "num_tokens": 61575631.0,
      "step": 29675
    },
    {
      "entropy": 5.566700077056884,
      "epoch": 2.6923076923076925,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042770105931715167,
      "loss": 4.914,
      "mean_token_accuracy": 0.2195332646369934,
      "num_tokens": 61586398.0,
      "step": 29680
    },
    {
      "entropy": 5.566110372543335,
      "epoch": 2.6927612481857763,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004276772938221054,
      "loss": 4.823,
      "mean_token_accuracy": 0.22573158740997315,
      "num_tokens": 61596630.0,
      "step": 29685
    },
    {
      "entropy": 5.611611366271973,
      "epoch": 2.6932148040638606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004276535251696537,
      "loss": 4.9612,
      "mean_token_accuracy": 0.21690455824136734,
      "num_tokens": 61605965.0,
      "step": 29690
    },
    {
      "entropy": 5.634567451477051,
      "epoch": 2.693668359941945,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004276297533602883,
      "loss": 4.9822,
      "mean_token_accuracy": 0.21563731878995895,
      "num_tokens": 61616973.0,
      "step": 29695
    },
    {
      "entropy": 5.679215717315674,
      "epoch": 2.694121915820029,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042760597839450057,
      "loss": 4.9602,
      "mean_token_accuracy": 0.21137825697660445,
      "num_tokens": 61627420.0,
      "step": 29700
    },
    {
      "entropy": 5.611460113525391,
      "epoch": 2.694575471698113,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004275822002727824,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2199986085295677,
      "num_tokens": 61638315.0,
      "step": 29705
    },
    {
      "entropy": 5.587549161911011,
      "epoch": 2.6950290275761972,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042755841899562544,
      "loss": 4.9165,
      "mean_token_accuracy": 0.2215933233499527,
      "num_tokens": 61649294.0,
      "step": 29710
    },
    {
      "entropy": 5.653368949890137,
      "epoch": 2.6954825834542815,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042753463456352155,
      "loss": 4.9711,
      "mean_token_accuracy": 0.21183942556381224,
      "num_tokens": 61660312.0,
      "step": 29715
    },
    {
      "entropy": 5.581273889541626,
      "epoch": 2.6959361393323658,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004275108469769625,
      "loss": 4.9317,
      "mean_token_accuracy": 0.21674729585647584,
      "num_tokens": 61670539.0,
      "step": 29720
    },
    {
      "entropy": 5.626755475997925,
      "epoch": 2.69638969521045,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042748705623644025,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2196592941880226,
      "num_tokens": 61679613.0,
      "step": 29725
    },
    {
      "entropy": 5.648359537124634,
      "epoch": 2.6968432510885343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004274632623424468,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21355932503938674,
      "num_tokens": 61689714.0,
      "step": 29730
    },
    {
      "entropy": 5.664665746688843,
      "epoch": 2.6972968069666186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042743946529547435,
      "loss": 5.0025,
      "mean_token_accuracy": 0.21372633129358293,
      "num_tokens": 61700701.0,
      "step": 29735
    },
    {
      "entropy": 5.6268946647644045,
      "epoch": 2.6977503628447024,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004274156650960148,
      "loss": 4.9762,
      "mean_token_accuracy": 0.21446214020252227,
      "num_tokens": 61711403.0,
      "step": 29740
    },
    {
      "entropy": 5.505466794967651,
      "epoch": 2.6982039187227866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004273918617445605,
      "loss": 4.7848,
      "mean_token_accuracy": 0.2230610355734825,
      "num_tokens": 61721787.0,
      "step": 29745
    },
    {
      "entropy": 5.682938814163208,
      "epoch": 2.698657474600871,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042736805524160363,
      "loss": 5.017,
      "mean_token_accuracy": 0.2149973765015602,
      "num_tokens": 61732702.0,
      "step": 29750
    },
    {
      "entropy": 5.570976400375367,
      "epoch": 2.6991110304789547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042734424558763656,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2231902226805687,
      "num_tokens": 61743477.0,
      "step": 29755
    },
    {
      "entropy": 5.61804428100586,
      "epoch": 2.699564586357039,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004273204327831516,
      "loss": 4.955,
      "mean_token_accuracy": 0.21172530204057693,
      "num_tokens": 61753620.0,
      "step": 29760
    },
    {
      "entropy": 5.650320148468017,
      "epoch": 2.7000181422351233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004272966168286414,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22440818697214127,
      "num_tokens": 61763678.0,
      "step": 29765
    },
    {
      "entropy": 5.6596893787384035,
      "epoch": 2.7004716981132075,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042727279772459837,
      "loss": 4.9567,
      "mean_token_accuracy": 0.2176760882139206,
      "num_tokens": 61774419.0,
      "step": 29770
    },
    {
      "entropy": 5.64282546043396,
      "epoch": 2.700925253991292,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000427248975471515,
      "loss": 4.9864,
      "mean_token_accuracy": 0.2102653369307518,
      "num_tokens": 61785546.0,
      "step": 29775
    },
    {
      "entropy": 5.624925947189331,
      "epoch": 2.701378809869376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042722515006988397,
      "loss": 4.8919,
      "mean_token_accuracy": 0.21678797900676727,
      "num_tokens": 61795846.0,
      "step": 29780
    },
    {
      "entropy": 5.608879089355469,
      "epoch": 2.7018323657474603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004272013215201982,
      "loss": 4.9484,
      "mean_token_accuracy": 0.2091482624411583,
      "num_tokens": 61806440.0,
      "step": 29785
    },
    {
      "entropy": 5.599849843978882,
      "epoch": 2.702285921625544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042717748982295015,
      "loss": 4.9074,
      "mean_token_accuracy": 0.21571664065122603,
      "num_tokens": 61815371.0,
      "step": 29790
    },
    {
      "entropy": 5.672802591323853,
      "epoch": 2.7027394775036284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004271536549786329,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22120023369789124,
      "num_tokens": 61825225.0,
      "step": 29795
    },
    {
      "entropy": 5.614306402206421,
      "epoch": 2.7031930333817127,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004271298169877393,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2165918156504631,
      "num_tokens": 61836705.0,
      "step": 29800
    },
    {
      "entropy": 5.5905663013458256,
      "epoch": 2.703646589259797,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004271059758507624,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21777166277170182,
      "num_tokens": 61847082.0,
      "step": 29805
    },
    {
      "entropy": 5.618663311004639,
      "epoch": 2.7041001451378808,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042708213156819505,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21548520475625993,
      "num_tokens": 61857130.0,
      "step": 29810
    },
    {
      "entropy": 5.677494621276855,
      "epoch": 2.704553701015965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004270582841405305,
      "loss": 5.0051,
      "mean_token_accuracy": 0.21440144181251525,
      "num_tokens": 61866796.0,
      "step": 29815
    },
    {
      "entropy": 5.642102766036987,
      "epoch": 2.7050072568940493,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004270344335682618,
      "loss": 4.9645,
      "mean_token_accuracy": 0.20686332434415816,
      "num_tokens": 61876356.0,
      "step": 29820
    },
    {
      "entropy": 5.614577007293701,
      "epoch": 2.7054608127721336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042701057985188235,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2088135987520218,
      "num_tokens": 61887374.0,
      "step": 29825
    },
    {
      "entropy": 5.658870077133178,
      "epoch": 2.705914368650218,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004269867229918854,
      "loss": 5.0005,
      "mean_token_accuracy": 0.2139015957713127,
      "num_tokens": 61897376.0,
      "step": 29830
    },
    {
      "entropy": 5.682542610168457,
      "epoch": 2.706367924528302,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042696286298876424,
      "loss": 5.0151,
      "mean_token_accuracy": 0.20650754570961,
      "num_tokens": 61908212.0,
      "step": 29835
    },
    {
      "entropy": 5.648695373535157,
      "epoch": 2.706821480406386,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004269389998430124,
      "loss": 5.0371,
      "mean_token_accuracy": 0.20157751739025115,
      "num_tokens": 61920699.0,
      "step": 29840
    },
    {
      "entropy": 5.609693098068237,
      "epoch": 2.70727503628447,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004269151335551233,
      "loss": 4.962,
      "mean_token_accuracy": 0.21183295398950577,
      "num_tokens": 61931556.0,
      "step": 29845
    },
    {
      "entropy": 5.726107025146485,
      "epoch": 2.7077285921625545,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004268912641255905,
      "loss": 5.0616,
      "mean_token_accuracy": 0.21557562202215194,
      "num_tokens": 61941483.0,
      "step": 29850
    },
    {
      "entropy": 5.686029672622681,
      "epoch": 2.7081821480406387,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042686739155490763,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2136765629053116,
      "num_tokens": 61951262.0,
      "step": 29855
    },
    {
      "entropy": 5.652771139144898,
      "epoch": 2.7086357039187225,
      "grad_norm": 1.125,
      "learning_rate": 0.0004268435158435684,
      "loss": 4.9892,
      "mean_token_accuracy": 0.2087709352374077,
      "num_tokens": 61962423.0,
      "step": 29860
    },
    {
      "entropy": 5.6843337059021,
      "epoch": 2.709089259796807,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042681963699206667,
      "loss": 5.0543,
      "mean_token_accuracy": 0.2027088612318039,
      "num_tokens": 61973276.0,
      "step": 29865
    },
    {
      "entropy": 5.596109914779663,
      "epoch": 2.709542815674891,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042679575500089605,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2200896143913269,
      "num_tokens": 61984369.0,
      "step": 29870
    },
    {
      "entropy": 5.573590803146362,
      "epoch": 2.7099963715529753,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004267718698705506,
      "loss": 4.8706,
      "mean_token_accuracy": 0.21173152476549148,
      "num_tokens": 61994224.0,
      "step": 29875
    },
    {
      "entropy": 5.599390983581543,
      "epoch": 2.7104499274310596,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004267479816015242,
      "loss": 4.9461,
      "mean_token_accuracy": 0.21422012895345688,
      "num_tokens": 62004546.0,
      "step": 29880
    },
    {
      "entropy": 5.674924707412719,
      "epoch": 2.710903483309144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004267240901943109,
      "loss": 4.9533,
      "mean_token_accuracy": 0.21763171702623368,
      "num_tokens": 62013924.0,
      "step": 29885
    },
    {
      "entropy": 5.622979307174683,
      "epoch": 2.7113570391872277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042670019564940474,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21537649780511856,
      "num_tokens": 62026150.0,
      "step": 29890
    },
    {
      "entropy": 5.63336820602417,
      "epoch": 2.711810595065312,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004266762979672998,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21427132040262223,
      "num_tokens": 62036203.0,
      "step": 29895
    },
    {
      "entropy": 5.612219333648682,
      "epoch": 2.7122641509433962,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042665239714849044,
      "loss": 4.9338,
      "mean_token_accuracy": 0.21643629521131516,
      "num_tokens": 62046961.0,
      "step": 29900
    },
    {
      "entropy": 5.547346401214599,
      "epoch": 2.7127177068214805,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042662849319347073,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21425059884786607,
      "num_tokens": 62057365.0,
      "step": 29905
    },
    {
      "entropy": 5.587605142593384,
      "epoch": 2.7131712626995643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042660458610273527,
      "loss": 4.8949,
      "mean_token_accuracy": 0.21007274836301804,
      "num_tokens": 62067934.0,
      "step": 29910
    },
    {
      "entropy": 5.692172574996948,
      "epoch": 2.7136248185776486,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004265806758767783,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2105989024043083,
      "num_tokens": 62077203.0,
      "step": 29915
    },
    {
      "entropy": 5.651901721954346,
      "epoch": 2.714078374455733,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042655676251609435,
      "loss": 4.9227,
      "mean_token_accuracy": 0.217285817861557,
      "num_tokens": 62087262.0,
      "step": 29920
    },
    {
      "entropy": 5.6052361011505125,
      "epoch": 2.714531930333817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042653284602117795,
      "loss": 4.931,
      "mean_token_accuracy": 0.21326714158058166,
      "num_tokens": 62097931.0,
      "step": 29925
    },
    {
      "entropy": 5.6357732772827145,
      "epoch": 2.7149854862119014,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004265089263925236,
      "loss": 5.0135,
      "mean_token_accuracy": 0.20617836564779282,
      "num_tokens": 62109478.0,
      "step": 29930
    },
    {
      "entropy": 5.67087721824646,
      "epoch": 2.7154390420899857,
      "grad_norm": 1.0625,
      "learning_rate": 0.000426485003630626,
      "loss": 5.0088,
      "mean_token_accuracy": 0.2102243572473526,
      "num_tokens": 62120735.0,
      "step": 29935
    },
    {
      "entropy": 5.61495885848999,
      "epoch": 2.71589259796807,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004264610777359801,
      "loss": 4.9348,
      "mean_token_accuracy": 0.21342699378728866,
      "num_tokens": 62130950.0,
      "step": 29940
    },
    {
      "entropy": 5.648841333389282,
      "epoch": 2.7163461538461537,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004264371487090804,
      "loss": 4.9536,
      "mean_token_accuracy": 0.21394813805818558,
      "num_tokens": 62141351.0,
      "step": 29945
    },
    {
      "entropy": 5.528857040405273,
      "epoch": 2.716799709724238,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042641321655042185,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2171203076839447,
      "num_tokens": 62151388.0,
      "step": 29950
    },
    {
      "entropy": 5.649282789230346,
      "epoch": 2.7172532656023223,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004263892812604995,
      "loss": 5.081,
      "mean_token_accuracy": 0.20694220662117005,
      "num_tokens": 62162809.0,
      "step": 29955
    },
    {
      "entropy": 5.607805585861206,
      "epoch": 2.7177068214804065,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042636534283980814,
      "loss": 4.9385,
      "mean_token_accuracy": 0.2116482675075531,
      "num_tokens": 62172738.0,
      "step": 29960
    },
    {
      "entropy": 5.706306028366089,
      "epoch": 2.7181603773584904,
      "grad_norm": 1.203125,
      "learning_rate": 0.000426341401288843,
      "loss": 5.0701,
      "mean_token_accuracy": 0.20600081086158753,
      "num_tokens": 62182508.0,
      "step": 29965
    },
    {
      "entropy": 5.664566087722778,
      "epoch": 2.7186139332365746,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042631745660809905,
      "loss": 5.0309,
      "mean_token_accuracy": 0.20101944208145142,
      "num_tokens": 62192873.0,
      "step": 29970
    },
    {
      "entropy": 5.669667339324951,
      "epoch": 2.719067489114659,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042629350879807155,
      "loss": 5.0485,
      "mean_token_accuracy": 0.20599255412817002,
      "num_tokens": 62204727.0,
      "step": 29975
    },
    {
      "entropy": 5.604229736328125,
      "epoch": 2.719521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042626955785925576,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21691482067108153,
      "num_tokens": 62214268.0,
      "step": 29980
    },
    {
      "entropy": 5.674219083786011,
      "epoch": 2.7199746008708274,
      "grad_norm": 1.125,
      "learning_rate": 0.00042624560379214693,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2162727102637291,
      "num_tokens": 62224756.0,
      "step": 29985
    },
    {
      "entropy": 5.708204650878907,
      "epoch": 2.7204281567489117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004262216465972405,
      "loss": 4.9272,
      "mean_token_accuracy": 0.21419196873903273,
      "num_tokens": 62234065.0,
      "step": 29990
    },
    {
      "entropy": 5.658069610595703,
      "epoch": 2.7208817126269955,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004261976862750318,
      "loss": 4.9389,
      "mean_token_accuracy": 0.21742093861103057,
      "num_tokens": 62243732.0,
      "step": 29995
    },
    {
      "entropy": 5.634906816482544,
      "epoch": 2.72133526850508,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004261737228260165,
      "loss": 4.9461,
      "mean_token_accuracy": 0.2169894829392433,
      "num_tokens": 62253461.0,
      "step": 30000
    },
    {
      "epoch": 2.72133526850508,
      "eval_entropy": 5.322430483032675,
      "eval_loss": 4.991870880126953,
      "eval_mean_token_accuracy": 0.22135150900692152,
      "eval_num_tokens": 62253461.0,
      "eval_runtime": 20.5124,
      "eval_samples_per_second": 1723.834,
      "eval_steps_per_second": 215.479,
      "step": 30000
    },
    {
      "entropy": 5.635187101364136,
      "epoch": 2.721788824383164,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004261497562506901,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21515329778194428,
      "num_tokens": 62263830.0,
      "step": 30005
    },
    {
      "entropy": 5.578507328033448,
      "epoch": 2.7222423802612483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004261257865495483,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22162572741508485,
      "num_tokens": 62274383.0,
      "step": 30010
    },
    {
      "entropy": 5.688109731674194,
      "epoch": 2.722695936139332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004261018137230865,
      "loss": 5.0803,
      "mean_token_accuracy": 0.19924449771642685,
      "num_tokens": 62286476.0,
      "step": 30015
    },
    {
      "entropy": 5.712141323089599,
      "epoch": 2.7231494920174164,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042607783777180085,
      "loss": 4.9895,
      "mean_token_accuracy": 0.21388456374406814,
      "num_tokens": 62296192.0,
      "step": 30020
    },
    {
      "entropy": 5.597686815261841,
      "epoch": 2.7236030478955007,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042605385869618693,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2227681025862694,
      "num_tokens": 62305300.0,
      "step": 30025
    },
    {
      "entropy": 5.6276061058044435,
      "epoch": 2.724056603773585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004260298764967408,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2132501110434532,
      "num_tokens": 62316802.0,
      "step": 30030
    },
    {
      "entropy": 5.5780035018920895,
      "epoch": 2.724510159651669,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004260058911739583,
      "loss": 4.9277,
      "mean_token_accuracy": 0.2125789999961853,
      "num_tokens": 62326444.0,
      "step": 30035
    },
    {
      "entropy": 5.625784158706665,
      "epoch": 2.7249637155297535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042598190272833547,
      "loss": 4.8647,
      "mean_token_accuracy": 0.21831411868333817,
      "num_tokens": 62336872.0,
      "step": 30040
    },
    {
      "entropy": 5.532861709594727,
      "epoch": 2.7254172714078373,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004259579111603685,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22044002413749694,
      "num_tokens": 62346771.0,
      "step": 30045
    },
    {
      "entropy": 5.616319084167481,
      "epoch": 2.7258708272859216,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042593391647055334,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21780407875776292,
      "num_tokens": 62357503.0,
      "step": 30050
    },
    {
      "entropy": 5.55256404876709,
      "epoch": 2.726324383164006,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004259099186593864,
      "loss": 4.8386,
      "mean_token_accuracy": 0.21823133528232574,
      "num_tokens": 62367664.0,
      "step": 30055
    },
    {
      "entropy": 5.620317077636718,
      "epoch": 2.72677793904209,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004258859177273639,
      "loss": 4.9108,
      "mean_token_accuracy": 0.21361131817102433,
      "num_tokens": 62376924.0,
      "step": 30060
    },
    {
      "entropy": 5.539564228057861,
      "epoch": 2.727231494920174,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004258619136749821,
      "loss": 4.8424,
      "mean_token_accuracy": 0.21896851658821107,
      "num_tokens": 62387529.0,
      "step": 30065
    },
    {
      "entropy": 5.577095031738281,
      "epoch": 2.727685050798258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004258379065027375,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21758293509483337,
      "num_tokens": 62397833.0,
      "step": 30070
    },
    {
      "entropy": 5.65999207496643,
      "epoch": 2.7281386066763424,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004258138962111266,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2107671931385994,
      "num_tokens": 62409022.0,
      "step": 30075
    },
    {
      "entropy": 5.642628145217896,
      "epoch": 2.7285921625544267,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004257898828006459,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20631146430969238,
      "num_tokens": 62419105.0,
      "step": 30080
    },
    {
      "entropy": 5.658534955978394,
      "epoch": 2.729045718432511,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004257658662717919,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22136663645505905,
      "num_tokens": 62428371.0,
      "step": 30085
    },
    {
      "entropy": 5.62534465789795,
      "epoch": 2.7294992743105952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004257418466250615,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2163520872592926,
      "num_tokens": 62439301.0,
      "step": 30090
    },
    {
      "entropy": 5.623083639144897,
      "epoch": 2.7299528301886795,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042571782386095124,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21385710090398788,
      "num_tokens": 62449450.0,
      "step": 30095
    },
    {
      "entropy": 5.61260666847229,
      "epoch": 2.7304063860667633,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00042569379797995793,
      "loss": 4.92,
      "mean_token_accuracy": 0.20734917223453522,
      "num_tokens": 62461051.0,
      "step": 30100
    },
    {
      "entropy": 5.613293027877807,
      "epoch": 2.7308599419448476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004256697689825785,
      "loss": 4.8713,
      "mean_token_accuracy": 0.21956653147935867,
      "num_tokens": 62471755.0,
      "step": 30105
    },
    {
      "entropy": 5.599010610580445,
      "epoch": 2.731313497822932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042564573686930987,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2142628774046898,
      "num_tokens": 62482060.0,
      "step": 30110
    },
    {
      "entropy": 5.628189754486084,
      "epoch": 2.7317670537010157,
      "grad_norm": 1.046875,
      "learning_rate": 0.000425621701640649,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21206493973731994,
      "num_tokens": 62493735.0,
      "step": 30115
    },
    {
      "entropy": 5.667255878448486,
      "epoch": 2.7322206095791,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000425597663297093,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21188863664865493,
      "num_tokens": 62504582.0,
      "step": 30120
    },
    {
      "entropy": 5.552774095535279,
      "epoch": 2.7326741654571842,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004255736218391389,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2082519993185997,
      "num_tokens": 62516665.0,
      "step": 30125
    },
    {
      "entropy": 5.597484683990478,
      "epoch": 2.7331277213352685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042554957726728394,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21525535434484483,
      "num_tokens": 62527745.0,
      "step": 30130
    },
    {
      "entropy": 5.640411281585694,
      "epoch": 2.7335812772133528,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004255255295820253,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21911579221487046,
      "num_tokens": 62538622.0,
      "step": 30135
    },
    {
      "entropy": 5.620315408706665,
      "epoch": 2.734034833091437,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042550147878386044,
      "loss": 4.9243,
      "mean_token_accuracy": 0.21643706411123276,
      "num_tokens": 62548302.0,
      "step": 30140
    },
    {
      "entropy": 5.564068174362182,
      "epoch": 2.7344883889695213,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004254774248732867,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2270464763045311,
      "num_tokens": 62557150.0,
      "step": 30145
    },
    {
      "entropy": 5.74721212387085,
      "epoch": 2.734941944847605,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042545336785080126,
      "loss": 5.1335,
      "mean_token_accuracy": 0.20262795686721802,
      "num_tokens": 62568291.0,
      "step": 30150
    },
    {
      "entropy": 5.639190340042115,
      "epoch": 2.7353955007256894,
      "grad_norm": 1.3125,
      "learning_rate": 0.000425429307716902,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21788402050733566,
      "num_tokens": 62579351.0,
      "step": 30155
    },
    {
      "entropy": 5.510387134552002,
      "epoch": 2.7358490566037736,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042540524447208625,
      "loss": 4.8014,
      "mean_token_accuracy": 0.22315987199544907,
      "num_tokens": 62588614.0,
      "step": 30160
    },
    {
      "entropy": 5.632987594604492,
      "epoch": 2.736302612481858,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042538117811685174,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2125912755727768,
      "num_tokens": 62598844.0,
      "step": 30165
    },
    {
      "entropy": 5.650135040283203,
      "epoch": 2.7367561683599417,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042535710865169616,
      "loss": 4.9603,
      "mean_token_accuracy": 0.20938429981470108,
      "num_tokens": 62608722.0,
      "step": 30170
    },
    {
      "entropy": 5.573190689086914,
      "epoch": 2.737209724238026,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004253330360771173,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2145365744829178,
      "num_tokens": 62618712.0,
      "step": 30175
    },
    {
      "entropy": 5.643289518356323,
      "epoch": 2.7376632801161103,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042530896039361284,
      "loss": 5.0504,
      "mean_token_accuracy": 0.19993704408407212,
      "num_tokens": 62629329.0,
      "step": 30180
    },
    {
      "entropy": 5.640778303146362,
      "epoch": 2.7381168359941945,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042528488160168077,
      "loss": 4.9718,
      "mean_token_accuracy": 0.21046598106622696,
      "num_tokens": 62639644.0,
      "step": 30185
    },
    {
      "entropy": 5.63208417892456,
      "epoch": 2.738570391872279,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042526079970181915,
      "loss": 5.0115,
      "mean_token_accuracy": 0.20216554254293442,
      "num_tokens": 62650979.0,
      "step": 30190
    },
    {
      "entropy": 5.695004081726074,
      "epoch": 2.739023947750363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004252367146945259,
      "loss": 5.0258,
      "mean_token_accuracy": 0.20917264223098755,
      "num_tokens": 62661163.0,
      "step": 30195
    },
    {
      "entropy": 5.649981546401977,
      "epoch": 2.739477503628447,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042521262658029905,
      "loss": 5.0179,
      "mean_token_accuracy": 0.20641657710075378,
      "num_tokens": 62672371.0,
      "step": 30200
    },
    {
      "entropy": 5.639853000640869,
      "epoch": 2.739931059506531,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042518853535963673,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21631878167390822,
      "num_tokens": 62682521.0,
      "step": 30205
    },
    {
      "entropy": 5.615931653976441,
      "epoch": 2.7403846153846154,
      "grad_norm": 1.671875,
      "learning_rate": 0.00042516444103303736,
      "loss": 4.8326,
      "mean_token_accuracy": 0.22980313152074813,
      "num_tokens": 62692420.0,
      "step": 30210
    },
    {
      "entropy": 5.566523170471191,
      "epoch": 2.7408381712626997,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042514034360099907,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22017002999782562,
      "num_tokens": 62702686.0,
      "step": 30215
    },
    {
      "entropy": 5.613425922393799,
      "epoch": 2.7412917271407835,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042511624306402015,
      "loss": 4.9518,
      "mean_token_accuracy": 0.21421704590320587,
      "num_tokens": 62713262.0,
      "step": 30220
    },
    {
      "entropy": 5.641849136352539,
      "epoch": 2.7417452830188678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004250921394225992,
      "loss": 4.891,
      "mean_token_accuracy": 0.2165065214037895,
      "num_tokens": 62724460.0,
      "step": 30225
    },
    {
      "entropy": 5.618679761886597,
      "epoch": 2.742198838896952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004250680326772343,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21132843047380448,
      "num_tokens": 62734190.0,
      "step": 30230
    },
    {
      "entropy": 5.6210884094238285,
      "epoch": 2.7426523947750363,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004250439228284244,
      "loss": 4.9084,
      "mean_token_accuracy": 0.21287071108818054,
      "num_tokens": 62744323.0,
      "step": 30235
    },
    {
      "entropy": 5.633791065216064,
      "epoch": 2.7431059506531206,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042501980987666807,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2136697769165039,
      "num_tokens": 62753868.0,
      "step": 30240
    },
    {
      "entropy": 5.64769229888916,
      "epoch": 2.743559506531205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004249956938224636,
      "loss": 5.0297,
      "mean_token_accuracy": 0.21195100694894792,
      "num_tokens": 62763954.0,
      "step": 30245
    },
    {
      "entropy": 5.602226257324219,
      "epoch": 2.7440130624092887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042497157466631015,
      "loss": 4.9093,
      "mean_token_accuracy": 0.222989721596241,
      "num_tokens": 62774116.0,
      "step": 30250
    },
    {
      "entropy": 5.667446088790894,
      "epoch": 2.744466618287373,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042494745240870626,
      "loss": 4.9843,
      "mean_token_accuracy": 0.20882740318775178,
      "num_tokens": 62784939.0,
      "step": 30255
    },
    {
      "entropy": 5.602467346191406,
      "epoch": 2.744920174165457,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042492332705015075,
      "loss": 4.9418,
      "mean_token_accuracy": 0.21344192773103715,
      "num_tokens": 62794988.0,
      "step": 30260
    },
    {
      "entropy": 5.637758779525757,
      "epoch": 2.7453737300435415,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042489919859114273,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21601847261190416,
      "num_tokens": 62804984.0,
      "step": 30265
    },
    {
      "entropy": 5.660616397857666,
      "epoch": 2.7458272859216253,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042487506703218104,
      "loss": 4.9177,
      "mean_token_accuracy": 0.20936132669448854,
      "num_tokens": 62816401.0,
      "step": 30270
    },
    {
      "entropy": 5.586795806884766,
      "epoch": 2.7462808417997095,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004248509323737648,
      "loss": 4.872,
      "mean_token_accuracy": 0.2210390254855156,
      "num_tokens": 62826411.0,
      "step": 30275
    },
    {
      "entropy": 5.645794773101807,
      "epoch": 2.746734397677794,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004248267946163931,
      "loss": 4.9702,
      "mean_token_accuracy": 0.2082474038004875,
      "num_tokens": 62837228.0,
      "step": 30280
    },
    {
      "entropy": 5.568917322158813,
      "epoch": 2.747187953555878,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042480265376056506,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22329940497875214,
      "num_tokens": 62847852.0,
      "step": 30285
    },
    {
      "entropy": 5.5985109329223635,
      "epoch": 2.7476415094339623,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042477850980678,
      "loss": 4.9777,
      "mean_token_accuracy": 0.2103733941912651,
      "num_tokens": 62859516.0,
      "step": 30290
    },
    {
      "entropy": 5.663137817382813,
      "epoch": 2.7480950653120466,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004247543627555372,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2140062466263771,
      "num_tokens": 62869757.0,
      "step": 30295
    },
    {
      "entropy": 5.598596000671387,
      "epoch": 2.748548621190131,
      "grad_norm": 1.296875,
      "learning_rate": 0.000424730212607336,
      "loss": 5.0152,
      "mean_token_accuracy": 0.21465295553207397,
      "num_tokens": 62879618.0,
      "step": 30300
    },
    {
      "entropy": 5.619273471832275,
      "epoch": 2.7490021770682147,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004247060593626758,
      "loss": 4.9602,
      "mean_token_accuracy": 0.2062534585595131,
      "num_tokens": 62890491.0,
      "step": 30305
    },
    {
      "entropy": 5.694744062423706,
      "epoch": 2.749455732946299,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004246819030220562,
      "loss": 4.974,
      "mean_token_accuracy": 0.22353752851486205,
      "num_tokens": 62899954.0,
      "step": 30310
    },
    {
      "entropy": 5.619402742385864,
      "epoch": 2.7499092888243832,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004246577435859767,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21618437618017197,
      "num_tokens": 62911857.0,
      "step": 30315
    },
    {
      "entropy": 5.572789192199707,
      "epoch": 2.7503628447024675,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004246335810549369,
      "loss": 4.9132,
      "mean_token_accuracy": 0.220131753385067,
      "num_tokens": 62922023.0,
      "step": 30320
    },
    {
      "entropy": 5.636546897888183,
      "epoch": 2.7508164005805513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042460941542943656,
      "loss": 4.975,
      "mean_token_accuracy": 0.21941631734371186,
      "num_tokens": 62932845.0,
      "step": 30325
    },
    {
      "entropy": 5.6449686050415036,
      "epoch": 2.7512699564586356,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004245852467099753,
      "loss": 4.9275,
      "mean_token_accuracy": 0.20911925435066223,
      "num_tokens": 62942881.0,
      "step": 30330
    },
    {
      "entropy": 5.600296258926392,
      "epoch": 2.75172351233672,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004245610748970531,
      "loss": 4.8887,
      "mean_token_accuracy": 0.218708373606205,
      "num_tokens": 62953957.0,
      "step": 30335
    },
    {
      "entropy": 5.539433383941651,
      "epoch": 2.752177068214804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042453689999116974,
      "loss": 4.8881,
      "mean_token_accuracy": 0.2184162750840187,
      "num_tokens": 62964803.0,
      "step": 30340
    },
    {
      "entropy": 5.57485408782959,
      "epoch": 2.7526306240928884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004245127219928252,
      "loss": 4.8971,
      "mean_token_accuracy": 0.21340616047382355,
      "num_tokens": 62975477.0,
      "step": 30345
    },
    {
      "entropy": 5.620100355148315,
      "epoch": 2.7530841799709727,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042448854090251944,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21272512525320053,
      "num_tokens": 62985430.0,
      "step": 30350
    },
    {
      "entropy": 5.622733211517334,
      "epoch": 2.7535377358490565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042446435672075254,
      "loss": 4.939,
      "mean_token_accuracy": 0.21676833927631378,
      "num_tokens": 62997493.0,
      "step": 30355
    },
    {
      "entropy": 5.6869219779968265,
      "epoch": 2.7539912917271407,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042444016944802467,
      "loss": 5.0164,
      "mean_token_accuracy": 0.21761286854743958,
      "num_tokens": 63007573.0,
      "step": 30360
    },
    {
      "entropy": 5.6685192584991455,
      "epoch": 2.754444847605225,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042441597908483603,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2181703642010689,
      "num_tokens": 63018372.0,
      "step": 30365
    },
    {
      "entropy": 5.607378005981445,
      "epoch": 2.7548984034833093,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004243917856316869,
      "loss": 4.923,
      "mean_token_accuracy": 0.21784745156764984,
      "num_tokens": 63028632.0,
      "step": 30370
    },
    {
      "entropy": 5.607122278213501,
      "epoch": 2.755351959361393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042436758908907747,
      "loss": 4.9381,
      "mean_token_accuracy": 0.20953853130340577,
      "num_tokens": 63038510.0,
      "step": 30375
    },
    {
      "entropy": 5.706074857711792,
      "epoch": 2.7558055152394774,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042434338945750833,
      "loss": 5.1047,
      "mean_token_accuracy": 0.19659190475940705,
      "num_tokens": 63049375.0,
      "step": 30380
    },
    {
      "entropy": 5.652553462982178,
      "epoch": 2.7562590711175616,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042431918673747974,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21329842358827591,
      "num_tokens": 63059422.0,
      "step": 30385
    },
    {
      "entropy": 5.668554830551147,
      "epoch": 2.756712626995646,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042429498092949245,
      "loss": 5.0166,
      "mean_token_accuracy": 0.2077143058180809,
      "num_tokens": 63071101.0,
      "step": 30390
    },
    {
      "entropy": 5.562238454818726,
      "epoch": 2.75716618287373,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004242707720340468,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2166548877954483,
      "num_tokens": 63081104.0,
      "step": 30395
    },
    {
      "entropy": 5.62669734954834,
      "epoch": 2.7576197387518144,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004242465600516436,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21130558252334594,
      "num_tokens": 63091123.0,
      "step": 30400
    },
    {
      "entropy": 5.574954032897949,
      "epoch": 2.7580732946298983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042422234498278353,
      "loss": 4.9178,
      "mean_token_accuracy": 0.21986084282398224,
      "num_tokens": 63101541.0,
      "step": 30405
    },
    {
      "entropy": 5.6344380378723145,
      "epoch": 2.7585268505079825,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042419812682796727,
      "loss": 5.0219,
      "mean_token_accuracy": 0.20091724246740342,
      "num_tokens": 63112012.0,
      "step": 30410
    },
    {
      "entropy": 5.657411575317383,
      "epoch": 2.758980406386067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004241739055876957,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2195555418729782,
      "num_tokens": 63122021.0,
      "step": 30415
    },
    {
      "entropy": 5.692275476455689,
      "epoch": 2.759433962264151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042414968126246985,
      "loss": 5.033,
      "mean_token_accuracy": 0.21552336663007737,
      "num_tokens": 63132286.0,
      "step": 30420
    },
    {
      "entropy": 5.581722640991211,
      "epoch": 2.759887518142235,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042412545385279055,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21918760389089584,
      "num_tokens": 63142159.0,
      "step": 30425
    },
    {
      "entropy": 5.581647825241089,
      "epoch": 2.760341074020319,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042410122335915883,
      "loss": 4.8765,
      "mean_token_accuracy": 0.21418473422527312,
      "num_tokens": 63152225.0,
      "step": 30430
    },
    {
      "entropy": 5.605106449127197,
      "epoch": 2.7607946298984034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004240769897820758,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21331572383642197,
      "num_tokens": 63162522.0,
      "step": 30435
    },
    {
      "entropy": 5.6426314353942875,
      "epoch": 2.7612481857764877,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042405275312204266,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2206499993801117,
      "num_tokens": 63172094.0,
      "step": 30440
    },
    {
      "entropy": 5.622878313064575,
      "epoch": 2.761701741654572,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042402851337956066,
      "loss": 4.95,
      "mean_token_accuracy": 0.21529112309217452,
      "num_tokens": 63182122.0,
      "step": 30445
    },
    {
      "entropy": 5.570079612731933,
      "epoch": 2.762155297532656,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042400427055513094,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21446115374565125,
      "num_tokens": 63193174.0,
      "step": 30450
    },
    {
      "entropy": 5.620320844650268,
      "epoch": 2.7626088534107405,
      "grad_norm": 1.140625,
      "learning_rate": 0.000423980024649255,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22202438861131668,
      "num_tokens": 63203698.0,
      "step": 30455
    },
    {
      "entropy": 5.527012872695923,
      "epoch": 2.7630624092888243,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004239557756624342,
      "loss": 4.8228,
      "mean_token_accuracy": 0.21544385701417923,
      "num_tokens": 63213343.0,
      "step": 30460
    },
    {
      "entropy": 5.638984441757202,
      "epoch": 2.7635159651669086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042393152359516994,
      "loss": 5.0026,
      "mean_token_accuracy": 0.21217646449804306,
      "num_tokens": 63223962.0,
      "step": 30465
    },
    {
      "entropy": 5.605218744277954,
      "epoch": 2.763969521044993,
      "grad_norm": 1.125,
      "learning_rate": 0.00042390726844796384,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21396935731172562,
      "num_tokens": 63234279.0,
      "step": 30470
    },
    {
      "entropy": 5.677052783966064,
      "epoch": 2.7644230769230766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004238830102213174,
      "loss": 5.0031,
      "mean_token_accuracy": 0.21241945326328276,
      "num_tokens": 63245567.0,
      "step": 30475
    },
    {
      "entropy": 5.615251779556274,
      "epoch": 2.764876632801161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004238587489157325,
      "loss": 4.897,
      "mean_token_accuracy": 0.22111137211322784,
      "num_tokens": 63256227.0,
      "step": 30480
    },
    {
      "entropy": 5.550910711288452,
      "epoch": 2.765330188679245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004238344845317107,
      "loss": 4.8411,
      "mean_token_accuracy": 0.21429300606250762,
      "num_tokens": 63266618.0,
      "step": 30485
    },
    {
      "entropy": 5.553992509841919,
      "epoch": 2.7657837445573294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004238102170697539,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22071612030267715,
      "num_tokens": 63277130.0,
      "step": 30490
    },
    {
      "entropy": 5.537805414199829,
      "epoch": 2.7662373004354137,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004237859465303637,
      "loss": 4.803,
      "mean_token_accuracy": 0.22024587243795396,
      "num_tokens": 63287351.0,
      "step": 30495
    },
    {
      "entropy": 5.536068248748779,
      "epoch": 2.766690856313498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042376167291404234,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21927444636821747,
      "num_tokens": 63298573.0,
      "step": 30500
    },
    {
      "entropy": 5.604266214370727,
      "epoch": 2.7671444121915822,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042373739622129166,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21442582905292512,
      "num_tokens": 63310740.0,
      "step": 30505
    },
    {
      "entropy": 5.60201678276062,
      "epoch": 2.767597968069666,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042371311645261376,
      "loss": 4.9393,
      "mean_token_accuracy": 0.21336130797863007,
      "num_tokens": 63321323.0,
      "step": 30510
    },
    {
      "entropy": 5.607174682617187,
      "epoch": 2.7680515239477503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004236888336085107,
      "loss": 4.9133,
      "mean_token_accuracy": 0.2210501626133919,
      "num_tokens": 63332322.0,
      "step": 30515
    },
    {
      "entropy": 5.571479415893554,
      "epoch": 2.7685050798258346,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004236645476894846,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22529702484607697,
      "num_tokens": 63342204.0,
      "step": 30520
    },
    {
      "entropy": 5.503850889205933,
      "epoch": 2.768958635703919,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004236402586960378,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22196262776851655,
      "num_tokens": 63352460.0,
      "step": 30525
    },
    {
      "entropy": 5.486683320999146,
      "epoch": 2.7694121915820027,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004236159666286726,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22235184162855148,
      "num_tokens": 63362474.0,
      "step": 30530
    },
    {
      "entropy": 5.6453734874725345,
      "epoch": 2.769865747460087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042359167148789135,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22348087877035142,
      "num_tokens": 63372537.0,
      "step": 30535
    },
    {
      "entropy": 5.581620693206787,
      "epoch": 2.7703193033381712,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004235673732741964,
      "loss": 4.8225,
      "mean_token_accuracy": 0.22849991917610168,
      "num_tokens": 63382985.0,
      "step": 30540
    },
    {
      "entropy": 5.583538818359375,
      "epoch": 2.7707728592162555,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042354307198809037,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21852828115224837,
      "num_tokens": 63393424.0,
      "step": 30545
    },
    {
      "entropy": 5.532370662689209,
      "epoch": 2.7712264150943398,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004235187676300757,
      "loss": 4.8399,
      "mean_token_accuracy": 0.21931068748235702,
      "num_tokens": 63403160.0,
      "step": 30550
    },
    {
      "entropy": 5.6277649879455565,
      "epoch": 2.771679970972424,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004234944602006552,
      "loss": 4.9504,
      "mean_token_accuracy": 0.21400920301675797,
      "num_tokens": 63413001.0,
      "step": 30555
    },
    {
      "entropy": 5.626135683059692,
      "epoch": 2.772133526850508,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042347014970033124,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22025052905082704,
      "num_tokens": 63422735.0,
      "step": 30560
    },
    {
      "entropy": 5.534072637557983,
      "epoch": 2.772587082728592,
      "grad_norm": 1.125,
      "learning_rate": 0.0004234458361296069,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21442594975233079,
      "num_tokens": 63432566.0,
      "step": 30565
    },
    {
      "entropy": 5.601199960708618,
      "epoch": 2.7730406386066764,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042342151948898476,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21695515960454942,
      "num_tokens": 63442412.0,
      "step": 30570
    },
    {
      "entropy": 5.591886901855469,
      "epoch": 2.7734941944847606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042339719977896784,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22086687237024308,
      "num_tokens": 63453018.0,
      "step": 30575
    },
    {
      "entropy": 5.6380339622497555,
      "epoch": 2.7739477503628445,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000423372877000059,
      "loss": 4.9657,
      "mean_token_accuracy": 0.21296003460884094,
      "num_tokens": 63464286.0,
      "step": 30580
    },
    {
      "entropy": 5.582695436477661,
      "epoch": 2.7744013062409287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042334855115276117,
      "loss": 4.9255,
      "mean_token_accuracy": 0.21609513014554976,
      "num_tokens": 63475138.0,
      "step": 30585
    },
    {
      "entropy": 5.665557098388672,
      "epoch": 2.774854862119013,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042332422223757756,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2145642712712288,
      "num_tokens": 63486104.0,
      "step": 30590
    },
    {
      "entropy": 5.558847522735595,
      "epoch": 2.7753084179970973,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042329989025501123,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2161530315876007,
      "num_tokens": 63496490.0,
      "step": 30595
    },
    {
      "entropy": 5.581520366668701,
      "epoch": 2.7757619738751815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042327555520556536,
      "loss": 4.8872,
      "mean_token_accuracy": 0.21411698907613755,
      "num_tokens": 63506562.0,
      "step": 30600
    },
    {
      "entropy": 5.645355319976806,
      "epoch": 2.776215529753266,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042325121708974324,
      "loss": 4.9202,
      "mean_token_accuracy": 0.22246306985616685,
      "num_tokens": 63516641.0,
      "step": 30605
    },
    {
      "entropy": 5.667929172515869,
      "epoch": 2.7766690856313496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004232268759080482,
      "loss": 4.9197,
      "mean_token_accuracy": 0.22113981246948242,
      "num_tokens": 63526864.0,
      "step": 30610
    },
    {
      "entropy": 5.61673755645752,
      "epoch": 2.777122641509434,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042320253166098354,
      "loss": 5.0041,
      "mean_token_accuracy": 0.20740817338228226,
      "num_tokens": 63537160.0,
      "step": 30615
    },
    {
      "entropy": 5.623918056488037,
      "epoch": 2.777576197387518,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042317818434905275,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21271833777427673,
      "num_tokens": 63547485.0,
      "step": 30620
    },
    {
      "entropy": 5.619379997253418,
      "epoch": 2.7780297532656024,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004231538339727594,
      "loss": 4.8909,
      "mean_token_accuracy": 0.2110832616686821,
      "num_tokens": 63559369.0,
      "step": 30625
    },
    {
      "entropy": 5.711053562164307,
      "epoch": 2.7784833091436862,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042312948053260704,
      "loss": 5.0807,
      "mean_token_accuracy": 0.20195436030626296,
      "num_tokens": 63570033.0,
      "step": 30630
    },
    {
      "entropy": 5.6182355880737305,
      "epoch": 2.7789368650217705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004231051240290992,
      "loss": 4.8991,
      "mean_token_accuracy": 0.21603305637836456,
      "num_tokens": 63581274.0,
      "step": 30635
    },
    {
      "entropy": 5.66331205368042,
      "epoch": 2.7793904208998548,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042308076446273963,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21470437943935394,
      "num_tokens": 63590903.0,
      "step": 30640
    },
    {
      "entropy": 5.590117645263672,
      "epoch": 2.779843976777939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042305640183403214,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2088554844260216,
      "num_tokens": 63600982.0,
      "step": 30645
    },
    {
      "entropy": 5.494660329818726,
      "epoch": 2.7802975326560233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042303203614348057,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2201700747013092,
      "num_tokens": 63612411.0,
      "step": 30650
    },
    {
      "entropy": 5.587432479858398,
      "epoch": 2.7807510885341076,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004230076673915888,
      "loss": 4.889,
      "mean_token_accuracy": 0.22291014194488526,
      "num_tokens": 63622128.0,
      "step": 30655
    },
    {
      "entropy": 5.623287105560303,
      "epoch": 2.781204644412192,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042298329557886074,
      "loss": 4.9505,
      "mean_token_accuracy": 0.21366265714168547,
      "num_tokens": 63632047.0,
      "step": 30660
    },
    {
      "entropy": 5.654997301101685,
      "epoch": 2.7816582002902757,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004229589207058003,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21421873867511748,
      "num_tokens": 63642490.0,
      "step": 30665
    },
    {
      "entropy": 5.585863780975342,
      "epoch": 2.78211175616836,
      "grad_norm": 1.125,
      "learning_rate": 0.00042293454277291177,
      "loss": 4.9116,
      "mean_token_accuracy": 0.21514899283647537,
      "num_tokens": 63653206.0,
      "step": 30670
    },
    {
      "entropy": 5.556608247756958,
      "epoch": 2.782565312046444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042291016178069916,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21656676828861238,
      "num_tokens": 63663427.0,
      "step": 30675
    },
    {
      "entropy": 5.543863534927368,
      "epoch": 2.7830188679245285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004228857777296668,
      "loss": 4.8984,
      "mean_token_accuracy": 0.2201876848936081,
      "num_tokens": 63674269.0,
      "step": 30680
    },
    {
      "entropy": 5.632609748840332,
      "epoch": 2.7834724238026123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004228613906203188,
      "loss": 4.8977,
      "mean_token_accuracy": 0.21720017790794371,
      "num_tokens": 63683989.0,
      "step": 30685
    },
    {
      "entropy": 5.593549394607544,
      "epoch": 2.7839259796806965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042283700045315955,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21415619254112245,
      "num_tokens": 63694540.0,
      "step": 30690
    },
    {
      "entropy": 5.615074396133423,
      "epoch": 2.784379535558781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042281260722869355,
      "loss": 4.8956,
      "mean_token_accuracy": 0.21547946333885193,
      "num_tokens": 63705149.0,
      "step": 30695
    },
    {
      "entropy": 5.58606481552124,
      "epoch": 2.784833091436865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042278821094742495,
      "loss": 4.872,
      "mean_token_accuracy": 0.2183101147413254,
      "num_tokens": 63714061.0,
      "step": 30700
    },
    {
      "entropy": 5.617989349365234,
      "epoch": 2.7852866473149493,
      "grad_norm": 1.125,
      "learning_rate": 0.00042276381160985874,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2174408480525017,
      "num_tokens": 63724732.0,
      "step": 30705
    },
    {
      "entropy": 5.619711780548096,
      "epoch": 2.7857402031930336,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004227394092164991,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21474839448928834,
      "num_tokens": 63734942.0,
      "step": 30710
    },
    {
      "entropy": 5.677192544937133,
      "epoch": 2.7861937590711174,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004227150037678509,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2185208097100258,
      "num_tokens": 63745326.0,
      "step": 30715
    },
    {
      "entropy": 5.6415215015411375,
      "epoch": 2.7866473149492017,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004226905952644188,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2165289729833603,
      "num_tokens": 63755355.0,
      "step": 30720
    },
    {
      "entropy": 5.615700101852417,
      "epoch": 2.787100870827286,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042266618370670743,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21593225449323655,
      "num_tokens": 63765112.0,
      "step": 30725
    },
    {
      "entropy": 5.574749374389649,
      "epoch": 2.7875544267053702,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042264176909522183,
      "loss": 4.8636,
      "mean_token_accuracy": 0.2189328044652939,
      "num_tokens": 63774680.0,
      "step": 30730
    },
    {
      "entropy": 5.651346158981323,
      "epoch": 2.788007982583454,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004226173514304669,
      "loss": 4.9802,
      "mean_token_accuracy": 0.21668965518474578,
      "num_tokens": 63784939.0,
      "step": 30735
    },
    {
      "entropy": 5.644754457473755,
      "epoch": 2.7884615384615383,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004225929307129474,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20335191637277603,
      "num_tokens": 63795836.0,
      "step": 30740
    },
    {
      "entropy": 5.599572324752808,
      "epoch": 2.7889150943396226,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042256850694316844,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2145790696144104,
      "num_tokens": 63805607.0,
      "step": 30745
    },
    {
      "entropy": 5.571775960922241,
      "epoch": 2.789368650217707,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004225440801216353,
      "loss": 4.8779,
      "mean_token_accuracy": 0.21612794548273087,
      "num_tokens": 63815081.0,
      "step": 30750
    },
    {
      "entropy": 5.539501476287842,
      "epoch": 2.789822206095791,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004225196502488529,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2155549019575119,
      "num_tokens": 63825932.0,
      "step": 30755
    },
    {
      "entropy": 5.57619833946228,
      "epoch": 2.7902757619738754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042249521732532657,
      "loss": 4.8529,
      "mean_token_accuracy": 0.21924570500850676,
      "num_tokens": 63835653.0,
      "step": 30760
    },
    {
      "entropy": 5.707841873168945,
      "epoch": 2.790729317851959,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004224707813515615,
      "loss": 5.0487,
      "mean_token_accuracy": 0.20518928170204162,
      "num_tokens": 63847657.0,
      "step": 30765
    },
    {
      "entropy": 5.683817577362061,
      "epoch": 2.7911828737300435,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042244634232806324,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21522964537143707,
      "num_tokens": 63858476.0,
      "step": 30770
    },
    {
      "entropy": 5.578890132904053,
      "epoch": 2.7916364296081277,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042242190025533694,
      "loss": 4.8101,
      "mean_token_accuracy": 0.22306279540061952,
      "num_tokens": 63867970.0,
      "step": 30775
    },
    {
      "entropy": 5.617247295379639,
      "epoch": 2.792089985486212,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004223974551338881,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22677446603775026,
      "num_tokens": 63878294.0,
      "step": 30780
    },
    {
      "entropy": 5.584589529037475,
      "epoch": 2.792543541364296,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004223730069642224,
      "loss": 4.9036,
      "mean_token_accuracy": 0.21328573673963547,
      "num_tokens": 63889315.0,
      "step": 30785
    },
    {
      "entropy": 5.5599555492401125,
      "epoch": 2.79299709724238,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004223485557468454,
      "loss": 4.8835,
      "mean_token_accuracy": 0.21495374739170076,
      "num_tokens": 63899923.0,
      "step": 30790
    },
    {
      "entropy": 5.564339733123779,
      "epoch": 2.7934506531204644,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042232410148226277,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22505258470773698,
      "num_tokens": 63909813.0,
      "step": 30795
    },
    {
      "entropy": 5.621263599395752,
      "epoch": 2.7939042089985486,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042229964417098007,
      "loss": 4.913,
      "mean_token_accuracy": 0.21303170174360275,
      "num_tokens": 63920931.0,
      "step": 30800
    },
    {
      "entropy": 5.5804585933685305,
      "epoch": 2.794357764876633,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042227518381350316,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22410394698381425,
      "num_tokens": 63930799.0,
      "step": 30805
    },
    {
      "entropy": 5.700989007949829,
      "epoch": 2.794811320754717,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042225072041033804,
      "loss": 5.0125,
      "mean_token_accuracy": 0.21102853566408158,
      "num_tokens": 63941763.0,
      "step": 30810
    },
    {
      "entropy": 5.679899740219116,
      "epoch": 2.7952648766328014,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042222625396199043,
      "loss": 4.9945,
      "mean_token_accuracy": 0.20874671787023544,
      "num_tokens": 63952775.0,
      "step": 30815
    },
    {
      "entropy": 5.635723400115967,
      "epoch": 2.7957184325108853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004222017844689663,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21501920819282533,
      "num_tokens": 63964223.0,
      "step": 30820
    },
    {
      "entropy": 5.657394027709961,
      "epoch": 2.7961719883889695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004221773119317719,
      "loss": 4.9967,
      "mean_token_accuracy": 0.21145854592323304,
      "num_tokens": 63976413.0,
      "step": 30825
    },
    {
      "entropy": 5.6927361488342285,
      "epoch": 2.796625544267054,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042215283635091313,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21410064697265624,
      "num_tokens": 63986578.0,
      "step": 30830
    },
    {
      "entropy": 5.655504274368286,
      "epoch": 2.7970791001451376,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004221283577268962,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21705943942070008,
      "num_tokens": 63996856.0,
      "step": 30835
    },
    {
      "entropy": 5.579367780685425,
      "epoch": 2.797532656023222,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042210387606022724,
      "loss": 4.9049,
      "mean_token_accuracy": 0.21957875192165374,
      "num_tokens": 64007700.0,
      "step": 30840
    },
    {
      "entropy": 5.654842329025269,
      "epoch": 2.797986211901306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004220793913514128,
      "loss": 5.0408,
      "mean_token_accuracy": 0.2083938390016556,
      "num_tokens": 64018771.0,
      "step": 30845
    },
    {
      "entropy": 5.623392248153687,
      "epoch": 2.7984397677793904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004220549036009589,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21311204582452775,
      "num_tokens": 64029381.0,
      "step": 30850
    },
    {
      "entropy": 5.560973501205444,
      "epoch": 2.7988933236574747,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042203041280937224,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2163993701338768,
      "num_tokens": 64039596.0,
      "step": 30855
    },
    {
      "entropy": 5.642567682266235,
      "epoch": 2.799346879535559,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004220059189771592,
      "loss": 4.9387,
      "mean_token_accuracy": 0.22403726130723953,
      "num_tokens": 64049123.0,
      "step": 30860
    },
    {
      "entropy": 5.629621505737305,
      "epoch": 2.799800435413643,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042198142210482623,
      "loss": 4.9753,
      "mean_token_accuracy": 0.2101119115948677,
      "num_tokens": 64059426.0,
      "step": 30865
    },
    {
      "entropy": 5.5925756931304935,
      "epoch": 2.800253991291727,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042195692219287997,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2274341717362404,
      "num_tokens": 64069227.0,
      "step": 30870
    },
    {
      "entropy": 5.650067234039307,
      "epoch": 2.8007075471698113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004219324192418272,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22078555673360825,
      "num_tokens": 64078482.0,
      "step": 30875
    },
    {
      "entropy": 5.590852546691894,
      "epoch": 2.8011611030478956,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042190791325217447,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2191412478685379,
      "num_tokens": 64088664.0,
      "step": 30880
    },
    {
      "entropy": 5.585071706771851,
      "epoch": 2.80161465892598,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004218834042244287,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22087487280368806,
      "num_tokens": 64099316.0,
      "step": 30885
    },
    {
      "entropy": 5.558380889892578,
      "epoch": 2.8020682148040637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042185889215909665,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21728195250034332,
      "num_tokens": 64109786.0,
      "step": 30890
    },
    {
      "entropy": 5.630571603775024,
      "epoch": 2.802521770682148,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042183437705668535,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21791791468858718,
      "num_tokens": 64120771.0,
      "step": 30895
    },
    {
      "entropy": 5.645556211471558,
      "epoch": 2.802975326560232,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042180985891770176,
      "loss": 4.9591,
      "mean_token_accuracy": 0.21070195883512496,
      "num_tokens": 64130833.0,
      "step": 30900
    },
    {
      "entropy": 5.568078947067261,
      "epoch": 2.8034288824383164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217853377426528,
      "loss": 4.8777,
      "mean_token_accuracy": 0.21844976395368576,
      "num_tokens": 64139672.0,
      "step": 30905
    },
    {
      "entropy": 5.698852157592773,
      "epoch": 2.8038824383164007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004217608135320456,
      "loss": 5.052,
      "mean_token_accuracy": 0.21011776626110076,
      "num_tokens": 64150383.0,
      "step": 30910
    },
    {
      "entropy": 5.595062017440796,
      "epoch": 2.804335994194485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004217362862863875,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22307278364896774,
      "num_tokens": 64160786.0,
      "step": 30915
    },
    {
      "entropy": 5.6643421173095705,
      "epoch": 2.804789550072569,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042171175600618546,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21249350011348725,
      "num_tokens": 64172057.0,
      "step": 30920
    },
    {
      "entropy": 5.626429462432862,
      "epoch": 2.805243105950653,
      "grad_norm": 1.25,
      "learning_rate": 0.00042168722269194704,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21921728253364564,
      "num_tokens": 64182599.0,
      "step": 30925
    },
    {
      "entropy": 5.61863112449646,
      "epoch": 2.8056966618287373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004216626863441794,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21093594282865524,
      "num_tokens": 64193356.0,
      "step": 30930
    },
    {
      "entropy": 5.685582828521729,
      "epoch": 2.8061502177068216,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042163814696339014,
      "loss": 5.0414,
      "mean_token_accuracy": 0.21075332164764404,
      "num_tokens": 64203625.0,
      "step": 30935
    },
    {
      "entropy": 5.635419654846191,
      "epoch": 2.8066037735849054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004216136045500866,
      "loss": 5.038,
      "mean_token_accuracy": 0.2122482806444168,
      "num_tokens": 64214742.0,
      "step": 30940
    },
    {
      "entropy": 5.663224172592163,
      "epoch": 2.8070573294629897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004215890591047763,
      "loss": 4.9439,
      "mean_token_accuracy": 0.2146236315369606,
      "num_tokens": 64223811.0,
      "step": 30945
    },
    {
      "entropy": 5.632388782501221,
      "epoch": 2.807510885341074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042156451062796686,
      "loss": 4.8472,
      "mean_token_accuracy": 0.22328245341777803,
      "num_tokens": 64234854.0,
      "step": 30950
    },
    {
      "entropy": 5.648586559295654,
      "epoch": 2.8079644412191582,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042153995912016606,
      "loss": 4.9524,
      "mean_token_accuracy": 0.2123437687754631,
      "num_tokens": 64245914.0,
      "step": 30955
    },
    {
      "entropy": 5.610034227371216,
      "epoch": 2.8084179970972425,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004215154045818816,
      "loss": 4.9996,
      "mean_token_accuracy": 0.20932998955249787,
      "num_tokens": 64256217.0,
      "step": 30960
    },
    {
      "entropy": 5.620008850097657,
      "epoch": 2.8088715529753268,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004214908470136212,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2039376363158226,
      "num_tokens": 64267211.0,
      "step": 30965
    },
    {
      "entropy": 5.697167158126831,
      "epoch": 2.8093251088534106,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004214662864158927,
      "loss": 4.9759,
      "mean_token_accuracy": 0.21304525285959244,
      "num_tokens": 64277770.0,
      "step": 30970
    },
    {
      "entropy": 5.610387563705444,
      "epoch": 2.809778664731495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042144172278920413,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22335876822471618,
      "num_tokens": 64287973.0,
      "step": 30975
    },
    {
      "entropy": 5.58164119720459,
      "epoch": 2.810232220609579,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004214171561340634,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21139176189899445,
      "num_tokens": 64297898.0,
      "step": 30980
    },
    {
      "entropy": 5.613345146179199,
      "epoch": 2.8106857764876634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042139258645097863,
      "loss": 4.8135,
      "mean_token_accuracy": 0.22891245484352113,
      "num_tokens": 64308114.0,
      "step": 30985
    },
    {
      "entropy": 5.574546194076538,
      "epoch": 2.811139332365747,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042136801374045775,
      "loss": 4.8957,
      "mean_token_accuracy": 0.21791755855083467,
      "num_tokens": 64319586.0,
      "step": 30990
    },
    {
      "entropy": 5.61319785118103,
      "epoch": 2.8115928882438315,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004213434380030091,
      "loss": 4.9818,
      "mean_token_accuracy": 0.21329596489667893,
      "num_tokens": 64330612.0,
      "step": 30995
    },
    {
      "entropy": 5.664735555648804,
      "epoch": 2.8120464441219157,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004213188592391409,
      "loss": 5.0106,
      "mean_token_accuracy": 0.21196627020835876,
      "num_tokens": 64341934.0,
      "step": 31000
    },
    {
      "entropy": 5.642933416366577,
      "epoch": 2.8125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004212942774493614,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2197117179632187,
      "num_tokens": 64352166.0,
      "step": 31005
    },
    {
      "entropy": 5.619105863571167,
      "epoch": 2.8129535558780843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042126969263417885,
      "loss": 4.9514,
      "mean_token_accuracy": 0.20967821776866913,
      "num_tokens": 64361995.0,
      "step": 31010
    },
    {
      "entropy": 5.596150159835815,
      "epoch": 2.8134071117561685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042124510479410183,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2137259691953659,
      "num_tokens": 64372336.0,
      "step": 31015
    },
    {
      "entropy": 5.658761119842529,
      "epoch": 2.813860667634253,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042122051392963885,
      "loss": 4.9494,
      "mean_token_accuracy": 0.2260981097817421,
      "num_tokens": 64382309.0,
      "step": 31020
    },
    {
      "entropy": 5.661155319213867,
      "epoch": 2.8143142235123366,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004211959200412984,
      "loss": 4.9697,
      "mean_token_accuracy": 0.21206053495407104,
      "num_tokens": 64393205.0,
      "step": 31025
    },
    {
      "entropy": 5.580132246017456,
      "epoch": 2.814767779390421,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004211713231295889,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21480776965618134,
      "num_tokens": 64403068.0,
      "step": 31030
    },
    {
      "entropy": 5.627785587310791,
      "epoch": 2.815221335268505,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004211467231950194,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21832402050495148,
      "num_tokens": 64414296.0,
      "step": 31035
    },
    {
      "entropy": 5.5854919910430905,
      "epoch": 2.8156748911465894,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042112212023809824,
      "loss": 4.9549,
      "mean_token_accuracy": 0.20738893151283264,
      "num_tokens": 64424748.0,
      "step": 31040
    },
    {
      "entropy": 5.671691513061523,
      "epoch": 2.8161284470246732,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004210975142593345,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2151064470410347,
      "num_tokens": 64434723.0,
      "step": 31045
    },
    {
      "entropy": 5.71134581565857,
      "epoch": 2.8165820029027575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004210729052592369,
      "loss": 5.0202,
      "mean_token_accuracy": 0.2091205358505249,
      "num_tokens": 64445864.0,
      "step": 31050
    },
    {
      "entropy": 5.5783933162689205,
      "epoch": 2.8170355587808418,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004210482932383144,
      "loss": 4.8677,
      "mean_token_accuracy": 0.21529244035482406,
      "num_tokens": 64456704.0,
      "step": 31055
    },
    {
      "entropy": 5.559114646911621,
      "epoch": 2.817489114658926,
      "grad_norm": 1.25,
      "learning_rate": 0.000421023678197076,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2209993988275528,
      "num_tokens": 64466870.0,
      "step": 31060
    },
    {
      "entropy": 5.644223833084107,
      "epoch": 2.8179426705370103,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004209990601360307,
      "loss": 4.936,
      "mean_token_accuracy": 0.21361759901046753,
      "num_tokens": 64477069.0,
      "step": 31065
    },
    {
      "entropy": 5.682915592193604,
      "epoch": 2.8183962264150946,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042097443905568765,
      "loss": 4.9427,
      "mean_token_accuracy": 0.2114916577935219,
      "num_tokens": 64487165.0,
      "step": 31070
    },
    {
      "entropy": 5.572510862350464,
      "epoch": 2.8188497822931784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042094981495655596,
      "loss": 4.8161,
      "mean_token_accuracy": 0.2176745906472206,
      "num_tokens": 64497977.0,
      "step": 31075
    },
    {
      "entropy": 5.605168294906616,
      "epoch": 2.8193033381712627,
      "grad_norm": 1.140625,
      "learning_rate": 0.000420925187839145,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21277851909399031,
      "num_tokens": 64508767.0,
      "step": 31080
    },
    {
      "entropy": 5.562963056564331,
      "epoch": 2.819756894049347,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042090055770396383,
      "loss": 4.8369,
      "mean_token_accuracy": 0.21968455314636232,
      "num_tokens": 64519040.0,
      "step": 31085
    },
    {
      "entropy": 5.587634897232055,
      "epoch": 2.820210449927431,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042087592455152206,
      "loss": 4.8931,
      "mean_token_accuracy": 0.21901083290576934,
      "num_tokens": 64529721.0,
      "step": 31090
    },
    {
      "entropy": 5.6160064220428465,
      "epoch": 2.820664005805515,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000420851288382329,
      "loss": 5.0348,
      "mean_token_accuracy": 0.2015969142317772,
      "num_tokens": 64541299.0,
      "step": 31095
    },
    {
      "entropy": 5.650126886367798,
      "epoch": 2.8211175616835993,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042082664919689406,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2132447898387909,
      "num_tokens": 64550831.0,
      "step": 31100
    },
    {
      "entropy": 5.62502155303955,
      "epoch": 2.8215711175616836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004208020069957269,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21360254734754563,
      "num_tokens": 64562076.0,
      "step": 31105
    },
    {
      "entropy": 5.559401321411133,
      "epoch": 2.822024673439768,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000420777361779337,
      "loss": 4.9142,
      "mean_token_accuracy": 0.21161176413297653,
      "num_tokens": 64572955.0,
      "step": 31110
    },
    {
      "entropy": 5.670926141738891,
      "epoch": 2.822478229317852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042075271354823427,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22100793570280075,
      "num_tokens": 64584376.0,
      "step": 31115
    },
    {
      "entropy": 5.703147506713867,
      "epoch": 2.8229317851959363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004207280623029281,
      "loss": 4.9646,
      "mean_token_accuracy": 0.21137995421886444,
      "num_tokens": 64594522.0,
      "step": 31120
    },
    {
      "entropy": 5.6463898658752445,
      "epoch": 2.82338534107402,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042070340804392853,
      "loss": 4.974,
      "mean_token_accuracy": 0.21526576727628707,
      "num_tokens": 64604625.0,
      "step": 31125
    },
    {
      "entropy": 5.606982564926147,
      "epoch": 2.8238388969521044,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004206787507717454,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21040987521409987,
      "num_tokens": 64614184.0,
      "step": 31130
    },
    {
      "entropy": 5.6795258045196535,
      "epoch": 2.8242924528301887,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042065409048688856,
      "loss": 4.9967,
      "mean_token_accuracy": 0.21107614487409593,
      "num_tokens": 64623019.0,
      "step": 31135
    },
    {
      "entropy": 5.633118915557861,
      "epoch": 2.824746008708273,
      "grad_norm": 1.25,
      "learning_rate": 0.000420629427189868,
      "loss": 4.9273,
      "mean_token_accuracy": 0.21277052909135818,
      "num_tokens": 64634017.0,
      "step": 31140
    },
    {
      "entropy": 5.616308641433716,
      "epoch": 2.825199564586357,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042060476088119376,
      "loss": 4.9154,
      "mean_token_accuracy": 0.21951388716697692,
      "num_tokens": 64645007.0,
      "step": 31145
    },
    {
      "entropy": 5.676414585113525,
      "epoch": 2.825653120464441,
      "grad_norm": 1.15625,
      "learning_rate": 0.000420580091561376,
      "loss": 4.9624,
      "mean_token_accuracy": 0.20871936976909639,
      "num_tokens": 64655705.0,
      "step": 31150
    },
    {
      "entropy": 5.711740446090698,
      "epoch": 2.8261066763425253,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042055541923092485,
      "loss": 5.0212,
      "mean_token_accuracy": 0.21520592123270035,
      "num_tokens": 64666094.0,
      "step": 31155
    },
    {
      "entropy": 5.697028017044067,
      "epoch": 2.8265602322206096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004205307438903505,
      "loss": 5.0159,
      "mean_token_accuracy": 0.2037213236093521,
      "num_tokens": 64676107.0,
      "step": 31160
    },
    {
      "entropy": 5.6362292766571045,
      "epoch": 2.827013788098694,
      "grad_norm": 1.125,
      "learning_rate": 0.0004205060655401633,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21696954369544982,
      "num_tokens": 64685846.0,
      "step": 31165
    },
    {
      "entropy": 5.569850635528565,
      "epoch": 2.827467343976778,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004204813841808736,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2152869537472725,
      "num_tokens": 64695959.0,
      "step": 31170
    },
    {
      "entropy": 5.617218255996704,
      "epoch": 2.8279208998548624,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004204566998129917,
      "loss": 4.8911,
      "mean_token_accuracy": 0.22237360179424287,
      "num_tokens": 64705594.0,
      "step": 31175
    },
    {
      "entropy": 5.641166162490845,
      "epoch": 2.828374455732946,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042043201243702834,
      "loss": 4.956,
      "mean_token_accuracy": 0.21581101715564727,
      "num_tokens": 64715974.0,
      "step": 31180
    },
    {
      "entropy": 5.588622760772705,
      "epoch": 2.8288280116110305,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004204073220534938,
      "loss": 4.9061,
      "mean_token_accuracy": 0.21878686398267747,
      "num_tokens": 64725846.0,
      "step": 31185
    },
    {
      "entropy": 5.604059934616089,
      "epoch": 2.8292815674891147,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004203826286628988,
      "loss": 4.8758,
      "mean_token_accuracy": 0.21943391263484954,
      "num_tokens": 64735789.0,
      "step": 31190
    },
    {
      "entropy": 5.574193239212036,
      "epoch": 2.8297351233671986,
      "grad_norm": 1.140625,
      "learning_rate": 0.000420357932265754,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2329629987478256,
      "num_tokens": 64745738.0,
      "step": 31195
    },
    {
      "entropy": 5.619892883300781,
      "epoch": 2.830188679245283,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004203332328625701,
      "loss": 4.9208,
      "mean_token_accuracy": 0.2084958389401436,
      "num_tokens": 64756523.0,
      "step": 31200
    },
    {
      "entropy": 5.703876972198486,
      "epoch": 2.830642235123367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042030853045385796,
      "loss": 5.0154,
      "mean_token_accuracy": 0.20294613689184188,
      "num_tokens": 64767367.0,
      "step": 31205
    },
    {
      "entropy": 5.661301708221435,
      "epoch": 2.8310957910014514,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042028382504012837,
      "loss": 4.9451,
      "mean_token_accuracy": 0.21805305927991867,
      "num_tokens": 64777729.0,
      "step": 31210
    },
    {
      "entropy": 5.699280214309693,
      "epoch": 2.8315493468795356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042025911662189224,
      "loss": 4.9712,
      "mean_token_accuracy": 0.2135970801115036,
      "num_tokens": 64788223.0,
      "step": 31215
    },
    {
      "entropy": 5.624813747406006,
      "epoch": 2.83200290275762,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042023440519966057,
      "loss": 4.9188,
      "mean_token_accuracy": 0.2183419182896614,
      "num_tokens": 64797367.0,
      "step": 31220
    },
    {
      "entropy": 5.63323712348938,
      "epoch": 2.832456458635704,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042020969077394434,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21575139611959457,
      "num_tokens": 64807179.0,
      "step": 31225
    },
    {
      "entropy": 5.66260838508606,
      "epoch": 2.832910014513788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042018497334525475,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2116038203239441,
      "num_tokens": 64817276.0,
      "step": 31230
    },
    {
      "entropy": 5.63536901473999,
      "epoch": 2.8333635703918723,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004201602529141029,
      "loss": 4.9387,
      "mean_token_accuracy": 0.2160851314663887,
      "num_tokens": 64827509.0,
      "step": 31235
    },
    {
      "entropy": 5.6317322731018065,
      "epoch": 2.8338171262699565,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042013552948100003,
      "loss": 4.9313,
      "mean_token_accuracy": 0.21394189745187758,
      "num_tokens": 64836578.0,
      "step": 31240
    },
    {
      "entropy": 5.5700640201568605,
      "epoch": 2.834270682148041,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004201108030464574,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22321531176567078,
      "num_tokens": 64846518.0,
      "step": 31245
    },
    {
      "entropy": 5.67998514175415,
      "epoch": 2.8347242380261246,
      "grad_norm": 1.25,
      "learning_rate": 0.00042008607361098646,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21569398641586304,
      "num_tokens": 64856304.0,
      "step": 31250
    },
    {
      "entropy": 5.610930585861206,
      "epoch": 2.835177793904209,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004200613411750985,
      "loss": 4.8412,
      "mean_token_accuracy": 0.23118834495544432,
      "num_tokens": 64867381.0,
      "step": 31255
    },
    {
      "entropy": 5.654486274719238,
      "epoch": 2.835631349782293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000420036605739305,
      "loss": 4.963,
      "mean_token_accuracy": 0.21422148793935775,
      "num_tokens": 64877306.0,
      "step": 31260
    },
    {
      "entropy": 5.549334096908569,
      "epoch": 2.8360849056603774,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004200118673041176,
      "loss": 4.8903,
      "mean_token_accuracy": 0.21658690124750138,
      "num_tokens": 64886932.0,
      "step": 31265
    },
    {
      "entropy": 5.650542116165161,
      "epoch": 2.8365384615384617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041998712587004776,
      "loss": 4.956,
      "mean_token_accuracy": 0.21156884729862213,
      "num_tokens": 64897387.0,
      "step": 31270
    },
    {
      "entropy": 5.645903873443603,
      "epoch": 2.836992017416546,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041996238143760726,
      "loss": 4.9551,
      "mean_token_accuracy": 0.21583790332078934,
      "num_tokens": 64907251.0,
      "step": 31275
    },
    {
      "entropy": 5.581920146942139,
      "epoch": 2.8374455732946298,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041993763400730776,
      "loss": 4.908,
      "mean_token_accuracy": 0.22165969014167786,
      "num_tokens": 64917300.0,
      "step": 31280
    },
    {
      "entropy": 5.563831901550293,
      "epoch": 2.837899129172714,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004199128835796611,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22365381121635436,
      "num_tokens": 64927133.0,
      "step": 31285
    },
    {
      "entropy": 5.602923250198364,
      "epoch": 2.8383526850507983,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000419888130155179,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2144875168800354,
      "num_tokens": 64937789.0,
      "step": 31290
    },
    {
      "entropy": 5.6876220703125,
      "epoch": 2.8388062409288826,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004198633737343735,
      "loss": 4.9483,
      "mean_token_accuracy": 0.21668002754449844,
      "num_tokens": 64948127.0,
      "step": 31295
    },
    {
      "entropy": 5.667296457290649,
      "epoch": 2.8392597968069664,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004198386143177565,
      "loss": 4.9896,
      "mean_token_accuracy": 0.21293671429157257,
      "num_tokens": 64959632.0,
      "step": 31300
    },
    {
      "entropy": 5.509975004196167,
      "epoch": 2.8397133526850507,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041981385190584003,
      "loss": 4.7808,
      "mean_token_accuracy": 0.224514701962471,
      "num_tokens": 64970411.0,
      "step": 31305
    },
    {
      "entropy": 5.562983989715576,
      "epoch": 2.840166908563135,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004197890864991362,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2174754485487938,
      "num_tokens": 64980748.0,
      "step": 31310
    },
    {
      "entropy": 5.586369466781616,
      "epoch": 2.840620464441219,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004197643180981572,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2163645178079605,
      "num_tokens": 64990818.0,
      "step": 31315
    },
    {
      "entropy": 5.634396171569824,
      "epoch": 2.8410740203193035,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004197395467034152,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22361039817333223,
      "num_tokens": 65001993.0,
      "step": 31320
    },
    {
      "entropy": 5.549083852767945,
      "epoch": 2.8415275761973877,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004197147723154225,
      "loss": 4.8846,
      "mean_token_accuracy": 0.222554612159729,
      "num_tokens": 65013007.0,
      "step": 31325
    },
    {
      "entropy": 5.668735933303833,
      "epoch": 2.8419811320754715,
      "grad_norm": 1.125,
      "learning_rate": 0.0004196899949346914,
      "loss": 4.8912,
      "mean_token_accuracy": 0.21653439551591874,
      "num_tokens": 65023595.0,
      "step": 31330
    },
    {
      "entropy": 5.602653551101684,
      "epoch": 2.842434687953556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004196652145617343,
      "loss": 4.9025,
      "mean_token_accuracy": 0.22129547446966172,
      "num_tokens": 65033631.0,
      "step": 31335
    },
    {
      "entropy": 5.645292091369629,
      "epoch": 2.84288824383164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004196404311970638,
      "loss": 4.945,
      "mean_token_accuracy": 0.21404220163822174,
      "num_tokens": 65044059.0,
      "step": 31340
    },
    {
      "entropy": 5.594947481155396,
      "epoch": 2.8433417997097243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004196156448411922,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21391836702823638,
      "num_tokens": 65053931.0,
      "step": 31345
    },
    {
      "entropy": 5.651429748535156,
      "epoch": 2.843795355587808,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041959085549463234,
      "loss": 4.9341,
      "mean_token_accuracy": 0.2207110345363617,
      "num_tokens": 65064011.0,
      "step": 31350
    },
    {
      "entropy": 5.6820303440094,
      "epoch": 2.8442489114658924,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004195660631578966,
      "loss": 4.9803,
      "mean_token_accuracy": 0.21971448361873627,
      "num_tokens": 65074417.0,
      "step": 31355
    },
    {
      "entropy": 5.582434940338135,
      "epoch": 2.8447024673439767,
      "grad_norm": 1.203125,
      "learning_rate": 0.000419541267831498,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22106510102748872,
      "num_tokens": 65084593.0,
      "step": 31360
    },
    {
      "entropy": 5.577517938613892,
      "epoch": 2.845156023222061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000419516469515949,
      "loss": 4.8993,
      "mean_token_accuracy": 0.21986448615789414,
      "num_tokens": 65095047.0,
      "step": 31365
    },
    {
      "entropy": 5.662061405181885,
      "epoch": 2.8456095791001452,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041949166821176264,
      "loss": 5.0004,
      "mean_token_accuracy": 0.2182724580168724,
      "num_tokens": 65104434.0,
      "step": 31370
    },
    {
      "entropy": 5.6851483345031735,
      "epoch": 2.8460631349782295,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041946686391945183,
      "loss": 4.9768,
      "mean_token_accuracy": 0.21837612688541413,
      "num_tokens": 65114039.0,
      "step": 31375
    },
    {
      "entropy": 5.712004661560059,
      "epoch": 2.8465166908563138,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004194420566395294,
      "loss": 5.0505,
      "mean_token_accuracy": 0.20623664408922196,
      "num_tokens": 65125387.0,
      "step": 31380
    },
    {
      "entropy": 5.679856204986573,
      "epoch": 2.8469702467343976,
      "grad_norm": 1.125,
      "learning_rate": 0.0004194172463725084,
      "loss": 4.9628,
      "mean_token_accuracy": 0.22171991169452668,
      "num_tokens": 65136351.0,
      "step": 31385
    },
    {
      "entropy": 5.609826469421387,
      "epoch": 2.847423802612482,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041939243311890195,
      "loss": 4.9241,
      "mean_token_accuracy": 0.21284692585468293,
      "num_tokens": 65147343.0,
      "step": 31390
    },
    {
      "entropy": 5.573555898666382,
      "epoch": 2.847877358490566,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041936761687922324,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22336629331111907,
      "num_tokens": 65157097.0,
      "step": 31395
    },
    {
      "entropy": 5.640733814239502,
      "epoch": 2.8483309143686504,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041934279765398535,
      "loss": 4.938,
      "mean_token_accuracy": 0.21219056248664855,
      "num_tokens": 65167546.0,
      "step": 31400
    },
    {
      "entropy": 5.67144775390625,
      "epoch": 2.848784470246734,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004193179754437017,
      "loss": 4.951,
      "mean_token_accuracy": 0.20674739629030228,
      "num_tokens": 65178497.0,
      "step": 31405
    },
    {
      "entropy": 5.57733793258667,
      "epoch": 2.8492380261248185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004192931502488855,
      "loss": 4.8465,
      "mean_token_accuracy": 0.22393180578947067,
      "num_tokens": 65189530.0,
      "step": 31410
    },
    {
      "entropy": 5.5836952209472654,
      "epoch": 2.8496915820029027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041926832207005025,
      "loss": 4.9231,
      "mean_token_accuracy": 0.21195393055677414,
      "num_tokens": 65199836.0,
      "step": 31415
    },
    {
      "entropy": 5.516938638687134,
      "epoch": 2.850145137880987,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004192434909077092,
      "loss": 4.7745,
      "mean_token_accuracy": 0.22483824342489242,
      "num_tokens": 65210208.0,
      "step": 31420
    },
    {
      "entropy": 5.62606725692749,
      "epoch": 2.8505986937590713,
      "grad_norm": 1.125,
      "learning_rate": 0.00041921865676237605,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21618360430002212,
      "num_tokens": 65220954.0,
      "step": 31425
    },
    {
      "entropy": 5.651532411575317,
      "epoch": 2.8510522496371555,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041919381963456437,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21500850468873978,
      "num_tokens": 65230870.0,
      "step": 31430
    },
    {
      "entropy": 5.655472326278686,
      "epoch": 2.8515058055152394,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004191689795247877,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2218406692147255,
      "num_tokens": 65241913.0,
      "step": 31435
    },
    {
      "entropy": 5.670316123962403,
      "epoch": 2.8519593613933236,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004191441364335598,
      "loss": 4.9215,
      "mean_token_accuracy": 0.21658070683479308,
      "num_tokens": 65252513.0,
      "step": 31440
    },
    {
      "entropy": 5.5814519882202145,
      "epoch": 2.852412917271408,
      "grad_norm": 1.125,
      "learning_rate": 0.00041911929036139457,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21429946273565292,
      "num_tokens": 65262695.0,
      "step": 31445
    },
    {
      "entropy": 5.666453170776367,
      "epoch": 2.852866473149492,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041909444130880553,
      "loss": 4.9909,
      "mean_token_accuracy": 0.2131132498383522,
      "num_tokens": 65274131.0,
      "step": 31450
    },
    {
      "entropy": 5.6249415397644045,
      "epoch": 2.853320029027576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041906958927630673,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21268677115440368,
      "num_tokens": 65285080.0,
      "step": 31455
    },
    {
      "entropy": 5.761142349243164,
      "epoch": 2.8537735849056602,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004190447342644122,
      "loss": 5.1121,
      "mean_token_accuracy": 0.21379054188728333,
      "num_tokens": 65296512.0,
      "step": 31460
    },
    {
      "entropy": 5.5873730182647705,
      "epoch": 2.8542271407837445,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004190198762736358,
      "loss": 4.8613,
      "mean_token_accuracy": 0.22401265501976014,
      "num_tokens": 65306054.0,
      "step": 31465
    },
    {
      "entropy": 5.529381322860718,
      "epoch": 2.8546806966618288,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004189950153044917,
      "loss": 4.9271,
      "mean_token_accuracy": 0.21471306532621384,
      "num_tokens": 65315046.0,
      "step": 31470
    },
    {
      "entropy": 5.5956275939941404,
      "epoch": 2.855134252539913,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004189701513574939,
      "loss": 4.8794,
      "mean_token_accuracy": 0.21540447026491166,
      "num_tokens": 65325482.0,
      "step": 31475
    },
    {
      "entropy": 5.579676103591919,
      "epoch": 2.8555878084179973,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004189452844331567,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2180931955575943,
      "num_tokens": 65335518.0,
      "step": 31480
    },
    {
      "entropy": 5.615693044662476,
      "epoch": 2.856041364296081,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041892041453199443,
      "loss": 4.9779,
      "mean_token_accuracy": 0.21862132251262664,
      "num_tokens": 65346079.0,
      "step": 31485
    },
    {
      "entropy": 5.56201548576355,
      "epoch": 2.8564949201741654,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004188955416545212,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2199193388223648,
      "num_tokens": 65356719.0,
      "step": 31490
    },
    {
      "entropy": 5.635035705566406,
      "epoch": 2.8569484760522497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004188706658012516,
      "loss": 4.9485,
      "mean_token_accuracy": 0.21020206063985825,
      "num_tokens": 65367877.0,
      "step": 31495
    },
    {
      "entropy": 5.697389459609985,
      "epoch": 2.857402031930334,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041884578697269994,
      "loss": 4.9944,
      "mean_token_accuracy": 0.2107715055346489,
      "num_tokens": 65379624.0,
      "step": 31500
    },
    {
      "entropy": 5.6691365242004395,
      "epoch": 2.8578555878084178,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004188209051693807,
      "loss": 5.0026,
      "mean_token_accuracy": 0.21422182619571686,
      "num_tokens": 65390417.0,
      "step": 31505
    },
    {
      "entropy": 5.528211688995361,
      "epoch": 2.858309143686502,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041879602039180856,
      "loss": 4.7555,
      "mean_token_accuracy": 0.22769322842359543,
      "num_tokens": 65401681.0,
      "step": 31510
    },
    {
      "entropy": 5.604157781600952,
      "epoch": 2.8587626995645863,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041877113264049805,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22010585367679597,
      "num_tokens": 65412520.0,
      "step": 31515
    },
    {
      "entropy": 5.608847570419312,
      "epoch": 2.8592162554426706,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041874624191596383,
      "loss": 4.9251,
      "mean_token_accuracy": 0.219710573554039,
      "num_tokens": 65423201.0,
      "step": 31520
    },
    {
      "entropy": 5.571673011779785,
      "epoch": 2.859669811320755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004187213482187207,
      "loss": 4.9422,
      "mean_token_accuracy": 0.21446287333965303,
      "num_tokens": 65432778.0,
      "step": 31525
    },
    {
      "entropy": 5.682842016220093,
      "epoch": 2.860123367198839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004186964515492835,
      "loss": 4.9373,
      "mean_token_accuracy": 0.21742546260356904,
      "num_tokens": 65442323.0,
      "step": 31530
    },
    {
      "entropy": 5.628350114822387,
      "epoch": 2.8605769230769234,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004186715519081671,
      "loss": 4.9275,
      "mean_token_accuracy": 0.21248687505722047,
      "num_tokens": 65451980.0,
      "step": 31535
    },
    {
      "entropy": 5.523522758483887,
      "epoch": 2.861030478955007,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004186466492958864,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21626941710710526,
      "num_tokens": 65463359.0,
      "step": 31540
    },
    {
      "entropy": 5.597301912307739,
      "epoch": 2.8614840348330914,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004186217437129562,
      "loss": 4.9267,
      "mean_token_accuracy": 0.21817844212055207,
      "num_tokens": 65474155.0,
      "step": 31545
    },
    {
      "entropy": 5.627587461471558,
      "epoch": 2.8619375907111757,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041859683515989196,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22138632982969284,
      "num_tokens": 65484884.0,
      "step": 31550
    },
    {
      "entropy": 5.667737340927124,
      "epoch": 2.8623911465892595,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041857192363720845,
      "loss": 4.9437,
      "mean_token_accuracy": 0.21947628408670425,
      "num_tokens": 65494380.0,
      "step": 31555
    },
    {
      "entropy": 5.621940040588379,
      "epoch": 2.862844702467344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004185470091454211,
      "loss": 4.9228,
      "mean_token_accuracy": 0.21476275473833084,
      "num_tokens": 65505735.0,
      "step": 31560
    },
    {
      "entropy": 5.637242174148559,
      "epoch": 2.863298258345428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004185220916850449,
      "loss": 4.8794,
      "mean_token_accuracy": 0.21687520444393157,
      "num_tokens": 65515956.0,
      "step": 31565
    },
    {
      "entropy": 5.581989574432373,
      "epoch": 2.8637518142235123,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041849717125659537,
      "loss": 4.8589,
      "mean_token_accuracy": 0.21592623740434647,
      "num_tokens": 65526369.0,
      "step": 31570
    },
    {
      "entropy": 5.610959386825561,
      "epoch": 2.8642053701015966,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004184722478605877,
      "loss": 4.8863,
      "mean_token_accuracy": 0.221421717107296,
      "num_tokens": 65536032.0,
      "step": 31575
    },
    {
      "entropy": 5.59868106842041,
      "epoch": 2.864658925979681,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004184473214975374,
      "loss": 4.9567,
      "mean_token_accuracy": 0.2114228069782257,
      "num_tokens": 65546121.0,
      "step": 31580
    },
    {
      "entropy": 5.607974004745484,
      "epoch": 2.865112481857765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041842239216795994,
      "loss": 4.907,
      "mean_token_accuracy": 0.21633208245038987,
      "num_tokens": 65555915.0,
      "step": 31585
    },
    {
      "entropy": 5.616454839706421,
      "epoch": 2.865566037735849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041839745987237086,
      "loss": 4.946,
      "mean_token_accuracy": 0.20695098042488097,
      "num_tokens": 65566267.0,
      "step": 31590
    },
    {
      "entropy": 5.6284112453460695,
      "epoch": 2.866019593613933,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004183725246112857,
      "loss": 4.898,
      "mean_token_accuracy": 0.22578234821558,
      "num_tokens": 65577126.0,
      "step": 31595
    },
    {
      "entropy": 5.659144592285156,
      "epoch": 2.8664731494920175,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041834758638522034,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20263938456773758,
      "num_tokens": 65587158.0,
      "step": 31600
    },
    {
      "entropy": 5.620755338668824,
      "epoch": 2.8669267053701017,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041832264519469026,
      "loss": 4.901,
      "mean_token_accuracy": 0.2108721911907196,
      "num_tokens": 65597720.0,
      "step": 31605
    },
    {
      "entropy": 5.6876058101654055,
      "epoch": 2.8673802612481856,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041829770104021143,
      "loss": 4.9446,
      "mean_token_accuracy": 0.2152615189552307,
      "num_tokens": 65607409.0,
      "step": 31610
    },
    {
      "entropy": 5.633953857421875,
      "epoch": 2.86783381712627,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041827275392229956,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2274174064397812,
      "num_tokens": 65616785.0,
      "step": 31615
    },
    {
      "entropy": 5.601583051681518,
      "epoch": 2.868287373004354,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004182478038414706,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21374938786029815,
      "num_tokens": 65625794.0,
      "step": 31620
    },
    {
      "entropy": 5.51266450881958,
      "epoch": 2.8687409288824384,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041822285079824056,
      "loss": 4.7989,
      "mean_token_accuracy": 0.22594573050737382,
      "num_tokens": 65635789.0,
      "step": 31625
    },
    {
      "entropy": 5.654076671600341,
      "epoch": 2.8691944847605226,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004181978947931256,
      "loss": 5.0388,
      "mean_token_accuracy": 0.20779097974300384,
      "num_tokens": 65646266.0,
      "step": 31630
    },
    {
      "entropy": 5.6497128963470455,
      "epoch": 2.869648040638607,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004181729358266415,
      "loss": 4.929,
      "mean_token_accuracy": 0.21519651114940644,
      "num_tokens": 65656538.0,
      "step": 31635
    },
    {
      "entropy": 5.686936235427856,
      "epoch": 2.8701015965166907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041814797389930474,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21691731959581376,
      "num_tokens": 65666926.0,
      "step": 31640
    },
    {
      "entropy": 5.635845708847046,
      "epoch": 2.870555152394775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004181230090116313,
      "loss": 4.9085,
      "mean_token_accuracy": 0.21366910636425018,
      "num_tokens": 65677219.0,
      "step": 31645
    },
    {
      "entropy": 5.615746116638183,
      "epoch": 2.8710087082728593,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041809804116413763,
      "loss": 4.9428,
      "mean_token_accuracy": 0.21356865614652634,
      "num_tokens": 65688281.0,
      "step": 31650
    },
    {
      "entropy": 5.741971254348755,
      "epoch": 2.8714622641509435,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041807307035734,
      "loss": 5.0311,
      "mean_token_accuracy": 0.20811943411827089,
      "num_tokens": 65698815.0,
      "step": 31655
    },
    {
      "entropy": 5.656602048873902,
      "epoch": 2.8719158200290273,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004180480965917548,
      "loss": 4.9463,
      "mean_token_accuracy": 0.2153574302792549,
      "num_tokens": 65709481.0,
      "step": 31660
    },
    {
      "entropy": 5.693263959884644,
      "epoch": 2.8723693759071116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041802311986789854,
      "loss": 4.9276,
      "mean_token_accuracy": 0.21124169677495958,
      "num_tokens": 65720561.0,
      "step": 31665
    },
    {
      "entropy": 5.6647398471832275,
      "epoch": 2.872822931785196,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041799814018628765,
      "loss": 4.9833,
      "mean_token_accuracy": 0.21254894137382507,
      "num_tokens": 65730384.0,
      "step": 31670
    },
    {
      "entropy": 5.61438627243042,
      "epoch": 2.87327648766328,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041797315754743883,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21776314228773117,
      "num_tokens": 65740711.0,
      "step": 31675
    },
    {
      "entropy": 5.724636650085449,
      "epoch": 2.8737300435413644,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041794817195186867,
      "loss": 5.0394,
      "mean_token_accuracy": 0.20853618234395982,
      "num_tokens": 65751033.0,
      "step": 31680
    },
    {
      "entropy": 5.70962347984314,
      "epoch": 2.8741835994194487,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004179231834000939,
      "loss": 5.0175,
      "mean_token_accuracy": 0.210690413415432,
      "num_tokens": 65761500.0,
      "step": 31685
    },
    {
      "entropy": 5.58393497467041,
      "epoch": 2.8746371552975325,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041789819189263117,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2188462033867836,
      "num_tokens": 65771095.0,
      "step": 31690
    },
    {
      "entropy": 5.580032110214233,
      "epoch": 2.8750907111756168,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041787319742999756,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21615469604730606,
      "num_tokens": 65782058.0,
      "step": 31695
    },
    {
      "entropy": 5.634597969055176,
      "epoch": 2.875544267053701,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004178482000127098,
      "loss": 4.9624,
      "mean_token_accuracy": 0.22178417593240737,
      "num_tokens": 65791710.0,
      "step": 31700
    },
    {
      "entropy": 5.668760681152344,
      "epoch": 2.8759978229317853,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004178231996412848,
      "loss": 5.0009,
      "mean_token_accuracy": 0.20415839403867722,
      "num_tokens": 65802180.0,
      "step": 31705
    },
    {
      "entropy": 5.593621873855591,
      "epoch": 2.876451378809869,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004177981963162396,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22886699289083481,
      "num_tokens": 65812998.0,
      "step": 31710
    },
    {
      "entropy": 5.638305568695069,
      "epoch": 2.8769049346879534,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004177731900380914,
      "loss": 4.94,
      "mean_token_accuracy": 0.2166304796934128,
      "num_tokens": 65824210.0,
      "step": 31715
    },
    {
      "entropy": 5.5992003917694095,
      "epoch": 2.8773584905660377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004177481808073572,
      "loss": 4.8831,
      "mean_token_accuracy": 0.22450800091028214,
      "num_tokens": 65834322.0,
      "step": 31720
    },
    {
      "entropy": 5.616180801391602,
      "epoch": 2.877812046444122,
      "grad_norm": 1.125,
      "learning_rate": 0.00041772316862455426,
      "loss": 4.8343,
      "mean_token_accuracy": 0.227716264128685,
      "num_tokens": 65845347.0,
      "step": 31725
    },
    {
      "entropy": 5.677000951766968,
      "epoch": 2.878265602322206,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041769815349019986,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21536764055490493,
      "num_tokens": 65855616.0,
      "step": 31730
    },
    {
      "entropy": 5.667771053314209,
      "epoch": 2.8787191582002905,
      "grad_norm": 1.640625,
      "learning_rate": 0.00041767313540481125,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2174030527472496,
      "num_tokens": 65865542.0,
      "step": 31735
    },
    {
      "entropy": 5.582042121887207,
      "epoch": 2.8791727140783747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004176481143689058,
      "loss": 4.9075,
      "mean_token_accuracy": 0.21429865807294846,
      "num_tokens": 65874478.0,
      "step": 31740
    },
    {
      "entropy": 5.667848300933838,
      "epoch": 2.8796262699564585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004176230903830011,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21303968429565429,
      "num_tokens": 65884609.0,
      "step": 31745
    },
    {
      "entropy": 5.6350995063781735,
      "epoch": 2.880079825834543,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004175980634476145,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21959954649209976,
      "num_tokens": 65894486.0,
      "step": 31750
    },
    {
      "entropy": 5.578351402282715,
      "epoch": 2.880533381712627,
      "grad_norm": 1.25,
      "learning_rate": 0.0004175730335632635,
      "loss": 4.8766,
      "mean_token_accuracy": 0.219723142683506,
      "num_tokens": 65904643.0,
      "step": 31755
    },
    {
      "entropy": 5.521179628372193,
      "epoch": 2.8809869375907113,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041754800073046593,
      "loss": 4.8858,
      "mean_token_accuracy": 0.21895606815814972,
      "num_tokens": 65914107.0,
      "step": 31760
    },
    {
      "entropy": 5.565024185180664,
      "epoch": 2.881440493468795,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004175229649497394,
      "loss": 4.8429,
      "mean_token_accuracy": 0.2212000772356987,
      "num_tokens": 65925112.0,
      "step": 31765
    },
    {
      "entropy": 5.662316799163818,
      "epoch": 2.8818940493468794,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004174979262216016,
      "loss": 4.9436,
      "mean_token_accuracy": 0.21651653051376343,
      "num_tokens": 65934624.0,
      "step": 31770
    },
    {
      "entropy": 5.674941492080689,
      "epoch": 2.8823476052249637,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004174728845465704,
      "loss": 4.902,
      "mean_token_accuracy": 0.21791282445192336,
      "num_tokens": 65945951.0,
      "step": 31775
    },
    {
      "entropy": 5.555245065689087,
      "epoch": 2.882801161103048,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004174478399251636,
      "loss": 4.8376,
      "mean_token_accuracy": 0.22578643411397933,
      "num_tokens": 65956871.0,
      "step": 31780
    },
    {
      "entropy": 5.70052547454834,
      "epoch": 2.8832547169811322,
      "grad_norm": 1.125,
      "learning_rate": 0.0004174227923578992,
      "loss": 5.0504,
      "mean_token_accuracy": 0.2095884695649147,
      "num_tokens": 65967814.0,
      "step": 31785
    },
    {
      "entropy": 5.669099140167236,
      "epoch": 2.8837082728592165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004173977418452952,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2096838042140007,
      "num_tokens": 65978118.0,
      "step": 31790
    },
    {
      "entropy": 5.5983318328857425,
      "epoch": 2.8841618287373003,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041737268838786955,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22294967621564865,
      "num_tokens": 65988240.0,
      "step": 31795
    },
    {
      "entropy": 5.619547128677368,
      "epoch": 2.8846153846153846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004173476319861404,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22018044739961623,
      "num_tokens": 65998126.0,
      "step": 31800
    },
    {
      "entropy": 5.647696685791016,
      "epoch": 2.885068940493469,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004173225726406261,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2179265409708023,
      "num_tokens": 66009118.0,
      "step": 31805
    },
    {
      "entropy": 5.669038438796997,
      "epoch": 2.885522496371553,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041729751035184457,
      "loss": 4.9893,
      "mean_token_accuracy": 0.2082985132932663,
      "num_tokens": 66019340.0,
      "step": 31810
    },
    {
      "entropy": 5.616820812225342,
      "epoch": 2.885976052249637,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041727244512031425,
      "loss": 4.8987,
      "mean_token_accuracy": 0.21786166727542877,
      "num_tokens": 66029784.0,
      "step": 31815
    },
    {
      "entropy": 5.674930095672607,
      "epoch": 2.886429608127721,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004172473769465536,
      "loss": 5.0498,
      "mean_token_accuracy": 0.20394045412540435,
      "num_tokens": 66039987.0,
      "step": 31820
    },
    {
      "entropy": 5.635810041427613,
      "epoch": 2.8868831640058055,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041722230583108097,
      "loss": 4.9512,
      "mean_token_accuracy": 0.22123308032751082,
      "num_tokens": 66049150.0,
      "step": 31825
    },
    {
      "entropy": 5.662168979644775,
      "epoch": 2.8873367198838897,
      "grad_norm": 1.125,
      "learning_rate": 0.00041719723177441467,
      "loss": 4.9604,
      "mean_token_accuracy": 0.21941876113414766,
      "num_tokens": 66060719.0,
      "step": 31830
    },
    {
      "entropy": 5.685385227203369,
      "epoch": 2.887790275761974,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004171721547770735,
      "loss": 4.9699,
      "mean_token_accuracy": 0.21484654694795607,
      "num_tokens": 66071126.0,
      "step": 31835
    },
    {
      "entropy": 5.6451198101043705,
      "epoch": 2.8882438316400583,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004171470748395759,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22343157678842546,
      "num_tokens": 66080928.0,
      "step": 31840
    },
    {
      "entropy": 5.654672479629516,
      "epoch": 2.888697387518142,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004171219919624406,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21406376659870147,
      "num_tokens": 66090900.0,
      "step": 31845
    },
    {
      "entropy": 5.630568313598633,
      "epoch": 2.8891509433962264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041709690614618624,
      "loss": 4.9519,
      "mean_token_accuracy": 0.21272231340408326,
      "num_tokens": 66100720.0,
      "step": 31850
    },
    {
      "entropy": 5.608840465545654,
      "epoch": 2.8896044992743106,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041707181739133163,
      "loss": 4.9656,
      "mean_token_accuracy": 0.20827713757753372,
      "num_tokens": 66110780.0,
      "step": 31855
    },
    {
      "entropy": 5.6194854259490965,
      "epoch": 2.890058055152395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004170467256983957,
      "loss": 4.9012,
      "mean_token_accuracy": 0.22411205172538756,
      "num_tokens": 66120976.0,
      "step": 31860
    },
    {
      "entropy": 5.649025964736938,
      "epoch": 2.8905116110304787,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004170216310678973,
      "loss": 4.9192,
      "mean_token_accuracy": 0.2179570972919464,
      "num_tokens": 66130221.0,
      "step": 31865
    },
    {
      "entropy": 5.59353404045105,
      "epoch": 2.890965166908563,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041699653350035527,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22496102303266524,
      "num_tokens": 66140060.0,
      "step": 31870
    },
    {
      "entropy": 5.6029860496521,
      "epoch": 2.8914187227866472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041697143299628886,
      "loss": 4.9216,
      "mean_token_accuracy": 0.2239190846681595,
      "num_tokens": 66151640.0,
      "step": 31875
    },
    {
      "entropy": 5.5633018016815186,
      "epoch": 2.8918722786647315,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041694632955621697,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22784023582935334,
      "num_tokens": 66162923.0,
      "step": 31880
    },
    {
      "entropy": 5.6306946754455565,
      "epoch": 2.8923258345428158,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041692122318065885,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21433405727148055,
      "num_tokens": 66173108.0,
      "step": 31885
    },
    {
      "entropy": 5.612142038345337,
      "epoch": 2.8927793904209,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041689611387013363,
      "loss": 4.8848,
      "mean_token_accuracy": 0.22003877758979798,
      "num_tokens": 66185280.0,
      "step": 31890
    },
    {
      "entropy": 5.603176927566528,
      "epoch": 2.8932329462989843,
      "grad_norm": 1.125,
      "learning_rate": 0.0004168710016251606,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21247122883796693,
      "num_tokens": 66196020.0,
      "step": 31895
    },
    {
      "entropy": 5.61596155166626,
      "epoch": 2.893686502177068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004168458864462592,
      "loss": 4.8489,
      "mean_token_accuracy": 0.21819493621587754,
      "num_tokens": 66205753.0,
      "step": 31900
    },
    {
      "entropy": 5.661674928665161,
      "epoch": 2.8941400580551524,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041682076833394864,
      "loss": 5.0223,
      "mean_token_accuracy": 0.21034709960222245,
      "num_tokens": 66218415.0,
      "step": 31905
    },
    {
      "entropy": 5.652190923690796,
      "epoch": 2.8945936139332367,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041679564728874845,
      "loss": 5.0076,
      "mean_token_accuracy": 0.2117157071828842,
      "num_tokens": 66228090.0,
      "step": 31910
    },
    {
      "entropy": 5.564500522613526,
      "epoch": 2.8950471698113205,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004167705233111782,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2234255254268646,
      "num_tokens": 66237587.0,
      "step": 31915
    },
    {
      "entropy": 5.601796293258667,
      "epoch": 2.8955007256894048,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041674539640175733,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21766695529222488,
      "num_tokens": 66248162.0,
      "step": 31920
    },
    {
      "entropy": 5.597803354263306,
      "epoch": 2.895954281567489,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004167202665610055,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22405479848384857,
      "num_tokens": 66258370.0,
      "step": 31925
    },
    {
      "entropy": 5.634462833404541,
      "epoch": 2.8964078374455733,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041669513378944254,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2120415061712265,
      "num_tokens": 66269267.0,
      "step": 31930
    },
    {
      "entropy": 5.683089113235473,
      "epoch": 2.8968613933236576,
      "grad_norm": 1.125,
      "learning_rate": 0.00041666999808758815,
      "loss": 4.9353,
      "mean_token_accuracy": 0.2169550195336342,
      "num_tokens": 66280077.0,
      "step": 31935
    },
    {
      "entropy": 5.650359058380127,
      "epoch": 2.897314949201742,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000416644859455962,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22299490123987198,
      "num_tokens": 66291678.0,
      "step": 31940
    },
    {
      "entropy": 5.599107694625855,
      "epoch": 2.897768505079826,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004166197178950841,
      "loss": 4.9125,
      "mean_token_accuracy": 0.2158701941370964,
      "num_tokens": 66302046.0,
      "step": 31945
    },
    {
      "entropy": 5.594408178329468,
      "epoch": 2.89822206095791,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041659457340547436,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21868760287761688,
      "num_tokens": 66311888.0,
      "step": 31950
    },
    {
      "entropy": 5.566522884368896,
      "epoch": 2.898675616835994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041656942598765275,
      "loss": 4.8745,
      "mean_token_accuracy": 0.21936390548944473,
      "num_tokens": 66321507.0,
      "step": 31955
    },
    {
      "entropy": 5.549931955337525,
      "epoch": 2.8991291727140784,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041654427564213924,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22292966842651368,
      "num_tokens": 66331458.0,
      "step": 31960
    },
    {
      "entropy": 5.5977123260498045,
      "epoch": 2.8995827285921627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041651912236945416,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22431049793958663,
      "num_tokens": 66340875.0,
      "step": 31965
    },
    {
      "entropy": 5.624931573867798,
      "epoch": 2.9000362844702465,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041649396617011753,
      "loss": 4.9808,
      "mean_token_accuracy": 0.21221574693918227,
      "num_tokens": 66350784.0,
      "step": 31970
    },
    {
      "entropy": 5.646308517456054,
      "epoch": 2.900489840348331,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004164688070446496,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21126966327428817,
      "num_tokens": 66361591.0,
      "step": 31975
    },
    {
      "entropy": 5.6111658096313475,
      "epoch": 2.900943396226415,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004164436449935707,
      "loss": 4.8731,
      "mean_token_accuracy": 0.2227505549788475,
      "num_tokens": 66371540.0,
      "step": 31980
    },
    {
      "entropy": 5.624380397796631,
      "epoch": 2.9013969521044993,
      "grad_norm": 1.25,
      "learning_rate": 0.00041641848001740126,
      "loss": 4.9313,
      "mean_token_accuracy": 0.21938995122909546,
      "num_tokens": 66382053.0,
      "step": 31985
    },
    {
      "entropy": 5.60094256401062,
      "epoch": 2.9018505079825836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004163933121166615,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21962404996156693,
      "num_tokens": 66392563.0,
      "step": 31990
    },
    {
      "entropy": 5.602155923843384,
      "epoch": 2.902304063860668,
      "grad_norm": 1.109375,
      "learning_rate": 0.000416368141291872,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2217767745256424,
      "num_tokens": 66403582.0,
      "step": 31995
    },
    {
      "entropy": 5.5652412414550785,
      "epoch": 2.9027576197387517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041634296754355346,
      "loss": 4.8405,
      "mean_token_accuracy": 0.21778323501348495,
      "num_tokens": 66413713.0,
      "step": 32000
    },
    {
      "entropy": 5.605284976959228,
      "epoch": 2.903211175616836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004163177908722262,
      "loss": 4.918,
      "mean_token_accuracy": 0.21381809562444687,
      "num_tokens": 66424802.0,
      "step": 32005
    },
    {
      "entropy": 5.617951345443726,
      "epoch": 2.90366473149492,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041629261127841105,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22058770209550857,
      "num_tokens": 66434716.0,
      "step": 32010
    },
    {
      "entropy": 5.614213752746582,
      "epoch": 2.9041182873730045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041626742876262876,
      "loss": 4.9343,
      "mean_token_accuracy": 0.2156784012913704,
      "num_tokens": 66445725.0,
      "step": 32015
    },
    {
      "entropy": 5.639075469970703,
      "epoch": 2.9045718432510883,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041624224332539997,
      "loss": 4.9778,
      "mean_token_accuracy": 0.2084185928106308,
      "num_tokens": 66457492.0,
      "step": 32020
    },
    {
      "entropy": 5.624993801116943,
      "epoch": 2.9050253991291726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004162170549672455,
      "loss": 4.8962,
      "mean_token_accuracy": 0.21812325119972228,
      "num_tokens": 66467490.0,
      "step": 32025
    },
    {
      "entropy": 5.650731945037842,
      "epoch": 2.905478955007257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004161918636886865,
      "loss": 5.005,
      "mean_token_accuracy": 0.20764007717370986,
      "num_tokens": 66478160.0,
      "step": 32030
    },
    {
      "entropy": 5.658470392227173,
      "epoch": 2.905932510885341,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004161666694902437,
      "loss": 4.918,
      "mean_token_accuracy": 0.21377474516630174,
      "num_tokens": 66488181.0,
      "step": 32035
    },
    {
      "entropy": 5.644133853912353,
      "epoch": 2.9063860667634254,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004161414723724383,
      "loss": 4.9597,
      "mean_token_accuracy": 0.217026250064373,
      "num_tokens": 66498507.0,
      "step": 32040
    },
    {
      "entropy": 5.538712215423584,
      "epoch": 2.9068396226415096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004161162723357912,
      "loss": 4.8498,
      "mean_token_accuracy": 0.21767690330743789,
      "num_tokens": 66507752.0,
      "step": 32045
    },
    {
      "entropy": 5.547859191894531,
      "epoch": 2.9072931785195935,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004160910693808236,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21349812895059586,
      "num_tokens": 66519360.0,
      "step": 32050
    },
    {
      "entropy": 5.6784350872039795,
      "epoch": 2.9077467343976777,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041606586350805686,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2138091057538986,
      "num_tokens": 66529790.0,
      "step": 32055
    },
    {
      "entropy": 5.61729998588562,
      "epoch": 2.908200290275762,
      "grad_norm": 1.140625,
      "learning_rate": 0.000416040654718012,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2244242712855339,
      "num_tokens": 66539617.0,
      "step": 32060
    },
    {
      "entropy": 5.602614545822144,
      "epoch": 2.9086538461538463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041601544301121045,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22266987711191177,
      "num_tokens": 66549861.0,
      "step": 32065
    },
    {
      "entropy": 5.684524869918823,
      "epoch": 2.90910740203193,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004159902283881737,
      "loss": 5.0253,
      "mean_token_accuracy": 0.21683745980262756,
      "num_tokens": 66560388.0,
      "step": 32070
    },
    {
      "entropy": 5.676492166519165,
      "epoch": 2.9095609579100143,
      "grad_norm": 1.15625,
      "learning_rate": 0.000415965010849423,
      "loss": 4.994,
      "mean_token_accuracy": 0.21124920397996902,
      "num_tokens": 66570795.0,
      "step": 32075
    },
    {
      "entropy": 5.6598851680755615,
      "epoch": 2.9100145137880986,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041593979039548,
      "loss": 4.9487,
      "mean_token_accuracy": 0.20902905017137527,
      "num_tokens": 66580268.0,
      "step": 32080
    },
    {
      "entropy": 5.649728679656983,
      "epoch": 2.910468069666183,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004159145670268662,
      "loss": 4.972,
      "mean_token_accuracy": 0.20908472836017608,
      "num_tokens": 66591192.0,
      "step": 32085
    },
    {
      "entropy": 5.629008483886719,
      "epoch": 2.910921625544267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004158893407441033,
      "loss": 5.0082,
      "mean_token_accuracy": 0.21610246151685714,
      "num_tokens": 66600768.0,
      "step": 32090
    },
    {
      "entropy": 5.739854717254639,
      "epoch": 2.9113751814223514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041586411154771285,
      "loss": 4.9735,
      "mean_token_accuracy": 0.2134055495262146,
      "num_tokens": 66612207.0,
      "step": 32095
    },
    {
      "entropy": 5.593867921829224,
      "epoch": 2.9118287373004357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041583887943821667,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22379922568798066,
      "num_tokens": 66622175.0,
      "step": 32100
    },
    {
      "entropy": 5.591528224945068,
      "epoch": 2.9122822931785195,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004158136444161367,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21095404475927354,
      "num_tokens": 66633936.0,
      "step": 32105
    },
    {
      "entropy": 5.6519874095916744,
      "epoch": 2.9127358490566038,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041578840648199455,
      "loss": 4.9773,
      "mean_token_accuracy": 0.20750181525945663,
      "num_tokens": 66643502.0,
      "step": 32110
    },
    {
      "entropy": 5.620491933822632,
      "epoch": 2.913189404934688,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004157631656363124,
      "loss": 4.9091,
      "mean_token_accuracy": 0.21942169815301896,
      "num_tokens": 66654890.0,
      "step": 32115
    },
    {
      "entropy": 5.636258602142334,
      "epoch": 2.9136429608127723,
      "grad_norm": 1.0625,
      "learning_rate": 0.000415737921879612,
      "loss": 5.0105,
      "mean_token_accuracy": 0.2002956375479698,
      "num_tokens": 66667037.0,
      "step": 32120
    },
    {
      "entropy": 5.62758092880249,
      "epoch": 2.914096516690856,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041571267521241554,
      "loss": 4.8873,
      "mean_token_accuracy": 0.21479302197694777,
      "num_tokens": 66677291.0,
      "step": 32125
    },
    {
      "entropy": 5.548876094818115,
      "epoch": 2.9145500725689404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004156874256352451,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2155035600066185,
      "num_tokens": 66687421.0,
      "step": 32130
    },
    {
      "entropy": 5.692036962509155,
      "epoch": 2.9150036284470247,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004156621731486228,
      "loss": 5.0043,
      "mean_token_accuracy": 0.208040988445282,
      "num_tokens": 66697190.0,
      "step": 32135
    },
    {
      "entropy": 5.642501401901245,
      "epoch": 2.915457184325109,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000415636917753071,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21206559985876083,
      "num_tokens": 66707255.0,
      "step": 32140
    },
    {
      "entropy": 5.571289014816284,
      "epoch": 2.915910740203193,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004156116594491119,
      "loss": 4.8466,
      "mean_token_accuracy": 0.21873885542154312,
      "num_tokens": 66716451.0,
      "step": 32145
    },
    {
      "entropy": 5.56790771484375,
      "epoch": 2.9163642960812775,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004155863982372678,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22132851779460908,
      "num_tokens": 66727157.0,
      "step": 32150
    },
    {
      "entropy": 5.655651569366455,
      "epoch": 2.9168178519593613,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041556113411806117,
      "loss": 4.9386,
      "mean_token_accuracy": 0.21709243059158326,
      "num_tokens": 66737523.0,
      "step": 32155
    },
    {
      "entropy": 5.6316324234008786,
      "epoch": 2.9172714078374455,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041553586709201453,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21690703779459,
      "num_tokens": 66748183.0,
      "step": 32160
    },
    {
      "entropy": 5.662561321258545,
      "epoch": 2.91772496371553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004155105971596503,
      "loss": 4.9034,
      "mean_token_accuracy": 0.21917158663272857,
      "num_tokens": 66758794.0,
      "step": 32165
    },
    {
      "entropy": 5.697680997848511,
      "epoch": 2.918178519593614,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041548532432149105,
      "loss": 5.02,
      "mean_token_accuracy": 0.2154909536242485,
      "num_tokens": 66769916.0,
      "step": 32170
    },
    {
      "entropy": 5.648323678970337,
      "epoch": 2.918632075471698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004154600485780595,
      "loss": 4.9237,
      "mean_token_accuracy": 0.2226147621870041,
      "num_tokens": 66779658.0,
      "step": 32175
    },
    {
      "entropy": 5.577060747146606,
      "epoch": 2.919085631349782,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041543476992987843,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22520497143268586,
      "num_tokens": 66788909.0,
      "step": 32180
    },
    {
      "entropy": 5.681152248382569,
      "epoch": 2.9195391872278664,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041540948837747056,
      "loss": 4.9705,
      "mean_token_accuracy": 0.2129875659942627,
      "num_tokens": 66799519.0,
      "step": 32185
    },
    {
      "entropy": 5.577448129653931,
      "epoch": 2.9199927431059507,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004153842039213586,
      "loss": 4.8804,
      "mean_token_accuracy": 0.21812061965465546,
      "num_tokens": 66808915.0,
      "step": 32190
    },
    {
      "entropy": 5.637794399261475,
      "epoch": 2.920446298984035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004153589165620656,
      "loss": 4.9689,
      "mean_token_accuracy": 0.20915681272745132,
      "num_tokens": 66818819.0,
      "step": 32195
    },
    {
      "entropy": 5.53039927482605,
      "epoch": 2.9208998548621192,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041533362630011435,
      "loss": 4.8114,
      "mean_token_accuracy": 0.2310406669974327,
      "num_tokens": 66829274.0,
      "step": 32200
    },
    {
      "entropy": 5.710303831100464,
      "epoch": 2.921353410740203,
      "grad_norm": 1.203125,
      "learning_rate": 0.000415308333136028,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21913468688726426,
      "num_tokens": 66839659.0,
      "step": 32205
    },
    {
      "entropy": 5.642279148101807,
      "epoch": 2.9218069666182873,
      "grad_norm": 1.390625,
      "learning_rate": 0.00041528303707032957,
      "loss": 4.9819,
      "mean_token_accuracy": 0.2132548525929451,
      "num_tokens": 66850415.0,
      "step": 32210
    },
    {
      "entropy": 5.621463203430176,
      "epoch": 2.9222605224963716,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004152577381035422,
      "loss": 4.881,
      "mean_token_accuracy": 0.22740049809217452,
      "num_tokens": 66860663.0,
      "step": 32215
    },
    {
      "entropy": 5.6980788230896,
      "epoch": 2.922714078374456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041523243623618907,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21582137495279313,
      "num_tokens": 66871473.0,
      "step": 32220
    },
    {
      "entropy": 5.59284234046936,
      "epoch": 2.9231676342525397,
      "grad_norm": 1.25,
      "learning_rate": 0.00041520713146879333,
      "loss": 4.8771,
      "mean_token_accuracy": 0.21921272724866867,
      "num_tokens": 66881609.0,
      "step": 32225
    },
    {
      "entropy": 5.654316139221192,
      "epoch": 2.923621190130624,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041518182380187853,
      "loss": 4.9392,
      "mean_token_accuracy": 0.2122931256890297,
      "num_tokens": 66891780.0,
      "step": 32230
    },
    {
      "entropy": 5.664815616607666,
      "epoch": 2.924074746008708,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041515651323596774,
      "loss": 5.0191,
      "mean_token_accuracy": 0.21146702021360397,
      "num_tokens": 66902251.0,
      "step": 32235
    },
    {
      "entropy": 5.632319736480713,
      "epoch": 2.9245283018867925,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041513119977158464,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21739636361598969,
      "num_tokens": 66912538.0,
      "step": 32240
    },
    {
      "entropy": 5.634463930130005,
      "epoch": 2.9249818577648767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004151058834092526,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2178140252828598,
      "num_tokens": 66922648.0,
      "step": 32245
    },
    {
      "entropy": 5.621490955352783,
      "epoch": 2.925435413642961,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041508056414949515,
      "loss": 4.8505,
      "mean_token_accuracy": 0.2177532583475113,
      "num_tokens": 66933115.0,
      "step": 32250
    },
    {
      "entropy": 5.690641212463379,
      "epoch": 2.9258889695210453,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000415055241992836,
      "loss": 5.0674,
      "mean_token_accuracy": 0.20501761883497238,
      "num_tokens": 66943727.0,
      "step": 32255
    },
    {
      "entropy": 5.6623735427856445,
      "epoch": 2.926342525399129,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004150299169397987,
      "loss": 4.8976,
      "mean_token_accuracy": 0.21861513406038285,
      "num_tokens": 66953488.0,
      "step": 32260
    },
    {
      "entropy": 5.739543867111206,
      "epoch": 2.9267960812772134,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004150045889909071,
      "loss": 5.0359,
      "mean_token_accuracy": 0.20360918641090392,
      "num_tokens": 66964586.0,
      "step": 32265
    },
    {
      "entropy": 5.644960880279541,
      "epoch": 2.9272496371552976,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004149792581466848,
      "loss": 4.9565,
      "mean_token_accuracy": 0.21176497489213944,
      "num_tokens": 66975032.0,
      "step": 32270
    },
    {
      "entropy": 5.633436918258667,
      "epoch": 2.9277031930333814,
      "grad_norm": 1.28125,
      "learning_rate": 0.00041495392440765586,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2188872441649437,
      "num_tokens": 66984910.0,
      "step": 32275
    },
    {
      "entropy": 5.631566429138184,
      "epoch": 2.9281567489114657,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041492858777434407,
      "loss": 4.7275,
      "mean_token_accuracy": 0.2310669168829918,
      "num_tokens": 66995238.0,
      "step": 32280
    },
    {
      "entropy": 5.6392073154449465,
      "epoch": 2.92861030478955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004149032482472734,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2279650390148163,
      "num_tokens": 67005596.0,
      "step": 32285
    },
    {
      "entropy": 5.744066524505615,
      "epoch": 2.9290638606676342,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041487790582696803,
      "loss": 5.0921,
      "mean_token_accuracy": 0.20277939289808272,
      "num_tokens": 67017312.0,
      "step": 32290
    },
    {
      "entropy": 5.6483276844024655,
      "epoch": 2.9295174165457185,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004148525605139518,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2208987608551979,
      "num_tokens": 67026692.0,
      "step": 32295
    },
    {
      "entropy": 5.601144123077392,
      "epoch": 2.9299709724238028,
      "grad_norm": 1.125,
      "learning_rate": 0.00041482721230874897,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21518077105283737,
      "num_tokens": 67036801.0,
      "step": 32300
    },
    {
      "entropy": 5.610764646530152,
      "epoch": 2.930424528301887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041480186121188383,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22005629390478135,
      "num_tokens": 67047262.0,
      "step": 32305
    },
    {
      "entropy": 5.65587887763977,
      "epoch": 2.930878084179971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004147765072238806,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2122058689594269,
      "num_tokens": 67058019.0,
      "step": 32310
    },
    {
      "entropy": 5.696411657333374,
      "epoch": 2.931331640058055,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004147511503452635,
      "loss": 4.9947,
      "mean_token_accuracy": 0.21138232797384263,
      "num_tokens": 67068844.0,
      "step": 32315
    },
    {
      "entropy": 5.681244564056397,
      "epoch": 2.9317851959361394,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041472579057655703,
      "loss": 4.8596,
      "mean_token_accuracy": 0.22963950037956238,
      "num_tokens": 67077400.0,
      "step": 32320
    },
    {
      "entropy": 5.639327430725098,
      "epoch": 2.9322387518142237,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004147004279182856,
      "loss": 5.039,
      "mean_token_accuracy": 0.20738574862480164,
      "num_tokens": 67088352.0,
      "step": 32325
    },
    {
      "entropy": 5.69011960029602,
      "epoch": 2.9326923076923075,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004146750623709737,
      "loss": 4.9819,
      "mean_token_accuracy": 0.21507629454135896,
      "num_tokens": 67097879.0,
      "step": 32330
    },
    {
      "entropy": 5.639441299438476,
      "epoch": 2.9331458635703918,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041464969393514595,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22310535460710526,
      "num_tokens": 67108355.0,
      "step": 32335
    },
    {
      "entropy": 5.6523185729980465,
      "epoch": 2.933599419448476,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004146243226113269,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21199088245630265,
      "num_tokens": 67118348.0,
      "step": 32340
    },
    {
      "entropy": 5.55693039894104,
      "epoch": 2.9340529753265603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041459894840004135,
      "loss": 4.854,
      "mean_token_accuracy": 0.22015868723392487,
      "num_tokens": 67128944.0,
      "step": 32345
    },
    {
      "entropy": 5.65935378074646,
      "epoch": 2.9345065312046446,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041457357130181396,
      "loss": 4.9691,
      "mean_token_accuracy": 0.21251435279846193,
      "num_tokens": 67139365.0,
      "step": 32350
    },
    {
      "entropy": 5.633871269226074,
      "epoch": 2.934960087082729,
      "grad_norm": 1.125,
      "learning_rate": 0.00041454819131716945,
      "loss": 4.9177,
      "mean_token_accuracy": 0.217627677321434,
      "num_tokens": 67150406.0,
      "step": 32355
    },
    {
      "entropy": 5.712287139892578,
      "epoch": 2.9354136429608126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041452280844663295,
      "loss": 5.0901,
      "mean_token_accuracy": 0.20792538970708846,
      "num_tokens": 67161631.0,
      "step": 32360
    },
    {
      "entropy": 5.624512147903443,
      "epoch": 2.935867198838897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004144974226907291,
      "loss": 4.9525,
      "mean_token_accuracy": 0.2217943102121353,
      "num_tokens": 67171735.0,
      "step": 32365
    },
    {
      "entropy": 5.647053384780884,
      "epoch": 2.936320754716981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000414472034049983,
      "loss": 4.971,
      "mean_token_accuracy": 0.21038858592510223,
      "num_tokens": 67181616.0,
      "step": 32370
    },
    {
      "entropy": 5.623701524734497,
      "epoch": 2.9367743105950654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004144466425249197,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21765595227479934,
      "num_tokens": 67192519.0,
      "step": 32375
    },
    {
      "entropy": 5.668231439590454,
      "epoch": 2.9372278664731493,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004144212481160643,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21166486591100692,
      "num_tokens": 67202805.0,
      "step": 32380
    },
    {
      "entropy": 5.652287912368775,
      "epoch": 2.9376814223512335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004143958508239419,
      "loss": 4.92,
      "mean_token_accuracy": 0.21630799770355225,
      "num_tokens": 67212778.0,
      "step": 32385
    },
    {
      "entropy": 5.606048822402954,
      "epoch": 2.938134978229318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004143704506490779,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21764194816350937,
      "num_tokens": 67222588.0,
      "step": 32390
    },
    {
      "entropy": 5.683053207397461,
      "epoch": 2.938588534107402,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004143450475919975,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2170100286602974,
      "num_tokens": 67233022.0,
      "step": 32395
    },
    {
      "entropy": 5.603223848342895,
      "epoch": 2.9390420899854863,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004143196416532259,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2183255895972252,
      "num_tokens": 67242631.0,
      "step": 32400
    },
    {
      "entropy": 5.545856285095215,
      "epoch": 2.9394956458635706,
      "grad_norm": 1.125,
      "learning_rate": 0.0004142942328332887,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22344223409891129,
      "num_tokens": 67252376.0,
      "step": 32405
    },
    {
      "entropy": 5.6171863079071045,
      "epoch": 2.9399492017416544,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004142688211327111,
      "loss": 4.8584,
      "mean_token_accuracy": 0.21825803965330123,
      "num_tokens": 67262209.0,
      "step": 32410
    },
    {
      "entropy": 5.700575971603394,
      "epoch": 2.9404027576197387,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004142434065520189,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21095220297574996,
      "num_tokens": 67272893.0,
      "step": 32415
    },
    {
      "entropy": 5.618619394302368,
      "epoch": 2.940856313497823,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004142179890917376,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21322965621948242,
      "num_tokens": 67284017.0,
      "step": 32420
    },
    {
      "entropy": 5.58260293006897,
      "epoch": 2.941309869375907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041419256875239275,
      "loss": 4.9187,
      "mean_token_accuracy": 0.2184085726737976,
      "num_tokens": 67293674.0,
      "step": 32425
    },
    {
      "entropy": 5.584957838058472,
      "epoch": 2.941763425253991,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004141671455345101,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21590945422649382,
      "num_tokens": 67303759.0,
      "step": 32430
    },
    {
      "entropy": 5.678482389450073,
      "epoch": 2.9422169811320753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041414171943861543,
      "loss": 4.9763,
      "mean_token_accuracy": 0.21104395836591722,
      "num_tokens": 67314962.0,
      "step": 32435
    },
    {
      "entropy": 5.69281702041626,
      "epoch": 2.9426705370101596,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004141162904652346,
      "loss": 5.0493,
      "mean_token_accuracy": 0.2060800611972809,
      "num_tokens": 67325210.0,
      "step": 32440
    },
    {
      "entropy": 5.626198101043701,
      "epoch": 2.943124092888244,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004140908586148934,
      "loss": 4.9631,
      "mean_token_accuracy": 0.2103262260556221,
      "num_tokens": 67335110.0,
      "step": 32445
    },
    {
      "entropy": 5.651398849487305,
      "epoch": 2.943577648766328,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004140654238881178,
      "loss": 5.0342,
      "mean_token_accuracy": 0.21087422370910644,
      "num_tokens": 67345872.0,
      "step": 32450
    },
    {
      "entropy": 5.6606560230255125,
      "epoch": 2.9440312046444124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041403998628543375,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21527948826551438,
      "num_tokens": 67356532.0,
      "step": 32455
    },
    {
      "entropy": 5.6151916027069095,
      "epoch": 2.9444847605224966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041401454580736744,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21767721176147461,
      "num_tokens": 67366603.0,
      "step": 32460
    },
    {
      "entropy": 5.717464780807495,
      "epoch": 2.9449383164005805,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004139891024544448,
      "loss": 5.0483,
      "mean_token_accuracy": 0.21272033005952834,
      "num_tokens": 67377487.0,
      "step": 32465
    },
    {
      "entropy": 5.71870379447937,
      "epoch": 2.9453918722786647,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004139636562271921,
      "loss": 5.0134,
      "mean_token_accuracy": 0.21258057355880738,
      "num_tokens": 67387704.0,
      "step": 32470
    },
    {
      "entropy": 5.668634748458862,
      "epoch": 2.945845428156749,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041393820712613567,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21034898310899736,
      "num_tokens": 67398409.0,
      "step": 32475
    },
    {
      "entropy": 5.624455785751342,
      "epoch": 2.9462989840348333,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041391275515180166,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21351640671491623,
      "num_tokens": 67408905.0,
      "step": 32480
    },
    {
      "entropy": 5.659879970550537,
      "epoch": 2.946752539912917,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004138873003047165,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22312901616096498,
      "num_tokens": 67418466.0,
      "step": 32485
    },
    {
      "entropy": 5.577113056182862,
      "epoch": 2.9472060957910013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004138618425854066,
      "loss": 4.9261,
      "mean_token_accuracy": 0.22014663517475128,
      "num_tokens": 67428597.0,
      "step": 32490
    },
    {
      "entropy": 5.648064041137696,
      "epoch": 2.9476596516690856,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041383638199439824,
      "loss": 4.9099,
      "mean_token_accuracy": 0.21465332210063934,
      "num_tokens": 67439158.0,
      "step": 32495
    },
    {
      "entropy": 5.678783559799195,
      "epoch": 2.94811320754717,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041381091853221835,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21027152836322785,
      "num_tokens": 67449540.0,
      "step": 32500
    },
    {
      "entropy": 5.618897438049316,
      "epoch": 2.948566763425254,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004137854521993932,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2194940432906151,
      "num_tokens": 67459385.0,
      "step": 32505
    },
    {
      "entropy": 5.628474521636963,
      "epoch": 2.9490203193033384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041375998299644946,
      "loss": 4.9878,
      "mean_token_accuracy": 0.2158135861158371,
      "num_tokens": 67469597.0,
      "step": 32510
    },
    {
      "entropy": 5.626009321212768,
      "epoch": 2.9494738751814222,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000413734510923914,
      "loss": 4.8575,
      "mean_token_accuracy": 0.21722062826156616,
      "num_tokens": 67479271.0,
      "step": 32515
    },
    {
      "entropy": 5.594732284545898,
      "epoch": 2.9499274310595065,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004137090359823135,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22307846546173096,
      "num_tokens": 67488997.0,
      "step": 32520
    },
    {
      "entropy": 5.614832019805908,
      "epoch": 2.9503809869375908,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041368355817217474,
      "loss": 4.9181,
      "mean_token_accuracy": 0.21649934053421022,
      "num_tokens": 67499507.0,
      "step": 32525
    },
    {
      "entropy": 5.647829532623291,
      "epoch": 2.950834542815675,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041365807749402467,
      "loss": 4.8761,
      "mean_token_accuracy": 0.21700233072042466,
      "num_tokens": 67509916.0,
      "step": 32530
    },
    {
      "entropy": 5.644301128387451,
      "epoch": 2.951288098693759,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004136325939483902,
      "loss": 4.9118,
      "mean_token_accuracy": 0.21198012083768844,
      "num_tokens": 67519759.0,
      "step": 32535
    },
    {
      "entropy": 5.642089605331421,
      "epoch": 2.951741654571843,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004136071075357984,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2172844350337982,
      "num_tokens": 67529261.0,
      "step": 32540
    },
    {
      "entropy": 5.659871006011963,
      "epoch": 2.9521952104499274,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004135816182567763,
      "loss": 4.9139,
      "mean_token_accuracy": 0.21320982426404952,
      "num_tokens": 67539084.0,
      "step": 32545
    },
    {
      "entropy": 5.669245815277099,
      "epoch": 2.9526487663280117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000413556126111851,
      "loss": 4.9746,
      "mean_token_accuracy": 0.21693313121795654,
      "num_tokens": 67548551.0,
      "step": 32550
    },
    {
      "entropy": 5.646106863021851,
      "epoch": 2.953102322206096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004135306311015497,
      "loss": 4.9153,
      "mean_token_accuracy": 0.21295546889305114,
      "num_tokens": 67558364.0,
      "step": 32555
    },
    {
      "entropy": 5.685067224502563,
      "epoch": 2.95355587808418,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004135051332263996,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22068729251623154,
      "num_tokens": 67569257.0,
      "step": 32560
    },
    {
      "entropy": 5.631531620025635,
      "epoch": 2.954009433962264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041347963248692814,
      "loss": 5.0133,
      "mean_token_accuracy": 0.20805506110191346,
      "num_tokens": 67580135.0,
      "step": 32565
    },
    {
      "entropy": 5.6100989818573,
      "epoch": 2.9544629898403483,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004134541288836626,
      "loss": 4.9225,
      "mean_token_accuracy": 0.2145952120423317,
      "num_tokens": 67589705.0,
      "step": 32570
    },
    {
      "entropy": 5.658466243743897,
      "epoch": 2.9549165457184325,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041342862241713034,
      "loss": 4.883,
      "mean_token_accuracy": 0.22025612592697144,
      "num_tokens": 67599648.0,
      "step": 32575
    },
    {
      "entropy": 5.710981655120849,
      "epoch": 2.955370101596517,
      "grad_norm": 1.25,
      "learning_rate": 0.00041340311308785886,
      "loss": 4.8994,
      "mean_token_accuracy": 0.21508888602256776,
      "num_tokens": 67609624.0,
      "step": 32580
    },
    {
      "entropy": 5.624305534362793,
      "epoch": 2.9558236574746006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041337760089637577,
      "loss": 4.9158,
      "mean_token_accuracy": 0.21643614023923874,
      "num_tokens": 67619596.0,
      "step": 32585
    },
    {
      "entropy": 5.6316124439239506,
      "epoch": 2.956277213352685,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041335208584320865,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21681285798549652,
      "num_tokens": 67629667.0,
      "step": 32590
    },
    {
      "entropy": 5.683869123458862,
      "epoch": 2.956730769230769,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004133265679288851,
      "loss": 4.9416,
      "mean_token_accuracy": 0.21449050605297088,
      "num_tokens": 67640048.0,
      "step": 32595
    },
    {
      "entropy": 5.711351776123047,
      "epoch": 2.9571843251088534,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004133010471539329,
      "loss": 5.0028,
      "mean_token_accuracy": 0.20918145328760146,
      "num_tokens": 67650731.0,
      "step": 32600
    },
    {
      "entropy": 5.630876731872559,
      "epoch": 2.9576378809869377,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041327552351887983,
      "loss": 4.923,
      "mean_token_accuracy": 0.21795839667320252,
      "num_tokens": 67661696.0,
      "step": 32605
    },
    {
      "entropy": 5.661803483963013,
      "epoch": 2.958091436865022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041324999702425366,
      "loss": 4.9548,
      "mean_token_accuracy": 0.2131459206342697,
      "num_tokens": 67671758.0,
      "step": 32610
    },
    {
      "entropy": 5.60401029586792,
      "epoch": 2.9585449927431062,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004132244676705823,
      "loss": 4.9208,
      "mean_token_accuracy": 0.215928290784359,
      "num_tokens": 67682488.0,
      "step": 32615
    },
    {
      "entropy": 5.668144845962525,
      "epoch": 2.95899854862119,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004131989354583938,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21717481166124344,
      "num_tokens": 67692945.0,
      "step": 32620
    },
    {
      "entropy": 5.653250074386596,
      "epoch": 2.9594521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000413173400388216,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22022566795349122,
      "num_tokens": 67702924.0,
      "step": 32625
    },
    {
      "entropy": 5.641571712493897,
      "epoch": 2.9599056603773586,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004131478624605771,
      "loss": 4.958,
      "mean_token_accuracy": 0.21174908727407454,
      "num_tokens": 67714284.0,
      "step": 32630
    },
    {
      "entropy": 5.6709541320800785,
      "epoch": 2.9603592162554424,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004131223216760052,
      "loss": 4.9906,
      "mean_token_accuracy": 0.208943210542202,
      "num_tokens": 67725902.0,
      "step": 32635
    },
    {
      "entropy": 5.682101440429688,
      "epoch": 2.9608127721335267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004130967780350286,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2131103605031967,
      "num_tokens": 67736268.0,
      "step": 32640
    },
    {
      "entropy": 5.714348602294922,
      "epoch": 2.961266328011611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004130712315381753,
      "loss": 4.9634,
      "mean_token_accuracy": 0.21993314772844313,
      "num_tokens": 67745709.0,
      "step": 32645
    },
    {
      "entropy": 5.663623857498169,
      "epoch": 2.961719883889695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041304568218597385,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21612116396427156,
      "num_tokens": 67756092.0,
      "step": 32650
    },
    {
      "entropy": 5.688958215713501,
      "epoch": 2.9621734397677795,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041302012997895246,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2126275986433029,
      "num_tokens": 67766224.0,
      "step": 32655
    },
    {
      "entropy": 5.764392328262329,
      "epoch": 2.9626269956458637,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004129945749176395,
      "loss": 5.0906,
      "mean_token_accuracy": 0.20831307470798494,
      "num_tokens": 67776787.0,
      "step": 32660
    },
    {
      "entropy": 5.633889055252075,
      "epoch": 2.963080551523948,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041296901700256375,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21655898690223693,
      "num_tokens": 67787924.0,
      "step": 32665
    },
    {
      "entropy": 5.661701250076294,
      "epoch": 2.963534107402032,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004129434562342535,
      "loss": 4.9424,
      "mean_token_accuracy": 0.20514094829559326,
      "num_tokens": 67798544.0,
      "step": 32670
    },
    {
      "entropy": 5.5932252407073975,
      "epoch": 2.963987663280116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041291789261323734,
      "loss": 4.9226,
      "mean_token_accuracy": 0.21156342327594757,
      "num_tokens": 67809528.0,
      "step": 32675
    },
    {
      "entropy": 5.53718638420105,
      "epoch": 2.9644412191582004,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041289232614004416,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2308894529938698,
      "num_tokens": 67820263.0,
      "step": 32680
    },
    {
      "entropy": 5.7219658374786375,
      "epoch": 2.9648947750362846,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004128667568152024,
      "loss": 5.0139,
      "mean_token_accuracy": 0.20956661254167558,
      "num_tokens": 67830687.0,
      "step": 32685
    },
    {
      "entropy": 5.760214853286743,
      "epoch": 2.9653483309143684,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041284118463924107,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20952886193990708,
      "num_tokens": 67841714.0,
      "step": 32690
    },
    {
      "entropy": 5.63067135810852,
      "epoch": 2.9658018867924527,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004128156096126889,
      "loss": 4.9232,
      "mean_token_accuracy": 0.21679938733577728,
      "num_tokens": 67851952.0,
      "step": 32695
    },
    {
      "entropy": 5.565825080871582,
      "epoch": 2.966255442670537,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041279003173607477,
      "loss": 4.8124,
      "mean_token_accuracy": 0.22083090990781784,
      "num_tokens": 67861958.0,
      "step": 32700
    },
    {
      "entropy": 5.590000963211059,
      "epoch": 2.9667089985486212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004127644510099278,
      "loss": 4.8142,
      "mean_token_accuracy": 0.22088842391967772,
      "num_tokens": 67871654.0,
      "step": 32705
    },
    {
      "entropy": 5.5833799839019775,
      "epoch": 2.9671625544267055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004127388674347767,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2217372477054596,
      "num_tokens": 67881678.0,
      "step": 32710
    },
    {
      "entropy": 5.605008888244629,
      "epoch": 2.96761611030479,
      "grad_norm": 1.125,
      "learning_rate": 0.00041271328101115075,
      "loss": 4.9526,
      "mean_token_accuracy": 0.21317858546972274,
      "num_tokens": 67892663.0,
      "step": 32715
    },
    {
      "entropy": 5.779286050796509,
      "epoch": 2.9680696661828736,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041268769173957915,
      "loss": 5.1001,
      "mean_token_accuracy": 0.2057110920548439,
      "num_tokens": 67903308.0,
      "step": 32720
    },
    {
      "entropy": 5.702057313919068,
      "epoch": 2.968523222060958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041266209962059097,
      "loss": 4.9089,
      "mean_token_accuracy": 0.22073094993829728,
      "num_tokens": 67913092.0,
      "step": 32725
    },
    {
      "entropy": 5.660963821411133,
      "epoch": 2.968976777939042,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004126365046547154,
      "loss": 4.9089,
      "mean_token_accuracy": 0.2293435201048851,
      "num_tokens": 67923708.0,
      "step": 32730
    },
    {
      "entropy": 5.555353450775146,
      "epoch": 2.9694303338171264,
      "grad_norm": 1.125,
      "learning_rate": 0.0004126109068424819,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21543344408273696,
      "num_tokens": 67934164.0,
      "step": 32735
    },
    {
      "entropy": 5.673551464080811,
      "epoch": 2.96988388969521,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004125853061844198,
      "loss": 5.0004,
      "mean_token_accuracy": 0.21204173415899277,
      "num_tokens": 67943162.0,
      "step": 32740
    },
    {
      "entropy": 5.703625011444092,
      "epoch": 2.9703374455732945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004125597026810585,
      "loss": 5.0104,
      "mean_token_accuracy": 0.21263764351606368,
      "num_tokens": 67953943.0,
      "step": 32745
    },
    {
      "entropy": 5.611979198455811,
      "epoch": 2.9707910014513788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041253409633292747,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21745888590812684,
      "num_tokens": 67964603.0,
      "step": 32750
    },
    {
      "entropy": 5.672776222229004,
      "epoch": 2.971244557329463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041250848714055626,
      "loss": 4.9877,
      "mean_token_accuracy": 0.218428435921669,
      "num_tokens": 67974541.0,
      "step": 32755
    },
    {
      "entropy": 5.683468818664551,
      "epoch": 2.9716981132075473,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004124828751044746,
      "loss": 4.9429,
      "mean_token_accuracy": 0.21311630457639694,
      "num_tokens": 67984968.0,
      "step": 32760
    },
    {
      "entropy": 5.65533857345581,
      "epoch": 2.9721516690856316,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004124572602252119,
      "loss": 4.8878,
      "mean_token_accuracy": 0.2201162576675415,
      "num_tokens": 67995567.0,
      "step": 32765
    },
    {
      "entropy": 5.696247100830078,
      "epoch": 2.9726052249637154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004124316425032981,
      "loss": 5.0443,
      "mean_token_accuracy": 0.2082052394747734,
      "num_tokens": 68008530.0,
      "step": 32770
    },
    {
      "entropy": 5.706831550598144,
      "epoch": 2.9730587808417996,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041240602193926294,
      "loss": 5.0063,
      "mean_token_accuracy": 0.21404547840356827,
      "num_tokens": 68018393.0,
      "step": 32775
    },
    {
      "entropy": 5.698282480239868,
      "epoch": 2.973512336719884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004123803985336362,
      "loss": 4.958,
      "mean_token_accuracy": 0.2079292804002762,
      "num_tokens": 68027677.0,
      "step": 32780
    },
    {
      "entropy": 5.633030271530151,
      "epoch": 2.973965892597968,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041235477228694786,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22678932994604112,
      "num_tokens": 68037157.0,
      "step": 32785
    },
    {
      "entropy": 5.627995824813842,
      "epoch": 2.974419448476052,
      "grad_norm": 1.125,
      "learning_rate": 0.00041232914319972775,
      "loss": 4.9172,
      "mean_token_accuracy": 0.22160340547561647,
      "num_tokens": 68047793.0,
      "step": 32790
    },
    {
      "entropy": 5.648998546600342,
      "epoch": 2.9748730043541363,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041230351127250603,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21816315203905107,
      "num_tokens": 68058155.0,
      "step": 32795
    },
    {
      "entropy": 5.69756293296814,
      "epoch": 2.9753265602322205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041227787650581266,
      "loss": 4.863,
      "mean_token_accuracy": 0.22036237716674806,
      "num_tokens": 68067440.0,
      "step": 32800
    },
    {
      "entropy": 5.6331467628479,
      "epoch": 2.975780116110305,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041225223890017785,
      "loss": 4.953,
      "mean_token_accuracy": 0.2165030851960182,
      "num_tokens": 68077815.0,
      "step": 32805
    },
    {
      "entropy": 5.6192214012146,
      "epoch": 2.976233671988389,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041222659845613173,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2265534892678261,
      "num_tokens": 68087692.0,
      "step": 32810
    },
    {
      "entropy": 5.602951860427856,
      "epoch": 2.9766872278664733,
      "grad_norm": 1.25,
      "learning_rate": 0.00041220095517420464,
      "loss": 4.8743,
      "mean_token_accuracy": 0.21783803254365922,
      "num_tokens": 68097774.0,
      "step": 32815
    },
    {
      "entropy": 5.62679352760315,
      "epoch": 2.9771407837445576,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004121753090549268,
      "loss": 4.951,
      "mean_token_accuracy": 0.2136034682393074,
      "num_tokens": 68107992.0,
      "step": 32820
    },
    {
      "entropy": 5.6071819305419925,
      "epoch": 2.9775943396226414,
      "grad_norm": 1.34375,
      "learning_rate": 0.00041214966009882857,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2273598715662956,
      "num_tokens": 68117103.0,
      "step": 32825
    },
    {
      "entropy": 5.720724439620971,
      "epoch": 2.9780478955007257,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004121240083064405,
      "loss": 4.9875,
      "mean_token_accuracy": 0.21425906270742417,
      "num_tokens": 68127951.0,
      "step": 32830
    },
    {
      "entropy": 5.7262262344360355,
      "epoch": 2.97850145137881,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041209835367829293,
      "loss": 4.9985,
      "mean_token_accuracy": 0.21389012038707733,
      "num_tokens": 68138229.0,
      "step": 32835
    },
    {
      "entropy": 5.578013181686401,
      "epoch": 2.9789550072568938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041207269621491654,
      "loss": 4.9074,
      "mean_token_accuracy": 0.2163704439997673,
      "num_tokens": 68148019.0,
      "step": 32840
    },
    {
      "entropy": 5.569364595413208,
      "epoch": 2.979408563134978,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004120470359168418,
      "loss": 4.9281,
      "mean_token_accuracy": 0.20953331142663956,
      "num_tokens": 68158704.0,
      "step": 32845
    },
    {
      "entropy": 5.704509210586548,
      "epoch": 2.9798621190130623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041202137278459946,
      "loss": 4.926,
      "mean_token_accuracy": 0.22298386693000793,
      "num_tokens": 68169105.0,
      "step": 32850
    },
    {
      "entropy": 5.681221437454224,
      "epoch": 2.9803156748911466,
      "grad_norm": 1.125,
      "learning_rate": 0.0004119957068187202,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2200848400592804,
      "num_tokens": 68179778.0,
      "step": 32855
    },
    {
      "entropy": 5.631128406524658,
      "epoch": 2.980769230769231,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004119700380197348,
      "loss": 4.8958,
      "mean_token_accuracy": 0.2159089118242264,
      "num_tokens": 68189405.0,
      "step": 32860
    },
    {
      "entropy": 5.577587461471557,
      "epoch": 2.981222786647315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041194436638817417,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22622825652360917,
      "num_tokens": 68199351.0,
      "step": 32865
    },
    {
      "entropy": 5.680678796768189,
      "epoch": 2.9816763425253994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041191869192456907,
      "loss": 5.0085,
      "mean_token_accuracy": 0.2117750957608223,
      "num_tokens": 68209904.0,
      "step": 32870
    },
    {
      "entropy": 5.715876674652099,
      "epoch": 2.982129898403483,
      "grad_norm": 1.125,
      "learning_rate": 0.0004118930146294506,
      "loss": 5.0072,
      "mean_token_accuracy": 0.2112344190478325,
      "num_tokens": 68221754.0,
      "step": 32875
    },
    {
      "entropy": 5.675957822799683,
      "epoch": 2.9825834542815675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004118673345033496,
      "loss": 4.9379,
      "mean_token_accuracy": 0.20875275433063506,
      "num_tokens": 68232328.0,
      "step": 32880
    },
    {
      "entropy": 5.610381269454956,
      "epoch": 2.9830370101596517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004118416515467974,
      "loss": 4.8574,
      "mean_token_accuracy": 0.21462386548519136,
      "num_tokens": 68242812.0,
      "step": 32885
    },
    {
      "entropy": 5.73562183380127,
      "epoch": 2.983490566037736,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004118159657603248,
      "loss": 5.0095,
      "mean_token_accuracy": 0.2113611564040184,
      "num_tokens": 68253888.0,
      "step": 32890
    },
    {
      "entropy": 5.688562393188477,
      "epoch": 2.98394412191582,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004117902771444633,
      "loss": 4.9266,
      "mean_token_accuracy": 0.21949954181909562,
      "num_tokens": 68264877.0,
      "step": 32895
    },
    {
      "entropy": 5.7436951160430905,
      "epoch": 2.984397677793904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000411764585699744,
      "loss": 5.0371,
      "mean_token_accuracy": 0.20340751409530639,
      "num_tokens": 68275863.0,
      "step": 32900
    },
    {
      "entropy": 5.6724865436553955,
      "epoch": 2.9848512336719883,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041173889142669806,
      "loss": 4.9367,
      "mean_token_accuracy": 0.21793349981307983,
      "num_tokens": 68286768.0,
      "step": 32905
    },
    {
      "entropy": 5.638830947875976,
      "epoch": 2.9853047895500726,
      "grad_norm": 1.125,
      "learning_rate": 0.0004117131943258572,
      "loss": 4.8834,
      "mean_token_accuracy": 0.2214945673942566,
      "num_tokens": 68296990.0,
      "step": 32910
    },
    {
      "entropy": 5.579110479354858,
      "epoch": 2.985758345428157,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041168749439775255,
      "loss": 4.8804,
      "mean_token_accuracy": 0.2154851496219635,
      "num_tokens": 68307646.0,
      "step": 32915
    },
    {
      "entropy": 5.686327838897705,
      "epoch": 2.986211901306241,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004116617916429156,
      "loss": 5.0234,
      "mean_token_accuracy": 0.20120131373405456,
      "num_tokens": 68319586.0,
      "step": 32920
    },
    {
      "entropy": 5.735942077636719,
      "epoch": 2.986665457184325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004116360860618781,
      "loss": 4.9839,
      "mean_token_accuracy": 0.2116524264216423,
      "num_tokens": 68329692.0,
      "step": 32925
    },
    {
      "entropy": 5.7310521602630615,
      "epoch": 2.9871190130624092,
      "grad_norm": 1.125,
      "learning_rate": 0.0004116103776551715,
      "loss": 4.9543,
      "mean_token_accuracy": 0.2143690124154091,
      "num_tokens": 68339589.0,
      "step": 32930
    },
    {
      "entropy": 5.683925533294678,
      "epoch": 2.9875725689404935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041158466642332743,
      "loss": 5.0091,
      "mean_token_accuracy": 0.20463186353445054,
      "num_tokens": 68349485.0,
      "step": 32935
    },
    {
      "entropy": 5.67258586883545,
      "epoch": 2.9880261248185778,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004115589523668777,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22407964766025543,
      "num_tokens": 68359695.0,
      "step": 32940
    },
    {
      "entropy": 5.608180284500122,
      "epoch": 2.9884796806966616,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000411533235486354,
      "loss": 4.8211,
      "mean_token_accuracy": 0.2210881754755974,
      "num_tokens": 68369591.0,
      "step": 32945
    },
    {
      "entropy": 5.6670667171478275,
      "epoch": 2.988933236574746,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004115075157822883,
      "loss": 4.958,
      "mean_token_accuracy": 0.2200934037566185,
      "num_tokens": 68380382.0,
      "step": 32950
    },
    {
      "entropy": 5.557702112197876,
      "epoch": 2.98938679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004114817932552122,
      "loss": 4.8164,
      "mean_token_accuracy": 0.22235634922981262,
      "num_tokens": 68389865.0,
      "step": 32955
    },
    {
      "entropy": 5.600421571731568,
      "epoch": 2.9898403483309144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041145606790565797,
      "loss": 4.8585,
      "mean_token_accuracy": 0.227959243953228,
      "num_tokens": 68399267.0,
      "step": 32960
    },
    {
      "entropy": 5.67458553314209,
      "epoch": 2.9902939042089987,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004114303397341575,
      "loss": 4.9642,
      "mean_token_accuracy": 0.22205188870429993,
      "num_tokens": 68409480.0,
      "step": 32965
    },
    {
      "entropy": 5.76980767250061,
      "epoch": 2.990747460087083,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004114046087412427,
      "loss": 5.0598,
      "mean_token_accuracy": 0.21033355742692947,
      "num_tokens": 68420019.0,
      "step": 32970
    },
    {
      "entropy": 5.592140054702758,
      "epoch": 2.9912010159651667,
      "grad_norm": 1.28125,
      "learning_rate": 0.000411378874927446,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22039477229118348,
      "num_tokens": 68429967.0,
      "step": 32975
    },
    {
      "entropy": 5.552520322799682,
      "epoch": 2.991654571843251,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004113531382932993,
      "loss": 4.8869,
      "mean_token_accuracy": 0.21943461745977402,
      "num_tokens": 68441639.0,
      "step": 32980
    },
    {
      "entropy": 5.649450302124023,
      "epoch": 2.9921081277213353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000411327398839335,
      "loss": 4.9503,
      "mean_token_accuracy": 0.20862580835819244,
      "num_tokens": 68452781.0,
      "step": 32985
    },
    {
      "entropy": 5.624648094177246,
      "epoch": 2.9925616835994195,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004113016565660854,
      "loss": 4.887,
      "mean_token_accuracy": 0.2202865868806839,
      "num_tokens": 68463353.0,
      "step": 32990
    },
    {
      "entropy": 5.602261304855347,
      "epoch": 2.9930152394775034,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041127591147408274,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21036900132894515,
      "num_tokens": 68473507.0,
      "step": 32995
    },
    {
      "entropy": 5.65443115234375,
      "epoch": 2.9934687953555876,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041125016356385945,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21953432708978654,
      "num_tokens": 68483737.0,
      "step": 33000
    },
    {
      "epoch": 2.9934687953555876,
      "eval_entropy": 5.387715898056375,
      "eval_loss": 4.981586933135986,
      "eval_mean_token_accuracy": 0.2226150446027787,
      "eval_num_tokens": 68483737.0,
      "eval_runtime": 20.5391,
      "eval_samples_per_second": 1721.592,
      "eval_steps_per_second": 215.199,
      "step": 33000
    },
    {
      "entropy": 5.660589122772217,
      "epoch": 2.993922351233672,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041122441283594817,
      "loss": 4.9551,
      "mean_token_accuracy": 0.22006729245185852,
      "num_tokens": 68494013.0,
      "step": 33005
    },
    {
      "entropy": 5.590733003616333,
      "epoch": 2.994375907111756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004111986592908813,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22423492521047592,
      "num_tokens": 68504361.0,
      "step": 33010
    },
    {
      "entropy": 5.608458185195923,
      "epoch": 2.9948294629898404,
      "grad_norm": 1.375,
      "learning_rate": 0.00041117290292919134,
      "loss": 4.8618,
      "mean_token_accuracy": 0.22024160623550415,
      "num_tokens": 68514825.0,
      "step": 33015
    },
    {
      "entropy": 5.642227029800415,
      "epoch": 2.9952830188679247,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004111471437514111,
      "loss": 4.8795,
      "mean_token_accuracy": 0.21597835868597032,
      "num_tokens": 68524858.0,
      "step": 33020
    },
    {
      "entropy": 5.6194559097290036,
      "epoch": 2.995736574746009,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004111213817580733,
      "loss": 4.8358,
      "mean_token_accuracy": 0.22545648962259293,
      "num_tokens": 68534481.0,
      "step": 33025
    },
    {
      "entropy": 5.586737155914307,
      "epoch": 2.996190130624093,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004110956169497106,
      "loss": 4.9297,
      "mean_token_accuracy": 0.2136531263589859,
      "num_tokens": 68545436.0,
      "step": 33030
    },
    {
      "entropy": 5.638039398193359,
      "epoch": 2.996643686502177,
      "grad_norm": 1.25,
      "learning_rate": 0.0004110698493268558,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22607792615890504,
      "num_tokens": 68555434.0,
      "step": 33035
    },
    {
      "entropy": 5.663933277130127,
      "epoch": 2.9970972423802613,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004110440788900419,
      "loss": 5.0302,
      "mean_token_accuracy": 0.20975909978151322,
      "num_tokens": 68566295.0,
      "step": 33040
    },
    {
      "entropy": 5.626414155960083,
      "epoch": 2.9975507982583456,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041101830563980174,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22308431267738343,
      "num_tokens": 68575927.0,
      "step": 33045
    },
    {
      "entropy": 5.724464082717896,
      "epoch": 2.9980043541364294,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041099252957666843,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21989697217941284,
      "num_tokens": 68586254.0,
      "step": 33050
    },
    {
      "entropy": 5.596921968460083,
      "epoch": 2.9984579100145137,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041096675070117496,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21485316008329391,
      "num_tokens": 68596881.0,
      "step": 33055
    },
    {
      "entropy": 5.554052352905273,
      "epoch": 2.998911465892598,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041094096901385434,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22131993174552916,
      "num_tokens": 68606875.0,
      "step": 33060
    },
    {
      "entropy": 5.685102653503418,
      "epoch": 2.999365021770682,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004109151845152399,
      "loss": 4.953,
      "mean_token_accuracy": 0.21720768213272096,
      "num_tokens": 68617011.0,
      "step": 33065
    },
    {
      "entropy": 5.714345693588257,
      "epoch": 2.9998185776487665,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004108893972058649,
      "loss": 4.9573,
      "mean_token_accuracy": 0.2159837156534195,
      "num_tokens": 68627764.0,
      "step": 33070
    },
    {
      "entropy": 5.70031509399414,
      "epoch": 3.0002721335268503,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041086360708626236,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21976932287216186,
      "num_tokens": 68637272.0,
      "step": 33075
    },
    {
      "entropy": 5.597321557998657,
      "epoch": 3.0007256894049346,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041083781415696585,
      "loss": 4.8833,
      "mean_token_accuracy": 0.21273597478866577,
      "num_tokens": 68648058.0,
      "step": 33080
    },
    {
      "entropy": 5.563616752624512,
      "epoch": 3.001179245283019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004108120184185088,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22532851248979568,
      "num_tokens": 68658174.0,
      "step": 33085
    },
    {
      "entropy": 5.651817035675049,
      "epoch": 3.001632801161103,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041078621987142457,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22987704873085021,
      "num_tokens": 68669281.0,
      "step": 33090
    },
    {
      "entropy": 5.622703647613525,
      "epoch": 3.0020863570391874,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004107604185162467,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21468985974788665,
      "num_tokens": 68679879.0,
      "step": 33095
    },
    {
      "entropy": 5.591121530532837,
      "epoch": 3.0025399129172716,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041073461435350877,
      "loss": 4.8088,
      "mean_token_accuracy": 0.2245236650109291,
      "num_tokens": 68689852.0,
      "step": 33100
    },
    {
      "entropy": 5.584159421920776,
      "epoch": 3.0029934687953554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041070880738374447,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22060276418924332,
      "num_tokens": 68700149.0,
      "step": 33105
    },
    {
      "entropy": 5.6279764652252195,
      "epoch": 3.0034470246734397,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041068299760748734,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2235507845878601,
      "num_tokens": 68710837.0,
      "step": 33110
    },
    {
      "entropy": 5.641936111450195,
      "epoch": 3.003900580551524,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041065718502527134,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21745191067457198,
      "num_tokens": 68721451.0,
      "step": 33115
    },
    {
      "entropy": 5.606793737411499,
      "epoch": 3.0043541364296082,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041063136963763013,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22340964674949645,
      "num_tokens": 68732911.0,
      "step": 33120
    },
    {
      "entropy": 5.606755256652832,
      "epoch": 3.0048076923076925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004106055514450977,
      "loss": 4.8276,
      "mean_token_accuracy": 0.22579917013645173,
      "num_tokens": 68742650.0,
      "step": 33125
    },
    {
      "entropy": 5.574276447296143,
      "epoch": 3.0052612481857763,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004105797304482079,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22119048982858658,
      "num_tokens": 68753660.0,
      "step": 33130
    },
    {
      "entropy": 5.571784591674804,
      "epoch": 3.0057148040638606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004105539066474947,
      "loss": 4.7567,
      "mean_token_accuracy": 0.2258976697921753,
      "num_tokens": 68763868.0,
      "step": 33135
    },
    {
      "entropy": 5.622152328491211,
      "epoch": 3.006168359941945,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041052808004349214,
      "loss": 4.8621,
      "mean_token_accuracy": 0.21943342089653015,
      "num_tokens": 68773911.0,
      "step": 33140
    },
    {
      "entropy": 5.668749094009399,
      "epoch": 3.006621915820029,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004105022506367344,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2139795497059822,
      "num_tokens": 68786423.0,
      "step": 33145
    },
    {
      "entropy": 5.647215032577515,
      "epoch": 3.0070754716981134,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004104764184277555,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21283783614635468,
      "num_tokens": 68797329.0,
      "step": 33150
    },
    {
      "entropy": 5.70860915184021,
      "epoch": 3.0075290275761972,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004104505834170898,
      "loss": 4.9415,
      "mean_token_accuracy": 0.21623314470052718,
      "num_tokens": 68808447.0,
      "step": 33155
    },
    {
      "entropy": 5.6651805400848385,
      "epoch": 3.0079825834542815,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041042474560527153,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22218254804611207,
      "num_tokens": 68819381.0,
      "step": 33160
    },
    {
      "entropy": 5.558456611633301,
      "epoch": 3.0084361393323658,
      "grad_norm": 1.140625,
      "learning_rate": 0.000410398904992835,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22530175596475602,
      "num_tokens": 68829319.0,
      "step": 33165
    },
    {
      "entropy": 5.5831138610839846,
      "epoch": 3.00888969521045,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041037306158031464,
      "loss": 4.9209,
      "mean_token_accuracy": 0.22112355679273604,
      "num_tokens": 68838640.0,
      "step": 33170
    },
    {
      "entropy": 5.656088924407959,
      "epoch": 3.0093432510885343,
      "grad_norm": 1.125,
      "learning_rate": 0.0004103472153682448,
      "loss": 4.862,
      "mean_token_accuracy": 0.2178482547402382,
      "num_tokens": 68847960.0,
      "step": 33175
    },
    {
      "entropy": 5.635356712341308,
      "epoch": 3.009796806966618,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041032136635716016,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2185422495007515,
      "num_tokens": 68858851.0,
      "step": 33180
    },
    {
      "entropy": 5.5657525062561035,
      "epoch": 3.0102503628447024,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004102955145475951,
      "loss": 4.8499,
      "mean_token_accuracy": 0.21637112498283387,
      "num_tokens": 68869006.0,
      "step": 33185
    },
    {
      "entropy": 5.706431245803833,
      "epoch": 3.0107039187227866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041026965994008435,
      "loss": 4.954,
      "mean_token_accuracy": 0.2118320271372795,
      "num_tokens": 68879606.0,
      "step": 33190
    },
    {
      "entropy": 5.5825831413269045,
      "epoch": 3.011157474600871,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041024380253516253,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23019160330295563,
      "num_tokens": 68889862.0,
      "step": 33195
    },
    {
      "entropy": 5.733935737609864,
      "epoch": 3.011611030478955,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041021794233336444,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21864238679409026,
      "num_tokens": 68900512.0,
      "step": 33200
    },
    {
      "entropy": 5.653041648864746,
      "epoch": 3.012064586357039,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004101920793352249,
      "loss": 4.8974,
      "mean_token_accuracy": 0.21337901651859284,
      "num_tokens": 68910475.0,
      "step": 33205
    },
    {
      "entropy": 5.683540391921997,
      "epoch": 3.0125181422351233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004101662135412786,
      "loss": 4.919,
      "mean_token_accuracy": 0.21979176998138428,
      "num_tokens": 68920583.0,
      "step": 33210
    },
    {
      "entropy": 5.562118577957153,
      "epoch": 3.0129716981132075,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004101403449520606,
      "loss": 4.8156,
      "mean_token_accuracy": 0.21956888735294341,
      "num_tokens": 68931016.0,
      "step": 33215
    },
    {
      "entropy": 5.585714435577392,
      "epoch": 3.013425253991292,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041011447356810584,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22134097963571547,
      "num_tokens": 68941489.0,
      "step": 33220
    },
    {
      "entropy": 5.662279367446899,
      "epoch": 3.013878809869376,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004100885993899493,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2130902513861656,
      "num_tokens": 68952322.0,
      "step": 33225
    },
    {
      "entropy": 5.66927137374878,
      "epoch": 3.01433236574746,
      "grad_norm": 1.125,
      "learning_rate": 0.0004100627224181261,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21905177980661392,
      "num_tokens": 68962570.0,
      "step": 33230
    },
    {
      "entropy": 5.688972473144531,
      "epoch": 3.014785921625544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041003684265317143,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2224120169878006,
      "num_tokens": 68973733.0,
      "step": 33235
    },
    {
      "entropy": 5.651996517181397,
      "epoch": 3.0152394775036284,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041001096009562045,
      "loss": 4.8716,
      "mean_token_accuracy": 0.22570587545633317,
      "num_tokens": 68983526.0,
      "step": 33240
    },
    {
      "entropy": 5.7207330703735355,
      "epoch": 3.0156930333817127,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004099850747460084,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21919573545455934,
      "num_tokens": 68993968.0,
      "step": 33245
    },
    {
      "entropy": 5.6513673782348635,
      "epoch": 3.016146589259797,
      "grad_norm": 1.25,
      "learning_rate": 0.00040995918660487055,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22989965379238128,
      "num_tokens": 69003599.0,
      "step": 33250
    },
    {
      "entropy": 5.608249092102051,
      "epoch": 3.0166001451378808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004099332956727424,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2143179252743721,
      "num_tokens": 69013686.0,
      "step": 33255
    },
    {
      "entropy": 5.635790920257568,
      "epoch": 3.017053701015965,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004099074019501593,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2163786768913269,
      "num_tokens": 69023803.0,
      "step": 33260
    },
    {
      "entropy": 5.594792270660401,
      "epoch": 3.0175072568940493,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004098815054376567,
      "loss": 4.7808,
      "mean_token_accuracy": 0.22405432015657425,
      "num_tokens": 69033885.0,
      "step": 33265
    },
    {
      "entropy": 5.626929187774659,
      "epoch": 3.0179608127721336,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004098556061357702,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22519037872552872,
      "num_tokens": 69044471.0,
      "step": 33270
    },
    {
      "entropy": 5.563388109207153,
      "epoch": 3.018414368650218,
      "grad_norm": 1.125,
      "learning_rate": 0.00040982970404503544,
      "loss": 4.7509,
      "mean_token_accuracy": 0.22540322542190552,
      "num_tokens": 69055109.0,
      "step": 33275
    },
    {
      "entropy": 5.562762069702148,
      "epoch": 3.018867924528302,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040980379916598803,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22227307856082917,
      "num_tokens": 69064692.0,
      "step": 33280
    },
    {
      "entropy": 5.616875267028808,
      "epoch": 3.019321480406386,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004097778914991637,
      "loss": 4.8491,
      "mean_token_accuracy": 0.2216279074549675,
      "num_tokens": 69075331.0,
      "step": 33285
    },
    {
      "entropy": 5.618284034729004,
      "epoch": 3.01977503628447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040975198104509827,
      "loss": 4.8491,
      "mean_token_accuracy": 0.21935871988534927,
      "num_tokens": 69085017.0,
      "step": 33290
    },
    {
      "entropy": 5.57413649559021,
      "epoch": 3.0202285921625545,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004097260678043275,
      "loss": 4.8674,
      "mean_token_accuracy": 0.21314176470041274,
      "num_tokens": 69096128.0,
      "step": 33295
    },
    {
      "entropy": 5.653529262542724,
      "epoch": 3.0206821480406387,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040970015177738725,
      "loss": 4.9717,
      "mean_token_accuracy": 0.21279960870742798,
      "num_tokens": 69106479.0,
      "step": 33300
    },
    {
      "entropy": 5.714112901687622,
      "epoch": 3.021135703918723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004096742329648136,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21783403754234315,
      "num_tokens": 69116346.0,
      "step": 33305
    },
    {
      "entropy": 5.618188142776489,
      "epoch": 3.021589259796807,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004096483113671425,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22162439823150634,
      "num_tokens": 69126489.0,
      "step": 33310
    },
    {
      "entropy": 5.620934915542603,
      "epoch": 3.022042815674891,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040962238698490994,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22120945006608964,
      "num_tokens": 69136618.0,
      "step": 33315
    },
    {
      "entropy": 5.688378143310547,
      "epoch": 3.0224963715529753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004095964598186522,
      "loss": 4.919,
      "mean_token_accuracy": 0.21332519203424455,
      "num_tokens": 69146242.0,
      "step": 33320
    },
    {
      "entropy": 5.637007999420166,
      "epoch": 3.0229499274310596,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004095705298689053,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22259001731872557,
      "num_tokens": 69157269.0,
      "step": 33325
    },
    {
      "entropy": 5.722749280929565,
      "epoch": 3.023403483309144,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040954459713620563,
      "loss": 4.9424,
      "mean_token_accuracy": 0.2150605797767639,
      "num_tokens": 69169260.0,
      "step": 33330
    },
    {
      "entropy": 5.743881177902222,
      "epoch": 3.0238570391872277,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004095186616210893,
      "loss": 4.9959,
      "mean_token_accuracy": 0.20425416082143782,
      "num_tokens": 69179815.0,
      "step": 33335
    },
    {
      "entropy": 5.702618026733399,
      "epoch": 3.024310595065312,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004094927233240928,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21368433386087418,
      "num_tokens": 69189959.0,
      "step": 33340
    },
    {
      "entropy": 5.656851577758789,
      "epoch": 3.0247641509433962,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040946678224575264,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21942167580127717,
      "num_tokens": 69199485.0,
      "step": 33345
    },
    {
      "entropy": 5.566434288024903,
      "epoch": 3.0252177068214805,
      "grad_norm": 1.359375,
      "learning_rate": 0.000409440838386605,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21972195208072662,
      "num_tokens": 69209180.0,
      "step": 33350
    },
    {
      "entropy": 5.657252931594849,
      "epoch": 3.0256712626995648,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004094148917471866,
      "loss": 4.9332,
      "mean_token_accuracy": 0.2158432736992836,
      "num_tokens": 69220648.0,
      "step": 33355
    },
    {
      "entropy": 5.577131175994873,
      "epoch": 3.0261248185776486,
      "grad_norm": 1.28125,
      "learning_rate": 0.000409388942328034,
      "loss": 4.7622,
      "mean_token_accuracy": 0.22943407595157622,
      "num_tokens": 69230685.0,
      "step": 33360
    },
    {
      "entropy": 5.581336641311646,
      "epoch": 3.026578374455733,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004093629901296838,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2172699123620987,
      "num_tokens": 69240407.0,
      "step": 33365
    },
    {
      "entropy": 5.717920446395874,
      "epoch": 3.027031930333817,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004093370351526727,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21414300799369812,
      "num_tokens": 69250624.0,
      "step": 33370
    },
    {
      "entropy": 5.735086393356323,
      "epoch": 3.0274854862119014,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004093110773975376,
      "loss": 4.9421,
      "mean_token_accuracy": 0.2098846450448036,
      "num_tokens": 69261131.0,
      "step": 33375
    },
    {
      "entropy": 5.62031626701355,
      "epoch": 3.0279390420899857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004092851168648152,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21156414449214936,
      "num_tokens": 69270668.0,
      "step": 33380
    },
    {
      "entropy": 5.646076202392578,
      "epoch": 3.0283925979680695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004092591535550422,
      "loss": 4.8233,
      "mean_token_accuracy": 0.22934465110301971,
      "num_tokens": 69280528.0,
      "step": 33385
    },
    {
      "entropy": 5.5961799144744875,
      "epoch": 3.0288461538461537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004092331874687558,
      "loss": 4.8438,
      "mean_token_accuracy": 0.2253849610686302,
      "num_tokens": 69290376.0,
      "step": 33390
    },
    {
      "entropy": 5.64811692237854,
      "epoch": 3.029299709724238,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040920721860649285,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22648802846670152,
      "num_tokens": 69301766.0,
      "step": 33395
    },
    {
      "entropy": 5.634732246398926,
      "epoch": 3.0297532656023223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004091812469687904,
      "loss": 4.847,
      "mean_token_accuracy": 0.21963511556386947,
      "num_tokens": 69312668.0,
      "step": 33400
    },
    {
      "entropy": 5.6247211456298825,
      "epoch": 3.0302068214804065,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004091552725561856,
      "loss": 4.911,
      "mean_token_accuracy": 0.21388354152441025,
      "num_tokens": 69323195.0,
      "step": 33405
    },
    {
      "entropy": 5.614585542678833,
      "epoch": 3.0306603773584904,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004091292953692156,
      "loss": 4.8703,
      "mean_token_accuracy": 0.21828990280628205,
      "num_tokens": 69333812.0,
      "step": 33410
    },
    {
      "entropy": 5.675078201293945,
      "epoch": 3.0311139332365746,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040910331540841755,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22610004395246505,
      "num_tokens": 69344378.0,
      "step": 33415
    },
    {
      "entropy": 5.640748739242554,
      "epoch": 3.031567489114659,
      "grad_norm": 1.125,
      "learning_rate": 0.00040907733267432877,
      "loss": 4.8813,
      "mean_token_accuracy": 0.2240175038576126,
      "num_tokens": 69354592.0,
      "step": 33420
    },
    {
      "entropy": 5.655765724182129,
      "epoch": 3.032021044992743,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040905134716748656,
      "loss": 4.8758,
      "mean_token_accuracy": 0.217238187789917,
      "num_tokens": 69365011.0,
      "step": 33425
    },
    {
      "entropy": 5.576217699050903,
      "epoch": 3.0324746008708274,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004090253588884284,
      "loss": 4.8738,
      "mean_token_accuracy": 0.21141038686037064,
      "num_tokens": 69377010.0,
      "step": 33430
    },
    {
      "entropy": 5.6457024097442625,
      "epoch": 3.0329281567489113,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040899936783769157,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22136796563863753,
      "num_tokens": 69387263.0,
      "step": 33435
    },
    {
      "entropy": 5.635177516937256,
      "epoch": 3.0333817126269955,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040897337401581374,
      "loss": 4.9106,
      "mean_token_accuracy": 0.22257238179445266,
      "num_tokens": 69399188.0,
      "step": 33440
    },
    {
      "entropy": 5.614470815658569,
      "epoch": 3.03383526850508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040894737742333244,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21780644357204437,
      "num_tokens": 69408271.0,
      "step": 33445
    },
    {
      "entropy": 5.691281938552857,
      "epoch": 3.034288824383164,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004089213780607851,
      "loss": 4.9868,
      "mean_token_accuracy": 0.21004820317029954,
      "num_tokens": 69418424.0,
      "step": 33450
    },
    {
      "entropy": 5.703288507461548,
      "epoch": 3.0347423802612483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004088953759287096,
      "loss": 4.8979,
      "mean_token_accuracy": 0.2254528969526291,
      "num_tokens": 69427802.0,
      "step": 33455
    },
    {
      "entropy": 5.6461645603179935,
      "epoch": 3.0351959361393326,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004088693710276436,
      "loss": 4.9193,
      "mean_token_accuracy": 0.2237873539328575,
      "num_tokens": 69437581.0,
      "step": 33460
    },
    {
      "entropy": 5.60552659034729,
      "epoch": 3.0356494920174164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040884336335812494,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22404371351003646,
      "num_tokens": 69447711.0,
      "step": 33465
    },
    {
      "entropy": 5.674390649795532,
      "epoch": 3.0361030478955007,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004088173529206914,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22031345516443251,
      "num_tokens": 69457692.0,
      "step": 33470
    },
    {
      "entropy": 5.62730712890625,
      "epoch": 3.036556603773585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004087913397158809,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22014811336994172,
      "num_tokens": 69468034.0,
      "step": 33475
    },
    {
      "entropy": 5.599385786056518,
      "epoch": 3.037010159651669,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040876532374423135,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22177741825580596,
      "num_tokens": 69477719.0,
      "step": 33480
    },
    {
      "entropy": 5.6568694591522215,
      "epoch": 3.0374637155297535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004087393050062809,
      "loss": 4.8811,
      "mean_token_accuracy": 0.21688098907470704,
      "num_tokens": 69487261.0,
      "step": 33485
    },
    {
      "entropy": 5.66193118095398,
      "epoch": 3.0379172714078373,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040871328350256745,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21805978566408157,
      "num_tokens": 69496658.0,
      "step": 33490
    },
    {
      "entropy": 5.735902595520019,
      "epoch": 3.0383708272859216,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040868725923362927,
      "loss": 5.0183,
      "mean_token_accuracy": 0.21095375418663026,
      "num_tokens": 69507197.0,
      "step": 33495
    },
    {
      "entropy": 5.63435959815979,
      "epoch": 3.038824383164006,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040866123220000455,
      "loss": 4.929,
      "mean_token_accuracy": 0.21919573098421097,
      "num_tokens": 69516905.0,
      "step": 33500
    },
    {
      "entropy": 5.651447868347168,
      "epoch": 3.03927793904209,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040863520240223144,
      "loss": 4.841,
      "mean_token_accuracy": 0.22511135190725326,
      "num_tokens": 69526869.0,
      "step": 33505
    },
    {
      "entropy": 5.660409021377563,
      "epoch": 3.0397314949201744,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004086091698408483,
      "loss": 4.9134,
      "mean_token_accuracy": 0.2196820467710495,
      "num_tokens": 69537058.0,
      "step": 33510
    },
    {
      "entropy": 5.707027816772461,
      "epoch": 3.040185050798258,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040858313451639343,
      "loss": 4.9818,
      "mean_token_accuracy": 0.2082364782691002,
      "num_tokens": 69547617.0,
      "step": 33515
    },
    {
      "entropy": 5.703563261032104,
      "epoch": 3.0406386066763424,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004085570964294053,
      "loss": 4.9271,
      "mean_token_accuracy": 0.20861987322568892,
      "num_tokens": 69558733.0,
      "step": 33520
    },
    {
      "entropy": 5.727454423904419,
      "epoch": 3.0410921625544267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004085310555804225,
      "loss": 4.98,
      "mean_token_accuracy": 0.2141845554113388,
      "num_tokens": 69570011.0,
      "step": 33525
    },
    {
      "entropy": 5.726199293136597,
      "epoch": 3.041545718432511,
      "grad_norm": 1.25,
      "learning_rate": 0.00040850501196998336,
      "loss": 4.9982,
      "mean_token_accuracy": 0.21202619671821593,
      "num_tokens": 69580894.0,
      "step": 33530
    },
    {
      "entropy": 5.6468220233917235,
      "epoch": 3.0419992743105952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040847896559862654,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2234838679432869,
      "num_tokens": 69591438.0,
      "step": 33535
    },
    {
      "entropy": 5.642037391662598,
      "epoch": 3.042452830188679,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004084529164668907,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21272068470716476,
      "num_tokens": 69601745.0,
      "step": 33540
    },
    {
      "entropy": 5.6669515609741214,
      "epoch": 3.0429063860667633,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040842686457531453,
      "loss": 4.8814,
      "mean_token_accuracy": 0.21904023140668868,
      "num_tokens": 69611872.0,
      "step": 33545
    },
    {
      "entropy": 5.670464611053466,
      "epoch": 3.0433599419448476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004084008099244368,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2135903537273407,
      "num_tokens": 69621673.0,
      "step": 33550
    },
    {
      "entropy": 5.584834957122803,
      "epoch": 3.043813497822932,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040837475251479633,
      "loss": 4.814,
      "mean_token_accuracy": 0.22037193179130554,
      "num_tokens": 69632090.0,
      "step": 33555
    },
    {
      "entropy": 5.571874380111694,
      "epoch": 3.044267053701016,
      "grad_norm": 1.171875,
      "learning_rate": 0.000408348692346932,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22599076479673386,
      "num_tokens": 69642920.0,
      "step": 33560
    },
    {
      "entropy": 5.644674301147461,
      "epoch": 3.0447206095791,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040832262942138277,
      "loss": 4.9561,
      "mean_token_accuracy": 0.2174089655280113,
      "num_tokens": 69653936.0,
      "step": 33565
    },
    {
      "entropy": 5.639392423629761,
      "epoch": 3.0451741654571842,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004082965637386875,
      "loss": 4.8246,
      "mean_token_accuracy": 0.22010339349508284,
      "num_tokens": 69663362.0,
      "step": 33570
    },
    {
      "entropy": 5.635706186294556,
      "epoch": 3.0456277213352685,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004082704952993854,
      "loss": 4.8519,
      "mean_token_accuracy": 0.2273235410451889,
      "num_tokens": 69673178.0,
      "step": 33575
    },
    {
      "entropy": 5.69794454574585,
      "epoch": 3.0460812772133528,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004082444241040155,
      "loss": 4.933,
      "mean_token_accuracy": 0.21650497168302535,
      "num_tokens": 69683627.0,
      "step": 33580
    },
    {
      "entropy": 5.6187183380126955,
      "epoch": 3.046534833091437,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004082183501531169,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2176191419363022,
      "num_tokens": 69694010.0,
      "step": 33585
    },
    {
      "entropy": 5.63181586265564,
      "epoch": 3.046988388969521,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040819227344722893,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2169242024421692,
      "num_tokens": 69704888.0,
      "step": 33590
    },
    {
      "entropy": 5.655230569839477,
      "epoch": 3.047441944847605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040816619398689074,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22310821264982222,
      "num_tokens": 69714718.0,
      "step": 33595
    },
    {
      "entropy": 5.657966184616089,
      "epoch": 3.0478955007256894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040814011177264174,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22074198722839355,
      "num_tokens": 69726076.0,
      "step": 33600
    },
    {
      "entropy": 5.593572807312012,
      "epoch": 3.0483490566037736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040811402680502137,
      "loss": 4.8318,
      "mean_token_accuracy": 0.222875015437603,
      "num_tokens": 69736507.0,
      "step": 33605
    },
    {
      "entropy": 5.580551481246948,
      "epoch": 3.048802612481858,
      "grad_norm": 1.375,
      "learning_rate": 0.00040808793908456885,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22002605944871903,
      "num_tokens": 69746061.0,
      "step": 33610
    },
    {
      "entropy": 5.6494669914245605,
      "epoch": 3.0492561683599417,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004080618486118241,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21888525038957596,
      "num_tokens": 69756493.0,
      "step": 33615
    },
    {
      "entropy": 5.662083339691162,
      "epoch": 3.049709724238026,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004080357553873261,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2186819702386856,
      "num_tokens": 69766937.0,
      "step": 33620
    },
    {
      "entropy": 5.586837577819824,
      "epoch": 3.0501632801161103,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000408009659411615,
      "loss": 4.8731,
      "mean_token_accuracy": 0.22092850059270858,
      "num_tokens": 69776872.0,
      "step": 33625
    },
    {
      "entropy": 5.622227334976197,
      "epoch": 3.0506168359941945,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004079835606852301,
      "loss": 4.8834,
      "mean_token_accuracy": 0.21830135434865952,
      "num_tokens": 69788145.0,
      "step": 33630
    },
    {
      "entropy": 5.673195552825928,
      "epoch": 3.051070391872279,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004079574592087114,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21575358659029006,
      "num_tokens": 69797353.0,
      "step": 33635
    },
    {
      "entropy": 5.59747462272644,
      "epoch": 3.051523947750363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040793135498259854,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22126159071922302,
      "num_tokens": 69806915.0,
      "step": 33640
    },
    {
      "entropy": 5.5858407497406,
      "epoch": 3.051977503628447,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040790524800743134,
      "loss": 4.8807,
      "mean_token_accuracy": 0.21785001903772355,
      "num_tokens": 69817695.0,
      "step": 33645
    },
    {
      "entropy": 5.574267625808716,
      "epoch": 3.052431059506531,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004078791382837498,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23208732008934022,
      "num_tokens": 69827451.0,
      "step": 33650
    },
    {
      "entropy": 5.676915645599365,
      "epoch": 3.0528846153846154,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040785302581209374,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22189147472381593,
      "num_tokens": 69837897.0,
      "step": 33655
    },
    {
      "entropy": 5.65137734413147,
      "epoch": 3.0533381712626997,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040782691059300326,
      "loss": 4.8578,
      "mean_token_accuracy": 0.22230204641819,
      "num_tokens": 69847378.0,
      "step": 33660
    },
    {
      "entropy": 5.615173768997193,
      "epoch": 3.053791727140784,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004078007926270185,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21882296204566956,
      "num_tokens": 69857522.0,
      "step": 33665
    },
    {
      "entropy": 5.634403848648072,
      "epoch": 3.0542452830188678,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040777467191467946,
      "loss": 4.8623,
      "mean_token_accuracy": 0.22449126243591308,
      "num_tokens": 69868384.0,
      "step": 33670
    },
    {
      "entropy": 5.628997039794922,
      "epoch": 3.054698838896952,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004077485484565264,
      "loss": 4.8803,
      "mean_token_accuracy": 0.22230271995067596,
      "num_tokens": 69878761.0,
      "step": 33675
    },
    {
      "entropy": 5.664788913726807,
      "epoch": 3.0551523947750363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004077224222530994,
      "loss": 4.848,
      "mean_token_accuracy": 0.2174297273159027,
      "num_tokens": 69888366.0,
      "step": 33680
    },
    {
      "entropy": 5.650530242919922,
      "epoch": 3.0556059506531206,
      "grad_norm": 1.15625,
      "learning_rate": 0.000407696293304939,
      "loss": 4.8979,
      "mean_token_accuracy": 0.21089494675397874,
      "num_tokens": 69898489.0,
      "step": 33685
    },
    {
      "entropy": 5.620934247970581,
      "epoch": 3.056059506531205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004076701616125854,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2176748976111412,
      "num_tokens": 69908425.0,
      "step": 33690
    },
    {
      "entropy": 5.710889196395874,
      "epoch": 3.0565130624092887,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004076440271765791,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2194310560822487,
      "num_tokens": 69918488.0,
      "step": 33695
    },
    {
      "entropy": 5.548961734771728,
      "epoch": 3.056966618287373,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004076178899974605,
      "loss": 4.7948,
      "mean_token_accuracy": 0.22928571105003356,
      "num_tokens": 69928831.0,
      "step": 33700
    },
    {
      "entropy": 5.62413067817688,
      "epoch": 3.057420174165457,
      "grad_norm": 1.125,
      "learning_rate": 0.00040759175007577017,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2176133379340172,
      "num_tokens": 69938974.0,
      "step": 33705
    },
    {
      "entropy": 5.602591466903687,
      "epoch": 3.0578737300435415,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004075656074120486,
      "loss": 4.8099,
      "mean_token_accuracy": 0.21968888491392136,
      "num_tokens": 69948804.0,
      "step": 33710
    },
    {
      "entropy": 5.687808227539063,
      "epoch": 3.0583272859216257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004075394620068365,
      "loss": 4.8602,
      "mean_token_accuracy": 0.2233629733324051,
      "num_tokens": 69958945.0,
      "step": 33715
    },
    {
      "entropy": 5.59665813446045,
      "epoch": 3.0587808417997095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040751331386067456,
      "loss": 4.8369,
      "mean_token_accuracy": 0.22046376168727874,
      "num_tokens": 69968826.0,
      "step": 33720
    },
    {
      "entropy": 5.659530973434448,
      "epoch": 3.059234397677794,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004074871629741035,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21942270398139954,
      "num_tokens": 69978961.0,
      "step": 33725
    },
    {
      "entropy": 5.591757440567017,
      "epoch": 3.059687953555878,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004074610093476641,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21980908811092376,
      "num_tokens": 69989848.0,
      "step": 33730
    },
    {
      "entropy": 5.618377542495727,
      "epoch": 3.0601415094339623,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040743485298189744,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22139830738306046,
      "num_tokens": 69999607.0,
      "step": 33735
    },
    {
      "entropy": 5.636200380325318,
      "epoch": 3.0605950653120466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040740869387734415,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21685971319675446,
      "num_tokens": 70010446.0,
      "step": 33740
    },
    {
      "entropy": 5.582016229629517,
      "epoch": 3.0610486211901304,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004073825320345453,
      "loss": 4.8079,
      "mean_token_accuracy": 0.22456979006528854,
      "num_tokens": 70021148.0,
      "step": 33745
    },
    {
      "entropy": 5.685440969467163,
      "epoch": 3.0615021770682147,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040735636745404196,
      "loss": 4.9582,
      "mean_token_accuracy": 0.21200707852840422,
      "num_tokens": 70030765.0,
      "step": 33750
    },
    {
      "entropy": 5.704784679412842,
      "epoch": 3.061955732946299,
      "grad_norm": 1.25,
      "learning_rate": 0.00040733020013637526,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2192843124270439,
      "num_tokens": 70041717.0,
      "step": 33755
    },
    {
      "entropy": 5.69542293548584,
      "epoch": 3.0624092888243832,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040730403008208633,
      "loss": 4.94,
      "mean_token_accuracy": 0.219637830555439,
      "num_tokens": 70052178.0,
      "step": 33760
    },
    {
      "entropy": 5.599521636962891,
      "epoch": 3.0628628447024675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004072778572917163,
      "loss": 4.8892,
      "mean_token_accuracy": 0.21789907664060593,
      "num_tokens": 70062819.0,
      "step": 33765
    },
    {
      "entropy": 5.564668083190918,
      "epoch": 3.0633164005805513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004072516817658065,
      "loss": 4.9005,
      "mean_token_accuracy": 0.22014685422182084,
      "num_tokens": 70073966.0,
      "step": 33770
    },
    {
      "entropy": 5.678152370452881,
      "epoch": 3.0637699564586356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040722550350489825,
      "loss": 4.8859,
      "mean_token_accuracy": 0.21854854375123978,
      "num_tokens": 70083884.0,
      "step": 33775
    },
    {
      "entropy": 5.69217619895935,
      "epoch": 3.06422351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004071993225095327,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2179158642888069,
      "num_tokens": 70094910.0,
      "step": 33780
    },
    {
      "entropy": 5.5887480735778805,
      "epoch": 3.064677068214804,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004071731387802517,
      "loss": 4.7159,
      "mean_token_accuracy": 0.23645797222852707,
      "num_tokens": 70104480.0,
      "step": 33785
    },
    {
      "entropy": 5.594952487945557,
      "epoch": 3.0651306240928884,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004071469523175964,
      "loss": 4.8613,
      "mean_token_accuracy": 0.22238890081644058,
      "num_tokens": 70114003.0,
      "step": 33790
    },
    {
      "entropy": 5.64472222328186,
      "epoch": 3.065584179970972,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040712076312210844,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21822520345449448,
      "num_tokens": 70123604.0,
      "step": 33795
    },
    {
      "entropy": 5.6272890090942385,
      "epoch": 3.0660377358490565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040709457119432945,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21594662964344025,
      "num_tokens": 70133895.0,
      "step": 33800
    },
    {
      "entropy": 5.679041576385498,
      "epoch": 3.0664912917271407,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040706837653480113,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22441833317279816,
      "num_tokens": 70145063.0,
      "step": 33805
    },
    {
      "entropy": 5.635238885879517,
      "epoch": 3.066944847605225,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000407042179144065,
      "loss": 4.9495,
      "mean_token_accuracy": 0.21667262315750122,
      "num_tokens": 70155456.0,
      "step": 33810
    },
    {
      "entropy": 5.537987422943115,
      "epoch": 3.0673984034833093,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040701597902266306,
      "loss": 4.8087,
      "mean_token_accuracy": 0.22616068720817567,
      "num_tokens": 70165427.0,
      "step": 33815
    },
    {
      "entropy": 5.695072650909424,
      "epoch": 3.067851959361393,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000406989776171137,
      "loss": 4.9384,
      "mean_token_accuracy": 0.2185006931424141,
      "num_tokens": 70175161.0,
      "step": 33820
    },
    {
      "entropy": 5.601235008239746,
      "epoch": 3.0683055152394774,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040696357059002877,
      "loss": 4.8869,
      "mean_token_accuracy": 0.2227461114525795,
      "num_tokens": 70185026.0,
      "step": 33825
    },
    {
      "entropy": 5.651657009124756,
      "epoch": 3.0687590711175616,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004069373622798802,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23188664019107819,
      "num_tokens": 70195645.0,
      "step": 33830
    },
    {
      "entropy": 5.650928974151611,
      "epoch": 3.069212626995646,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004069111512412334,
      "loss": 4.955,
      "mean_token_accuracy": 0.21665334701538086,
      "num_tokens": 70206591.0,
      "step": 33835
    },
    {
      "entropy": 5.679553413391114,
      "epoch": 3.06966618287373,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040688493747463036,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22084722965955733,
      "num_tokens": 70216945.0,
      "step": 33840
    },
    {
      "entropy": 5.589774179458618,
      "epoch": 3.0701197387518144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004068587209806132,
      "loss": 4.7756,
      "mean_token_accuracy": 0.23484664261341096,
      "num_tokens": 70226983.0,
      "step": 33845
    },
    {
      "entropy": 5.591967153549194,
      "epoch": 3.0705732946298983,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040683250175972414,
      "loss": 4.8564,
      "mean_token_accuracy": 0.21944043636322022,
      "num_tokens": 70237730.0,
      "step": 33850
    },
    {
      "entropy": 5.636760997772217,
      "epoch": 3.0710268505079825,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004068062798125053,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21438830494880676,
      "num_tokens": 70248499.0,
      "step": 33855
    },
    {
      "entropy": 5.606294775009156,
      "epoch": 3.071480406386067,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000406780055139499,
      "loss": 4.8832,
      "mean_token_accuracy": 0.21702604740858078,
      "num_tokens": 70259294.0,
      "step": 33860
    },
    {
      "entropy": 5.685557222366333,
      "epoch": 3.071933962264151,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004067538277412476,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21512585133314133,
      "num_tokens": 70269400.0,
      "step": 33865
    },
    {
      "entropy": 5.7115343570709225,
      "epoch": 3.0723875181422353,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004067275976182935,
      "loss": 4.969,
      "mean_token_accuracy": 0.21931826919317246,
      "num_tokens": 70279380.0,
      "step": 33870
    },
    {
      "entropy": 5.681621026992798,
      "epoch": 3.072841074020319,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040670136477117906,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2218307927250862,
      "num_tokens": 70289919.0,
      "step": 33875
    },
    {
      "entropy": 5.702227926254272,
      "epoch": 3.0732946298984034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040667512920044684,
      "loss": 4.9848,
      "mean_token_accuracy": 0.2199021264910698,
      "num_tokens": 70300136.0,
      "step": 33880
    },
    {
      "entropy": 5.573601722717285,
      "epoch": 3.0737481857764877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040664889090663944,
      "loss": 4.7828,
      "mean_token_accuracy": 0.23343218863010406,
      "num_tokens": 70310072.0,
      "step": 33885
    },
    {
      "entropy": 5.663991785049438,
      "epoch": 3.074201741654572,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040662264989029933,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22599176019430162,
      "num_tokens": 70320185.0,
      "step": 33890
    },
    {
      "entropy": 5.637185382843017,
      "epoch": 3.074655297532656,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004065964061519694,
      "loss": 4.8795,
      "mean_token_accuracy": 0.21689095050096513,
      "num_tokens": 70330211.0,
      "step": 33895
    },
    {
      "entropy": 5.644785451889038,
      "epoch": 3.07510885341074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004065701596921922,
      "loss": 4.9571,
      "mean_token_accuracy": 0.21332648098468782,
      "num_tokens": 70341044.0,
      "step": 33900
    },
    {
      "entropy": 5.660111618041992,
      "epoch": 3.0755624092888243,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004065439105115106,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2186955764889717,
      "num_tokens": 70351320.0,
      "step": 33905
    },
    {
      "entropy": 5.6077038764953615,
      "epoch": 3.0760159651669086,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040651765861046734,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2264741376042366,
      "num_tokens": 70361214.0,
      "step": 33910
    },
    {
      "entropy": 5.701736164093018,
      "epoch": 3.076469521044993,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00040649140398960537,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22358901500701905,
      "num_tokens": 70371083.0,
      "step": 33915
    },
    {
      "entropy": 5.667549657821655,
      "epoch": 3.076923076923077,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040646514664946775,
      "loss": 5.0107,
      "mean_token_accuracy": 0.21717533022165297,
      "num_tokens": 70381987.0,
      "step": 33920
    },
    {
      "entropy": 5.5945886135101315,
      "epoch": 3.077376632801161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004064388865905974,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22062086760997773,
      "num_tokens": 70393462.0,
      "step": 33925
    },
    {
      "entropy": 5.581050539016724,
      "epoch": 3.077830188679245,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004064126238135373,
      "loss": 4.7556,
      "mean_token_accuracy": 0.23209473639726638,
      "num_tokens": 70403664.0,
      "step": 33930
    },
    {
      "entropy": 5.620469808578491,
      "epoch": 3.0782837445573294,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004063863583188308,
      "loss": 4.8997,
      "mean_token_accuracy": 0.21510847955942153,
      "num_tokens": 70414588.0,
      "step": 33935
    },
    {
      "entropy": 5.607044982910156,
      "epoch": 3.0787373004354137,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004063600901070207,
      "loss": 4.9434,
      "mean_token_accuracy": 0.21932426244020461,
      "num_tokens": 70425572.0,
      "step": 33940
    },
    {
      "entropy": 5.6360632419586185,
      "epoch": 3.079190856313498,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040633381917865057,
      "loss": 4.854,
      "mean_token_accuracy": 0.21611116230487823,
      "num_tokens": 70436140.0,
      "step": 33945
    },
    {
      "entropy": 5.599466228485108,
      "epoch": 3.079644412191582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004063075455342636,
      "loss": 4.8584,
      "mean_token_accuracy": 0.22500063627958297,
      "num_tokens": 70447117.0,
      "step": 33950
    },
    {
      "entropy": 5.643934202194214,
      "epoch": 3.080097968069666,
      "grad_norm": 1.125,
      "learning_rate": 0.0004062812691744032,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2246602803468704,
      "num_tokens": 70457756.0,
      "step": 33955
    },
    {
      "entropy": 5.620090579986572,
      "epoch": 3.0805515239477503,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004062549900996126,
      "loss": 4.9382,
      "mean_token_accuracy": 0.21645656675100328,
      "num_tokens": 70468671.0,
      "step": 33960
    },
    {
      "entropy": 5.641111564636231,
      "epoch": 3.0810050798258346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004062287083104354,
      "loss": 4.834,
      "mean_token_accuracy": 0.22732939720153808,
      "num_tokens": 70478212.0,
      "step": 33965
    },
    {
      "entropy": 5.573458051681518,
      "epoch": 3.081458635703919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040620242380741504,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22254758775234224,
      "num_tokens": 70489062.0,
      "step": 33970
    },
    {
      "entropy": 5.634737777709961,
      "epoch": 3.0819121915820027,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004061761365910951,
      "loss": 4.9361,
      "mean_token_accuracy": 0.2210964396595955,
      "num_tokens": 70499732.0,
      "step": 33975
    },
    {
      "entropy": 5.6545978546142575,
      "epoch": 3.082365747460087,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040614984666201927,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21345306038856507,
      "num_tokens": 70509777.0,
      "step": 33980
    },
    {
      "entropy": 5.683950424194336,
      "epoch": 3.0828193033381712,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004061235540207312,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21619537472724915,
      "num_tokens": 70520674.0,
      "step": 33985
    },
    {
      "entropy": 5.655579662322998,
      "epoch": 3.0832728592162555,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004060972586677746,
      "loss": 4.8788,
      "mean_token_accuracy": 0.2200765758752823,
      "num_tokens": 70531115.0,
      "step": 33990
    },
    {
      "entropy": 5.600628614425659,
      "epoch": 3.0837264150943398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004060709606036933,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22344197779893876,
      "num_tokens": 70540726.0,
      "step": 33995
    },
    {
      "entropy": 5.648413419723511,
      "epoch": 3.084179970972424,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004060446598290311,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22651210725307463,
      "num_tokens": 70551227.0,
      "step": 34000
    },
    {
      "entropy": 5.593591356277466,
      "epoch": 3.084633526850508,
      "grad_norm": 1.234375,
      "learning_rate": 0.000406018356344332,
      "loss": 4.8645,
      "mean_token_accuracy": 0.21862539649009705,
      "num_tokens": 70560815.0,
      "step": 34005
    },
    {
      "entropy": 5.67796311378479,
      "epoch": 3.085087082728592,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004059920501501398,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21048105210065843,
      "num_tokens": 70570247.0,
      "step": 34010
    },
    {
      "entropy": 5.67289981842041,
      "epoch": 3.0855406386066764,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004059657412469987,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2208164557814598,
      "num_tokens": 70581497.0,
      "step": 34015
    },
    {
      "entropy": 5.6354621887207035,
      "epoch": 3.0859941944847606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004059394296354527,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21745911985635757,
      "num_tokens": 70593133.0,
      "step": 34020
    },
    {
      "entropy": 5.636033916473389,
      "epoch": 3.086447750362845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004059131153160459,
      "loss": 4.8841,
      "mean_token_accuracy": 0.227353473007679,
      "num_tokens": 70603953.0,
      "step": 34025
    },
    {
      "entropy": 5.598169374465942,
      "epoch": 3.0869013062409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004058867982893225,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2273429214954376,
      "num_tokens": 70615292.0,
      "step": 34030
    },
    {
      "entropy": 5.682684469223022,
      "epoch": 3.087354862119013,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004058604785558268,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20723785012960433,
      "num_tokens": 70626681.0,
      "step": 34035
    },
    {
      "entropy": 5.692091655731201,
      "epoch": 3.0878084179970973,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040583415611610305,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2127999484539032,
      "num_tokens": 70637339.0,
      "step": 34040
    },
    {
      "entropy": 5.719115591049194,
      "epoch": 3.0882619738751815,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004058078309706956,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2234075501561165,
      "num_tokens": 70647980.0,
      "step": 34045
    },
    {
      "entropy": 5.613509178161621,
      "epoch": 3.088715529753266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004057815031201488,
      "loss": 4.7133,
      "mean_token_accuracy": 0.23890773504972457,
      "num_tokens": 70657042.0,
      "step": 34050
    },
    {
      "entropy": 5.646336412429809,
      "epoch": 3.0891690856313496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004057551725650073,
      "loss": 4.9233,
      "mean_token_accuracy": 0.2168408915400505,
      "num_tokens": 70668542.0,
      "step": 34055
    },
    {
      "entropy": 5.637628555297852,
      "epoch": 3.089622641509434,
      "grad_norm": 1.25,
      "learning_rate": 0.00040572883930581544,
      "loss": 4.9573,
      "mean_token_accuracy": 0.21869581639766694,
      "num_tokens": 70679332.0,
      "step": 34060
    },
    {
      "entropy": 5.758890295028687,
      "epoch": 3.090076197387518,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004057025033431179,
      "loss": 5.029,
      "mean_token_accuracy": 0.2133374273777008,
      "num_tokens": 70689897.0,
      "step": 34065
    },
    {
      "entropy": 5.6512048721313475,
      "epoch": 3.0905297532656024,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040567616467745933,
      "loss": 4.8395,
      "mean_token_accuracy": 0.22347125858068467,
      "num_tokens": 70699965.0,
      "step": 34070
    },
    {
      "entropy": 5.647654056549072,
      "epoch": 3.0909833091436867,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004056498233093843,
      "loss": 4.9554,
      "mean_token_accuracy": 0.219772107899189,
      "num_tokens": 70710583.0,
      "step": 34075
    },
    {
      "entropy": 5.575544309616089,
      "epoch": 3.0914368650217705,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040562347923943766,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23709533661603927,
      "num_tokens": 70721800.0,
      "step": 34080
    },
    {
      "entropy": 5.600347661972046,
      "epoch": 3.0918904208998548,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004055971324681642,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21650346070528032,
      "num_tokens": 70731527.0,
      "step": 34085
    },
    {
      "entropy": 5.698315620422363,
      "epoch": 3.092343976777939,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004055707829961087,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22603444904088973,
      "num_tokens": 70740821.0,
      "step": 34090
    },
    {
      "entropy": 5.652254962921143,
      "epoch": 3.0927975326560233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040554443082381627,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22231314182281495,
      "num_tokens": 70750506.0,
      "step": 34095
    },
    {
      "entropy": 5.617465400695801,
      "epoch": 3.0932510885341076,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004055180759518315,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22190633714199065,
      "num_tokens": 70762009.0,
      "step": 34100
    },
    {
      "entropy": 5.569088459014893,
      "epoch": 3.0937046444121914,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004054917183806999,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22388028502464294,
      "num_tokens": 70772030.0,
      "step": 34105
    },
    {
      "entropy": 5.65435757637024,
      "epoch": 3.0941582002902757,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040546535811096626,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2122175946831703,
      "num_tokens": 70782539.0,
      "step": 34110
    },
    {
      "entropy": 5.811967229843139,
      "epoch": 3.09461175616836,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040543899514317564,
      "loss": 5.0432,
      "mean_token_accuracy": 0.2077189639210701,
      "num_tokens": 70793433.0,
      "step": 34115
    },
    {
      "entropy": 5.678083705902099,
      "epoch": 3.095065312046444,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004054126294778735,
      "loss": 4.9312,
      "mean_token_accuracy": 0.2127001866698265,
      "num_tokens": 70802919.0,
      "step": 34120
    },
    {
      "entropy": 5.595437097549438,
      "epoch": 3.0955188679245285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040538626111560487,
      "loss": 4.8373,
      "mean_token_accuracy": 0.22360349744558333,
      "num_tokens": 70813827.0,
      "step": 34125
    },
    {
      "entropy": 5.678167486190796,
      "epoch": 3.0959724238026123,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040535989005691515,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22078392058610916,
      "num_tokens": 70824117.0,
      "step": 34130
    },
    {
      "entropy": 5.569955444335937,
      "epoch": 3.0964259796806965,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040533351630234985,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2252984508872032,
      "num_tokens": 70833387.0,
      "step": 34135
    },
    {
      "entropy": 5.602465343475342,
      "epoch": 3.096879535558781,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000405307139852454,
      "loss": 4.8553,
      "mean_token_accuracy": 0.21835729032754897,
      "num_tokens": 70844363.0,
      "step": 34140
    },
    {
      "entropy": 5.6697197437286375,
      "epoch": 3.097333091436865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040528076070777335,
      "loss": 4.9221,
      "mean_token_accuracy": 0.215872161090374,
      "num_tokens": 70854759.0,
      "step": 34145
    },
    {
      "entropy": 5.5396815776824955,
      "epoch": 3.0977866473149493,
      "grad_norm": 1.125,
      "learning_rate": 0.0004052543788688534,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2178231045603752,
      "num_tokens": 70865682.0,
      "step": 34150
    },
    {
      "entropy": 5.7179527759552,
      "epoch": 3.098240203193033,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004052279943362396,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2154765710234642,
      "num_tokens": 70875449.0,
      "step": 34155
    },
    {
      "entropy": 5.698051500320434,
      "epoch": 3.0986937590711174,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004052016071104779,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22106702774763107,
      "num_tokens": 70885885.0,
      "step": 34160
    },
    {
      "entropy": 5.642788219451904,
      "epoch": 3.0991473149492017,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004051752171921136,
      "loss": 4.8553,
      "mean_token_accuracy": 0.219012251496315,
      "num_tokens": 70896320.0,
      "step": 34165
    },
    {
      "entropy": 5.590386533737183,
      "epoch": 3.099600870827286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040514882458169267,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22594895362854003,
      "num_tokens": 70907074.0,
      "step": 34170
    },
    {
      "entropy": 5.6268908977508545,
      "epoch": 3.1000544267053702,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004051224292797609,
      "loss": 4.9149,
      "mean_token_accuracy": 0.20936949104070662,
      "num_tokens": 70916567.0,
      "step": 34175
    },
    {
      "entropy": 5.6432532787323,
      "epoch": 3.100507982583454,
      "grad_norm": 1.375,
      "learning_rate": 0.00040509603128686413,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22425558269023896,
      "num_tokens": 70926163.0,
      "step": 34180
    },
    {
      "entropy": 5.685975360870361,
      "epoch": 3.1009615384615383,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004050696306035483,
      "loss": 4.8808,
      "mean_token_accuracy": 0.21428776979446412,
      "num_tokens": 70937358.0,
      "step": 34185
    },
    {
      "entropy": 5.600000095367432,
      "epoch": 3.1014150943396226,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040504322723035937,
      "loss": 4.8721,
      "mean_token_accuracy": 0.2204276368021965,
      "num_tokens": 70949305.0,
      "step": 34190
    },
    {
      "entropy": 5.582901859283448,
      "epoch": 3.101868650217707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00040501682116784326,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22548231184482576,
      "num_tokens": 70959146.0,
      "step": 34195
    },
    {
      "entropy": 5.602471542358399,
      "epoch": 3.102322206095791,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040499041241654626,
      "loss": 4.9239,
      "mean_token_accuracy": 0.21245956271886826,
      "num_tokens": 70968828.0,
      "step": 34200
    },
    {
      "entropy": 5.6074143409729,
      "epoch": 3.1027757619738754,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004049640009770144,
      "loss": 4.8757,
      "mean_token_accuracy": 0.220449398458004,
      "num_tokens": 70979733.0,
      "step": 34205
    },
    {
      "entropy": 5.679778051376343,
      "epoch": 3.103229317851959,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004049375868497939,
      "loss": 4.9285,
      "mean_token_accuracy": 0.223466919362545,
      "num_tokens": 70989726.0,
      "step": 34210
    },
    {
      "entropy": 5.689506387710571,
      "epoch": 3.1036828737300435,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040491117003543087,
      "loss": 4.9461,
      "mean_token_accuracy": 0.2188631072640419,
      "num_tokens": 71000715.0,
      "step": 34215
    },
    {
      "entropy": 5.588877630233765,
      "epoch": 3.1041364296081277,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040488475053447186,
      "loss": 4.7218,
      "mean_token_accuracy": 0.23317428976297377,
      "num_tokens": 71011453.0,
      "step": 34220
    },
    {
      "entropy": 5.643509721755981,
      "epoch": 3.104589985486212,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040485832834746306,
      "loss": 4.918,
      "mean_token_accuracy": 0.21931473463773726,
      "num_tokens": 71021477.0,
      "step": 34225
    },
    {
      "entropy": 5.587640619277954,
      "epoch": 3.1050435413642963,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040483190347495097,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22006329149007797,
      "num_tokens": 71031706.0,
      "step": 34230
    },
    {
      "entropy": 5.641937875747681,
      "epoch": 3.10549709724238,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000404805475917482,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22420570999383926,
      "num_tokens": 71042882.0,
      "step": 34235
    },
    {
      "entropy": 5.624765110015869,
      "epoch": 3.1059506531204644,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040477904567560275,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22554701417684556,
      "num_tokens": 71052804.0,
      "step": 34240
    },
    {
      "entropy": 5.686054515838623,
      "epoch": 3.1064042089985486,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004047526127498597,
      "loss": 4.9721,
      "mean_token_accuracy": 0.2081589311361313,
      "num_tokens": 71064349.0,
      "step": 34245
    },
    {
      "entropy": 5.640777254104615,
      "epoch": 3.106857764876633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040472617714079964,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22418843656778337,
      "num_tokens": 71075754.0,
      "step": 34250
    },
    {
      "entropy": 5.664772939682007,
      "epoch": 3.107311320754717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004046997388489691,
      "loss": 4.8373,
      "mean_token_accuracy": 0.22224117070436478,
      "num_tokens": 71085338.0,
      "step": 34255
    },
    {
      "entropy": 5.584049415588379,
      "epoch": 3.107764876632801,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004046732978749149,
      "loss": 4.8456,
      "mean_token_accuracy": 0.2197646200656891,
      "num_tokens": 71097089.0,
      "step": 34260
    },
    {
      "entropy": 5.555621910095215,
      "epoch": 3.1082184325108853,
      "grad_norm": 1.078125,
      "learning_rate": 0.000404646854219184,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2302098348736763,
      "num_tokens": 71106990.0,
      "step": 34265
    },
    {
      "entropy": 5.60454216003418,
      "epoch": 3.1086719883889695,
      "grad_norm": 1.109375,
      "learning_rate": 0.000404620407882323,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22267862111330033,
      "num_tokens": 71117634.0,
      "step": 34270
    },
    {
      "entropy": 5.578424453735352,
      "epoch": 3.109125544267054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040459395886487885,
      "loss": 4.7991,
      "mean_token_accuracy": 0.2192835882306099,
      "num_tokens": 71127717.0,
      "step": 34275
    },
    {
      "entropy": 5.640869808197022,
      "epoch": 3.109579100145138,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004045675071673988,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22645338028669357,
      "num_tokens": 71137172.0,
      "step": 34280
    },
    {
      "entropy": 5.681245136260986,
      "epoch": 3.110032656023222,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004045410527904296,
      "loss": 4.9219,
      "mean_token_accuracy": 0.21089086979627608,
      "num_tokens": 71146843.0,
      "step": 34285
    },
    {
      "entropy": 5.565523481369018,
      "epoch": 3.110486211901306,
      "grad_norm": 1.25,
      "learning_rate": 0.0004045145957345184,
      "loss": 4.7591,
      "mean_token_accuracy": 0.23147310614585875,
      "num_tokens": 71156619.0,
      "step": 34290
    },
    {
      "entropy": 5.618736457824707,
      "epoch": 3.1109397677793904,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004044881360002123,
      "loss": 4.869,
      "mean_token_accuracy": 0.21954719871282577,
      "num_tokens": 71167234.0,
      "step": 34295
    },
    {
      "entropy": 5.6318083763122555,
      "epoch": 3.1113933236574747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004044616735880586,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22432273775339126,
      "num_tokens": 71178049.0,
      "step": 34300
    },
    {
      "entropy": 5.602747917175293,
      "epoch": 3.111846879535559,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040443520849860453,
      "loss": 4.8068,
      "mean_token_accuracy": 0.22295840233564376,
      "num_tokens": 71188568.0,
      "step": 34305
    },
    {
      "entropy": 5.645314073562622,
      "epoch": 3.1123004354136428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040440874073239733,
      "loss": 4.8968,
      "mean_token_accuracy": 0.21727484315633774,
      "num_tokens": 71199004.0,
      "step": 34310
    },
    {
      "entropy": 5.56238694190979,
      "epoch": 3.112753991291727,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040438227028998436,
      "loss": 4.861,
      "mean_token_accuracy": 0.22123947292566298,
      "num_tokens": 71209239.0,
      "step": 34315
    },
    {
      "entropy": 5.618282890319824,
      "epoch": 3.1132075471698113,
      "grad_norm": 1.25,
      "learning_rate": 0.00040435579717191313,
      "loss": 4.9377,
      "mean_token_accuracy": 0.21733635365962983,
      "num_tokens": 71219469.0,
      "step": 34320
    },
    {
      "entropy": 5.6560159683227536,
      "epoch": 3.1136611030478956,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040432932137873107,
      "loss": 4.9348,
      "mean_token_accuracy": 0.2126480832695961,
      "num_tokens": 71230670.0,
      "step": 34325
    },
    {
      "entropy": 5.721735191345215,
      "epoch": 3.11411465892598,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004043028429109857,
      "loss": 4.9851,
      "mean_token_accuracy": 0.2140109807252884,
      "num_tokens": 71241512.0,
      "step": 34330
    },
    {
      "entropy": 5.619463539123535,
      "epoch": 3.1145682148040637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004042763617692245,
      "loss": 4.8237,
      "mean_token_accuracy": 0.22331147491931916,
      "num_tokens": 71250869.0,
      "step": 34335
    },
    {
      "entropy": 5.592361783981323,
      "epoch": 3.115021770682148,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040424987795399523,
      "loss": 4.8348,
      "mean_token_accuracy": 0.22232449799776077,
      "num_tokens": 71260577.0,
      "step": 34340
    },
    {
      "entropy": 5.643589830398559,
      "epoch": 3.115475326560232,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004042233914658455,
      "loss": 4.9007,
      "mean_token_accuracy": 0.21975219994783401,
      "num_tokens": 71270185.0,
      "step": 34345
    },
    {
      "entropy": 5.579405641555786,
      "epoch": 3.1159288824383164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040419690230532324,
      "loss": 4.8463,
      "mean_token_accuracy": 0.22394930571317673,
      "num_tokens": 71279427.0,
      "step": 34350
    },
    {
      "entropy": 5.6149085521697994,
      "epoch": 3.1163824383164007,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040417041047297606,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22412284165620805,
      "num_tokens": 71290857.0,
      "step": 34355
    },
    {
      "entropy": 5.666221857070923,
      "epoch": 3.116835994194485,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004041439159693518,
      "loss": 4.9223,
      "mean_token_accuracy": 0.2188919499516487,
      "num_tokens": 71300971.0,
      "step": 34360
    },
    {
      "entropy": 5.672841596603393,
      "epoch": 3.117289550072569,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004041174187949985,
      "loss": 4.98,
      "mean_token_accuracy": 0.21181755363941193,
      "num_tokens": 71311017.0,
      "step": 34365
    },
    {
      "entropy": 5.667797565460205,
      "epoch": 3.117743105950653,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040409091895046413,
      "loss": 4.9794,
      "mean_token_accuracy": 0.21333476155996323,
      "num_tokens": 71321546.0,
      "step": 34370
    },
    {
      "entropy": 5.625165939331055,
      "epoch": 3.1181966618287373,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040406441643629656,
      "loss": 4.901,
      "mean_token_accuracy": 0.2174267962574959,
      "num_tokens": 71332345.0,
      "step": 34375
    },
    {
      "entropy": 5.708079481124878,
      "epoch": 3.1186502177068216,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040403791125304405,
      "loss": 4.9678,
      "mean_token_accuracy": 0.22510417699813842,
      "num_tokens": 71342670.0,
      "step": 34380
    },
    {
      "entropy": 5.64798674583435,
      "epoch": 3.119103773584906,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004040114034012546,
      "loss": 4.8256,
      "mean_token_accuracy": 0.21764114648103713,
      "num_tokens": 71353440.0,
      "step": 34385
    },
    {
      "entropy": 5.661004161834716,
      "epoch": 3.1195573294629897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040398489288147646,
      "loss": 4.9516,
      "mean_token_accuracy": 0.2206189066171646,
      "num_tokens": 71363199.0,
      "step": 34390
    },
    {
      "entropy": 5.646486282348633,
      "epoch": 3.120010885341074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040395837969425786,
      "loss": 4.878,
      "mean_token_accuracy": 0.22108954489231109,
      "num_tokens": 71372925.0,
      "step": 34395
    },
    {
      "entropy": 5.679801368713379,
      "epoch": 3.1204644412191582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004039318638401471,
      "loss": 4.9053,
      "mean_token_accuracy": 0.2170091837644577,
      "num_tokens": 71383350.0,
      "step": 34400
    },
    {
      "entropy": 5.620130157470703,
      "epoch": 3.1209179970972425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004039053453196926,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23016337305307388,
      "num_tokens": 71394006.0,
      "step": 34405
    },
    {
      "entropy": 5.583880519866943,
      "epoch": 3.1213715529753268,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004038788241334427,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22084433287382127,
      "num_tokens": 71403945.0,
      "step": 34410
    },
    {
      "entropy": 5.664468908309937,
      "epoch": 3.1218251088534106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004038523002819458,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21915849298238754,
      "num_tokens": 71414238.0,
      "step": 34415
    },
    {
      "entropy": 5.596865034103393,
      "epoch": 3.122278664731495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040382577376575055,
      "loss": 4.9183,
      "mean_token_accuracy": 0.21341052204370498,
      "num_tokens": 71423586.0,
      "step": 34420
    },
    {
      "entropy": 5.634012985229492,
      "epoch": 3.122732220609579,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004037992445854054,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22419103533029555,
      "num_tokens": 71434261.0,
      "step": 34425
    },
    {
      "entropy": 5.642942333221436,
      "epoch": 3.1231857764876634,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040377271274145916,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2194810315966606,
      "num_tokens": 71443952.0,
      "step": 34430
    },
    {
      "entropy": 5.682459592819214,
      "epoch": 3.1236393323657476,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004037461782344603,
      "loss": 4.8547,
      "mean_token_accuracy": 0.225861120223999,
      "num_tokens": 71454051.0,
      "step": 34435
    },
    {
      "entropy": 5.615170478820801,
      "epoch": 3.1240928882438315,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040371964106495777,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22322461605072022,
      "num_tokens": 71464579.0,
      "step": 34440
    },
    {
      "entropy": 5.613699531555175,
      "epoch": 3.1245464441219157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040369310123350015,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22144591808319092,
      "num_tokens": 71474631.0,
      "step": 34445
    },
    {
      "entropy": 5.719875144958496,
      "epoch": 3.125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040366655874063645,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2226760357618332,
      "num_tokens": 71485266.0,
      "step": 34450
    },
    {
      "entropy": 5.606971788406372,
      "epoch": 3.1254535558780843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004036400135869156,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23296214640140533,
      "num_tokens": 71495248.0,
      "step": 34455
    },
    {
      "entropy": 5.622581243515015,
      "epoch": 3.1259071117561685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040361346577288645,
      "loss": 4.8615,
      "mean_token_accuracy": 0.22479719370603563,
      "num_tokens": 71505600.0,
      "step": 34460
    },
    {
      "entropy": 5.6418421268463135,
      "epoch": 3.1263606676342524,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040358691529909796,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22595479637384414,
      "num_tokens": 71516634.0,
      "step": 34465
    },
    {
      "entropy": 5.690576791763306,
      "epoch": 3.1268142235123366,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040356036216609943,
      "loss": 4.8823,
      "mean_token_accuracy": 0.21582386493682862,
      "num_tokens": 71526745.0,
      "step": 34470
    },
    {
      "entropy": 5.723757982254028,
      "epoch": 3.127267779390421,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040353380637443974,
      "loss": 5.0039,
      "mean_token_accuracy": 0.20903454273939132,
      "num_tokens": 71537015.0,
      "step": 34475
    },
    {
      "entropy": 5.6166853427886965,
      "epoch": 3.127721335268505,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040350724792466826,
      "loss": 4.7678,
      "mean_token_accuracy": 0.22553042471408843,
      "num_tokens": 71546660.0,
      "step": 34480
    },
    {
      "entropy": 5.599078559875489,
      "epoch": 3.1281748911465894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004034806868173341,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22059424370527267,
      "num_tokens": 71557309.0,
      "step": 34485
    },
    {
      "entropy": 5.553763389587402,
      "epoch": 3.1286284470246732,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004034541230529866,
      "loss": 4.7781,
      "mean_token_accuracy": 0.2237554520368576,
      "num_tokens": 71567651.0,
      "step": 34490
    },
    {
      "entropy": 5.602401065826416,
      "epoch": 3.1290820029027575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040342755663217507,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22586155235767363,
      "num_tokens": 71577538.0,
      "step": 34495
    },
    {
      "entropy": 5.605494594573974,
      "epoch": 3.1295355587808418,
      "grad_norm": 1.203125,
      "learning_rate": 0.000403400987555449,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22852900326251985,
      "num_tokens": 71588299.0,
      "step": 34500
    },
    {
      "entropy": 5.649336624145508,
      "epoch": 3.129989114658926,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004033744158233577,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21785535365343095,
      "num_tokens": 71598989.0,
      "step": 34505
    },
    {
      "entropy": 5.648256015777588,
      "epoch": 3.1304426705370103,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040334784143645087,
      "loss": 4.8893,
      "mean_token_accuracy": 0.21701750606298448,
      "num_tokens": 71608596.0,
      "step": 34510
    },
    {
      "entropy": 5.566717672348022,
      "epoch": 3.1308962264150946,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004033212643952779,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22374419122934341,
      "num_tokens": 71619182.0,
      "step": 34515
    },
    {
      "entropy": 5.647668933868408,
      "epoch": 3.1313497822931784,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040329468470038857,
      "loss": 4.9092,
      "mean_token_accuracy": 0.21866436302661896,
      "num_tokens": 71629526.0,
      "step": 34520
    },
    {
      "entropy": 5.6823893070220945,
      "epoch": 3.1318033381712627,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004032681023523324,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21634445041418077,
      "num_tokens": 71640146.0,
      "step": 34525
    },
    {
      "entropy": 5.629496669769287,
      "epoch": 3.132256894049347,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004032415173516592,
      "loss": 4.7993,
      "mean_token_accuracy": 0.22431942522525788,
      "num_tokens": 71649652.0,
      "step": 34530
    },
    {
      "entropy": 5.657601165771484,
      "epoch": 3.132710449927431,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004032149296989187,
      "loss": 4.875,
      "mean_token_accuracy": 0.2248346045613289,
      "num_tokens": 71659264.0,
      "step": 34535
    },
    {
      "entropy": 5.658296203613281,
      "epoch": 3.133164005805515,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040318833939466083,
      "loss": 4.8883,
      "mean_token_accuracy": 0.21746332049369813,
      "num_tokens": 71669175.0,
      "step": 34540
    },
    {
      "entropy": 5.598744058609009,
      "epoch": 3.1336175616835993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004031617464394354,
      "loss": 4.8961,
      "mean_token_accuracy": 0.21986704021692277,
      "num_tokens": 71679537.0,
      "step": 34545
    },
    {
      "entropy": 5.657726716995239,
      "epoch": 3.1340711175616836,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004031351508337925,
      "loss": 4.9238,
      "mean_token_accuracy": 0.22306990325450898,
      "num_tokens": 71690024.0,
      "step": 34550
    },
    {
      "entropy": 5.725058937072754,
      "epoch": 3.134524673439768,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004031085525782819,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22586895525455475,
      "num_tokens": 71700019.0,
      "step": 34555
    },
    {
      "entropy": 5.695723390579223,
      "epoch": 3.134978229317852,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004030819516734539,
      "loss": 4.9345,
      "mean_token_accuracy": 0.2098943531513214,
      "num_tokens": 71710570.0,
      "step": 34560
    },
    {
      "entropy": 5.651312971115113,
      "epoch": 3.1354317851959363,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040305534811985844,
      "loss": 4.9254,
      "mean_token_accuracy": 0.2238598048686981,
      "num_tokens": 71720619.0,
      "step": 34565
    },
    {
      "entropy": 5.603794765472412,
      "epoch": 3.13588534107402,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004030287419180457,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22368565797805787,
      "num_tokens": 71730201.0,
      "step": 34570
    },
    {
      "entropy": 5.646978712081909,
      "epoch": 3.1363388969521044,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040300213306856606,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22063202261924744,
      "num_tokens": 71741368.0,
      "step": 34575
    },
    {
      "entropy": 5.580033874511718,
      "epoch": 3.1367924528301887,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004029755215719696,
      "loss": 4.8116,
      "mean_token_accuracy": 0.22565243542194366,
      "num_tokens": 71751361.0,
      "step": 34580
    },
    {
      "entropy": 5.677279615402222,
      "epoch": 3.137246008708273,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040294890742880677,
      "loss": 4.8546,
      "mean_token_accuracy": 0.22193023115396499,
      "num_tokens": 71761164.0,
      "step": 34585
    },
    {
      "entropy": 5.597966575622559,
      "epoch": 3.1376995645863572,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040292229063962787,
      "loss": 4.8487,
      "mean_token_accuracy": 0.2306194305419922,
      "num_tokens": 71770726.0,
      "step": 34590
    },
    {
      "entropy": 5.657065534591675,
      "epoch": 3.138153120464441,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004028956712049834,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22643675804138183,
      "num_tokens": 71781124.0,
      "step": 34595
    },
    {
      "entropy": 5.627010488510132,
      "epoch": 3.1386066763425253,
      "grad_norm": 1.140625,
      "learning_rate": 0.000402869049125424,
      "loss": 4.8646,
      "mean_token_accuracy": 0.21795458495616912,
      "num_tokens": 71791218.0,
      "step": 34600
    },
    {
      "entropy": 5.641670083999633,
      "epoch": 3.1390602322206096,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004028424244015,
      "loss": 4.881,
      "mean_token_accuracy": 0.22243322432041168,
      "num_tokens": 71801013.0,
      "step": 34605
    },
    {
      "entropy": 5.661030197143555,
      "epoch": 3.139513788098694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402815797033762,
      "loss": 4.9599,
      "mean_token_accuracy": 0.20959967970848084,
      "num_tokens": 71811742.0,
      "step": 34610
    },
    {
      "entropy": 5.570937299728394,
      "epoch": 3.139967343976778,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004027891670227609,
      "loss": 4.8678,
      "mean_token_accuracy": 0.21754186898469924,
      "num_tokens": 71822265.0,
      "step": 34615
    },
    {
      "entropy": 5.637689733505249,
      "epoch": 3.140420899854862,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004027625343690472,
      "loss": 4.8418,
      "mean_token_accuracy": 0.22411544919013976,
      "num_tokens": 71832090.0,
      "step": 34620
    },
    {
      "entropy": 5.619313430786133,
      "epoch": 3.140874455732946,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004027358990731717,
      "loss": 4.8611,
      "mean_token_accuracy": 0.22226228415966034,
      "num_tokens": 71841381.0,
      "step": 34625
    },
    {
      "entropy": 5.620007848739624,
      "epoch": 3.1413280116110305,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040270926113568524,
      "loss": 4.8856,
      "mean_token_accuracy": 0.21865382492542268,
      "num_tokens": 71851383.0,
      "step": 34630
    },
    {
      "entropy": 5.607455825805664,
      "epoch": 3.1417815674891147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040268262055713873,
      "loss": 4.86,
      "mean_token_accuracy": 0.2236224055290222,
      "num_tokens": 71862588.0,
      "step": 34635
    },
    {
      "entropy": 5.591871500015259,
      "epoch": 3.142235123367199,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004026559773380831,
      "loss": 4.8453,
      "mean_token_accuracy": 0.22121569514274597,
      "num_tokens": 71873975.0,
      "step": 34640
    },
    {
      "entropy": 5.69539623260498,
      "epoch": 3.142688679245283,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004026293314790694,
      "loss": 4.9365,
      "mean_token_accuracy": 0.21123400032520295,
      "num_tokens": 71884084.0,
      "step": 34645
    },
    {
      "entropy": 5.629751968383789,
      "epoch": 3.143142235123367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004026026829806485,
      "loss": 4.8856,
      "mean_token_accuracy": 0.2175264000892639,
      "num_tokens": 71894643.0,
      "step": 34650
    },
    {
      "entropy": 5.549112176895141,
      "epoch": 3.1435957910014514,
      "grad_norm": 1.25,
      "learning_rate": 0.00040257603184337167,
      "loss": 4.8164,
      "mean_token_accuracy": 0.22330582588911058,
      "num_tokens": 71905245.0,
      "step": 34655
    },
    {
      "entropy": 5.66940975189209,
      "epoch": 3.1440493468795356,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040254937806778995,
      "loss": 4.9305,
      "mean_token_accuracy": 0.21012532263994216,
      "num_tokens": 71916473.0,
      "step": 34660
    },
    {
      "entropy": 5.64020562171936,
      "epoch": 3.14450290275762,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040252272165445456,
      "loss": 4.887,
      "mean_token_accuracy": 0.22013067752122878,
      "num_tokens": 71927465.0,
      "step": 34665
    },
    {
      "entropy": 5.634952163696289,
      "epoch": 3.1449564586357037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004024960626039168,
      "loss": 4.893,
      "mean_token_accuracy": 0.2251055896282196,
      "num_tokens": 71937380.0,
      "step": 34670
    },
    {
      "entropy": 5.6282249927520756,
      "epoch": 3.145410014513788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040246940091672806,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22478660196065903,
      "num_tokens": 71948088.0,
      "step": 34675
    },
    {
      "entropy": 5.641007804870606,
      "epoch": 3.1458635703918723,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004024427365934395,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2287004753947258,
      "num_tokens": 71958284.0,
      "step": 34680
    },
    {
      "entropy": 5.640749597549439,
      "epoch": 3.1463171262699565,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040241606963460283,
      "loss": 4.9369,
      "mean_token_accuracy": 0.21499605774879454,
      "num_tokens": 71968138.0,
      "step": 34685
    },
    {
      "entropy": 5.679515838623047,
      "epoch": 3.146770682148041,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040238940004076924,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21536802053451537,
      "num_tokens": 71978275.0,
      "step": 34690
    },
    {
      "entropy": 5.605430269241333,
      "epoch": 3.1472242380261246,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040236272781249047,
      "loss": 4.9073,
      "mean_token_accuracy": 0.21896383464336394,
      "num_tokens": 71988875.0,
      "step": 34695
    },
    {
      "entropy": 5.66218934059143,
      "epoch": 3.147677793904209,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040233605295031795,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2237221822142601,
      "num_tokens": 71999560.0,
      "step": 34700
    },
    {
      "entropy": 5.663083744049072,
      "epoch": 3.148131349782293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004023093754548034,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22820204198360444,
      "num_tokens": 72009970.0,
      "step": 34705
    },
    {
      "entropy": 5.65379581451416,
      "epoch": 3.1485849056603774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004022826953264986,
      "loss": 4.9034,
      "mean_token_accuracy": 0.2280860051512718,
      "num_tokens": 72020196.0,
      "step": 34710
    },
    {
      "entropy": 5.605658435821534,
      "epoch": 3.1490384615384617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040225601256595525,
      "loss": 4.8071,
      "mean_token_accuracy": 0.22312479615211486,
      "num_tokens": 72031456.0,
      "step": 34715
    },
    {
      "entropy": 5.667160415649414,
      "epoch": 3.149492017416546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000402229327173725,
      "loss": 4.9328,
      "mean_token_accuracy": 0.21889615952968597,
      "num_tokens": 72042260.0,
      "step": 34720
    },
    {
      "entropy": 5.699582195281982,
      "epoch": 3.1499455732946298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040220263915035997,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20393963903188705,
      "num_tokens": 72053463.0,
      "step": 34725
    },
    {
      "entropy": 5.741630029678345,
      "epoch": 3.150399129172714,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040217594849641195,
      "loss": 4.9989,
      "mean_token_accuracy": 0.21108173429965973,
      "num_tokens": 72064508.0,
      "step": 34730
    },
    {
      "entropy": 5.593961620330811,
      "epoch": 3.1508526850507983,
      "grad_norm": 1.25,
      "learning_rate": 0.00040214925521243287,
      "loss": 4.8874,
      "mean_token_accuracy": 0.21538327485322953,
      "num_tokens": 72074758.0,
      "step": 34735
    },
    {
      "entropy": 5.618295574188233,
      "epoch": 3.1513062409288826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040212255929897474,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22247523963451385,
      "num_tokens": 72085548.0,
      "step": 34740
    },
    {
      "entropy": 5.653174543380738,
      "epoch": 3.1517597968069664,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004020958607565897,
      "loss": 4.9696,
      "mean_token_accuracy": 0.2121471107006073,
      "num_tokens": 72096773.0,
      "step": 34745
    },
    {
      "entropy": 5.630391025543213,
      "epoch": 3.1522133526850507,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004020691595858299,
      "loss": 4.9366,
      "mean_token_accuracy": 0.21827808916568756,
      "num_tokens": 72106832.0,
      "step": 34750
    },
    {
      "entropy": 5.634949827194214,
      "epoch": 3.152666908563135,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040204245578724754,
      "loss": 4.9308,
      "mean_token_accuracy": 0.2228184938430786,
      "num_tokens": 72117758.0,
      "step": 34755
    },
    {
      "entropy": 5.673316097259521,
      "epoch": 3.153120464441219,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040201574936139475,
      "loss": 4.8381,
      "mean_token_accuracy": 0.2179548978805542,
      "num_tokens": 72127522.0,
      "step": 34760
    },
    {
      "entropy": 5.605249834060669,
      "epoch": 3.1535740203193035,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004019890403088239,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22111939936876296,
      "num_tokens": 72138310.0,
      "step": 34765
    },
    {
      "entropy": 5.6006762981414795,
      "epoch": 3.1540275761973877,
      "grad_norm": 1.25,
      "learning_rate": 0.00040196232863008735,
      "loss": 4.9176,
      "mean_token_accuracy": 0.22710379511117934,
      "num_tokens": 72148958.0,
      "step": 34770
    },
    {
      "entropy": 5.649778223037719,
      "epoch": 3.1544811320754715,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040193561432573754,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22144597321748732,
      "num_tokens": 72158594.0,
      "step": 34775
    },
    {
      "entropy": 5.702221775054932,
      "epoch": 3.154934687953556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004019088973963268,
      "loss": 4.9673,
      "mean_token_accuracy": 0.2173070177435875,
      "num_tokens": 72168552.0,
      "step": 34780
    },
    {
      "entropy": 5.643583345413208,
      "epoch": 3.15538824383164,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040188217784240785,
      "loss": 4.843,
      "mean_token_accuracy": 0.22206685543060303,
      "num_tokens": 72178584.0,
      "step": 34785
    },
    {
      "entropy": 5.59626259803772,
      "epoch": 3.1558417997097243,
      "grad_norm": 1.15625,
      "learning_rate": 0.000401855455664533,
      "loss": 4.869,
      "mean_token_accuracy": 0.2235589787364006,
      "num_tokens": 72189144.0,
      "step": 34790
    },
    {
      "entropy": 5.667062902450562,
      "epoch": 3.1562953555878086,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004018287308632551,
      "loss": 4.9089,
      "mean_token_accuracy": 0.2191287621855736,
      "num_tokens": 72198326.0,
      "step": 34795
    },
    {
      "entropy": 5.662862014770508,
      "epoch": 3.1567489114658924,
      "grad_norm": 1.375,
      "learning_rate": 0.0004018020034391266,
      "loss": 4.8743,
      "mean_token_accuracy": 0.22026202082633972,
      "num_tokens": 72207916.0,
      "step": 34800
    },
    {
      "entropy": 5.70360631942749,
      "epoch": 3.1572024673439767,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004017752733927004,
      "loss": 4.9101,
      "mean_token_accuracy": 0.22381116598844528,
      "num_tokens": 72219111.0,
      "step": 34805
    },
    {
      "entropy": 5.635295438766479,
      "epoch": 3.157656023222061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004017485407245293,
      "loss": 4.909,
      "mean_token_accuracy": 0.22153260409832,
      "num_tokens": 72230081.0,
      "step": 34810
    },
    {
      "entropy": 5.560833311080932,
      "epoch": 3.1581095791001452,
      "grad_norm": 1.265625,
      "learning_rate": 0.000401721805435166,
      "loss": 4.8453,
      "mean_token_accuracy": 0.22652087211608887,
      "num_tokens": 72240693.0,
      "step": 34815
    },
    {
      "entropy": 5.558881759643555,
      "epoch": 3.1585631349782295,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040169506752516356,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23199326246976854,
      "num_tokens": 72250925.0,
      "step": 34820
    },
    {
      "entropy": 5.567978239059448,
      "epoch": 3.1590166908563133,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004016683269950748,
      "loss": 4.8163,
      "mean_token_accuracy": 0.23164866268634796,
      "num_tokens": 72261027.0,
      "step": 34825
    },
    {
      "entropy": 5.642466688156128,
      "epoch": 3.1594702467343976,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040164158384545267,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23070315271615982,
      "num_tokens": 72269430.0,
      "step": 34830
    },
    {
      "entropy": 5.567317533493042,
      "epoch": 3.159923802612482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004016148380768504,
      "loss": 4.8133,
      "mean_token_accuracy": 0.22541503310203553,
      "num_tokens": 72280207.0,
      "step": 34835
    },
    {
      "entropy": 5.648813199996948,
      "epoch": 3.160377358490566,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040158808968982103,
      "loss": 4.9817,
      "mean_token_accuracy": 0.22365449517965316,
      "num_tokens": 72290093.0,
      "step": 34840
    },
    {
      "entropy": 5.608896970748901,
      "epoch": 3.1608309143686504,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040156133868491766,
      "loss": 4.7996,
      "mean_token_accuracy": 0.2220901757478714,
      "num_tokens": 72300203.0,
      "step": 34845
    },
    {
      "entropy": 5.63458251953125,
      "epoch": 3.161284470246734,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040153458506269365,
      "loss": 4.8776,
      "mean_token_accuracy": 0.21886882185935974,
      "num_tokens": 72311778.0,
      "step": 34850
    },
    {
      "entropy": 5.6488481044769285,
      "epoch": 3.1617380261248185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000401507828823702,
      "loss": 4.8702,
      "mean_token_accuracy": 0.21621656119823457,
      "num_tokens": 72322142.0,
      "step": 34855
    },
    {
      "entropy": 5.6816082954406735,
      "epoch": 3.1621915820029027,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040148106996849636,
      "loss": 4.9468,
      "mean_token_accuracy": 0.2160783126950264,
      "num_tokens": 72332312.0,
      "step": 34860
    },
    {
      "entropy": 5.537344217300415,
      "epoch": 3.162645137880987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040145430849762993,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22452087551355362,
      "num_tokens": 72342157.0,
      "step": 34865
    },
    {
      "entropy": 5.64272723197937,
      "epoch": 3.1630986937590713,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040142754441165615,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2209317833185196,
      "num_tokens": 72353417.0,
      "step": 34870
    },
    {
      "entropy": 5.675842142105102,
      "epoch": 3.1635522496371555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004014007777111285,
      "loss": 4.863,
      "mean_token_accuracy": 0.22777371257543563,
      "num_tokens": 72363744.0,
      "step": 34875
    },
    {
      "entropy": 5.656932020187378,
      "epoch": 3.1640058055152394,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040137400839660066,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21974178403615952,
      "num_tokens": 72375241.0,
      "step": 34880
    },
    {
      "entropy": 5.642176008224487,
      "epoch": 3.1644593613933236,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040134723646862607,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2205859661102295,
      "num_tokens": 72385475.0,
      "step": 34885
    },
    {
      "entropy": 5.622985887527466,
      "epoch": 3.164912917271408,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040132046192775835,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21726237386465072,
      "num_tokens": 72396417.0,
      "step": 34890
    },
    {
      "entropy": 5.637712907791138,
      "epoch": 3.165366473149492,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004012936847745514,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21803796887397767,
      "num_tokens": 72406169.0,
      "step": 34895
    },
    {
      "entropy": 5.656631183624268,
      "epoch": 3.165820029027576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004012669050095589,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2148038312792778,
      "num_tokens": 72416522.0,
      "step": 34900
    },
    {
      "entropy": 5.6881077766418455,
      "epoch": 3.1662735849056602,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004012401226333345,
      "loss": 4.9332,
      "mean_token_accuracy": 0.21814776062965394,
      "num_tokens": 72426000.0,
      "step": 34905
    },
    {
      "entropy": 5.666016101837158,
      "epoch": 3.1667271407837445,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004012133376464322,
      "loss": 4.985,
      "mean_token_accuracy": 0.21452099680900574,
      "num_tokens": 72435952.0,
      "step": 34910
    },
    {
      "entropy": 5.648260307312012,
      "epoch": 3.1671806966618288,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00040118655004940597,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21772934794425963,
      "num_tokens": 72449957.0,
      "step": 34915
    },
    {
      "entropy": 5.660250473022461,
      "epoch": 3.167634252539913,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004011597598428097,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22428590953350067,
      "num_tokens": 72459956.0,
      "step": 34920
    },
    {
      "entropy": 5.627175569534302,
      "epoch": 3.1680878084179973,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004011329670271975,
      "loss": 4.828,
      "mean_token_accuracy": 0.22735811471939088,
      "num_tokens": 72471107.0,
      "step": 34925
    },
    {
      "entropy": 5.608335161209107,
      "epoch": 3.168541364296081,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040110617160312334,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22755054533481597,
      "num_tokens": 72481428.0,
      "step": 34930
    },
    {
      "entropy": 5.684324836730957,
      "epoch": 3.1689949201741654,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004010793735711415,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21034547686576843,
      "num_tokens": 72491133.0,
      "step": 34935
    },
    {
      "entropy": 5.604334783554077,
      "epoch": 3.1694484760522497,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040105257293180605,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22791304886341096,
      "num_tokens": 72502047.0,
      "step": 34940
    },
    {
      "entropy": 5.669875621795654,
      "epoch": 3.169902031930334,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004010257696856712,
      "loss": 4.886,
      "mean_token_accuracy": 0.21414785981178283,
      "num_tokens": 72513120.0,
      "step": 34945
    },
    {
      "entropy": 5.675568628311157,
      "epoch": 3.170355587808418,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040099896383329135,
      "loss": 4.9975,
      "mean_token_accuracy": 0.2245553106069565,
      "num_tokens": 72523118.0,
      "step": 34950
    },
    {
      "entropy": 5.6680426597595215,
      "epoch": 3.170809143686502,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040097215537522086,
      "loss": 4.9799,
      "mean_token_accuracy": 0.21648518294095992,
      "num_tokens": 72533697.0,
      "step": 34955
    },
    {
      "entropy": 5.573967599868775,
      "epoch": 3.1712626995645863,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040094534431201406,
      "loss": 4.838,
      "mean_token_accuracy": 0.22607991099357605,
      "num_tokens": 72543661.0,
      "step": 34960
    },
    {
      "entropy": 5.657451725006103,
      "epoch": 3.1717162554426706,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004009185306442255,
      "loss": 4.9564,
      "mean_token_accuracy": 0.21851294934749604,
      "num_tokens": 72554551.0,
      "step": 34965
    },
    {
      "entropy": 5.667049264907837,
      "epoch": 3.172169811320755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040089171437240954,
      "loss": 4.9862,
      "mean_token_accuracy": 0.2125397577881813,
      "num_tokens": 72565134.0,
      "step": 34970
    },
    {
      "entropy": 5.659738540649414,
      "epoch": 3.172623367198839,
      "grad_norm": 1.34375,
      "learning_rate": 0.00040086489549712097,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22232003808021544,
      "num_tokens": 72574981.0,
      "step": 34975
    },
    {
      "entropy": 5.632892370223999,
      "epoch": 3.173076923076923,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004008380740189142,
      "loss": 4.9021,
      "mean_token_accuracy": 0.21998363882303237,
      "num_tokens": 72585073.0,
      "step": 34980
    },
    {
      "entropy": 5.664550065994263,
      "epoch": 3.173530478955007,
      "grad_norm": 1.234375,
      "learning_rate": 0.000400811249938344,
      "loss": 4.906,
      "mean_token_accuracy": 0.2194871723651886,
      "num_tokens": 72595719.0,
      "step": 34985
    },
    {
      "entropy": 5.709740066528321,
      "epoch": 3.1739840348330914,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004007844232559652,
      "loss": 4.8901,
      "mean_token_accuracy": 0.21767390370368958,
      "num_tokens": 72605276.0,
      "step": 34990
    },
    {
      "entropy": 5.710125398635864,
      "epoch": 3.1744375907111757,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004007575939723324,
      "loss": 4.93,
      "mean_token_accuracy": 0.2184334248304367,
      "num_tokens": 72614964.0,
      "step": 34995
    },
    {
      "entropy": 5.688166570663452,
      "epoch": 3.17489114658926,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004007307620880004,
      "loss": 4.9089,
      "mean_token_accuracy": 0.223485267162323,
      "num_tokens": 72625461.0,
      "step": 35000
    },
    {
      "entropy": 5.628107118606567,
      "epoch": 3.175344702467344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040070392760352424,
      "loss": 4.9289,
      "mean_token_accuracy": 0.22304186671972276,
      "num_tokens": 72636192.0,
      "step": 35005
    },
    {
      "entropy": 5.678062295913696,
      "epoch": 3.175798258345428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040067709051945895,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21775807291269303,
      "num_tokens": 72648432.0,
      "step": 35010
    },
    {
      "entropy": 5.654417419433594,
      "epoch": 3.1762518142235123,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004006502508363593,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21412013620138168,
      "num_tokens": 72658678.0,
      "step": 35015
    },
    {
      "entropy": 5.761430406570435,
      "epoch": 3.1767053701015966,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040062340855478046,
      "loss": 5.0616,
      "mean_token_accuracy": 0.2071332409977913,
      "num_tokens": 72669191.0,
      "step": 35020
    },
    {
      "entropy": 5.663685083389282,
      "epoch": 3.177158925979681,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040059656367527754,
      "loss": 4.8641,
      "mean_token_accuracy": 0.2226305827498436,
      "num_tokens": 72679327.0,
      "step": 35025
    },
    {
      "entropy": 5.656969404220581,
      "epoch": 3.1776124818577647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004005697161984056,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21541935056447983,
      "num_tokens": 72689673.0,
      "step": 35030
    },
    {
      "entropy": 5.615091753005982,
      "epoch": 3.178066037735849,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040054286612472006,
      "loss": 4.9668,
      "mean_token_accuracy": 0.20662260204553604,
      "num_tokens": 72700606.0,
      "step": 35035
    },
    {
      "entropy": 5.693896341323852,
      "epoch": 3.178519593613933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040051601345477593,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2167987957596779,
      "num_tokens": 72711195.0,
      "step": 35040
    },
    {
      "entropy": 5.660331535339355,
      "epoch": 3.1789731494920175,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040048915818912875,
      "loss": 4.9155,
      "mean_token_accuracy": 0.21899165511131286,
      "num_tokens": 72721160.0,
      "step": 35045
    },
    {
      "entropy": 5.608598423004151,
      "epoch": 3.1794267053701017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040046230032833374,
      "loss": 4.9215,
      "mean_token_accuracy": 0.2172451213002205,
      "num_tokens": 72730867.0,
      "step": 35050
    },
    {
      "entropy": 5.609422206878662,
      "epoch": 3.1798802612481856,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004004354398729464,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21694985330104827,
      "num_tokens": 72740453.0,
      "step": 35055
    },
    {
      "entropy": 5.708440256118775,
      "epoch": 3.18033381712627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004004085768235222,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21265968829393386,
      "num_tokens": 72751395.0,
      "step": 35060
    },
    {
      "entropy": 5.662721157073975,
      "epoch": 3.180787373004354,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040038171118061673,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2200608730316162,
      "num_tokens": 72761300.0,
      "step": 35065
    },
    {
      "entropy": 5.618417596817016,
      "epoch": 3.1812409288824384,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004003548429447855,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22641838043928147,
      "num_tokens": 72772011.0,
      "step": 35070
    },
    {
      "entropy": 5.631150960922241,
      "epoch": 3.1816944847605226,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004003279721165841,
      "loss": 4.904,
      "mean_token_accuracy": 0.21885887533426285,
      "num_tokens": 72782521.0,
      "step": 35075
    },
    {
      "entropy": 5.592150068283081,
      "epoch": 3.182148040638607,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004003010986965683,
      "loss": 4.8058,
      "mean_token_accuracy": 0.2263212114572525,
      "num_tokens": 72793125.0,
      "step": 35080
    },
    {
      "entropy": 5.722436952590942,
      "epoch": 3.1826015965166907,
      "grad_norm": 1.1875,
      "learning_rate": 0.000400274222685294,
      "loss": 4.937,
      "mean_token_accuracy": 0.2178398996591568,
      "num_tokens": 72804541.0,
      "step": 35085
    },
    {
      "entropy": 5.7848927021026615,
      "epoch": 3.183055152394775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040024734408331664,
      "loss": 4.9941,
      "mean_token_accuracy": 0.22293094098567962,
      "num_tokens": 72814447.0,
      "step": 35090
    },
    {
      "entropy": 5.678209590911865,
      "epoch": 3.1835087082728593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040022046289119246,
      "loss": 4.9352,
      "mean_token_accuracy": 0.22801483571529388,
      "num_tokens": 72823545.0,
      "step": 35095
    },
    {
      "entropy": 5.637990236282349,
      "epoch": 3.1839622641509435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040019357910947714,
      "loss": 4.7775,
      "mean_token_accuracy": 0.22704705893993377,
      "num_tokens": 72833803.0,
      "step": 35100
    },
    {
      "entropy": 5.6992815971374515,
      "epoch": 3.1844158200290273,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040016669273872666,
      "loss": 4.9306,
      "mean_token_accuracy": 0.21763599067926406,
      "num_tokens": 72843383.0,
      "step": 35105
    },
    {
      "entropy": 5.7135937213897705,
      "epoch": 3.1848693759071116,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040013980377949716,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21702520847320556,
      "num_tokens": 72853285.0,
      "step": 35110
    },
    {
      "entropy": 5.650406789779663,
      "epoch": 3.185322931785196,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040011291223234456,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21380579024553298,
      "num_tokens": 72863556.0,
      "step": 35115
    },
    {
      "entropy": 5.653562068939209,
      "epoch": 3.18577648766328,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040008601809782507,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2249584048986435,
      "num_tokens": 72873296.0,
      "step": 35120
    },
    {
      "entropy": 5.680861186981201,
      "epoch": 3.1862300435413644,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004000591213764948,
      "loss": 4.9817,
      "mean_token_accuracy": 0.21199764609336852,
      "num_tokens": 72884738.0,
      "step": 35125
    },
    {
      "entropy": 5.721563053131104,
      "epoch": 3.1866835994194487,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040003222206891017,
      "loss": 4.9615,
      "mean_token_accuracy": 0.20880554914474486,
      "num_tokens": 72894573.0,
      "step": 35130
    },
    {
      "entropy": 5.606723403930664,
      "epoch": 3.1871371552975325,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004000053201756272,
      "loss": 4.8674,
      "mean_token_accuracy": 0.222569839656353,
      "num_tokens": 72904976.0,
      "step": 35135
    },
    {
      "entropy": 5.6478386402130125,
      "epoch": 3.1875907111756168,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003999784156972024,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21562621891498565,
      "num_tokens": 72915152.0,
      "step": 35140
    },
    {
      "entropy": 5.707026720046997,
      "epoch": 3.188044267053701,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003999515086341921,
      "loss": 4.9903,
      "mean_token_accuracy": 0.2125471755862236,
      "num_tokens": 72925450.0,
      "step": 35145
    },
    {
      "entropy": 5.616119861602783,
      "epoch": 3.1884978229317853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003999245989871527,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22242685109376908,
      "num_tokens": 72935990.0,
      "step": 35150
    },
    {
      "entropy": 5.599294424057007,
      "epoch": 3.1889513788098696,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039989768675664084,
      "loss": 4.7672,
      "mean_token_accuracy": 0.22641829699277877,
      "num_tokens": 72945662.0,
      "step": 35155
    },
    {
      "entropy": 5.571331214904785,
      "epoch": 3.1894049346879534,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039987077194321305,
      "loss": 4.7467,
      "mean_token_accuracy": 0.22900557816028594,
      "num_tokens": 72955313.0,
      "step": 35160
    },
    {
      "entropy": 5.697544527053833,
      "epoch": 3.1898584905660377,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039984385454742576,
      "loss": 5.0094,
      "mean_token_accuracy": 0.21341952681541443,
      "num_tokens": 72965347.0,
      "step": 35165
    },
    {
      "entropy": 5.619875717163086,
      "epoch": 3.190312046444122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039981693456983574,
      "loss": 4.8194,
      "mean_token_accuracy": 0.2230594888329506,
      "num_tokens": 72976794.0,
      "step": 35170
    },
    {
      "entropy": 5.633504486083984,
      "epoch": 3.190765602322206,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003997900120109998,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2272687330842018,
      "num_tokens": 72986837.0,
      "step": 35175
    },
    {
      "entropy": 5.6312840461730955,
      "epoch": 3.1912191582002905,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003997630868714745,
      "loss": 4.8942,
      "mean_token_accuracy": 0.2179254561662674,
      "num_tokens": 72996819.0,
      "step": 35180
    },
    {
      "entropy": 5.583497524261475,
      "epoch": 3.1916727140783743,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039973615915181687,
      "loss": 4.778,
      "mean_token_accuracy": 0.23091619610786437,
      "num_tokens": 73005963.0,
      "step": 35185
    },
    {
      "entropy": 5.621833086013794,
      "epoch": 3.1921262699564585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039970922885258363,
      "loss": 4.9421,
      "mean_token_accuracy": 0.2179371014237404,
      "num_tokens": 73015912.0,
      "step": 35190
    },
    {
      "entropy": 5.589060640335083,
      "epoch": 3.192579825834543,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003996822959743318,
      "loss": 4.8013,
      "mean_token_accuracy": 0.2317497044801712,
      "num_tokens": 73026528.0,
      "step": 35195
    },
    {
      "entropy": 5.662904214859009,
      "epoch": 3.193033381712627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003996553605176183,
      "loss": 4.9413,
      "mean_token_accuracy": 0.21369361877441406,
      "num_tokens": 73036255.0,
      "step": 35200
    },
    {
      "entropy": 5.678701686859131,
      "epoch": 3.1934869375907113,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003996284224830002,
      "loss": 5.004,
      "mean_token_accuracy": 0.2200263813138008,
      "num_tokens": 73046686.0,
      "step": 35205
    },
    {
      "entropy": 5.632307004928589,
      "epoch": 3.193940493468795,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039960148187103453,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22521648108959197,
      "num_tokens": 73056070.0,
      "step": 35210
    },
    {
      "entropy": 5.691978693008423,
      "epoch": 3.1943940493468794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039957453868227845,
      "loss": 4.9399,
      "mean_token_accuracy": 0.22403661161661148,
      "num_tokens": 73066288.0,
      "step": 35215
    },
    {
      "entropy": 5.674900102615356,
      "epoch": 3.1948476052249637,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039954759291728915,
      "loss": 4.8989,
      "mean_token_accuracy": 0.220204396545887,
      "num_tokens": 73076297.0,
      "step": 35220
    },
    {
      "entropy": 5.695418739318848,
      "epoch": 3.195301161103048,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039952064457662395,
      "loss": 4.988,
      "mean_token_accuracy": 0.2139391779899597,
      "num_tokens": 73086249.0,
      "step": 35225
    },
    {
      "entropy": 5.624527215957642,
      "epoch": 3.1957547169811322,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039949369366084,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2225458174943924,
      "num_tokens": 73096358.0,
      "step": 35230
    },
    {
      "entropy": 5.59328293800354,
      "epoch": 3.196208272859216,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039946674017049484,
      "loss": 4.8305,
      "mean_token_accuracy": 0.21974304616451262,
      "num_tokens": 73106891.0,
      "step": 35235
    },
    {
      "entropy": 5.574156093597412,
      "epoch": 3.1966618287373003,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039943978410614565,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22132344990968705,
      "num_tokens": 73117106.0,
      "step": 35240
    },
    {
      "entropy": 5.6115944385528564,
      "epoch": 3.1971153846153846,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039941282546835017,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22445307970046996,
      "num_tokens": 73127112.0,
      "step": 35245
    },
    {
      "entropy": 5.626039981842041,
      "epoch": 3.197568940493469,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039938586425766564,
      "loss": 4.8915,
      "mean_token_accuracy": 0.21814250200986862,
      "num_tokens": 73137751.0,
      "step": 35250
    },
    {
      "entropy": 5.682536363601685,
      "epoch": 3.198022496371553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003993589004746498,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22314903885126114,
      "num_tokens": 73149303.0,
      "step": 35255
    },
    {
      "entropy": 5.6274830341339115,
      "epoch": 3.198476052249637,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039933193411986024,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2225891351699829,
      "num_tokens": 73160025.0,
      "step": 35260
    },
    {
      "entropy": 5.621498012542725,
      "epoch": 3.198929608127721,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039930496519385454,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22356652915477754,
      "num_tokens": 73170721.0,
      "step": 35265
    },
    {
      "entropy": 5.667210960388184,
      "epoch": 3.1993831640058055,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039927799369719047,
      "loss": 4.9262,
      "mean_token_accuracy": 0.2160688117146492,
      "num_tokens": 73180902.0,
      "step": 35270
    },
    {
      "entropy": 5.561063957214356,
      "epoch": 3.1998367198838897,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003992510196304258,
      "loss": 4.7457,
      "mean_token_accuracy": 0.22781597226858138,
      "num_tokens": 73191259.0,
      "step": 35275
    },
    {
      "entropy": 5.631057262420654,
      "epoch": 3.200290275761974,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039922404299411846,
      "loss": 4.888,
      "mean_token_accuracy": 0.22873584181070328,
      "num_tokens": 73201654.0,
      "step": 35280
    },
    {
      "entropy": 5.657769107818604,
      "epoch": 3.2007438316400583,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003991970637888262,
      "loss": 4.9135,
      "mean_token_accuracy": 0.21260999292135238,
      "num_tokens": 73212658.0,
      "step": 35285
    },
    {
      "entropy": 5.685163831710815,
      "epoch": 3.201197387518142,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000399170082015107,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2187635138630867,
      "num_tokens": 73222883.0,
      "step": 35290
    },
    {
      "entropy": 5.625154256820679,
      "epoch": 3.2016509433962264,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000399143097673519,
      "loss": 4.8699,
      "mean_token_accuracy": 0.22039635926485063,
      "num_tokens": 73233731.0,
      "step": 35295
    },
    {
      "entropy": 5.631882619857788,
      "epoch": 3.2021044992743106,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039911611076461995,
      "loss": 4.8039,
      "mean_token_accuracy": 0.23064013719558715,
      "num_tokens": 73243794.0,
      "step": 35300
    },
    {
      "entropy": 5.6359117984771725,
      "epoch": 3.202558055152395,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039908912128896817,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21604838818311692,
      "num_tokens": 73254708.0,
      "step": 35305
    },
    {
      "entropy": 5.587806177139282,
      "epoch": 3.203011611030479,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003990621292471217,
      "loss": 4.8739,
      "mean_token_accuracy": 0.21345995664596557,
      "num_tokens": 73265311.0,
      "step": 35310
    },
    {
      "entropy": 5.754697704315186,
      "epoch": 3.203465166908563,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039903513463963874,
      "loss": 5.0488,
      "mean_token_accuracy": 0.21618608981370926,
      "num_tokens": 73275886.0,
      "step": 35315
    },
    {
      "entropy": 5.69934573173523,
      "epoch": 3.2039187227866472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003990081374670777,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22190704494714736,
      "num_tokens": 73285903.0,
      "step": 35320
    },
    {
      "entropy": 5.620176124572754,
      "epoch": 3.2043722786647315,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039898113772999665,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22238509654998778,
      "num_tokens": 73296571.0,
      "step": 35325
    },
    {
      "entropy": 5.652886486053466,
      "epoch": 3.2048258345428158,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039895413542895427,
      "loss": 4.933,
      "mean_token_accuracy": 0.21276887953281404,
      "num_tokens": 73307105.0,
      "step": 35330
    },
    {
      "entropy": 5.6077470779418945,
      "epoch": 3.2052793904209,
      "grad_norm": 1.375,
      "learning_rate": 0.0003989271305645086,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22340649068355561,
      "num_tokens": 73316664.0,
      "step": 35335
    },
    {
      "entropy": 5.631734037399292,
      "epoch": 3.205732946298984,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039890012313721835,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22388483732938766,
      "num_tokens": 73327438.0,
      "step": 35340
    },
    {
      "entropy": 5.661161088943482,
      "epoch": 3.206186502177068,
      "grad_norm": 1.28125,
      "learning_rate": 0.000398873113147642,
      "loss": 4.8982,
      "mean_token_accuracy": 0.22399021983146666,
      "num_tokens": 73337635.0,
      "step": 35345
    },
    {
      "entropy": 5.645139980316162,
      "epoch": 3.2066400580551524,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039884610059633807,
      "loss": 4.8883,
      "mean_token_accuracy": 0.22374069392681123,
      "num_tokens": 73347032.0,
      "step": 35350
    },
    {
      "entropy": 5.71831374168396,
      "epoch": 3.2070936139332367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003988190854838652,
      "loss": 4.973,
      "mean_token_accuracy": 0.2151808947324753,
      "num_tokens": 73357780.0,
      "step": 35355
    },
    {
      "entropy": 5.694021129608155,
      "epoch": 3.207547169811321,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039879206781078214,
      "loss": 4.9279,
      "mean_token_accuracy": 0.22107236832380295,
      "num_tokens": 73367831.0,
      "step": 35360
    },
    {
      "entropy": 5.647748231887817,
      "epoch": 3.2080007256894048,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003987650475776476,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22143419831991196,
      "num_tokens": 73378588.0,
      "step": 35365
    },
    {
      "entropy": 5.596695852279663,
      "epoch": 3.208454281567489,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039873802478502027,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22246301770210267,
      "num_tokens": 73388414.0,
      "step": 35370
    },
    {
      "entropy": 5.642647409439087,
      "epoch": 3.2089078374455733,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003987109994334591,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22339046895503997,
      "num_tokens": 73398891.0,
      "step": 35375
    },
    {
      "entropy": 5.656751108169556,
      "epoch": 3.2093613933236576,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003986839715235229,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2181890681385994,
      "num_tokens": 73409605.0,
      "step": 35380
    },
    {
      "entropy": 5.724397087097168,
      "epoch": 3.209814949201742,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039865694105577063,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2112596496939659,
      "num_tokens": 73419103.0,
      "step": 35385
    },
    {
      "entropy": 5.590491390228271,
      "epoch": 3.2102685050798256,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003986299080307614,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22329264581203462,
      "num_tokens": 73429134.0,
      "step": 35390
    },
    {
      "entropy": 5.632700490951538,
      "epoch": 3.21072206095791,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003986028724490541,
      "loss": 4.9815,
      "mean_token_accuracy": 0.21087060123682022,
      "num_tokens": 73440382.0,
      "step": 35395
    },
    {
      "entropy": 5.675074291229248,
      "epoch": 3.211175616835994,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003985758343112079,
      "loss": 4.8916,
      "mean_token_accuracy": 0.22479851245880128,
      "num_tokens": 73451300.0,
      "step": 35400
    },
    {
      "entropy": 5.637781620025635,
      "epoch": 3.2116291727140784,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039854879361778193,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2203120082616806,
      "num_tokens": 73461339.0,
      "step": 35405
    },
    {
      "entropy": 5.660482549667359,
      "epoch": 3.2120827285921627,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039852175036933545,
      "loss": 4.903,
      "mean_token_accuracy": 0.22388493567705153,
      "num_tokens": 73470961.0,
      "step": 35410
    },
    {
      "entropy": 5.7217240810394285,
      "epoch": 3.2125362844702465,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039849470456642777,
      "loss": 4.9718,
      "mean_token_accuracy": 0.21162750869989394,
      "num_tokens": 73481492.0,
      "step": 35415
    },
    {
      "entropy": 5.624343061447144,
      "epoch": 3.212989840348331,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039846765620961806,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22535259127616883,
      "num_tokens": 73492351.0,
      "step": 35420
    },
    {
      "entropy": 5.662600803375244,
      "epoch": 3.213443396226415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003984406052994658,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2222810670733452,
      "num_tokens": 73503265.0,
      "step": 35425
    },
    {
      "entropy": 5.685121536254883,
      "epoch": 3.2138969521044993,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003984135518365303,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21596424877643586,
      "num_tokens": 73513804.0,
      "step": 35430
    },
    {
      "entropy": 5.676623773574829,
      "epoch": 3.2143505079825836,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039838649582137113,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21769585758447646,
      "num_tokens": 73523453.0,
      "step": 35435
    },
    {
      "entropy": 5.723464632034302,
      "epoch": 3.214804063860668,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039835943725454777,
      "loss": 5.0164,
      "mean_token_accuracy": 0.20870174169540406,
      "num_tokens": 73533931.0,
      "step": 35440
    },
    {
      "entropy": 5.733582592010498,
      "epoch": 3.2152576197387517,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039833237613661984,
      "loss": 4.957,
      "mean_token_accuracy": 0.21903835237026215,
      "num_tokens": 73544562.0,
      "step": 35445
    },
    {
      "entropy": 5.681792545318603,
      "epoch": 3.215711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.0003983053124681469,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22375929057598115,
      "num_tokens": 73555149.0,
      "step": 35450
    },
    {
      "entropy": 5.574533557891845,
      "epoch": 3.21616473149492,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039827824624968873,
      "loss": 4.8218,
      "mean_token_accuracy": 0.21809957474470137,
      "num_tokens": 73565131.0,
      "step": 35455
    },
    {
      "entropy": 5.6377702236175535,
      "epoch": 3.2166182873730045,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039825117748180493,
      "loss": 4.864,
      "mean_token_accuracy": 0.22489175647497178,
      "num_tokens": 73574185.0,
      "step": 35460
    },
    {
      "entropy": 5.718073034286499,
      "epoch": 3.2170718432510883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039822410616505545,
      "loss": 4.9721,
      "mean_token_accuracy": 0.21349977999925612,
      "num_tokens": 73585733.0,
      "step": 35465
    },
    {
      "entropy": 5.744461154937744,
      "epoch": 3.2175253991291726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039819703230000006,
      "loss": 4.9812,
      "mean_token_accuracy": 0.211832857131958,
      "num_tokens": 73597077.0,
      "step": 35470
    },
    {
      "entropy": 5.622053432464599,
      "epoch": 3.217978955007257,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003981699558871986,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2208957478404045,
      "num_tokens": 73607885.0,
      "step": 35475
    },
    {
      "entropy": 5.672189617156983,
      "epoch": 3.218432510885341,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000398142876927211,
      "loss": 4.8894,
      "mean_token_accuracy": 0.2207530379295349,
      "num_tokens": 73618769.0,
      "step": 35480
    },
    {
      "entropy": 5.704537200927734,
      "epoch": 3.2188860667634254,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039811579542059744,
      "loss": 5.0076,
      "mean_token_accuracy": 0.2125947043299675,
      "num_tokens": 73629345.0,
      "step": 35485
    },
    {
      "entropy": 5.676772594451904,
      "epoch": 3.2193396226415096,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039808871136791786,
      "loss": 4.8672,
      "mean_token_accuracy": 0.21944503635168075,
      "num_tokens": 73639698.0,
      "step": 35490
    },
    {
      "entropy": 5.632308864593506,
      "epoch": 3.2197931785195935,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039806162476973227,
      "loss": 4.904,
      "mean_token_accuracy": 0.2181965231895447,
      "num_tokens": 73650954.0,
      "step": 35495
    },
    {
      "entropy": 5.694629335403443,
      "epoch": 3.2202467343976777,
      "grad_norm": 1.140625,
      "learning_rate": 0.000398034535626601,
      "loss": 4.9874,
      "mean_token_accuracy": 0.21623368859291076,
      "num_tokens": 73661195.0,
      "step": 35500
    },
    {
      "entropy": 5.7846475601196286,
      "epoch": 3.220700290275762,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039800744393908414,
      "loss": 5.0244,
      "mean_token_accuracy": 0.21160681545734406,
      "num_tokens": 73670512.0,
      "step": 35505
    },
    {
      "entropy": 5.714835119247437,
      "epoch": 3.2211538461538463,
      "grad_norm": 1.1875,
      "learning_rate": 0.000397980349707742,
      "loss": 4.9399,
      "mean_token_accuracy": 0.21603664010763168,
      "num_tokens": 73681927.0,
      "step": 35510
    },
    {
      "entropy": 5.699832582473755,
      "epoch": 3.2216074020319305,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039795325293313493,
      "loss": 4.9556,
      "mean_token_accuracy": 0.21593974381685258,
      "num_tokens": 73692025.0,
      "step": 35515
    },
    {
      "entropy": 5.6747016429901125,
      "epoch": 3.2220609579100143,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039792615361582324,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2130510166287422,
      "num_tokens": 73702278.0,
      "step": 35520
    },
    {
      "entropy": 5.757747602462769,
      "epoch": 3.2225145137880986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039789905175636726,
      "loss": 4.9917,
      "mean_token_accuracy": 0.2105499416589737,
      "num_tokens": 73713511.0,
      "step": 35525
    },
    {
      "entropy": 5.593490314483643,
      "epoch": 3.222968069666183,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039787194735532765,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22481331676244737,
      "num_tokens": 73723683.0,
      "step": 35530
    },
    {
      "entropy": 5.6559240341186525,
      "epoch": 3.223421625544267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003978448404132649,
      "loss": 4.8716,
      "mean_token_accuracy": 0.22651073336601257,
      "num_tokens": 73734191.0,
      "step": 35535
    },
    {
      "entropy": 5.70579423904419,
      "epoch": 3.2238751814223514,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039781773093073956,
      "loss": 4.9275,
      "mean_token_accuracy": 0.2158797040581703,
      "num_tokens": 73743747.0,
      "step": 35540
    },
    {
      "entropy": 5.631256055831909,
      "epoch": 3.2243287373004352,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003977906189083122,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22213522344827652,
      "num_tokens": 73754466.0,
      "step": 35545
    },
    {
      "entropy": 5.644285202026367,
      "epoch": 3.2247822931785195,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003977635043465435,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21791046112775803,
      "num_tokens": 73765948.0,
      "step": 35550
    },
    {
      "entropy": 5.6123292446136475,
      "epoch": 3.2252358490566038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003977363872459943,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22073015868663787,
      "num_tokens": 73776003.0,
      "step": 35555
    },
    {
      "entropy": 5.73852767944336,
      "epoch": 3.225689404934688,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003977092676072254,
      "loss": 5.0618,
      "mean_token_accuracy": 0.21750193685293198,
      "num_tokens": 73787821.0,
      "step": 35560
    },
    {
      "entropy": 5.670309734344483,
      "epoch": 3.2261429608127723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039768214543079746,
      "loss": 4.9086,
      "mean_token_accuracy": 0.2167245179414749,
      "num_tokens": 73799207.0,
      "step": 35565
    },
    {
      "entropy": 5.69982647895813,
      "epoch": 3.226596516690856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003976550207172716,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22242486774921416,
      "num_tokens": 73809987.0,
      "step": 35570
    },
    {
      "entropy": 5.612677431106567,
      "epoch": 3.2270500725689404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039762789346720854,
      "loss": 4.8668,
      "mean_token_accuracy": 0.21594077795743943,
      "num_tokens": 73819075.0,
      "step": 35575
    },
    {
      "entropy": 5.645557165145874,
      "epoch": 3.2275036284470247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003976007636811694,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22525837570428847,
      "num_tokens": 73829831.0,
      "step": 35580
    },
    {
      "entropy": 5.682927513122559,
      "epoch": 3.227957184325109,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003975736313597153,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2201295092701912,
      "num_tokens": 73839826.0,
      "step": 35585
    },
    {
      "entropy": 5.662317991256714,
      "epoch": 3.228410740203193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003975464965034072,
      "loss": 4.8305,
      "mean_token_accuracy": 0.22775400429964066,
      "num_tokens": 73850006.0,
      "step": 35590
    },
    {
      "entropy": 5.660862064361572,
      "epoch": 3.228864296081277,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003975193591128064,
      "loss": 4.8895,
      "mean_token_accuracy": 0.21874599754810334,
      "num_tokens": 73860309.0,
      "step": 35595
    },
    {
      "entropy": 5.656627607345581,
      "epoch": 3.2293178519593613,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039749221918847406,
      "loss": 4.9826,
      "mean_token_accuracy": 0.21137904226779938,
      "num_tokens": 73871613.0,
      "step": 35600
    },
    {
      "entropy": 5.569852590560913,
      "epoch": 3.2297714078374455,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003974650767309713,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22539979815483094,
      "num_tokens": 73882303.0,
      "step": 35605
    },
    {
      "entropy": 5.716156101226806,
      "epoch": 3.23022496371553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039743793174085965,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21033623218536376,
      "num_tokens": 73893548.0,
      "step": 35610
    },
    {
      "entropy": 5.696897554397583,
      "epoch": 3.230678519593614,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039741078421870024,
      "loss": 4.903,
      "mean_token_accuracy": 0.2179935619235039,
      "num_tokens": 73903934.0,
      "step": 35615
    },
    {
      "entropy": 5.708308601379395,
      "epoch": 3.231132075471698,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039738363416505474,
      "loss": 5.0172,
      "mean_token_accuracy": 0.2060520425438881,
      "num_tokens": 73913813.0,
      "step": 35620
    },
    {
      "entropy": 5.694481945037841,
      "epoch": 3.231585631349782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039735648158048444,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21358578354120256,
      "num_tokens": 73924339.0,
      "step": 35625
    },
    {
      "entropy": 5.679987668991089,
      "epoch": 3.2320391872278664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003973293264655509,
      "loss": 4.903,
      "mean_token_accuracy": 0.21622582972049714,
      "num_tokens": 73935318.0,
      "step": 35630
    },
    {
      "entropy": 5.6549663066864015,
      "epoch": 3.2324927431059507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003973021688208158,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22620645314455032,
      "num_tokens": 73945547.0,
      "step": 35635
    },
    {
      "entropy": 5.6736119270324705,
      "epoch": 3.232946298984035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003972750086468406,
      "loss": 4.9055,
      "mean_token_accuracy": 0.21593448966741563,
      "num_tokens": 73956195.0,
      "step": 35640
    },
    {
      "entropy": 5.671016216278076,
      "epoch": 3.2333998548621192,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039724784594418714,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22840246856212615,
      "num_tokens": 73966843.0,
      "step": 35645
    },
    {
      "entropy": 5.690511226654053,
      "epoch": 3.233853410740203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000397220680713417,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2211485892534256,
      "num_tokens": 73978516.0,
      "step": 35650
    },
    {
      "entropy": 5.601669788360596,
      "epoch": 3.2343069666182873,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039719351295509203,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22811519503593444,
      "num_tokens": 73988376.0,
      "step": 35655
    },
    {
      "entropy": 5.64025764465332,
      "epoch": 3.2347605224963716,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039716634266977413,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22719595283269883,
      "num_tokens": 73998232.0,
      "step": 35660
    },
    {
      "entropy": 5.624588680267334,
      "epoch": 3.235214078374456,
      "grad_norm": 1.125,
      "learning_rate": 0.0003971391698580252,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2200999528169632,
      "num_tokens": 74008726.0,
      "step": 35665
    },
    {
      "entropy": 5.649100923538208,
      "epoch": 3.23566763425254,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039711199452040696,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22007654458284379,
      "num_tokens": 74018746.0,
      "step": 35670
    },
    {
      "entropy": 5.64105076789856,
      "epoch": 3.236121190130624,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039708481665748157,
      "loss": 4.882,
      "mean_token_accuracy": 0.22190751731395722,
      "num_tokens": 74028615.0,
      "step": 35675
    },
    {
      "entropy": 5.659426021575928,
      "epoch": 3.236574746008708,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003970576362698111,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21838831007480622,
      "num_tokens": 74039219.0,
      "step": 35680
    },
    {
      "entropy": 5.683657646179199,
      "epoch": 3.2370283018867925,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039703045335795766,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22249771207571029,
      "num_tokens": 74049385.0,
      "step": 35685
    },
    {
      "entropy": 5.750092077255249,
      "epoch": 3.2374818577648767,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003970032679224833,
      "loss": 5.0358,
      "mean_token_accuracy": 0.20470202714204788,
      "num_tokens": 74059437.0,
      "step": 35690
    },
    {
      "entropy": 5.698085641860962,
      "epoch": 3.237935413642961,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039697607996395023,
      "loss": 4.9706,
      "mean_token_accuracy": 0.20968195796012878,
      "num_tokens": 74068834.0,
      "step": 35695
    },
    {
      "entropy": 5.682476806640625,
      "epoch": 3.238388969521045,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003969488894829208,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21189266443252563,
      "num_tokens": 74079757.0,
      "step": 35700
    },
    {
      "entropy": 5.7008191585540775,
      "epoch": 3.238842525399129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003969216964799573,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22583719789981843,
      "num_tokens": 74091092.0,
      "step": 35705
    },
    {
      "entropy": 5.627274990081787,
      "epoch": 3.2392960812772134,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003968945009556219,
      "loss": 4.8738,
      "mean_token_accuracy": 0.22194982916116715,
      "num_tokens": 74102272.0,
      "step": 35710
    },
    {
      "entropy": 5.631210899353027,
      "epoch": 3.2397496371552976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003968673029104773,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22760780900716782,
      "num_tokens": 74113101.0,
      "step": 35715
    },
    {
      "entropy": 5.715741777420044,
      "epoch": 3.240203193033382,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003968401023450858,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22306034713983536,
      "num_tokens": 74123180.0,
      "step": 35720
    },
    {
      "entropy": 5.622351789474488,
      "epoch": 3.2406567489114657,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039681289926000985,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22457147985696793,
      "num_tokens": 74133675.0,
      "step": 35725
    },
    {
      "entropy": 5.613019704818726,
      "epoch": 3.24111030478955,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003967856936558122,
      "loss": 4.8284,
      "mean_token_accuracy": 0.22165256142616271,
      "num_tokens": 74143472.0,
      "step": 35730
    },
    {
      "entropy": 5.6705546379089355,
      "epoch": 3.2415638606676342,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039675848553305526,
      "loss": 4.897,
      "mean_token_accuracy": 0.21510822623968123,
      "num_tokens": 74154682.0,
      "step": 35735
    },
    {
      "entropy": 5.672407102584839,
      "epoch": 3.2420174165457185,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003967312748923019,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22182407081127167,
      "num_tokens": 74164745.0,
      "step": 35740
    },
    {
      "entropy": 5.628395843505859,
      "epoch": 3.2424709724238028,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039670406173411473,
      "loss": 4.8322,
      "mean_token_accuracy": 0.22223233580589294,
      "num_tokens": 74175388.0,
      "step": 35745
    },
    {
      "entropy": 5.654885101318359,
      "epoch": 3.2429245283018866,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003966768460590566,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22196344435214996,
      "num_tokens": 74185929.0,
      "step": 35750
    },
    {
      "entropy": 5.684422731399536,
      "epoch": 3.243378084179971,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039664962786769013,
      "loss": 4.868,
      "mean_token_accuracy": 0.21964912116527557,
      "num_tokens": 74196165.0,
      "step": 35755
    },
    {
      "entropy": 5.659111022949219,
      "epoch": 3.243831640058055,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039662240716057854,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2240724116563797,
      "num_tokens": 74206843.0,
      "step": 35760
    },
    {
      "entropy": 5.672952175140381,
      "epoch": 3.2442851959361394,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003965951839382845,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22548126429319382,
      "num_tokens": 74216854.0,
      "step": 35765
    },
    {
      "entropy": 5.649486112594604,
      "epoch": 3.2447387518142237,
      "grad_norm": 1.140625,
      "learning_rate": 0.000396567958201371,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22194210588932037,
      "num_tokens": 74226158.0,
      "step": 35770
    },
    {
      "entropy": 5.652727222442627,
      "epoch": 3.2451923076923075,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003965407299504013,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2240336209535599,
      "num_tokens": 74236860.0,
      "step": 35775
    },
    {
      "entropy": 5.66900863647461,
      "epoch": 3.2456458635703918,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003965134991859382,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2252551555633545,
      "num_tokens": 74247706.0,
      "step": 35780
    },
    {
      "entropy": 5.71504774093628,
      "epoch": 3.246099419448476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000396486265908545,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2225152939558029,
      "num_tokens": 74259479.0,
      "step": 35785
    },
    {
      "entropy": 5.641512775421143,
      "epoch": 3.2465529753265603,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003964590301187849,
      "loss": 4.9385,
      "mean_token_accuracy": 0.219942107796669,
      "num_tokens": 74270371.0,
      "step": 35790
    },
    {
      "entropy": 5.636530351638794,
      "epoch": 3.2470065312046446,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003964317918172211,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23139665871858597,
      "num_tokens": 74279637.0,
      "step": 35795
    },
    {
      "entropy": 5.684038209915161,
      "epoch": 3.247460087082729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039640455100441684,
      "loss": 4.9898,
      "mean_token_accuracy": 0.21270882487297058,
      "num_tokens": 74289889.0,
      "step": 35800
    },
    {
      "entropy": 5.686616230010986,
      "epoch": 3.2479136429608126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039637730768093564,
      "loss": 4.9142,
      "mean_token_accuracy": 0.22774902135133743,
      "num_tokens": 74300949.0,
      "step": 35805
    },
    {
      "entropy": 5.6544688701629635,
      "epoch": 3.248367198838897,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003963500618473407,
      "loss": 4.8665,
      "mean_token_accuracy": 0.21937722563743592,
      "num_tokens": 74312077.0,
      "step": 35810
    },
    {
      "entropy": 5.626886987686158,
      "epoch": 3.248820754716981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003963228135041956,
      "loss": 4.9603,
      "mean_token_accuracy": 0.21689776629209517,
      "num_tokens": 74322610.0,
      "step": 35815
    },
    {
      "entropy": 5.587293243408203,
      "epoch": 3.2492743105950654,
      "grad_norm": 1.25,
      "learning_rate": 0.00039629556265206373,
      "loss": 4.814,
      "mean_token_accuracy": 0.22133294939994813,
      "num_tokens": 74331970.0,
      "step": 35820
    },
    {
      "entropy": 5.6503407001495365,
      "epoch": 3.2497278664731493,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003962683092915088,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22468665540218352,
      "num_tokens": 74342812.0,
      "step": 35825
    },
    {
      "entropy": 5.650708723068237,
      "epoch": 3.2501814223512335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003962410534230943,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22577900290489197,
      "num_tokens": 74353786.0,
      "step": 35830
    },
    {
      "entropy": 5.697111177444458,
      "epoch": 3.250634978229318,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039621379504738385,
      "loss": 4.9407,
      "mean_token_accuracy": 0.22278930842876435,
      "num_tokens": 74363732.0,
      "step": 35835
    },
    {
      "entropy": 5.664627504348755,
      "epoch": 3.251088534107402,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039618653416494137,
      "loss": 4.9262,
      "mean_token_accuracy": 0.21945639103651046,
      "num_tokens": 74374077.0,
      "step": 35840
    },
    {
      "entropy": 5.6513121128082275,
      "epoch": 3.2515420899854863,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039615927077633043,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2153422564268112,
      "num_tokens": 74384758.0,
      "step": 35845
    },
    {
      "entropy": 5.7302481174469,
      "epoch": 3.2519956458635706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039613200488211483,
      "loss": 4.9476,
      "mean_token_accuracy": 0.22317588627338408,
      "num_tokens": 74395811.0,
      "step": 35850
    },
    {
      "entropy": 5.692087268829345,
      "epoch": 3.2524492017416544,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039610473648285857,
      "loss": 4.9323,
      "mean_token_accuracy": 0.2221641570329666,
      "num_tokens": 74407830.0,
      "step": 35855
    },
    {
      "entropy": 5.628623533248901,
      "epoch": 3.2529027576197387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003960774655791255,
      "loss": 4.8517,
      "mean_token_accuracy": 0.21962568610906602,
      "num_tokens": 74417565.0,
      "step": 35860
    },
    {
      "entropy": 5.7060933113098145,
      "epoch": 3.253356313497823,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039605019217147947,
      "loss": 4.9123,
      "mean_token_accuracy": 0.2220562681555748,
      "num_tokens": 74428015.0,
      "step": 35865
    },
    {
      "entropy": 5.621052885055542,
      "epoch": 3.253809869375907,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039602291626048475,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2246765673160553,
      "num_tokens": 74439523.0,
      "step": 35870
    },
    {
      "entropy": 5.720399379730225,
      "epoch": 3.2542634252539915,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039599563784670526,
      "loss": 4.9912,
      "mean_token_accuracy": 0.2063504248857498,
      "num_tokens": 74449897.0,
      "step": 35875
    },
    {
      "entropy": 5.658713722229004,
      "epoch": 3.2547169811320753,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003959683569307051,
      "loss": 4.8205,
      "mean_token_accuracy": 0.2228706732392311,
      "num_tokens": 74459888.0,
      "step": 35880
    },
    {
      "entropy": 5.618401145935058,
      "epoch": 3.2551705370101596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003959410735130485,
      "loss": 4.8111,
      "mean_token_accuracy": 0.21698644906282424,
      "num_tokens": 74470894.0,
      "step": 35885
    },
    {
      "entropy": 5.618698453903198,
      "epoch": 3.255624092888244,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039591378759429974,
      "loss": 4.8086,
      "mean_token_accuracy": 0.22761537730693818,
      "num_tokens": 74481306.0,
      "step": 35890
    },
    {
      "entropy": 5.675455522537232,
      "epoch": 3.256077648766328,
      "grad_norm": 1.203125,
      "learning_rate": 0.000395886499175023,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22331613451242446,
      "num_tokens": 74491048.0,
      "step": 35895
    },
    {
      "entropy": 5.608415317535401,
      "epoch": 3.2565312046444124,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039585920825578266,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2218456208705902,
      "num_tokens": 74501656.0,
      "step": 35900
    },
    {
      "entropy": 5.720268487930298,
      "epoch": 3.256984760522496,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003958319148371431,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21802135705947875,
      "num_tokens": 74513280.0,
      "step": 35905
    },
    {
      "entropy": 5.638188219070434,
      "epoch": 3.2574383164005805,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003958046189196687,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21897765696048738,
      "num_tokens": 74522898.0,
      "step": 35910
    },
    {
      "entropy": 5.750848007202149,
      "epoch": 3.2578918722786647,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039577732050392395,
      "loss": 4.9411,
      "mean_token_accuracy": 0.21608620136976242,
      "num_tokens": 74532551.0,
      "step": 35915
    },
    {
      "entropy": 5.70573058128357,
      "epoch": 3.258345428156749,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003957500195904735,
      "loss": 4.912,
      "mean_token_accuracy": 0.22299078106880188,
      "num_tokens": 74542285.0,
      "step": 35920
    },
    {
      "entropy": 5.695182847976684,
      "epoch": 3.2587989840348333,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000395722716179882,
      "loss": 4.9514,
      "mean_token_accuracy": 0.22052583694458008,
      "num_tokens": 74551982.0,
      "step": 35925
    },
    {
      "entropy": 5.658247089385986,
      "epoch": 3.259252539912917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003956954102727138,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22178051769733428,
      "num_tokens": 74562190.0,
      "step": 35930
    },
    {
      "entropy": 5.620672750473022,
      "epoch": 3.2597060957910013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039566810186953376,
      "loss": 4.8434,
      "mean_token_accuracy": 0.223619481921196,
      "num_tokens": 74572673.0,
      "step": 35935
    },
    {
      "entropy": 5.597012567520141,
      "epoch": 3.2601596516690856,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003956407909709068,
      "loss": 4.8579,
      "mean_token_accuracy": 0.22504730075597762,
      "num_tokens": 74582686.0,
      "step": 35940
    },
    {
      "entropy": 5.690758419036865,
      "epoch": 3.26061320754717,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039561347757739734,
      "loss": 4.9767,
      "mean_token_accuracy": 0.21953113824129106,
      "num_tokens": 74593658.0,
      "step": 35945
    },
    {
      "entropy": 5.727089691162109,
      "epoch": 3.261066763425254,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039558616168957044,
      "loss": 4.9844,
      "mean_token_accuracy": 0.21975499242544175,
      "num_tokens": 74603585.0,
      "step": 35950
    },
    {
      "entropy": 5.661065578460693,
      "epoch": 3.2615203193033384,
      "grad_norm": 1.25,
      "learning_rate": 0.0003955588433079911,
      "loss": 4.866,
      "mean_token_accuracy": 0.2182395949959755,
      "num_tokens": 74612806.0,
      "step": 35955
    },
    {
      "entropy": 5.673149871826172,
      "epoch": 3.2619738751814222,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00039553152243322414,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22659615874290467,
      "num_tokens": 74623252.0,
      "step": 35960
    },
    {
      "entropy": 5.6325641632080075,
      "epoch": 3.2624274310595065,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003955041990658346,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22452736049890518,
      "num_tokens": 74633555.0,
      "step": 35965
    },
    {
      "entropy": 5.642836952209473,
      "epoch": 3.2628809869375908,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003954768732063874,
      "loss": 4.9071,
      "mean_token_accuracy": 0.2220741331577301,
      "num_tokens": 74643322.0,
      "step": 35970
    },
    {
      "entropy": 5.63678879737854,
      "epoch": 3.263334542815675,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003954495448554478,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22943014949560164,
      "num_tokens": 74653461.0,
      "step": 35975
    },
    {
      "entropy": 5.674677038192749,
      "epoch": 3.263788098693759,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039542221401358104,
      "loss": 4.92,
      "mean_token_accuracy": 0.21900758743286133,
      "num_tokens": 74663470.0,
      "step": 35980
    },
    {
      "entropy": 5.620491409301758,
      "epoch": 3.264241654571843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000395394880681352,
      "loss": 4.849,
      "mean_token_accuracy": 0.2268689602613449,
      "num_tokens": 74674074.0,
      "step": 35985
    },
    {
      "entropy": 5.712144565582276,
      "epoch": 3.2646952104499274,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039536754485932626,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21093817949295043,
      "num_tokens": 74684968.0,
      "step": 35990
    },
    {
      "entropy": 5.602517080307007,
      "epoch": 3.2651487663280117,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039534020654806906,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23351232409477235,
      "num_tokens": 74694776.0,
      "step": 35995
    },
    {
      "entropy": 5.625758409500122,
      "epoch": 3.265602322206096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039531286574814563,
      "loss": 4.833,
      "mean_token_accuracy": 0.2199109524488449,
      "num_tokens": 74705506.0,
      "step": 36000
    },
    {
      "epoch": 3.265602322206096,
      "eval_entropy": 5.361749818314254,
      "eval_loss": 4.977652549743652,
      "eval_mean_token_accuracy": 0.22397300535096573,
      "eval_num_tokens": 74705506.0,
      "eval_runtime": 20.5625,
      "eval_samples_per_second": 1719.632,
      "eval_steps_per_second": 214.954,
      "step": 36000
    },
    {
      "entropy": 5.610892629623413,
      "epoch": 3.26605587808418,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003952855224601215,
      "loss": 4.8786,
      "mean_token_accuracy": 0.2217954710125923,
      "num_tokens": 74715677.0,
      "step": 36005
    },
    {
      "entropy": 5.691538667678833,
      "epoch": 3.266509433962264,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003952581766845622,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2259572982788086,
      "num_tokens": 74725296.0,
      "step": 36010
    },
    {
      "entropy": 5.639406347274781,
      "epoch": 3.2669629898403483,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003952308284220331,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22929448932409285,
      "num_tokens": 74736124.0,
      "step": 36015
    },
    {
      "entropy": 5.564616107940674,
      "epoch": 3.2674165457184325,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039520347767309975,
      "loss": 4.9153,
      "mean_token_accuracy": 0.2192339852452278,
      "num_tokens": 74746447.0,
      "step": 36020
    },
    {
      "entropy": 5.629374647140503,
      "epoch": 3.267870101596517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000395176124438328,
      "loss": 4.8929,
      "mean_token_accuracy": 0.22023245841264724,
      "num_tokens": 74756947.0,
      "step": 36025
    },
    {
      "entropy": 5.652753782272339,
      "epoch": 3.2683236574746006,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00039514876871828326,
      "loss": 4.8179,
      "mean_token_accuracy": 0.2295978844165802,
      "num_tokens": 74767856.0,
      "step": 36030
    },
    {
      "entropy": 5.627947759628296,
      "epoch": 3.268777213352685,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039512141051353144,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22303730249404907,
      "num_tokens": 74778335.0,
      "step": 36035
    },
    {
      "entropy": 5.686800909042359,
      "epoch": 3.269230769230769,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003950940498246382,
      "loss": 4.877,
      "mean_token_accuracy": 0.2260562613606453,
      "num_tokens": 74788430.0,
      "step": 36040
    },
    {
      "entropy": 5.714208316802979,
      "epoch": 3.2696843251088534,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039506668665216944,
      "loss": 4.9976,
      "mean_token_accuracy": 0.2164226084947586,
      "num_tokens": 74800193.0,
      "step": 36045
    },
    {
      "entropy": 5.694384241104126,
      "epoch": 3.2701378809869377,
      "grad_norm": 1.21875,
      "learning_rate": 0.000395039320996691,
      "loss": 4.9266,
      "mean_token_accuracy": 0.20994631350040435,
      "num_tokens": 74810368.0,
      "step": 36050
    },
    {
      "entropy": 5.695540237426758,
      "epoch": 3.270591436865022,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039501195285876874,
      "loss": 4.8584,
      "mean_token_accuracy": 0.23109374195337296,
      "num_tokens": 74820687.0,
      "step": 36055
    },
    {
      "entropy": 5.633235883712769,
      "epoch": 3.271044992743106,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003949845822389688,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22457571625709533,
      "num_tokens": 74830489.0,
      "step": 36060
    },
    {
      "entropy": 5.581903266906738,
      "epoch": 3.27149854862119,
      "grad_norm": 1.109375,
      "learning_rate": 0.000394957209137857,
      "loss": 4.83,
      "mean_token_accuracy": 0.22166192531585693,
      "num_tokens": 74840638.0,
      "step": 36065
    },
    {
      "entropy": 5.700983905792237,
      "epoch": 3.2719521044992743,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003949298335559997,
      "loss": 5.0304,
      "mean_token_accuracy": 0.20710984766483306,
      "num_tokens": 74850600.0,
      "step": 36070
    },
    {
      "entropy": 5.698383569717407,
      "epoch": 3.2724056603773586,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003949024554939628,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22655698657035828,
      "num_tokens": 74861177.0,
      "step": 36075
    },
    {
      "entropy": 5.706443023681641,
      "epoch": 3.272859216255443,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039487507495231263,
      "loss": 4.9567,
      "mean_token_accuracy": 0.21684776544570922,
      "num_tokens": 74871747.0,
      "step": 36080
    },
    {
      "entropy": 5.631390523910523,
      "epoch": 3.2733127721335267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003948476919316153,
      "loss": 4.9172,
      "mean_token_accuracy": 0.21550982147455217,
      "num_tokens": 74882796.0,
      "step": 36085
    },
    {
      "entropy": 5.683108472824097,
      "epoch": 3.273766328011611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003948203064324371,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2205028235912323,
      "num_tokens": 74894123.0,
      "step": 36090
    },
    {
      "entropy": 5.657931995391846,
      "epoch": 3.274219883889695,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039479291845534453,
      "loss": 4.9403,
      "mean_token_accuracy": 0.21633604019880295,
      "num_tokens": 74904787.0,
      "step": 36095
    },
    {
      "entropy": 5.689729499816894,
      "epoch": 3.2746734397677795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039476552800090384,
      "loss": 4.9621,
      "mean_token_accuracy": 0.21248901188373565,
      "num_tokens": 74915041.0,
      "step": 36100
    },
    {
      "entropy": 5.717803955078125,
      "epoch": 3.2751269956458637,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039473813506968146,
      "loss": 4.8707,
      "mean_token_accuracy": 0.21151673048734665,
      "num_tokens": 74926217.0,
      "step": 36105
    },
    {
      "entropy": 5.718538379669189,
      "epoch": 3.2755805515239476,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039471073966224404,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2176157057285309,
      "num_tokens": 74936616.0,
      "step": 36110
    },
    {
      "entropy": 5.688256168365479,
      "epoch": 3.276034107402032,
      "grad_norm": 1.21875,
      "learning_rate": 0.000394683341779158,
      "loss": 4.9432,
      "mean_token_accuracy": 0.2147058442234993,
      "num_tokens": 74947443.0,
      "step": 36115
    },
    {
      "entropy": 5.632800722122193,
      "epoch": 3.276487663280116,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003946559414209899,
      "loss": 4.8668,
      "mean_token_accuracy": 0.2225779950618744,
      "num_tokens": 74957618.0,
      "step": 36120
    },
    {
      "entropy": 5.632121133804321,
      "epoch": 3.2769412191582004,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003946285385883065,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2197408765554428,
      "num_tokens": 74967982.0,
      "step": 36125
    },
    {
      "entropy": 5.623415088653564,
      "epoch": 3.2773947750362846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039460113328167434,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22264011353254318,
      "num_tokens": 74978801.0,
      "step": 36130
    },
    {
      "entropy": 5.624350070953369,
      "epoch": 3.2778483309143684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003945737255016603,
      "loss": 4.843,
      "mean_token_accuracy": 0.22074114382267,
      "num_tokens": 74989000.0,
      "step": 36135
    },
    {
      "entropy": 5.687885189056397,
      "epoch": 3.2783018867924527,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039454631524883123,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21587615758180617,
      "num_tokens": 74999664.0,
      "step": 36140
    },
    {
      "entropy": 5.661758470535278,
      "epoch": 3.278755442670537,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003945189025237538,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2234593316912651,
      "num_tokens": 75010187.0,
      "step": 36145
    },
    {
      "entropy": 5.751517963409424,
      "epoch": 3.2792089985486212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000394491487326995,
      "loss": 4.9588,
      "mean_token_accuracy": 0.21394787728786469,
      "num_tokens": 75021222.0,
      "step": 36150
    },
    {
      "entropy": 5.654802465438843,
      "epoch": 3.2796625544267055,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003944640696591219,
      "loss": 4.9027,
      "mean_token_accuracy": 0.22316223978996277,
      "num_tokens": 75031136.0,
      "step": 36155
    },
    {
      "entropy": 5.6088825225830075,
      "epoch": 3.28011611030479,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039443664952070135,
      "loss": 4.8736,
      "mean_token_accuracy": 0.226972796022892,
      "num_tokens": 75041545.0,
      "step": 36160
    },
    {
      "entropy": 5.649092721939087,
      "epoch": 3.2805696661828736,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039440922691230044,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22535209655761718,
      "num_tokens": 75051805.0,
      "step": 36165
    },
    {
      "entropy": 5.636212015151978,
      "epoch": 3.281023222060958,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039438180183448625,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22479672878980636,
      "num_tokens": 75062551.0,
      "step": 36170
    },
    {
      "entropy": 5.6396430969238285,
      "epoch": 3.281476777939042,
      "grad_norm": 1.21875,
      "learning_rate": 0.000394354374287826,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2200111597776413,
      "num_tokens": 75073040.0,
      "step": 36175
    },
    {
      "entropy": 5.675415754318237,
      "epoch": 3.2819303338171264,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039432694427288695,
      "loss": 4.8747,
      "mean_token_accuracy": 0.22565992921590805,
      "num_tokens": 75082543.0,
      "step": 36180
    },
    {
      "entropy": 5.718313789367675,
      "epoch": 3.28238388969521,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039429951179023614,
      "loss": 4.9453,
      "mean_token_accuracy": 0.21063097566366196,
      "num_tokens": 75092480.0,
      "step": 36185
    },
    {
      "entropy": 5.645403957366943,
      "epoch": 3.2828374455732945,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039427207684044116,
      "loss": 4.7944,
      "mean_token_accuracy": 0.21954277157783508,
      "num_tokens": 75102445.0,
      "step": 36190
    },
    {
      "entropy": 5.680039596557617,
      "epoch": 3.2832910014513788,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039424463942406917,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21217927485704421,
      "num_tokens": 75112793.0,
      "step": 36195
    },
    {
      "entropy": 5.751112747192383,
      "epoch": 3.283744557329463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003942171995416877,
      "loss": 4.9715,
      "mean_token_accuracy": 0.21384888887405396,
      "num_tokens": 75123312.0,
      "step": 36200
    },
    {
      "entropy": 5.658623266220093,
      "epoch": 3.2841981132075473,
      "grad_norm": 1.125,
      "learning_rate": 0.00039418975719386406,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22371156364679337,
      "num_tokens": 75133584.0,
      "step": 36205
    },
    {
      "entropy": 5.634584188461304,
      "epoch": 3.2846516690856316,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003941623123811659,
      "loss": 4.8926,
      "mean_token_accuracy": 0.2187499538064003,
      "num_tokens": 75144344.0,
      "step": 36210
    },
    {
      "entropy": 5.720593452453613,
      "epoch": 3.2851052249637154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039413486510416087,
      "loss": 4.9258,
      "mean_token_accuracy": 0.21624652445316314,
      "num_tokens": 75155086.0,
      "step": 36215
    },
    {
      "entropy": 5.666658353805542,
      "epoch": 3.2855587808417996,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039410741536341634,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22131350338459016,
      "num_tokens": 75166046.0,
      "step": 36220
    },
    {
      "entropy": 5.606917810440064,
      "epoch": 3.286012336719884,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039407996315950016,
      "loss": 4.8247,
      "mean_token_accuracy": 0.2269529029726982,
      "num_tokens": 75175388.0,
      "step": 36225
    },
    {
      "entropy": 5.64459490776062,
      "epoch": 3.286465892597968,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039405250849297997,
      "loss": 4.9367,
      "mean_token_accuracy": 0.22083232998847963,
      "num_tokens": 75186213.0,
      "step": 36230
    },
    {
      "entropy": 5.632690334320069,
      "epoch": 3.2869194484760524,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003940250513644236,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22192759215831756,
      "num_tokens": 75196141.0,
      "step": 36235
    },
    {
      "entropy": 5.604740905761719,
      "epoch": 3.2873730043541363,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003939975917743988,
      "loss": 4.7564,
      "mean_token_accuracy": 0.23138513714075087,
      "num_tokens": 75205414.0,
      "step": 36240
    },
    {
      "entropy": 5.703023910522461,
      "epoch": 3.2878265602322205,
      "grad_norm": 1.25,
      "learning_rate": 0.0003939701297234735,
      "loss": 4.9586,
      "mean_token_accuracy": 0.21892226934432985,
      "num_tokens": 75215860.0,
      "step": 36245
    },
    {
      "entropy": 5.667322492599487,
      "epoch": 3.288280116110305,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039394266521221565,
      "loss": 4.9188,
      "mean_token_accuracy": 0.21709271818399428,
      "num_tokens": 75226358.0,
      "step": 36250
    },
    {
      "entropy": 5.593117523193359,
      "epoch": 3.288733671988389,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00039391519824119315,
      "loss": 4.7509,
      "mean_token_accuracy": 0.23532009571790696,
      "num_tokens": 75235779.0,
      "step": 36255
    },
    {
      "entropy": 5.714635181427002,
      "epoch": 3.2891872278664733,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039388772881097396,
      "loss": 5.0044,
      "mean_token_accuracy": 0.2138289228081703,
      "num_tokens": 75246417.0,
      "step": 36260
    },
    {
      "entropy": 5.669832038879394,
      "epoch": 3.289640783744557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039386025692212637,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22328677475452424,
      "num_tokens": 75257094.0,
      "step": 36265
    },
    {
      "entropy": 5.63443341255188,
      "epoch": 3.2900943396226414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003938327825752183,
      "loss": 4.8416,
      "mean_token_accuracy": 0.22122082114219666,
      "num_tokens": 75267374.0,
      "step": 36270
    },
    {
      "entropy": 5.640927505493164,
      "epoch": 3.2905478955007257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000393805305770818,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2210882470011711,
      "num_tokens": 75276936.0,
      "step": 36275
    },
    {
      "entropy": 5.6239818096160885,
      "epoch": 3.29100145137881,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003937778265094937,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2238938108086586,
      "num_tokens": 75286403.0,
      "step": 36280
    },
    {
      "entropy": 5.688798475265503,
      "epoch": 3.291455007256894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003937503447918137,
      "loss": 4.9201,
      "mean_token_accuracy": 0.21869975626468657,
      "num_tokens": 75297421.0,
      "step": 36285
    },
    {
      "entropy": 5.686753749847412,
      "epoch": 3.291908563134978,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039372286061834633,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2145334392786026,
      "num_tokens": 75307955.0,
      "step": 36290
    },
    {
      "entropy": 5.668632173538208,
      "epoch": 3.2923621190130623,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039369537398966,
      "loss": 4.9388,
      "mean_token_accuracy": 0.21726780384778976,
      "num_tokens": 75318739.0,
      "step": 36295
    },
    {
      "entropy": 5.627797079086304,
      "epoch": 3.2928156748911466,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039366788490632286,
      "loss": 4.7827,
      "mean_token_accuracy": 0.22951727956533433,
      "num_tokens": 75328629.0,
      "step": 36300
    },
    {
      "entropy": 5.582014560699463,
      "epoch": 3.293269230769231,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039364039336890386,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22953133285045624,
      "num_tokens": 75337986.0,
      "step": 36305
    },
    {
      "entropy": 5.6177849769592285,
      "epoch": 3.293722786647315,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003936128993779712,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22597640752792358,
      "num_tokens": 75346169.0,
      "step": 36310
    },
    {
      "entropy": 5.688351345062256,
      "epoch": 3.2941763425253994,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003935854029340936,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22353022396564484,
      "num_tokens": 75356111.0,
      "step": 36315
    },
    {
      "entropy": 5.660332155227661,
      "epoch": 3.294629898403483,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039355790403783954,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22915925532579423,
      "num_tokens": 75366679.0,
      "step": 36320
    },
    {
      "entropy": 5.58781590461731,
      "epoch": 3.2950834542815675,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039353040268977785,
      "loss": 4.7707,
      "mean_token_accuracy": 0.23370709270238876,
      "num_tokens": 75377500.0,
      "step": 36325
    },
    {
      "entropy": 5.6220794200897215,
      "epoch": 3.2955370101596517,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003935028988904773,
      "loss": 4.8545,
      "mean_token_accuracy": 0.22094108313322067,
      "num_tokens": 75387115.0,
      "step": 36330
    },
    {
      "entropy": 5.6581378936767575,
      "epoch": 3.295990566037736,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039347539264050654,
      "loss": 4.9669,
      "mean_token_accuracy": 0.21484380662441255,
      "num_tokens": 75396330.0,
      "step": 36335
    },
    {
      "entropy": 5.651205968856812,
      "epoch": 3.29644412191582,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003934478839404345,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22515249848365784,
      "num_tokens": 75407168.0,
      "step": 36340
    },
    {
      "entropy": 5.6921299457550045,
      "epoch": 3.296897677793904,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039342037279083006,
      "loss": 4.9269,
      "mean_token_accuracy": 0.2141771823167801,
      "num_tokens": 75418117.0,
      "step": 36345
    },
    {
      "entropy": 5.6916783332824705,
      "epoch": 3.2973512336719883,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003933928591922621,
      "loss": 4.9948,
      "mean_token_accuracy": 0.2115752413868904,
      "num_tokens": 75429026.0,
      "step": 36350
    },
    {
      "entropy": 5.682325649261474,
      "epoch": 3.2978047895500726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003933653431452996,
      "loss": 4.8478,
      "mean_token_accuracy": 0.22262220382690429,
      "num_tokens": 75439710.0,
      "step": 36355
    },
    {
      "entropy": 5.672190189361572,
      "epoch": 3.298258345428157,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039333782465051173,
      "loss": 4.8669,
      "mean_token_accuracy": 0.22541048228740693,
      "num_tokens": 75450086.0,
      "step": 36360
    },
    {
      "entropy": 5.63658127784729,
      "epoch": 3.298711901306241,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003933103037084674,
      "loss": 4.8786,
      "mean_token_accuracy": 0.21531790494918823,
      "num_tokens": 75460988.0,
      "step": 36365
    },
    {
      "entropy": 5.69922776222229,
      "epoch": 3.299165457184325,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039328278031973586,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21082618832588196,
      "num_tokens": 75471560.0,
      "step": 36370
    },
    {
      "entropy": 5.668506669998169,
      "epoch": 3.2996190130624092,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003932552544848863,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21501551419496537,
      "num_tokens": 75482124.0,
      "step": 36375
    },
    {
      "entropy": 5.670668697357177,
      "epoch": 3.3000725689404935,
      "grad_norm": 1.296875,
      "learning_rate": 0.000393227726204488,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2223208263516426,
      "num_tokens": 75492351.0,
      "step": 36380
    },
    {
      "entropy": 5.668900442123413,
      "epoch": 3.3005261248185778,
      "grad_norm": 1.125,
      "learning_rate": 0.00039320019547911013,
      "loss": 4.8063,
      "mean_token_accuracy": 0.22544361352920533,
      "num_tokens": 75502475.0,
      "step": 36385
    },
    {
      "entropy": 5.609433889389038,
      "epoch": 3.3009796806966616,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003931726623093221,
      "loss": 4.8191,
      "mean_token_accuracy": 0.22827287763357162,
      "num_tokens": 75512345.0,
      "step": 36390
    },
    {
      "entropy": 5.580686712265015,
      "epoch": 3.301433236574746,
      "grad_norm": 1.25,
      "learning_rate": 0.00039314512669569323,
      "loss": 4.7972,
      "mean_token_accuracy": 0.22901195734739305,
      "num_tokens": 75523011.0,
      "step": 36395
    },
    {
      "entropy": 5.649140119552612,
      "epoch": 3.30188679245283,
      "grad_norm": 1.296875,
      "learning_rate": 0.00039311758863879316,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2181934267282486,
      "num_tokens": 75533775.0,
      "step": 36400
    },
    {
      "entropy": 5.664285039901733,
      "epoch": 3.3023403483309144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039309004813919117,
      "loss": 4.9679,
      "mean_token_accuracy": 0.2088213413953781,
      "num_tokens": 75544558.0,
      "step": 36405
    },
    {
      "entropy": 5.731262922286987,
      "epoch": 3.3027939042089987,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039306250519745696,
      "loss": 4.9881,
      "mean_token_accuracy": 0.2085910677909851,
      "num_tokens": 75555599.0,
      "step": 36410
    },
    {
      "entropy": 5.669095039367676,
      "epoch": 3.303247460087083,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039303495981416007,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22352191805839539,
      "num_tokens": 75566320.0,
      "step": 36415
    },
    {
      "entropy": 5.691562795639038,
      "epoch": 3.3037010159651667,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003930074119898701,
      "loss": 4.9191,
      "mean_token_accuracy": 0.2190934181213379,
      "num_tokens": 75576069.0,
      "step": 36420
    },
    {
      "entropy": 5.6472399711608885,
      "epoch": 3.304154571843251,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003929798617251567,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22200439870357513,
      "num_tokens": 75586012.0,
      "step": 36425
    },
    {
      "entropy": 5.680771112442017,
      "epoch": 3.3046081277213353,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003929523090205898,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21144813299179077,
      "num_tokens": 75596499.0,
      "step": 36430
    },
    {
      "entropy": 5.626072597503662,
      "epoch": 3.3050616835994195,
      "grad_norm": 1.28125,
      "learning_rate": 0.000392924753876739,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22322619408369065,
      "num_tokens": 75606329.0,
      "step": 36435
    },
    {
      "entropy": 5.665460729598999,
      "epoch": 3.305515239477504,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003928971962941743,
      "loss": 4.8942,
      "mean_token_accuracy": 0.21500070840120317,
      "num_tokens": 75617371.0,
      "step": 36440
    },
    {
      "entropy": 5.674899911880493,
      "epoch": 3.3059687953555876,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003928696362734655,
      "loss": 4.8768,
      "mean_token_accuracy": 0.21811844557523727,
      "num_tokens": 75627949.0,
      "step": 36445
    },
    {
      "entropy": 5.662122344970703,
      "epoch": 3.306422351233672,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039284207381518266,
      "loss": 4.9465,
      "mean_token_accuracy": 0.21947485953569412,
      "num_tokens": 75638642.0,
      "step": 36450
    },
    {
      "entropy": 5.6275004863739015,
      "epoch": 3.306875907111756,
      "grad_norm": 1.25,
      "learning_rate": 0.00039281450891989555,
      "loss": 4.8841,
      "mean_token_accuracy": 0.22198915481567383,
      "num_tokens": 75649997.0,
      "step": 36455
    },
    {
      "entropy": 5.630640888214112,
      "epoch": 3.3073294629898404,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003927869415881745,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22810237109661102,
      "num_tokens": 75660223.0,
      "step": 36460
    },
    {
      "entropy": 5.6481372833251955,
      "epoch": 3.3077830188679247,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003927593718205895,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2311721310019493,
      "num_tokens": 75670550.0,
      "step": 36465
    },
    {
      "entropy": 5.700517845153809,
      "epoch": 3.3082365747460085,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039273179961771065,
      "loss": 5.0029,
      "mean_token_accuracy": 0.2087605834007263,
      "num_tokens": 75681631.0,
      "step": 36470
    },
    {
      "entropy": 5.6566084861755375,
      "epoch": 3.308690130624093,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003927042249801082,
      "loss": 4.8407,
      "mean_token_accuracy": 0.2269186407327652,
      "num_tokens": 75691561.0,
      "step": 36475
    },
    {
      "entropy": 5.6630878925323485,
      "epoch": 3.309143686502177,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039267664790835235,
      "loss": 4.8691,
      "mean_token_accuracy": 0.2231012165546417,
      "num_tokens": 75701639.0,
      "step": 36480
    },
    {
      "entropy": 5.684294319152832,
      "epoch": 3.3095972423802613,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003926490684030134,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22791920006275176,
      "num_tokens": 75712823.0,
      "step": 36485
    },
    {
      "entropy": 5.638049697875976,
      "epoch": 3.3100507982583456,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039262148646466185,
      "loss": 4.9292,
      "mean_token_accuracy": 0.217800073325634,
      "num_tokens": 75723221.0,
      "step": 36490
    },
    {
      "entropy": 5.644089651107788,
      "epoch": 3.3105043541364294,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003925939020938679,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22390365302562715,
      "num_tokens": 75733565.0,
      "step": 36495
    },
    {
      "entropy": 5.7175843715667725,
      "epoch": 3.3109579100145137,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039256631529120206,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22599894255399705,
      "num_tokens": 75743511.0,
      "step": 36500
    },
    {
      "entropy": 5.7341443538665775,
      "epoch": 3.311411465892598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003925387260572349,
      "loss": 4.9597,
      "mean_token_accuracy": 0.2163417249917984,
      "num_tokens": 75754593.0,
      "step": 36505
    },
    {
      "entropy": 5.57831392288208,
      "epoch": 3.311865021770682,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039251113439253693,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23206037878990174,
      "num_tokens": 75764534.0,
      "step": 36510
    },
    {
      "entropy": 5.605554485321045,
      "epoch": 3.3123185776487665,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003924835402976788,
      "loss": 4.8303,
      "mean_token_accuracy": 0.22252101302146912,
      "num_tokens": 75774707.0,
      "step": 36515
    },
    {
      "entropy": 5.57983980178833,
      "epoch": 3.3127721335268507,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039245594377323106,
      "loss": 4.7979,
      "mean_token_accuracy": 0.2214741513133049,
      "num_tokens": 75784900.0,
      "step": 36520
    },
    {
      "entropy": 5.6884894371032715,
      "epoch": 3.3132256894049346,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003924283448197645,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21305411905050278,
      "num_tokens": 75795711.0,
      "step": 36525
    },
    {
      "entropy": 5.676147937774658,
      "epoch": 3.313679245283019,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003924007434378498,
      "loss": 4.8788,
      "mean_token_accuracy": 0.22075696289539337,
      "num_tokens": 75805951.0,
      "step": 36530
    },
    {
      "entropy": 5.610408544540405,
      "epoch": 3.314132801161103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003923731396280579,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22183863520622255,
      "num_tokens": 75816023.0,
      "step": 36535
    },
    {
      "entropy": 5.695167303085327,
      "epoch": 3.3145863570391874,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039234553339095944,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21697598695755005,
      "num_tokens": 75827395.0,
      "step": 36540
    },
    {
      "entropy": 5.714741992950439,
      "epoch": 3.315039912917271,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039231792472712554,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22679941952228547,
      "num_tokens": 75836663.0,
      "step": 36545
    },
    {
      "entropy": 5.678356409072876,
      "epoch": 3.3154934687953554,
      "grad_norm": 1.171875,
      "learning_rate": 0.000392290313637127,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22103452533483506,
      "num_tokens": 75846686.0,
      "step": 36550
    },
    {
      "entropy": 5.577677917480469,
      "epoch": 3.3159470246734397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039226270012153485,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21955255120992662,
      "num_tokens": 75857386.0,
      "step": 36555
    },
    {
      "entropy": 5.622571659088135,
      "epoch": 3.316400580551524,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003922350841809203,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22086475938558578,
      "num_tokens": 75867003.0,
      "step": 36560
    },
    {
      "entropy": 5.649987363815308,
      "epoch": 3.3168541364296082,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003922074658158543,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22323533445596694,
      "num_tokens": 75878155.0,
      "step": 36565
    },
    {
      "entropy": 5.624025535583496,
      "epoch": 3.3173076923076925,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039217984502690797,
      "loss": 4.891,
      "mean_token_accuracy": 0.22092503756284715,
      "num_tokens": 75888478.0,
      "step": 36570
    },
    {
      "entropy": 5.612895965576172,
      "epoch": 3.3177612481857763,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039215222181465263,
      "loss": 4.8331,
      "mean_token_accuracy": 0.22204791605472565,
      "num_tokens": 75899541.0,
      "step": 36575
    },
    {
      "entropy": 5.6683248519897464,
      "epoch": 3.3182148040638606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039212459617965956,
      "loss": 4.8753,
      "mean_token_accuracy": 0.22363875210285186,
      "num_tokens": 75910175.0,
      "step": 36580
    },
    {
      "entropy": 5.618531370162964,
      "epoch": 3.318668359941945,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003920969681224999,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21194512993097306,
      "num_tokens": 75919713.0,
      "step": 36585
    },
    {
      "entropy": 5.655419492721558,
      "epoch": 3.319121915820029,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039206933764374515,
      "loss": 4.9616,
      "mean_token_accuracy": 0.21352998912334442,
      "num_tokens": 75929366.0,
      "step": 36590
    },
    {
      "entropy": 5.7693907737731935,
      "epoch": 3.3195754716981134,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039204170474396656,
      "loss": 5.0029,
      "mean_token_accuracy": 0.20589322745800018,
      "num_tokens": 75940284.0,
      "step": 36595
    },
    {
      "entropy": 5.614499092102051,
      "epoch": 3.3200290275761972,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003920140694237358,
      "loss": 4.865,
      "mean_token_accuracy": 0.22291573286056518,
      "num_tokens": 75950752.0,
      "step": 36600
    },
    {
      "entropy": 5.676446151733399,
      "epoch": 3.3204825834542815,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003919864316836242,
      "loss": 4.9864,
      "mean_token_accuracy": 0.20611478239297867,
      "num_tokens": 75961308.0,
      "step": 36605
    },
    {
      "entropy": 5.677382326126098,
      "epoch": 3.3209361393323658,
      "grad_norm": 1.25,
      "learning_rate": 0.0003919587915242034,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22200298458337783,
      "num_tokens": 75972317.0,
      "step": 36610
    },
    {
      "entropy": 5.638326597213745,
      "epoch": 3.32138969521045,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039193114894604504,
      "loss": 4.8846,
      "mean_token_accuracy": 0.22921693474054336,
      "num_tokens": 75982769.0,
      "step": 36615
    },
    {
      "entropy": 5.640643692016601,
      "epoch": 3.3218432510885343,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039190350394972074,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2234386384487152,
      "num_tokens": 75992209.0,
      "step": 36620
    },
    {
      "entropy": 5.7000847339630125,
      "epoch": 3.322296806966618,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003918758565358021,
      "loss": 4.9656,
      "mean_token_accuracy": 0.22461982667446137,
      "num_tokens": 76001749.0,
      "step": 36625
    },
    {
      "entropy": 5.689386606216431,
      "epoch": 3.3227503628447024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000391848206704861,
      "loss": 4.8812,
      "mean_token_accuracy": 0.21831069588661195,
      "num_tokens": 76011326.0,
      "step": 36630
    },
    {
      "entropy": 5.584316396713257,
      "epoch": 3.3232039187227866,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039182055445746917,
      "loss": 4.8162,
      "mean_token_accuracy": 0.22590443342924119,
      "num_tokens": 76021527.0,
      "step": 36635
    },
    {
      "entropy": 5.629241704940796,
      "epoch": 3.323657474600871,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039179289979419853,
      "loss": 4.7573,
      "mean_token_accuracy": 0.23409700244665146,
      "num_tokens": 76031217.0,
      "step": 36640
    },
    {
      "entropy": 5.653891134262085,
      "epoch": 3.324111030478955,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039176524271562096,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21239077746868135,
      "num_tokens": 76041936.0,
      "step": 36645
    },
    {
      "entropy": 5.658161687850952,
      "epoch": 3.324564586357039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039173758322230847,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21971891224384307,
      "num_tokens": 76052431.0,
      "step": 36650
    },
    {
      "entropy": 5.645455360412598,
      "epoch": 3.3250181422351233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003917099213148329,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23003082126379013,
      "num_tokens": 76061589.0,
      "step": 36655
    },
    {
      "entropy": 5.699673986434936,
      "epoch": 3.3254716981132075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039168225699376646,
      "loss": 4.9247,
      "mean_token_accuracy": 0.22755897492170335,
      "num_tokens": 76071642.0,
      "step": 36660
    },
    {
      "entropy": 5.631250953674316,
      "epoch": 3.325925253991292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039165459025968126,
      "loss": 4.8284,
      "mean_token_accuracy": 0.22926007062196732,
      "num_tokens": 76081084.0,
      "step": 36665
    },
    {
      "entropy": 5.652356863021851,
      "epoch": 3.326378809869376,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003916269211131494,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2177904099225998,
      "num_tokens": 76091125.0,
      "step": 36670
    },
    {
      "entropy": 5.736379623413086,
      "epoch": 3.3268323657474603,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003915992495547431,
      "loss": 4.9609,
      "mean_token_accuracy": 0.21855230927467345,
      "num_tokens": 76101452.0,
      "step": 36675
    },
    {
      "entropy": 5.646273899078369,
      "epoch": 3.327285921625544,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039157157558503456,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21204464435577391,
      "num_tokens": 76112034.0,
      "step": 36680
    },
    {
      "entropy": 5.649657249450684,
      "epoch": 3.3277394775036284,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039154389920459615,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2227475881576538,
      "num_tokens": 76122395.0,
      "step": 36685
    },
    {
      "entropy": 5.693719482421875,
      "epoch": 3.3281930333817127,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003915162204140002,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22109930217266083,
      "num_tokens": 76133622.0,
      "step": 36690
    },
    {
      "entropy": 5.7071928024292,
      "epoch": 3.328646589259797,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039148853921381916,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22066134810447693,
      "num_tokens": 76143757.0,
      "step": 36695
    },
    {
      "entropy": 5.794562768936157,
      "epoch": 3.3291001451378808,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003914608556046254,
      "loss": 5.1261,
      "mean_token_accuracy": 0.20570924878120422,
      "num_tokens": 76154943.0,
      "step": 36700
    },
    {
      "entropy": 5.60697193145752,
      "epoch": 3.329553701015965,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003914331695869914,
      "loss": 4.8965,
      "mean_token_accuracy": 0.2189105272293091,
      "num_tokens": 76165227.0,
      "step": 36705
    },
    {
      "entropy": 5.6779335021972654,
      "epoch": 3.3300072568940493,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003914054811614899,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2119307190179825,
      "num_tokens": 76174127.0,
      "step": 36710
    },
    {
      "entropy": 5.733678865432739,
      "epoch": 3.3304608127721336,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003913777903286934,
      "loss": 5.0019,
      "mean_token_accuracy": 0.21379439234733583,
      "num_tokens": 76183518.0,
      "step": 36715
    },
    {
      "entropy": 5.659891605377197,
      "epoch": 3.330914368650218,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003913500970891745,
      "loss": 4.9296,
      "mean_token_accuracy": 0.21287260353565216,
      "num_tokens": 76193865.0,
      "step": 36720
    },
    {
      "entropy": 5.69427981376648,
      "epoch": 3.331367924528302,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003913224014435058,
      "loss": 4.9488,
      "mean_token_accuracy": 0.21583737581968307,
      "num_tokens": 76204273.0,
      "step": 36725
    },
    {
      "entropy": 5.621817302703858,
      "epoch": 3.331821480406386,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039129470339226044,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2346658691763878,
      "num_tokens": 76213706.0,
      "step": 36730
    },
    {
      "entropy": 5.670356130599975,
      "epoch": 3.33227503628447,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039126700293601086,
      "loss": 4.9549,
      "mean_token_accuracy": 0.22304135859012603,
      "num_tokens": 76224484.0,
      "step": 36735
    },
    {
      "entropy": 5.668298292160034,
      "epoch": 3.3327285921625545,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003912393000753299,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21329005658626557,
      "num_tokens": 76236004.0,
      "step": 36740
    },
    {
      "entropy": 5.721057987213134,
      "epoch": 3.3331821480406387,
      "grad_norm": 1.390625,
      "learning_rate": 0.00039121159481079077,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22428331673145294,
      "num_tokens": 76245109.0,
      "step": 36745
    },
    {
      "entropy": 5.691926574707031,
      "epoch": 3.3336357039187225,
      "grad_norm": 1.25,
      "learning_rate": 0.00039118388714296626,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22019402235746383,
      "num_tokens": 76255154.0,
      "step": 36750
    },
    {
      "entropy": 5.664948225021362,
      "epoch": 3.334089259796807,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003911561770724292,
      "loss": 4.848,
      "mean_token_accuracy": 0.21639056801795958,
      "num_tokens": 76266014.0,
      "step": 36755
    },
    {
      "entropy": 5.550282526016235,
      "epoch": 3.334542815674891,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000391128464599753,
      "loss": 4.8298,
      "mean_token_accuracy": 0.22408221364021302,
      "num_tokens": 76277593.0,
      "step": 36760
    },
    {
      "entropy": 5.585718441009521,
      "epoch": 3.3349963715529753,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003911007497255104,
      "loss": 4.7948,
      "mean_token_accuracy": 0.2299378275871277,
      "num_tokens": 76287816.0,
      "step": 36765
    },
    {
      "entropy": 5.6677734375,
      "epoch": 3.3354499274310596,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003910730324502748,
      "loss": 4.9411,
      "mean_token_accuracy": 0.22017153948545456,
      "num_tokens": 76298299.0,
      "step": 36770
    },
    {
      "entropy": 5.655438375473023,
      "epoch": 3.335903483309144,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003910453127746192,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21176403313875197,
      "num_tokens": 76308747.0,
      "step": 36775
    },
    {
      "entropy": 5.636774969100952,
      "epoch": 3.3363570391872277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003910175906991171,
      "loss": 4.8312,
      "mean_token_accuracy": 0.22091300189495086,
      "num_tokens": 76318923.0,
      "step": 36780
    },
    {
      "entropy": 5.7159336566925045,
      "epoch": 3.336810595065312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003909898662243415,
      "loss": 4.9482,
      "mean_token_accuracy": 0.21607514172792436,
      "num_tokens": 76327985.0,
      "step": 36785
    },
    {
      "entropy": 5.69345874786377,
      "epoch": 3.3372641509433962,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039096213935086604,
      "loss": 5.0028,
      "mean_token_accuracy": 0.2100082352757454,
      "num_tokens": 76338933.0,
      "step": 36790
    },
    {
      "entropy": 5.7899494647979735,
      "epoch": 3.3377177068214805,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039093441007926396,
      "loss": 5.0192,
      "mean_token_accuracy": 0.21192287653684616,
      "num_tokens": 76349601.0,
      "step": 36795
    },
    {
      "entropy": 5.675997591018676,
      "epoch": 3.3381712626995648,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039090667841010874,
      "loss": 4.8611,
      "mean_token_accuracy": 0.2271156594157219,
      "num_tokens": 76359968.0,
      "step": 36800
    },
    {
      "entropy": 5.589706611633301,
      "epoch": 3.3386248185776486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003908789443439737,
      "loss": 4.8186,
      "mean_token_accuracy": 0.22875311970710754,
      "num_tokens": 76369864.0,
      "step": 36805
    },
    {
      "entropy": 5.597778797149658,
      "epoch": 3.339078374455733,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003908512078814329,
      "loss": 4.8113,
      "mean_token_accuracy": 0.22541472017765046,
      "num_tokens": 76379158.0,
      "step": 36810
    },
    {
      "entropy": 5.627007865905762,
      "epoch": 3.339531930333817,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039082346902305937,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22010435312986373,
      "num_tokens": 76388798.0,
      "step": 36815
    },
    {
      "entropy": 5.593943929672241,
      "epoch": 3.3399854862119014,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000390795727769427,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22420567274093628,
      "num_tokens": 76399315.0,
      "step": 36820
    },
    {
      "entropy": 5.647522878646851,
      "epoch": 3.3404390420899857,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003907679841211095,
      "loss": 4.9062,
      "mean_token_accuracy": 0.21984747648239136,
      "num_tokens": 76408841.0,
      "step": 36825
    },
    {
      "entropy": 5.69876184463501,
      "epoch": 3.3408925979680695,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003907402380786806,
      "loss": 4.9585,
      "mean_token_accuracy": 0.2095290467143059,
      "num_tokens": 76420055.0,
      "step": 36830
    },
    {
      "entropy": 5.607536745071411,
      "epoch": 3.3413461538461537,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003907124896427141,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23015290796756743,
      "num_tokens": 76430921.0,
      "step": 36835
    },
    {
      "entropy": 5.6265613555908205,
      "epoch": 3.341799709724238,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003906847388137837,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22123363316059114,
      "num_tokens": 76441216.0,
      "step": 36840
    },
    {
      "entropy": 5.666194725036621,
      "epoch": 3.3422532656023223,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003906569855924636,
      "loss": 4.9053,
      "mean_token_accuracy": 0.222189000248909,
      "num_tokens": 76451940.0,
      "step": 36845
    },
    {
      "entropy": 5.676105785369873,
      "epoch": 3.3427068214804065,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003906292299793275,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2273963913321495,
      "num_tokens": 76461701.0,
      "step": 36850
    },
    {
      "entropy": 5.703907823562622,
      "epoch": 3.3431603773584904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039060147197494943,
      "loss": 5.0013,
      "mean_token_accuracy": 0.2098950758576393,
      "num_tokens": 76473658.0,
      "step": 36855
    },
    {
      "entropy": 5.687448930740357,
      "epoch": 3.3436139332365746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039057371157990347,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22016143947839736,
      "num_tokens": 76483821.0,
      "step": 36860
    },
    {
      "entropy": 5.6640924453735355,
      "epoch": 3.344067489114659,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039054594879476374,
      "loss": 4.9168,
      "mean_token_accuracy": 0.21166012585163116,
      "num_tokens": 76494699.0,
      "step": 36865
    },
    {
      "entropy": 5.6471892356872555,
      "epoch": 3.344521044992743,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039051818362010433,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22508293092250825,
      "num_tokens": 76504299.0,
      "step": 36870
    },
    {
      "entropy": 5.680266904830932,
      "epoch": 3.3449746008708274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003904904160564994,
      "loss": 4.9469,
      "mean_token_accuracy": 0.21361702531576157,
      "num_tokens": 76514469.0,
      "step": 36875
    },
    {
      "entropy": 5.661553192138672,
      "epoch": 3.3454281567489117,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003904626461045232,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2170775219798088,
      "num_tokens": 76524666.0,
      "step": 36880
    },
    {
      "entropy": 5.5723975658416744,
      "epoch": 3.3458817126269955,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00039043487376475017,
      "loss": 4.8728,
      "mean_token_accuracy": 0.21813237816095352,
      "num_tokens": 76534787.0,
      "step": 36885
    },
    {
      "entropy": 5.632811498641968,
      "epoch": 3.34633526850508,
      "grad_norm": 1.40625,
      "learning_rate": 0.00039040709903775446,
      "loss": 4.9093,
      "mean_token_accuracy": 0.21465326249599456,
      "num_tokens": 76545055.0,
      "step": 36890
    },
    {
      "entropy": 5.625395822525024,
      "epoch": 3.346788824383164,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003903793219241106,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22735338062047958,
      "num_tokens": 76555458.0,
      "step": 36895
    },
    {
      "entropy": 5.635955858230591,
      "epoch": 3.3472423802612483,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039035154242439294,
      "loss": 4.8311,
      "mean_token_accuracy": 0.23249824345111847,
      "num_tokens": 76566272.0,
      "step": 36900
    },
    {
      "entropy": 5.692262840270996,
      "epoch": 3.347695936139332,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003903237605391759,
      "loss": 4.8994,
      "mean_token_accuracy": 0.2201116606593132,
      "num_tokens": 76575656.0,
      "step": 36905
    },
    {
      "entropy": 5.738354206085205,
      "epoch": 3.3481494920174164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003902959762690342,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22220069468021392,
      "num_tokens": 76586481.0,
      "step": 36910
    },
    {
      "entropy": 5.654278039932251,
      "epoch": 3.3486030478955007,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003902681896145423,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22853632271289825,
      "num_tokens": 76596023.0,
      "step": 36915
    },
    {
      "entropy": 5.556289482116699,
      "epoch": 3.349056603773585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039024040057627487,
      "loss": 4.7945,
      "mean_token_accuracy": 0.22894588559865953,
      "num_tokens": 76605887.0,
      "step": 36920
    },
    {
      "entropy": 5.604000663757324,
      "epoch": 3.349510159651669,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003902126091548066,
      "loss": 4.8103,
      "mean_token_accuracy": 0.23095384538173674,
      "num_tokens": 76616007.0,
      "step": 36925
    },
    {
      "entropy": 5.676253414154052,
      "epoch": 3.3499637155297535,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039018481535071215,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21675093173980714,
      "num_tokens": 76626253.0,
      "step": 36930
    },
    {
      "entropy": 5.664085769653321,
      "epoch": 3.3504172714078373,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003901570191645665,
      "loss": 4.832,
      "mean_token_accuracy": 0.22442505210638047,
      "num_tokens": 76636341.0,
      "step": 36935
    },
    {
      "entropy": 5.647389030456543,
      "epoch": 3.3508708272859216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039012922059694416,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2246784672141075,
      "num_tokens": 76647616.0,
      "step": 36940
    },
    {
      "entropy": 5.63981146812439,
      "epoch": 3.351324383164006,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003901014196484203,
      "loss": 4.885,
      "mean_token_accuracy": 0.2271030455827713,
      "num_tokens": 76658261.0,
      "step": 36945
    },
    {
      "entropy": 5.594466972351074,
      "epoch": 3.35177793904209,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039007361631956976,
      "loss": 4.885,
      "mean_token_accuracy": 0.2193126916885376,
      "num_tokens": 76667985.0,
      "step": 36950
    },
    {
      "entropy": 5.62105975151062,
      "epoch": 3.3522314949201744,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039004581061096745,
      "loss": 4.8429,
      "mean_token_accuracy": 0.22684272080659867,
      "num_tokens": 76678538.0,
      "step": 36955
    },
    {
      "entropy": 5.662672328948974,
      "epoch": 3.352685050798258,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003900180025231886,
      "loss": 4.8028,
      "mean_token_accuracy": 0.22590315192937852,
      "num_tokens": 76688052.0,
      "step": 36960
    },
    {
      "entropy": 5.606722354888916,
      "epoch": 3.3531386066763424,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038999019205680796,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22506275326013564,
      "num_tokens": 76699241.0,
      "step": 36965
    },
    {
      "entropy": 5.619092273712158,
      "epoch": 3.3535921625544267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000389962379212401,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22068145275115966,
      "num_tokens": 76708979.0,
      "step": 36970
    },
    {
      "entropy": 5.671380138397216,
      "epoch": 3.354045718432511,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003899345639905426,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22392962127923965,
      "num_tokens": 76719330.0,
      "step": 36975
    },
    {
      "entropy": 5.705723571777344,
      "epoch": 3.3544992743105952,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003899067463918083,
      "loss": 4.9915,
      "mean_token_accuracy": 0.21215396970510483,
      "num_tokens": 76730210.0,
      "step": 36980
    },
    {
      "entropy": 5.753650712966919,
      "epoch": 3.354952830188679,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038987892641677307,
      "loss": 4.9653,
      "mean_token_accuracy": 0.2179285317659378,
      "num_tokens": 76739977.0,
      "step": 36985
    },
    {
      "entropy": 5.593807601928711,
      "epoch": 3.3554063860667633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003898511040660124,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22308100908994674,
      "num_tokens": 76750200.0,
      "step": 36990
    },
    {
      "entropy": 5.708875846862793,
      "epoch": 3.3558599419448476,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003898232793401017,
      "loss": 4.98,
      "mean_token_accuracy": 0.2099011167883873,
      "num_tokens": 76761420.0,
      "step": 36995
    },
    {
      "entropy": 5.652713680267334,
      "epoch": 3.356313497822932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038979545223961627,
      "loss": 4.867,
      "mean_token_accuracy": 0.22103002071380615,
      "num_tokens": 76771872.0,
      "step": 37000
    },
    {
      "entropy": 5.604804563522339,
      "epoch": 3.356767053701016,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038976762276513163,
      "loss": 4.8425,
      "mean_token_accuracy": 0.22331880331039428,
      "num_tokens": 76782303.0,
      "step": 37005
    },
    {
      "entropy": 5.6894979000091555,
      "epoch": 3.3572206095791,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003897397909172232,
      "loss": 4.9656,
      "mean_token_accuracy": 0.2123808816075325,
      "num_tokens": 76792595.0,
      "step": 37010
    },
    {
      "entropy": 5.684130144119263,
      "epoch": 3.3576741654571842,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003897119566964669,
      "loss": 4.9231,
      "mean_token_accuracy": 0.2205803468823433,
      "num_tokens": 76802577.0,
      "step": 37015
    },
    {
      "entropy": 5.658988523483276,
      "epoch": 3.3581277213352685,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003896841201034379,
      "loss": 4.8523,
      "mean_token_accuracy": 0.23045842349529266,
      "num_tokens": 76812686.0,
      "step": 37020
    },
    {
      "entropy": 5.584048175811768,
      "epoch": 3.3585812772133528,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003896562811387121,
      "loss": 4.778,
      "mean_token_accuracy": 0.23157900869846343,
      "num_tokens": 76823641.0,
      "step": 37025
    },
    {
      "entropy": 5.694049406051636,
      "epoch": 3.359034833091437,
      "grad_norm": 1.25,
      "learning_rate": 0.00038962843980286523,
      "loss": 4.9189,
      "mean_token_accuracy": 0.2203715994954109,
      "num_tokens": 76834060.0,
      "step": 37030
    },
    {
      "entropy": 5.743376731872559,
      "epoch": 3.3594883889695213,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038960059609647306,
      "loss": 4.9789,
      "mean_token_accuracy": 0.22213102728128434,
      "num_tokens": 76845078.0,
      "step": 37035
    },
    {
      "entropy": 5.704052209854126,
      "epoch": 3.359941944847605,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003895727500201113,
      "loss": 4.989,
      "mean_token_accuracy": 0.21332751214504242,
      "num_tokens": 76856539.0,
      "step": 37040
    },
    {
      "entropy": 5.727875566482544,
      "epoch": 3.3603955007256894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038954490157435583,
      "loss": 4.9905,
      "mean_token_accuracy": 0.21328065395355225,
      "num_tokens": 76866941.0,
      "step": 37045
    },
    {
      "entropy": 5.699187755584717,
      "epoch": 3.3608490566037736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038951705075978264,
      "loss": 4.905,
      "mean_token_accuracy": 0.2227076917886734,
      "num_tokens": 76878205.0,
      "step": 37050
    },
    {
      "entropy": 5.694589281082154,
      "epoch": 3.361302612481858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003894891975769677,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21085700541734695,
      "num_tokens": 76888609.0,
      "step": 37055
    },
    {
      "entropy": 5.614452981948853,
      "epoch": 3.3617561683599417,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003894613420264869,
      "loss": 4.9016,
      "mean_token_accuracy": 0.2165390744805336,
      "num_tokens": 76899637.0,
      "step": 37060
    },
    {
      "entropy": 5.670291185379028,
      "epoch": 3.362209724238026,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038943348410891643,
      "loss": 4.917,
      "mean_token_accuracy": 0.21379295587539673,
      "num_tokens": 76911186.0,
      "step": 37065
    },
    {
      "entropy": 5.658308696746826,
      "epoch": 3.3626632801161103,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003894056238248324,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22104533314704894,
      "num_tokens": 76921704.0,
      "step": 37070
    },
    {
      "entropy": 5.585191917419434,
      "epoch": 3.3631168359941945,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003893777611748108,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22796611934900285,
      "num_tokens": 76932362.0,
      "step": 37075
    },
    {
      "entropy": 5.675242280960083,
      "epoch": 3.363570391872279,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038934989615942807,
      "loss": 5.0149,
      "mean_token_accuracy": 0.20818758010864258,
      "num_tokens": 76942581.0,
      "step": 37080
    },
    {
      "entropy": 5.666801881790161,
      "epoch": 3.364023947750363,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003893220287792603,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22089650183916093,
      "num_tokens": 76952912.0,
      "step": 37085
    },
    {
      "entropy": 5.684109830856324,
      "epoch": 3.364477503628447,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038929415903488383,
      "loss": 4.9275,
      "mean_token_accuracy": 0.22022507935762406,
      "num_tokens": 76963320.0,
      "step": 37090
    },
    {
      "entropy": 5.640007448196411,
      "epoch": 3.364931059506531,
      "grad_norm": 1.25,
      "learning_rate": 0.000389266286926875,
      "loss": 4.7842,
      "mean_token_accuracy": 0.22773223668336867,
      "num_tokens": 76973448.0,
      "step": 37095
    },
    {
      "entropy": 5.605854892730713,
      "epoch": 3.3653846153846154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038923841245581026,
      "loss": 4.8371,
      "mean_token_accuracy": 0.2303074985742569,
      "num_tokens": 76983671.0,
      "step": 37100
    },
    {
      "entropy": 5.679612684249878,
      "epoch": 3.3658381712626997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038921053562226607,
      "loss": 4.9985,
      "mean_token_accuracy": 0.2068542405962944,
      "num_tokens": 76994345.0,
      "step": 37105
    },
    {
      "entropy": 5.636000537872315,
      "epoch": 3.3662917271407835,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038918265642681884,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2248877078294754,
      "num_tokens": 77004809.0,
      "step": 37110
    },
    {
      "entropy": 5.650054264068603,
      "epoch": 3.3667452830188678,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038915477487004525,
      "loss": 4.8744,
      "mean_token_accuracy": 0.22323116213083266,
      "num_tokens": 77015505.0,
      "step": 37115
    },
    {
      "entropy": 5.5719428062438965,
      "epoch": 3.367198838896952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038912689095252183,
      "loss": 4.7259,
      "mean_token_accuracy": 0.23410991579294205,
      "num_tokens": 77025572.0,
      "step": 37120
    },
    {
      "entropy": 5.697070598602295,
      "epoch": 3.3676523947750363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003890990046748251,
      "loss": 5.0259,
      "mean_token_accuracy": 0.206975819170475,
      "num_tokens": 77037273.0,
      "step": 37125
    },
    {
      "entropy": 5.7397599697113035,
      "epoch": 3.3681059506531206,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038907111603753196,
      "loss": 4.9702,
      "mean_token_accuracy": 0.21269505321979523,
      "num_tokens": 77047724.0,
      "step": 37130
    },
    {
      "entropy": 5.689666843414306,
      "epoch": 3.368559506531205,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038904322504121916,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22216669619083404,
      "num_tokens": 77057839.0,
      "step": 37135
    },
    {
      "entropy": 5.64613447189331,
      "epoch": 3.3690130624092887,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003890153316864632,
      "loss": 4.8549,
      "mean_token_accuracy": 0.21288612186908723,
      "num_tokens": 77068505.0,
      "step": 37140
    },
    {
      "entropy": 5.719022750854492,
      "epoch": 3.369466618287373,
      "grad_norm": 1.25,
      "learning_rate": 0.00038898743597384126,
      "loss": 4.9823,
      "mean_token_accuracy": 0.21557415872812272,
      "num_tokens": 77078126.0,
      "step": 37145
    },
    {
      "entropy": 5.681842279434204,
      "epoch": 3.369920174165457,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038895953790393014,
      "loss": 4.8987,
      "mean_token_accuracy": 0.2220318630337715,
      "num_tokens": 77087802.0,
      "step": 37150
    },
    {
      "entropy": 5.671897983551025,
      "epoch": 3.3703737300435415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038893163747730664,
      "loss": 4.912,
      "mean_token_accuracy": 0.22389960587024688,
      "num_tokens": 77097901.0,
      "step": 37155
    },
    {
      "entropy": 5.65239577293396,
      "epoch": 3.3708272859216257,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038890373469454794,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22164526283740998,
      "num_tokens": 77107604.0,
      "step": 37160
    },
    {
      "entropy": 5.672668886184693,
      "epoch": 3.3712808417997095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038887582955623084,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22696020007133483,
      "num_tokens": 77117410.0,
      "step": 37165
    },
    {
      "entropy": 5.720358514785767,
      "epoch": 3.371734397677794,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003888479220629326,
      "loss": 4.9008,
      "mean_token_accuracy": 0.224274879693985,
      "num_tokens": 77129546.0,
      "step": 37170
    },
    {
      "entropy": 5.648906469345093,
      "epoch": 3.372187953555878,
      "grad_norm": 1.125,
      "learning_rate": 0.0003888200122152304,
      "loss": 4.8379,
      "mean_token_accuracy": 0.22750701308250426,
      "num_tokens": 77139217.0,
      "step": 37175
    },
    {
      "entropy": 5.7034704208374025,
      "epoch": 3.3726415094339623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003887921000137013,
      "loss": 4.9913,
      "mean_token_accuracy": 0.2130390077829361,
      "num_tokens": 77149694.0,
      "step": 37180
    },
    {
      "entropy": 5.605927753448486,
      "epoch": 3.3730950653120466,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003887641854589226,
      "loss": 4.873,
      "mean_token_accuracy": 0.2249233528971672,
      "num_tokens": 77159291.0,
      "step": 37185
    },
    {
      "entropy": 5.645128631591797,
      "epoch": 3.3735486211901304,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003887362685514716,
      "loss": 4.8645,
      "mean_token_accuracy": 0.21839075833559035,
      "num_tokens": 77170507.0,
      "step": 37190
    },
    {
      "entropy": 5.676050186157227,
      "epoch": 3.3740021770682147,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003887083492919254,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22073099315166472,
      "num_tokens": 77179673.0,
      "step": 37195
    },
    {
      "entropy": 5.7013720035552975,
      "epoch": 3.374455732946299,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003886804276808617,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22848888039588927,
      "num_tokens": 77190163.0,
      "step": 37200
    },
    {
      "entropy": 5.682889127731324,
      "epoch": 3.3749092888243832,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003886525037188578,
      "loss": 4.8964,
      "mean_token_accuracy": 0.22601418942213058,
      "num_tokens": 77201498.0,
      "step": 37205
    },
    {
      "entropy": 5.675180387496948,
      "epoch": 3.3753628447024675,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003886245774064911,
      "loss": 4.8471,
      "mean_token_accuracy": 0.22367338985204696,
      "num_tokens": 77211475.0,
      "step": 37210
    },
    {
      "entropy": 5.654070663452148,
      "epoch": 3.3758164005805513,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038859664874433923,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22185786217451095,
      "num_tokens": 77221972.0,
      "step": 37215
    },
    {
      "entropy": 5.662895202636719,
      "epoch": 3.3762699564586356,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038856871773297965,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22149742394685745,
      "num_tokens": 77232701.0,
      "step": 37220
    },
    {
      "entropy": 5.671054363250732,
      "epoch": 3.37672351233672,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038854078437299005,
      "loss": 4.846,
      "mean_token_accuracy": 0.23028052002191543,
      "num_tokens": 77243242.0,
      "step": 37225
    },
    {
      "entropy": 5.643370151519775,
      "epoch": 3.377177068214804,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003885128486649482,
      "loss": 4.8323,
      "mean_token_accuracy": 0.2205931842327118,
      "num_tokens": 77254279.0,
      "step": 37230
    },
    {
      "entropy": 5.64086503982544,
      "epoch": 3.3776306240928884,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038848491060943164,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22233050018548967,
      "num_tokens": 77265483.0,
      "step": 37235
    },
    {
      "entropy": 5.6088789939880375,
      "epoch": 3.3780841799709727,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003884569702070182,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22421638518571854,
      "num_tokens": 77275848.0,
      "step": 37240
    },
    {
      "entropy": 5.607960176467896,
      "epoch": 3.3785377358490565,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038842902745828573,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22379076331853867,
      "num_tokens": 77285943.0,
      "step": 37245
    },
    {
      "entropy": 5.601777601242065,
      "epoch": 3.3789912917271407,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003884010823638121,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2228484570980072,
      "num_tokens": 77295276.0,
      "step": 37250
    },
    {
      "entropy": 5.703586292266846,
      "epoch": 3.379444847605225,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003883731349241751,
      "loss": 4.8749,
      "mean_token_accuracy": 0.2233763173222542,
      "num_tokens": 77305713.0,
      "step": 37255
    },
    {
      "entropy": 5.6744149684906,
      "epoch": 3.3798984034833093,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003883451851399528,
      "loss": 4.9604,
      "mean_token_accuracy": 0.2193122237920761,
      "num_tokens": 77316717.0,
      "step": 37260
    },
    {
      "entropy": 5.6145774841308596,
      "epoch": 3.380351959361393,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038831723301172325,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2225159451365471,
      "num_tokens": 77327763.0,
      "step": 37265
    },
    {
      "entropy": 5.685797071456909,
      "epoch": 3.3808055152394774,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003882892785400644,
      "loss": 4.9607,
      "mean_token_accuracy": 0.2123486265540123,
      "num_tokens": 77337495.0,
      "step": 37270
    },
    {
      "entropy": 5.707500696182251,
      "epoch": 3.3812590711175616,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038826132172555444,
      "loss": 4.958,
      "mean_token_accuracy": 0.21458590030670166,
      "num_tokens": 77348849.0,
      "step": 37275
    },
    {
      "entropy": 5.662078619003296,
      "epoch": 3.381712626995646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003882333625687714,
      "loss": 4.9431,
      "mean_token_accuracy": 0.2149022027850151,
      "num_tokens": 77359634.0,
      "step": 37280
    },
    {
      "entropy": 5.684861135482788,
      "epoch": 3.38216618287373,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038820540107029365,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22502207458019258,
      "num_tokens": 77369342.0,
      "step": 37285
    },
    {
      "entropy": 5.713677215576172,
      "epoch": 3.3826197387518144,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003881774372306993,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22382436841726303,
      "num_tokens": 77379963.0,
      "step": 37290
    },
    {
      "entropy": 5.627891492843628,
      "epoch": 3.3830732946298983,
      "grad_norm": 1.375,
      "learning_rate": 0.0003881494710505667,
      "loss": 4.827,
      "mean_token_accuracy": 0.22233303487300873,
      "num_tokens": 77390135.0,
      "step": 37295
    },
    {
      "entropy": 5.743196439743042,
      "epoch": 3.3835268505079825,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038812150253047427,
      "loss": 4.9799,
      "mean_token_accuracy": 0.22070779800415039,
      "num_tokens": 77400151.0,
      "step": 37300
    },
    {
      "entropy": 5.622131109237671,
      "epoch": 3.383980406386067,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003880935316710003,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22685371786355973,
      "num_tokens": 77410723.0,
      "step": 37305
    },
    {
      "entropy": 5.596622610092163,
      "epoch": 3.384433962264151,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003880655584727233,
      "loss": 4.8172,
      "mean_token_accuracy": 0.227056285738945,
      "num_tokens": 77421213.0,
      "step": 37310
    },
    {
      "entropy": 5.666327667236328,
      "epoch": 3.384887518142235,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003880375829362217,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22231413424015045,
      "num_tokens": 77431705.0,
      "step": 37315
    },
    {
      "entropy": 5.655713033676148,
      "epoch": 3.385341074020319,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003880096050620741,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22523622810840607,
      "num_tokens": 77442120.0,
      "step": 37320
    },
    {
      "entropy": 5.686656379699707,
      "epoch": 3.3857946298984034,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003879816248508591,
      "loss": 4.8793,
      "mean_token_accuracy": 0.2224569633603096,
      "num_tokens": 77451992.0,
      "step": 37325
    },
    {
      "entropy": 5.726354598999023,
      "epoch": 3.3862481857764877,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038795364230315523,
      "loss": 5.0837,
      "mean_token_accuracy": 0.20227451920509337,
      "num_tokens": 77462830.0,
      "step": 37330
    },
    {
      "entropy": 5.702050304412841,
      "epoch": 3.386701741654572,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003879256574195414,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22665280252695083,
      "num_tokens": 77472617.0,
      "step": 37335
    },
    {
      "entropy": 5.583208274841309,
      "epoch": 3.387155297532656,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038789767020059614,
      "loss": 4.8121,
      "mean_token_accuracy": 0.2265278548002243,
      "num_tokens": 77483430.0,
      "step": 37340
    },
    {
      "entropy": 5.572708511352539,
      "epoch": 3.38760885341074,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038786968064689816,
      "loss": 4.8168,
      "mean_token_accuracy": 0.2298482120037079,
      "num_tokens": 77493038.0,
      "step": 37345
    },
    {
      "entropy": 5.672232818603516,
      "epoch": 3.3880624092888243,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003878416887590265,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2102325975894928,
      "num_tokens": 77502816.0,
      "step": 37350
    },
    {
      "entropy": 5.606935834884643,
      "epoch": 3.3885159651669086,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038781369453756004,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22541743516921997,
      "num_tokens": 77513026.0,
      "step": 37355
    },
    {
      "entropy": 5.675303745269775,
      "epoch": 3.388969521044993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003877856979830775,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2166341409087181,
      "num_tokens": 77523217.0,
      "step": 37360
    },
    {
      "entropy": 5.624972152709961,
      "epoch": 3.389423076923077,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038775769909615805,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21927741914987564,
      "num_tokens": 77533339.0,
      "step": 37365
    },
    {
      "entropy": 5.64515528678894,
      "epoch": 3.389876632801161,
      "grad_norm": 1.25,
      "learning_rate": 0.00038772969787738074,
      "loss": 4.8557,
      "mean_token_accuracy": 0.2244590237736702,
      "num_tokens": 77543438.0,
      "step": 37370
    },
    {
      "entropy": 5.694658279418945,
      "epoch": 3.390330188679245,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003877016943273244,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21672084629535676,
      "num_tokens": 77552825.0,
      "step": 37375
    },
    {
      "entropy": 5.666503047943115,
      "epoch": 3.3907837445573294,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038767368844656836,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21386485546827316,
      "num_tokens": 77563107.0,
      "step": 37380
    },
    {
      "entropy": 5.716397905349732,
      "epoch": 3.3912373004354137,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038764568023569177,
      "loss": 4.9407,
      "mean_token_accuracy": 0.21293231546878816,
      "num_tokens": 77573312.0,
      "step": 37385
    },
    {
      "entropy": 5.649836015701294,
      "epoch": 3.391690856313498,
      "grad_norm": 1.25,
      "learning_rate": 0.0003876176696952737,
      "loss": 4.9157,
      "mean_token_accuracy": 0.21642308682203293,
      "num_tokens": 77583742.0,
      "step": 37390
    },
    {
      "entropy": 5.746195507049561,
      "epoch": 3.3921444121915822,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038758965682589357,
      "loss": 4.9448,
      "mean_token_accuracy": 0.21306571662425994,
      "num_tokens": 77593667.0,
      "step": 37395
    },
    {
      "entropy": 5.643326139450073,
      "epoch": 3.392597968069666,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003875616416281307,
      "loss": 4.8404,
      "mean_token_accuracy": 0.22089378535747528,
      "num_tokens": 77604249.0,
      "step": 37400
    },
    {
      "entropy": 5.625752019882202,
      "epoch": 3.3930515239477503,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038753362410256435,
      "loss": 4.8476,
      "mean_token_accuracy": 0.226080459356308,
      "num_tokens": 77615095.0,
      "step": 37405
    },
    {
      "entropy": 5.629147052764893,
      "epoch": 3.3935050798258346,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003875056042497739,
      "loss": 4.8549,
      "mean_token_accuracy": 0.22314595580101013,
      "num_tokens": 77625182.0,
      "step": 37410
    },
    {
      "entropy": 5.680332326889038,
      "epoch": 3.393958635703919,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038747758207033903,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22302542328834535,
      "num_tokens": 77635423.0,
      "step": 37415
    },
    {
      "entropy": 5.559545850753784,
      "epoch": 3.3944121915820027,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003874495575648391,
      "loss": 4.7665,
      "mean_token_accuracy": 0.22729846090078354,
      "num_tokens": 77645262.0,
      "step": 37420
    },
    {
      "entropy": 5.626524591445923,
      "epoch": 3.394865747460087,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038742153073385356,
      "loss": 4.9481,
      "mean_token_accuracy": 0.211332505941391,
      "num_tokens": 77657647.0,
      "step": 37425
    },
    {
      "entropy": 5.6801848888397215,
      "epoch": 3.3953193033381712,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038739350157796214,
      "loss": 4.9067,
      "mean_token_accuracy": 0.21953675597906114,
      "num_tokens": 77668074.0,
      "step": 37430
    },
    {
      "entropy": 5.733698558807373,
      "epoch": 3.3957728592162555,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038736547009774445,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2224317342042923,
      "num_tokens": 77678661.0,
      "step": 37435
    },
    {
      "entropy": 5.681566762924194,
      "epoch": 3.3962264150943398,
      "grad_norm": 1.25,
      "learning_rate": 0.0003873374362937803,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21618083119392395,
      "num_tokens": 77689206.0,
      "step": 37440
    },
    {
      "entropy": 5.681355905532837,
      "epoch": 3.396679970972424,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003873094001666493,
      "loss": 4.9644,
      "mean_token_accuracy": 0.22022362649440766,
      "num_tokens": 77699254.0,
      "step": 37445
    },
    {
      "entropy": 5.680988550186157,
      "epoch": 3.397133526850508,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003872813617169312,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22406539171934128,
      "num_tokens": 77709595.0,
      "step": 37450
    },
    {
      "entropy": 5.611234378814697,
      "epoch": 3.397587082728592,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038725332094520607,
      "loss": 4.8238,
      "mean_token_accuracy": 0.22151446342468262,
      "num_tokens": 77719446.0,
      "step": 37455
    },
    {
      "entropy": 5.6023729801177975,
      "epoch": 3.3980406386066764,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003872252778520536,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22817627638578414,
      "num_tokens": 77729723.0,
      "step": 37460
    },
    {
      "entropy": 5.6007133483886715,
      "epoch": 3.3984941944847606,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003871972324380537,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22117896378040314,
      "num_tokens": 77739380.0,
      "step": 37465
    },
    {
      "entropy": 5.671479368209839,
      "epoch": 3.3989477503628445,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003871691847037865,
      "loss": 4.8982,
      "mean_token_accuracy": 0.2255598098039627,
      "num_tokens": 77750683.0,
      "step": 37470
    },
    {
      "entropy": 5.675751209259033,
      "epoch": 3.3994013062409287,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038714113464983205,
      "loss": 4.9595,
      "mean_token_accuracy": 0.20942157804965972,
      "num_tokens": 77761951.0,
      "step": 37475
    },
    {
      "entropy": 5.707545471191406,
      "epoch": 3.399854862119013,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003871130822767702,
      "loss": 4.9252,
      "mean_token_accuracy": 0.21728074699640273,
      "num_tokens": 77772165.0,
      "step": 37480
    },
    {
      "entropy": 5.735255146026612,
      "epoch": 3.4003084179970973,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003870850275851814,
      "loss": 4.9675,
      "mean_token_accuracy": 0.21130006909370422,
      "num_tokens": 77781916.0,
      "step": 37485
    },
    {
      "entropy": 5.682441663742066,
      "epoch": 3.4007619738751815,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038705697057564556,
      "loss": 4.8796,
      "mean_token_accuracy": 0.21401282995939255,
      "num_tokens": 77791397.0,
      "step": 37490
    },
    {
      "entropy": 5.609546279907226,
      "epoch": 3.401215529753266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000387028911248743,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22183045893907546,
      "num_tokens": 77801694.0,
      "step": 37495
    },
    {
      "entropy": 5.69904408454895,
      "epoch": 3.4016690856313496,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038700084960505414,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22001601457595826,
      "num_tokens": 77813199.0,
      "step": 37500
    },
    {
      "entropy": 5.752951002120971,
      "epoch": 3.402122641509434,
      "grad_norm": 1.25,
      "learning_rate": 0.0003869727856451591,
      "loss": 5.0149,
      "mean_token_accuracy": 0.21459803879261016,
      "num_tokens": 77823689.0,
      "step": 37505
    },
    {
      "entropy": 5.681849384307862,
      "epoch": 3.402576197387518,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038694471936963825,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2315855070948601,
      "num_tokens": 77834331.0,
      "step": 37510
    },
    {
      "entropy": 5.65829029083252,
      "epoch": 3.4030297532656024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038691665077907213,
      "loss": 4.81,
      "mean_token_accuracy": 0.22342237383127211,
      "num_tokens": 77843988.0,
      "step": 37515
    },
    {
      "entropy": 5.624342012405395,
      "epoch": 3.4034833091436867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003868885798740412,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22668687552213668,
      "num_tokens": 77854553.0,
      "step": 37520
    },
    {
      "entropy": 5.600727605819702,
      "epoch": 3.4039368650217705,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038686050665512583,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22489483654499054,
      "num_tokens": 77864629.0,
      "step": 37525
    },
    {
      "entropy": 5.682349348068238,
      "epoch": 3.4043904208998548,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038683243112290673,
      "loss": 4.9529,
      "mean_token_accuracy": 0.21327644288539888,
      "num_tokens": 77875465.0,
      "step": 37530
    },
    {
      "entropy": 5.724593877792358,
      "epoch": 3.404843976777939,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003868043532779645,
      "loss": 4.9119,
      "mean_token_accuracy": 0.21864134967327117,
      "num_tokens": 77885337.0,
      "step": 37535
    },
    {
      "entropy": 5.67777304649353,
      "epoch": 3.4052975326560233,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038677627312087967,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22256972938776015,
      "num_tokens": 77896465.0,
      "step": 37540
    },
    {
      "entropy": 5.651318550109863,
      "epoch": 3.4057510885341076,
      "grad_norm": 1.25,
      "learning_rate": 0.0003867481906522331,
      "loss": 4.924,
      "mean_token_accuracy": 0.2127584770321846,
      "num_tokens": 77906214.0,
      "step": 37545
    },
    {
      "entropy": 5.695660400390625,
      "epoch": 3.4062046444121914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003867201058726054,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2069516211748123,
      "num_tokens": 77917466.0,
      "step": 37550
    },
    {
      "entropy": 5.757440233230591,
      "epoch": 3.4066582002902757,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003866920187825775,
      "loss": 4.9544,
      "mean_token_accuracy": 0.21870570927858352,
      "num_tokens": 77927648.0,
      "step": 37555
    },
    {
      "entropy": 5.623804521560669,
      "epoch": 3.40711175616836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038666392938273015,
      "loss": 4.8031,
      "mean_token_accuracy": 0.22679403573274612,
      "num_tokens": 77938028.0,
      "step": 37560
    },
    {
      "entropy": 5.6339939594268795,
      "epoch": 3.407565312046444,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003866358376736443,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22143707424402237,
      "num_tokens": 77947276.0,
      "step": 37565
    },
    {
      "entropy": 5.705109596252441,
      "epoch": 3.4080188679245285,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003866077436559009,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2228158950805664,
      "num_tokens": 77956791.0,
      "step": 37570
    },
    {
      "entropy": 5.7097245216369625,
      "epoch": 3.4084724238026123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003865796473300809,
      "loss": 4.9205,
      "mean_token_accuracy": 0.2197381228208542,
      "num_tokens": 77967402.0,
      "step": 37575
    },
    {
      "entropy": 5.754017734527588,
      "epoch": 3.4089259796806965,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003865515486967653,
      "loss": 5.0068,
      "mean_token_accuracy": 0.2147762730717659,
      "num_tokens": 77978000.0,
      "step": 37580
    },
    {
      "entropy": 5.616274976730347,
      "epoch": 3.409379535558781,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038652344775653535,
      "loss": 4.8143,
      "mean_token_accuracy": 0.23118745535612106,
      "num_tokens": 77987959.0,
      "step": 37585
    },
    {
      "entropy": 5.6466327667236325,
      "epoch": 3.409833091436865,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003864953445099721,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22591675221920013,
      "num_tokens": 77997990.0,
      "step": 37590
    },
    {
      "entropy": 5.691438865661621,
      "epoch": 3.4102866473149493,
      "grad_norm": 1.25,
      "learning_rate": 0.00038646723895765654,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22029051780700684,
      "num_tokens": 78008831.0,
      "step": 37595
    },
    {
      "entropy": 5.737888669967651,
      "epoch": 3.4107402031930336,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003864391311001702,
      "loss": 4.9781,
      "mean_token_accuracy": 0.21321150064468383,
      "num_tokens": 78019338.0,
      "step": 37600
    },
    {
      "entropy": 5.717546892166138,
      "epoch": 3.4111937590711174,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003864110209380942,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21511536985635757,
      "num_tokens": 78030700.0,
      "step": 37605
    },
    {
      "entropy": 5.669732570648193,
      "epoch": 3.4116473149492017,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038638290847200997,
      "loss": 4.847,
      "mean_token_accuracy": 0.23010757863521575,
      "num_tokens": 78040190.0,
      "step": 37610
    },
    {
      "entropy": 5.651835107803345,
      "epoch": 3.412100870827286,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003863547937024987,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22936272025108337,
      "num_tokens": 78050847.0,
      "step": 37615
    },
    {
      "entropy": 5.719991731643677,
      "epoch": 3.4125544267053702,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000386326676630142,
      "loss": 4.9,
      "mean_token_accuracy": 0.2278301775455475,
      "num_tokens": 78060584.0,
      "step": 37620
    },
    {
      "entropy": 5.6226318359375,
      "epoch": 3.413007982583454,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003862985572555212,
      "loss": 4.8141,
      "mean_token_accuracy": 0.2246528759598732,
      "num_tokens": 78071189.0,
      "step": 37625
    },
    {
      "entropy": 5.677136039733886,
      "epoch": 3.4134615384615383,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038627043557921785,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22015422731637954,
      "num_tokens": 78082054.0,
      "step": 37630
    },
    {
      "entropy": 5.613542985916138,
      "epoch": 3.4139150943396226,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003862423116018136,
      "loss": 4.8683,
      "mean_token_accuracy": 0.2167477622628212,
      "num_tokens": 78094032.0,
      "step": 37635
    },
    {
      "entropy": 5.6631272315979,
      "epoch": 3.414368650217707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038621418532389004,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2225535735487938,
      "num_tokens": 78104724.0,
      "step": 37640
    },
    {
      "entropy": 5.72815170288086,
      "epoch": 3.414822206095791,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038618605674602867,
      "loss": 4.9713,
      "mean_token_accuracy": 0.21340191811323167,
      "num_tokens": 78115927.0,
      "step": 37645
    },
    {
      "entropy": 5.704930019378662,
      "epoch": 3.4152757619738754,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038615792586881145,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22237690538167953,
      "num_tokens": 78126744.0,
      "step": 37650
    },
    {
      "entropy": 5.636739492416382,
      "epoch": 3.415729317851959,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003861297926928199,
      "loss": 4.8998,
      "mean_token_accuracy": 0.21860082745552062,
      "num_tokens": 78137625.0,
      "step": 37655
    },
    {
      "entropy": 5.599440336227417,
      "epoch": 3.4161828737300435,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003861016572186359,
      "loss": 4.824,
      "mean_token_accuracy": 0.2256937876343727,
      "num_tokens": 78147709.0,
      "step": 37660
    },
    {
      "entropy": 5.7300104141235355,
      "epoch": 3.4166364296081277,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038607351944684135,
      "loss": 5.0283,
      "mean_token_accuracy": 0.2183113783597946,
      "num_tokens": 78158634.0,
      "step": 37665
    },
    {
      "entropy": 5.749880695343018,
      "epoch": 3.417089985486212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003860453793780181,
      "loss": 4.9874,
      "mean_token_accuracy": 0.21051167994737624,
      "num_tokens": 78170279.0,
      "step": 37670
    },
    {
      "entropy": 5.6917126178741455,
      "epoch": 3.417543541364296,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038601723701274813,
      "loss": 4.908,
      "mean_token_accuracy": 0.2182104393839836,
      "num_tokens": 78180209.0,
      "step": 37675
    },
    {
      "entropy": 5.666525030136109,
      "epoch": 3.41799709724238,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003859890923516134,
      "loss": 4.945,
      "mean_token_accuracy": 0.2184756278991699,
      "num_tokens": 78189937.0,
      "step": 37680
    },
    {
      "entropy": 5.674356603622437,
      "epoch": 3.4184506531204644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000385960945395196,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22057731002569197,
      "num_tokens": 78199224.0,
      "step": 37685
    },
    {
      "entropy": 5.6816318988800045,
      "epoch": 3.4189042089985486,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003859327961440779,
      "loss": 4.8451,
      "mean_token_accuracy": 0.22224436402320863,
      "num_tokens": 78209912.0,
      "step": 37690
    },
    {
      "entropy": 5.671547269821167,
      "epoch": 3.419357764876633,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038590464459884133,
      "loss": 4.9281,
      "mean_token_accuracy": 0.2219547748565674,
      "num_tokens": 78220388.0,
      "step": 37695
    },
    {
      "entropy": 5.672980880737304,
      "epoch": 3.419811320754717,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038587649076006844,
      "loss": 5.027,
      "mean_token_accuracy": 0.21138158589601516,
      "num_tokens": 78231094.0,
      "step": 37700
    },
    {
      "entropy": 5.6724021434783936,
      "epoch": 3.420264876632801,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003858483346283415,
      "loss": 4.8737,
      "mean_token_accuracy": 0.21757461726665497,
      "num_tokens": 78241111.0,
      "step": 37705
    },
    {
      "entropy": 5.7311185836792,
      "epoch": 3.4207184325108853,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003858201762042428,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22171858698129654,
      "num_tokens": 78251036.0,
      "step": 37710
    },
    {
      "entropy": 5.6556333065032955,
      "epoch": 3.4211719883889695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003857920154883545,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2130550131201744,
      "num_tokens": 78261840.0,
      "step": 37715
    },
    {
      "entropy": 5.673561763763428,
      "epoch": 3.421625544267054,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003857638524812591,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21812672764062882,
      "num_tokens": 78272539.0,
      "step": 37720
    },
    {
      "entropy": 5.70460090637207,
      "epoch": 3.422079100145138,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038573568718353914,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2177668109536171,
      "num_tokens": 78283849.0,
      "step": 37725
    },
    {
      "entropy": 5.685485029220581,
      "epoch": 3.422532656023222,
      "grad_norm": 1.25,
      "learning_rate": 0.0003857075195957768,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21818661838769912,
      "num_tokens": 78293640.0,
      "step": 37730
    },
    {
      "entropy": 5.650167798995971,
      "epoch": 3.422986211901306,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003856793497185548,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22120181620121002,
      "num_tokens": 78303980.0,
      "step": 37735
    },
    {
      "entropy": 5.660114002227783,
      "epoch": 3.4234397677793904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003856511775524556,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22153453528881073,
      "num_tokens": 78314233.0,
      "step": 37740
    },
    {
      "entropy": 5.673801422119141,
      "epoch": 3.4238933236574747,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038562300309806193,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2216591715812683,
      "num_tokens": 78324449.0,
      "step": 37745
    },
    {
      "entropy": 5.715111207962036,
      "epoch": 3.424346879535559,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003855948263559563,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22024814784526825,
      "num_tokens": 78334357.0,
      "step": 37750
    },
    {
      "entropy": 5.704714393615722,
      "epoch": 3.424800435413643,
      "grad_norm": 1.328125,
      "learning_rate": 0.00038556664732672147,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2255471721291542,
      "num_tokens": 78344221.0,
      "step": 37755
    },
    {
      "entropy": 5.616919803619385,
      "epoch": 3.425253991291727,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038553846601094033,
      "loss": 4.8407,
      "mean_token_accuracy": 0.2266216903924942,
      "num_tokens": 78354097.0,
      "step": 37760
    },
    {
      "entropy": 5.7797502994537355,
      "epoch": 3.4257075471698113,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003855102824091954,
      "loss": 4.9633,
      "mean_token_accuracy": 0.22227668166160583,
      "num_tokens": 78363797.0,
      "step": 37765
    },
    {
      "entropy": 5.680646705627441,
      "epoch": 3.4261611030478956,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003854820965220697,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22004495859146117,
      "num_tokens": 78374084.0,
      "step": 37770
    },
    {
      "entropy": 5.6733451843261715,
      "epoch": 3.42661465892598,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038545390835014614,
      "loss": 4.8824,
      "mean_token_accuracy": 0.2188339501619339,
      "num_tokens": 78385346.0,
      "step": 37775
    },
    {
      "entropy": 5.651067686080933,
      "epoch": 3.4270682148040637,
      "grad_norm": 1.25,
      "learning_rate": 0.00038542571789400754,
      "loss": 4.864,
      "mean_token_accuracy": 0.22111764550209045,
      "num_tokens": 78395498.0,
      "step": 37780
    },
    {
      "entropy": 5.626645708084107,
      "epoch": 3.427521770682148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038539752515423704,
      "loss": 4.849,
      "mean_token_accuracy": 0.21533217132091523,
      "num_tokens": 78405792.0,
      "step": 37785
    },
    {
      "entropy": 5.718142509460449,
      "epoch": 3.427975326560232,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003853693301314175,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2163153573870659,
      "num_tokens": 78415672.0,
      "step": 37790
    },
    {
      "entropy": 5.639187955856324,
      "epoch": 3.4284288824383164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003853411328261321,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22452535331249238,
      "num_tokens": 78426937.0,
      "step": 37795
    },
    {
      "entropy": 5.645718145370483,
      "epoch": 3.4288824383164007,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000385312933238964,
      "loss": 4.925,
      "mean_token_accuracy": 0.2212074652314186,
      "num_tokens": 78436817.0,
      "step": 37800
    },
    {
      "entropy": 5.70794267654419,
      "epoch": 3.429335994194485,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003852847313704963,
      "loss": 4.9714,
      "mean_token_accuracy": 0.21779149174690246,
      "num_tokens": 78447348.0,
      "step": 37805
    },
    {
      "entropy": 5.69586215019226,
      "epoch": 3.429789550072569,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003852565272213123,
      "loss": 4.929,
      "mean_token_accuracy": 0.2188565492630005,
      "num_tokens": 78457631.0,
      "step": 37810
    },
    {
      "entropy": 5.609340572357178,
      "epoch": 3.430243105950653,
      "grad_norm": 1.25,
      "learning_rate": 0.0003852283207919951,
      "loss": 4.8273,
      "mean_token_accuracy": 0.2343819633126259,
      "num_tokens": 78468542.0,
      "step": 37815
    },
    {
      "entropy": 5.703023195266724,
      "epoch": 3.4306966618287373,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003852001120831282,
      "loss": 4.9321,
      "mean_token_accuracy": 0.22177557796239852,
      "num_tokens": 78478580.0,
      "step": 37820
    },
    {
      "entropy": 5.709153652191162,
      "epoch": 3.4311502177068216,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038517190109529497,
      "loss": 4.8988,
      "mean_token_accuracy": 0.2159486636519432,
      "num_tokens": 78488477.0,
      "step": 37825
    },
    {
      "entropy": 5.691554260253906,
      "epoch": 3.4316037735849054,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038514368782907864,
      "loss": 4.9301,
      "mean_token_accuracy": 0.2217624619603157,
      "num_tokens": 78498887.0,
      "step": 37830
    },
    {
      "entropy": 5.662042856216431,
      "epoch": 3.4320573294629897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038511547228506284,
      "loss": 4.8623,
      "mean_token_accuracy": 0.2234494522213936,
      "num_tokens": 78509030.0,
      "step": 37835
    },
    {
      "entropy": 5.554905223846435,
      "epoch": 3.432510885341074,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038508725446383097,
      "loss": 4.7697,
      "mean_token_accuracy": 0.2274718090891838,
      "num_tokens": 78518254.0,
      "step": 37840
    },
    {
      "entropy": 5.656516981124878,
      "epoch": 3.4329644412191582,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003850590343659667,
      "loss": 4.8695,
      "mean_token_accuracy": 0.22219008803367615,
      "num_tokens": 78527966.0,
      "step": 37845
    },
    {
      "entropy": 5.674332904815674,
      "epoch": 3.4334179970972425,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038503081199205353,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2196284279227257,
      "num_tokens": 78537932.0,
      "step": 37850
    },
    {
      "entropy": 5.74263334274292,
      "epoch": 3.4338715529753268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038500258734267505,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2177103042602539,
      "num_tokens": 78549213.0,
      "step": 37855
    },
    {
      "entropy": 5.641011810302734,
      "epoch": 3.4343251088534106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038497436041841517,
      "loss": 4.8238,
      "mean_token_accuracy": 0.2223459541797638,
      "num_tokens": 78558990.0,
      "step": 37860
    },
    {
      "entropy": 5.65340633392334,
      "epoch": 3.434778664731495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038494613121985735,
      "loss": 4.917,
      "mean_token_accuracy": 0.2153643101453781,
      "num_tokens": 78568496.0,
      "step": 37865
    },
    {
      "entropy": 5.631075143814087,
      "epoch": 3.435232220609579,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003849178997475856,
      "loss": 4.7822,
      "mean_token_accuracy": 0.23183700293302537,
      "num_tokens": 78578386.0,
      "step": 37870
    },
    {
      "entropy": 5.615643548965454,
      "epoch": 3.4356857764876634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038488966600218365,
      "loss": 4.8209,
      "mean_token_accuracy": 0.2239080235362053,
      "num_tokens": 78588855.0,
      "step": 37875
    },
    {
      "entropy": 5.6164830207824705,
      "epoch": 3.4361393323657476,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038486142998423535,
      "loss": 4.7809,
      "mean_token_accuracy": 0.23170846402645112,
      "num_tokens": 78599573.0,
      "step": 37880
    },
    {
      "entropy": 5.6793585300445555,
      "epoch": 3.4365928882438315,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003848331916943248,
      "loss": 4.9304,
      "mean_token_accuracy": 0.21864656805992128,
      "num_tokens": 78609793.0,
      "step": 37885
    },
    {
      "entropy": 5.578705596923828,
      "epoch": 3.4370464441219157,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003848049511330358,
      "loss": 4.837,
      "mean_token_accuracy": 0.2231828659772873,
      "num_tokens": 78619637.0,
      "step": 37890
    },
    {
      "entropy": 5.721876335144043,
      "epoch": 3.4375,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003847767083009525,
      "loss": 4.9121,
      "mean_token_accuracy": 0.2277653008699417,
      "num_tokens": 78629889.0,
      "step": 37895
    },
    {
      "entropy": 5.608948755264282,
      "epoch": 3.4379535558780843,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038474846319865876,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22613652050495148,
      "num_tokens": 78639853.0,
      "step": 37900
    },
    {
      "entropy": 5.628299045562744,
      "epoch": 3.4384071117561685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003847202158267389,
      "loss": 4.8392,
      "mean_token_accuracy": 0.2249119609594345,
      "num_tokens": 78649732.0,
      "step": 37905
    },
    {
      "entropy": 5.56564507484436,
      "epoch": 3.4388606676342524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038469196618577694,
      "loss": 4.7995,
      "mean_token_accuracy": 0.2232474684715271,
      "num_tokens": 78659794.0,
      "step": 37910
    },
    {
      "entropy": 5.746888160705566,
      "epoch": 3.4393142235123366,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003846637142763573,
      "loss": 4.9931,
      "mean_token_accuracy": 0.21394072622060775,
      "num_tokens": 78670526.0,
      "step": 37915
    },
    {
      "entropy": 5.689065074920654,
      "epoch": 3.439767779390421,
      "grad_norm": 1.265625,
      "learning_rate": 0.000384635460099064,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2263006865978241,
      "num_tokens": 78680177.0,
      "step": 37920
    },
    {
      "entropy": 5.715207958221436,
      "epoch": 3.440221335268505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038460720365448154,
      "loss": 5.0459,
      "mean_token_accuracy": 0.21093593984842302,
      "num_tokens": 78690958.0,
      "step": 37925
    },
    {
      "entropy": 5.725939559936523,
      "epoch": 3.4406748911465894,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003845789449431941,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21796528697013856,
      "num_tokens": 78700904.0,
      "step": 37930
    },
    {
      "entropy": 5.764712190628051,
      "epoch": 3.4411284470246732,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003845506839657862,
      "loss": 4.9793,
      "mean_token_accuracy": 0.21967338919639587,
      "num_tokens": 78712085.0,
      "step": 37935
    },
    {
      "entropy": 5.641454744338989,
      "epoch": 3.4415820029027575,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003845224207228422,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22283228486776352,
      "num_tokens": 78722161.0,
      "step": 37940
    },
    {
      "entropy": 5.709270811080932,
      "epoch": 3.4420355587808418,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038449415521494667,
      "loss": 4.889,
      "mean_token_accuracy": 0.21646807342767715,
      "num_tokens": 78732485.0,
      "step": 37945
    },
    {
      "entropy": 5.6481188297271725,
      "epoch": 3.442489114658926,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003844658874426841,
      "loss": 4.9095,
      "mean_token_accuracy": 0.21588062345981598,
      "num_tokens": 78744564.0,
      "step": 37950
    },
    {
      "entropy": 5.655069065093994,
      "epoch": 3.4429426705370103,
      "grad_norm": 1.125,
      "learning_rate": 0.000384437617406639,
      "loss": 4.883,
      "mean_token_accuracy": 0.22296066731214523,
      "num_tokens": 78755227.0,
      "step": 37955
    },
    {
      "entropy": 5.682131910324097,
      "epoch": 3.4433962264150946,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003844093451073962,
      "loss": 4.8729,
      "mean_token_accuracy": 0.21504031866788864,
      "num_tokens": 78765393.0,
      "step": 37960
    },
    {
      "entropy": 5.6994225025177006,
      "epoch": 3.4438497822931784,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038438107054554027,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22238041311502457,
      "num_tokens": 78775587.0,
      "step": 37965
    },
    {
      "entropy": 5.615812349319458,
      "epoch": 3.4443033381712627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003843527937216558,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23369012922048568,
      "num_tokens": 78785098.0,
      "step": 37970
    },
    {
      "entropy": 5.615220022201538,
      "epoch": 3.444756894049347,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003843245146363278,
      "loss": 4.849,
      "mean_token_accuracy": 0.22889381051063537,
      "num_tokens": 78795348.0,
      "step": 37975
    },
    {
      "entropy": 5.741587781906128,
      "epoch": 3.445210449927431,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038429623329014096,
      "loss": 4.9854,
      "mean_token_accuracy": 0.21596275717020036,
      "num_tokens": 78805348.0,
      "step": 37980
    },
    {
      "entropy": 5.638742113113404,
      "epoch": 3.445664005805515,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038426794968368015,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2226831004023552,
      "num_tokens": 78814765.0,
      "step": 37985
    },
    {
      "entropy": 5.7044840335845945,
      "epoch": 3.4461175616835993,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003842396638175303,
      "loss": 4.8624,
      "mean_token_accuracy": 0.2234186202287674,
      "num_tokens": 78824745.0,
      "step": 37990
    },
    {
      "entropy": 5.640507936477661,
      "epoch": 3.4465711175616836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003842113756922763,
      "loss": 4.9375,
      "mean_token_accuracy": 0.2211385264992714,
      "num_tokens": 78835101.0,
      "step": 37995
    },
    {
      "entropy": 5.733924150466919,
      "epoch": 3.447024673439768,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038418308530850333,
      "loss": 4.9633,
      "mean_token_accuracy": 0.2101848229765892,
      "num_tokens": 78845881.0,
      "step": 38000
    },
    {
      "entropy": 5.701048088073731,
      "epoch": 3.447478229317852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038415479266679624,
      "loss": 4.9052,
      "mean_token_accuracy": 0.21506670564413072,
      "num_tokens": 78856536.0,
      "step": 38005
    },
    {
      "entropy": 5.660458564758301,
      "epoch": 3.4479317851959363,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003841264977677403,
      "loss": 4.8863,
      "mean_token_accuracy": 0.2225690260529518,
      "num_tokens": 78866887.0,
      "step": 38010
    },
    {
      "entropy": 5.637609767913818,
      "epoch": 3.44838534107402,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038409820061192045,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22925398498773575,
      "num_tokens": 78877032.0,
      "step": 38015
    },
    {
      "entropy": 5.628288459777832,
      "epoch": 3.4488388969521044,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00038406990119992215,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22906812727451326,
      "num_tokens": 78887237.0,
      "step": 38020
    },
    {
      "entropy": 5.579012298583985,
      "epoch": 3.4492924528301887,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038404159953233047,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23342738747596742,
      "num_tokens": 78897436.0,
      "step": 38025
    },
    {
      "entropy": 5.691679525375366,
      "epoch": 3.449746008708273,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003840132956097307,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22103057950735092,
      "num_tokens": 78907691.0,
      "step": 38030
    },
    {
      "entropy": 5.653240060806274,
      "epoch": 3.450199564586357,
      "grad_norm": 1.25,
      "learning_rate": 0.00038398498943270816,
      "loss": 4.9157,
      "mean_token_accuracy": 0.22126582264900208,
      "num_tokens": 78917517.0,
      "step": 38035
    },
    {
      "entropy": 5.736886024475098,
      "epoch": 3.450653120464441,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003839566810018483,
      "loss": 4.8834,
      "mean_token_accuracy": 0.223101706802845,
      "num_tokens": 78927835.0,
      "step": 38040
    },
    {
      "entropy": 5.637302494049072,
      "epoch": 3.4511066763425253,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038392837031773656,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2215951919555664,
      "num_tokens": 78938853.0,
      "step": 38045
    },
    {
      "entropy": 5.627586698532104,
      "epoch": 3.4515602322206096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003839000573809583,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22822683453559875,
      "num_tokens": 78948508.0,
      "step": 38050
    },
    {
      "entropy": 5.6713666915893555,
      "epoch": 3.452013788098694,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038387174219209916,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22525650709867479,
      "num_tokens": 78959002.0,
      "step": 38055
    },
    {
      "entropy": 5.659831428527832,
      "epoch": 3.452467343976778,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003838434247517446,
      "loss": 4.878,
      "mean_token_accuracy": 0.2235659033060074,
      "num_tokens": 78970358.0,
      "step": 38060
    },
    {
      "entropy": 5.684469175338745,
      "epoch": 3.452920899854862,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003838151050604803,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23101768791675567,
      "num_tokens": 78979408.0,
      "step": 38065
    },
    {
      "entropy": 5.581966972351074,
      "epoch": 3.453374455732946,
      "grad_norm": 1.125,
      "learning_rate": 0.00038378678311889184,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2299179196357727,
      "num_tokens": 78990411.0,
      "step": 38070
    },
    {
      "entropy": 5.654430675506592,
      "epoch": 3.4538280116110305,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000383758458927565,
      "loss": 4.9477,
      "mean_token_accuracy": 0.22710520327091216,
      "num_tokens": 79001734.0,
      "step": 38075
    },
    {
      "entropy": 5.7028978824615475,
      "epoch": 3.4542815674891147,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003837301324870856,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22582774460315705,
      "num_tokens": 79011381.0,
      "step": 38080
    },
    {
      "entropy": 5.69885573387146,
      "epoch": 3.454735123367199,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038370180379803927,
      "loss": 4.9542,
      "mean_token_accuracy": 0.2189398929476738,
      "num_tokens": 79021124.0,
      "step": 38085
    },
    {
      "entropy": 5.710270977020263,
      "epoch": 3.455188679245283,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038367347286101197,
      "loss": 5.0073,
      "mean_token_accuracy": 0.20242873728275299,
      "num_tokens": 79033105.0,
      "step": 38090
    },
    {
      "entropy": 5.672910547256469,
      "epoch": 3.455642235123367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003836451396765896,
      "loss": 4.9173,
      "mean_token_accuracy": 0.2193543419241905,
      "num_tokens": 79043458.0,
      "step": 38095
    },
    {
      "entropy": 5.674237871170044,
      "epoch": 3.4560957910014514,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038361680424535795,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22289104461669923,
      "num_tokens": 79055186.0,
      "step": 38100
    },
    {
      "entropy": 5.689803218841552,
      "epoch": 3.4565493468795356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003835884665679032,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21835786551237107,
      "num_tokens": 79065743.0,
      "step": 38105
    },
    {
      "entropy": 5.683727693557739,
      "epoch": 3.45700290275762,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003835601266448112,
      "loss": 5.0002,
      "mean_token_accuracy": 0.20894202589988708,
      "num_tokens": 79077285.0,
      "step": 38110
    },
    {
      "entropy": 5.660084390640259,
      "epoch": 3.457456458635704,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003835317844766682,
      "loss": 4.8934,
      "mean_token_accuracy": 0.22260490655899048,
      "num_tokens": 79087885.0,
      "step": 38115
    },
    {
      "entropy": 5.714477682113648,
      "epoch": 3.457910014513788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038350344006406016,
      "loss": 4.9428,
      "mean_token_accuracy": 0.22318515926599503,
      "num_tokens": 79097907.0,
      "step": 38120
    },
    {
      "entropy": 5.756539726257325,
      "epoch": 3.4583635703918723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038347509340757343,
      "loss": 4.9987,
      "mean_token_accuracy": 0.2109343647956848,
      "num_tokens": 79108478.0,
      "step": 38125
    },
    {
      "entropy": 5.7112795352935795,
      "epoch": 3.4588171262699565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003834467445077941,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21225277781486512,
      "num_tokens": 79117988.0,
      "step": 38130
    },
    {
      "entropy": 5.722300815582275,
      "epoch": 3.459270682148041,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003834183933653084,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22180714309215546,
      "num_tokens": 79127786.0,
      "step": 38135
    },
    {
      "entropy": 5.656023550033569,
      "epoch": 3.4597242380261246,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003833900399807027,
      "loss": 4.8542,
      "mean_token_accuracy": 0.2294543206691742,
      "num_tokens": 79137775.0,
      "step": 38140
    },
    {
      "entropy": 5.593526315689087,
      "epoch": 3.460177793904209,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00038336168435456345,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2254769966006279,
      "num_tokens": 79147803.0,
      "step": 38145
    },
    {
      "entropy": 5.6052642345428465,
      "epoch": 3.460631349782293,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038333332648747684,
      "loss": 4.8092,
      "mean_token_accuracy": 0.22695891857147216,
      "num_tokens": 79158377.0,
      "step": 38150
    },
    {
      "entropy": 5.632893085479736,
      "epoch": 3.4610849056603774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003833049663800295,
      "loss": 4.8955,
      "mean_token_accuracy": 0.21827621906995773,
      "num_tokens": 79168514.0,
      "step": 38155
    },
    {
      "entropy": 5.713897705078125,
      "epoch": 3.4615384615384617,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003832766040328078,
      "loss": 5.0111,
      "mean_token_accuracy": 0.21202844381332397,
      "num_tokens": 79177588.0,
      "step": 38160
    },
    {
      "entropy": 5.667891025543213,
      "epoch": 3.461992017416546,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003832482394463984,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22351831793785096,
      "num_tokens": 79188128.0,
      "step": 38165
    },
    {
      "entropy": 5.61800971031189,
      "epoch": 3.4624455732946298,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003832198726213878,
      "loss": 4.8446,
      "mean_token_accuracy": 0.22358585000038148,
      "num_tokens": 79198928.0,
      "step": 38170
    },
    {
      "entropy": 5.588633298873901,
      "epoch": 3.462899129172714,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038319150355836266,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23361784368753433,
      "num_tokens": 79209751.0,
      "step": 38175
    },
    {
      "entropy": 5.6503729820251465,
      "epoch": 3.4633526850507983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038316313225790965,
      "loss": 4.9439,
      "mean_token_accuracy": 0.21502481400966644,
      "num_tokens": 79220210.0,
      "step": 38180
    },
    {
      "entropy": 5.6550792217254635,
      "epoch": 3.4638062409288826,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003831347587206155,
      "loss": 4.924,
      "mean_token_accuracy": 0.22276600897312165,
      "num_tokens": 79231530.0,
      "step": 38185
    },
    {
      "entropy": 5.633825445175171,
      "epoch": 3.4642597968069664,
      "grad_norm": 1.234375,
      "learning_rate": 0.000383106382947067,
      "loss": 4.8475,
      "mean_token_accuracy": 0.2198939487338066,
      "num_tokens": 79242771.0,
      "step": 38190
    },
    {
      "entropy": 5.6581581115722654,
      "epoch": 3.4647133526850507,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038307800493785093,
      "loss": 4.8528,
      "mean_token_accuracy": 0.22423576712608337,
      "num_tokens": 79253713.0,
      "step": 38195
    },
    {
      "entropy": 5.650691318511963,
      "epoch": 3.465166908563135,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038304962469355415,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2185642883181572,
      "num_tokens": 79264087.0,
      "step": 38200
    },
    {
      "entropy": 5.693933010101318,
      "epoch": 3.465620464441219,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003830212422147637,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22591497004032135,
      "num_tokens": 79274859.0,
      "step": 38205
    },
    {
      "entropy": 5.691009664535523,
      "epoch": 3.4660740203193035,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003829928575020663,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22495942264795304,
      "num_tokens": 79284778.0,
      "step": 38210
    },
    {
      "entropy": 5.708033752441406,
      "epoch": 3.4665275761973877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038296447055604916,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21413782387971877,
      "num_tokens": 79295034.0,
      "step": 38215
    },
    {
      "entropy": 5.628081941604615,
      "epoch": 3.4669811320754715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003829360813772992,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23043530881404878,
      "num_tokens": 79306013.0,
      "step": 38220
    },
    {
      "entropy": 5.679046535491944,
      "epoch": 3.467434687953556,
      "grad_norm": 1.125,
      "learning_rate": 0.00038290768996640366,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21370957642793656,
      "num_tokens": 79317590.0,
      "step": 38225
    },
    {
      "entropy": 5.718540954589844,
      "epoch": 3.46788824383164,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038287929632394957,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21658695191144944,
      "num_tokens": 79328669.0,
      "step": 38230
    },
    {
      "entropy": 5.581774139404297,
      "epoch": 3.4683417997097243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003828509004505241,
      "loss": 4.8088,
      "mean_token_accuracy": 0.23165071308612822,
      "num_tokens": 79339336.0,
      "step": 38235
    },
    {
      "entropy": 5.699062299728394,
      "epoch": 3.4687953555878086,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003828225023467146,
      "loss": 4.9507,
      "mean_token_accuracy": 0.21593518704175949,
      "num_tokens": 79350199.0,
      "step": 38240
    },
    {
      "entropy": 5.662779092788696,
      "epoch": 3.4692489114658924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003827941020131082,
      "loss": 4.8688,
      "mean_token_accuracy": 0.23303865492343903,
      "num_tokens": 79359906.0,
      "step": 38245
    },
    {
      "entropy": 5.639037466049194,
      "epoch": 3.4697024673439767,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038276569945029225,
      "loss": 4.8219,
      "mean_token_accuracy": 0.22069914042949676,
      "num_tokens": 79371395.0,
      "step": 38250
    },
    {
      "entropy": 5.668844509124756,
      "epoch": 3.470156023222061,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003827372946588543,
      "loss": 4.8839,
      "mean_token_accuracy": 0.2255866438150406,
      "num_tokens": 79381808.0,
      "step": 38255
    },
    {
      "entropy": 5.646842813491821,
      "epoch": 3.4706095791001452,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003827088876393816,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22579353600740432,
      "num_tokens": 79391906.0,
      "step": 38260
    },
    {
      "entropy": 5.644829654693604,
      "epoch": 3.4710631349782295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038268047839246165,
      "loss": 4.8448,
      "mean_token_accuracy": 0.22851358354091644,
      "num_tokens": 79402522.0,
      "step": 38265
    },
    {
      "entropy": 5.612664461135864,
      "epoch": 3.4715166908563133,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038265206691868195,
      "loss": 4.8068,
      "mean_token_accuracy": 0.22028758525848388,
      "num_tokens": 79412751.0,
      "step": 38270
    },
    {
      "entropy": 5.731999826431275,
      "epoch": 3.4719702467343976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003826236532186301,
      "loss": 4.9825,
      "mean_token_accuracy": 0.2159061998128891,
      "num_tokens": 79424512.0,
      "step": 38275
    },
    {
      "entropy": 5.798962831497192,
      "epoch": 3.472423802612482,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003825952372928936,
      "loss": 5.0673,
      "mean_token_accuracy": 0.20833854377269745,
      "num_tokens": 79434082.0,
      "step": 38280
    },
    {
      "entropy": 5.72457365989685,
      "epoch": 3.472877358490566,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038256681914206026,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22173282951116563,
      "num_tokens": 79444469.0,
      "step": 38285
    },
    {
      "entropy": 5.6188617706298825,
      "epoch": 3.4733309143686504,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038253839876671765,
      "loss": 4.8831,
      "mean_token_accuracy": 0.21797277331352233,
      "num_tokens": 79455036.0,
      "step": 38290
    },
    {
      "entropy": 5.669348812103271,
      "epoch": 3.473784470246734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003825099761674535,
      "loss": 4.8824,
      "mean_token_accuracy": 0.21737934201955794,
      "num_tokens": 79464600.0,
      "step": 38295
    },
    {
      "entropy": 5.6810894966125485,
      "epoch": 3.4742380261248185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003824815513448558,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22153210937976836,
      "num_tokens": 79474912.0,
      "step": 38300
    },
    {
      "entropy": 5.563897562026978,
      "epoch": 3.4746915820029027,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038245312429951214,
      "loss": 4.7714,
      "mean_token_accuracy": 0.22737380862236023,
      "num_tokens": 79485283.0,
      "step": 38305
    },
    {
      "entropy": 5.535518169403076,
      "epoch": 3.475145137880987,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003824246950320105,
      "loss": 4.7635,
      "mean_token_accuracy": 0.22254651486873628,
      "num_tokens": 79495323.0,
      "step": 38310
    },
    {
      "entropy": 5.658719301223755,
      "epoch": 3.4755986937590713,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003823962635429388,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22086288928985595,
      "num_tokens": 79505144.0,
      "step": 38315
    },
    {
      "entropy": 5.738939714431763,
      "epoch": 3.4760522496371555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003823678298328849,
      "loss": 4.9417,
      "mean_token_accuracy": 0.21781156063079835,
      "num_tokens": 79515402.0,
      "step": 38320
    },
    {
      "entropy": 5.6894818305969235,
      "epoch": 3.4765058055152394,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038233939390243697,
      "loss": 4.8169,
      "mean_token_accuracy": 0.23597838580608368,
      "num_tokens": 79525528.0,
      "step": 38325
    },
    {
      "entropy": 5.688918972015381,
      "epoch": 3.4769593613933236,
      "grad_norm": 1.25,
      "learning_rate": 0.00038231095575218303,
      "loss": 4.9114,
      "mean_token_accuracy": 0.22160641849040985,
      "num_tokens": 79535991.0,
      "step": 38330
    },
    {
      "entropy": 5.715981817245483,
      "epoch": 3.477412917271408,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003822825153827112,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21177517473697663,
      "num_tokens": 79545890.0,
      "step": 38335
    },
    {
      "entropy": 5.703035926818847,
      "epoch": 3.477866473149492,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038225407279460964,
      "loss": 4.9491,
      "mean_token_accuracy": 0.22066331803798675,
      "num_tokens": 79556363.0,
      "step": 38340
    },
    {
      "entropy": 5.682284927368164,
      "epoch": 3.478320029027576,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003822256279884665,
      "loss": 4.9008,
      "mean_token_accuracy": 0.22384336739778518,
      "num_tokens": 79566924.0,
      "step": 38345
    },
    {
      "entropy": 5.6439543724060055,
      "epoch": 3.4787735849056602,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038219718096487007,
      "loss": 4.8803,
      "mean_token_accuracy": 0.22242375314235688,
      "num_tokens": 79577322.0,
      "step": 38350
    },
    {
      "entropy": 5.771783685684204,
      "epoch": 3.4792271407837445,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038216873172440855,
      "loss": 4.9957,
      "mean_token_accuracy": 0.21211498528718947,
      "num_tokens": 79588436.0,
      "step": 38355
    },
    {
      "entropy": 5.746347904205322,
      "epoch": 3.4796806966618288,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003821402802676704,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22331545054912566,
      "num_tokens": 79599015.0,
      "step": 38360
    },
    {
      "entropy": 5.659091997146606,
      "epoch": 3.480134252539913,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003821118265952439,
      "loss": 4.8596,
      "mean_token_accuracy": 0.22707050442695617,
      "num_tokens": 79608443.0,
      "step": 38365
    },
    {
      "entropy": 5.616668748855591,
      "epoch": 3.4805878084179973,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003820833707077176,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2212822198867798,
      "num_tokens": 79619010.0,
      "step": 38370
    },
    {
      "entropy": 5.695712041854859,
      "epoch": 3.481041364296081,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003820549126056799,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22206550538539888,
      "num_tokens": 79629665.0,
      "step": 38375
    },
    {
      "entropy": 5.594873237609863,
      "epoch": 3.4814949201741654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003820264522897193,
      "loss": 4.7781,
      "mean_token_accuracy": 0.22759151458740234,
      "num_tokens": 79640578.0,
      "step": 38380
    },
    {
      "entropy": 5.691264009475708,
      "epoch": 3.4819484760522497,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003819979897604243,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2166321322321892,
      "num_tokens": 79649576.0,
      "step": 38385
    },
    {
      "entropy": 5.664264917373657,
      "epoch": 3.482402031930334,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038196952501838376,
      "loss": 4.9115,
      "mean_token_accuracy": 0.21286836564540862,
      "num_tokens": 79659926.0,
      "step": 38390
    },
    {
      "entropy": 5.723902702331543,
      "epoch": 3.4828555878084178,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00038194105806418617,
      "loss": 5.0232,
      "mean_token_accuracy": 0.2100686937570572,
      "num_tokens": 79669969.0,
      "step": 38395
    },
    {
      "entropy": 5.664978265762329,
      "epoch": 3.483309143686502,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003819125888984202,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22419894486665726,
      "num_tokens": 79679706.0,
      "step": 38400
    },
    {
      "entropy": 5.646444606781006,
      "epoch": 3.4837626995645863,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038188411752167466,
      "loss": 4.877,
      "mean_token_accuracy": 0.21980006396770477,
      "num_tokens": 79691543.0,
      "step": 38405
    },
    {
      "entropy": 5.677611494064331,
      "epoch": 3.4842162554426706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003818556439345383,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22034358084201813,
      "num_tokens": 79701657.0,
      "step": 38410
    },
    {
      "entropy": 5.646751499176025,
      "epoch": 3.484669811320755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038182716813760004,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22066344022750856,
      "num_tokens": 79711615.0,
      "step": 38415
    },
    {
      "entropy": 5.660435485839844,
      "epoch": 3.485123367198839,
      "grad_norm": 1.25,
      "learning_rate": 0.0003817986901314487,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22642081826925278,
      "num_tokens": 79721427.0,
      "step": 38420
    },
    {
      "entropy": 5.666409206390381,
      "epoch": 3.485576923076923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003817702099166732,
      "loss": 4.9175,
      "mean_token_accuracy": 0.22145477831363677,
      "num_tokens": 79732565.0,
      "step": 38425
    },
    {
      "entropy": 5.616452550888061,
      "epoch": 3.486030478955007,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003817417274938626,
      "loss": 4.7603,
      "mean_token_accuracy": 0.22668077945709228,
      "num_tokens": 79744496.0,
      "step": 38430
    },
    {
      "entropy": 5.6936869621276855,
      "epoch": 3.4864840348330914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003817132428636058,
      "loss": 4.963,
      "mean_token_accuracy": 0.2191098153591156,
      "num_tokens": 79755626.0,
      "step": 38435
    },
    {
      "entropy": 5.646067667007446,
      "epoch": 3.4869375907111757,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038168475602649203,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22019525170326232,
      "num_tokens": 79765447.0,
      "step": 38440
    },
    {
      "entropy": 5.652263832092285,
      "epoch": 3.48739114658926,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003816562669831103,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22075747698545456,
      "num_tokens": 79776047.0,
      "step": 38445
    },
    {
      "entropy": 5.672064399719238,
      "epoch": 3.487844702467344,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038162777573404975,
      "loss": 4.8574,
      "mean_token_accuracy": 0.22839194983243943,
      "num_tokens": 79786061.0,
      "step": 38450
    },
    {
      "entropy": 5.633407020568848,
      "epoch": 3.488298258345428,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038159928227989964,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2193361207842827,
      "num_tokens": 79796844.0,
      "step": 38455
    },
    {
      "entropy": 5.659973287582398,
      "epoch": 3.4887518142235123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003815707866212493,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22897655367851258,
      "num_tokens": 79808120.0,
      "step": 38460
    },
    {
      "entropy": 5.6969414234161375,
      "epoch": 3.4892053701015966,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003815422887586878,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22686137557029723,
      "num_tokens": 79818901.0,
      "step": 38465
    },
    {
      "entropy": 5.715629959106446,
      "epoch": 3.489658925979681,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003815137886928047,
      "loss": 4.9263,
      "mean_token_accuracy": 0.21924593448638915,
      "num_tokens": 79829090.0,
      "step": 38470
    },
    {
      "entropy": 5.747639989852905,
      "epoch": 3.490112481857765,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003814852864241892,
      "loss": 4.9445,
      "mean_token_accuracy": 0.2187946319580078,
      "num_tokens": 79838751.0,
      "step": 38475
    },
    {
      "entropy": 5.621283102035522,
      "epoch": 3.490566037735849,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038145678195343095,
      "loss": 4.9909,
      "mean_token_accuracy": 0.20928405970335007,
      "num_tokens": 79849410.0,
      "step": 38480
    },
    {
      "entropy": 5.754877376556396,
      "epoch": 3.491019593613933,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003814282752811193,
      "loss": 4.9221,
      "mean_token_accuracy": 0.21881479322910308,
      "num_tokens": 79859279.0,
      "step": 38485
    },
    {
      "entropy": 5.727818393707276,
      "epoch": 3.4914731494920175,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003813997664078438,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22467607706785203,
      "num_tokens": 79869535.0,
      "step": 38490
    },
    {
      "entropy": 5.725682544708252,
      "epoch": 3.4919267053701017,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038137125533419403,
      "loss": 5.0425,
      "mean_token_accuracy": 0.20266993045806886,
      "num_tokens": 79881396.0,
      "step": 38495
    },
    {
      "entropy": 5.651137638092041,
      "epoch": 3.4923802612481856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003813427420607596,
      "loss": 4.8462,
      "mean_token_accuracy": 0.21909057796001435,
      "num_tokens": 79891250.0,
      "step": 38500
    },
    {
      "entropy": 5.656732130050659,
      "epoch": 3.49283381712627,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038131422658813004,
      "loss": 4.8419,
      "mean_token_accuracy": 0.221185539662838,
      "num_tokens": 79901982.0,
      "step": 38505
    },
    {
      "entropy": 5.659779977798462,
      "epoch": 3.493287373004354,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038128570891689527,
      "loss": 4.959,
      "mean_token_accuracy": 0.21513214707374573,
      "num_tokens": 79912480.0,
      "step": 38510
    },
    {
      "entropy": 5.68430643081665,
      "epoch": 3.4937409288824384,
      "grad_norm": 1.15625,
      "learning_rate": 0.000381257189047645,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22824358940124512,
      "num_tokens": 79923793.0,
      "step": 38515
    },
    {
      "entropy": 5.715128993988037,
      "epoch": 3.4941944847605226,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038122866698096894,
      "loss": 4.9058,
      "mean_token_accuracy": 0.2185749128460884,
      "num_tokens": 79933858.0,
      "step": 38520
    },
    {
      "entropy": 5.745144367218018,
      "epoch": 3.494648040638607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038120014271745693,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2169090613722801,
      "num_tokens": 79944093.0,
      "step": 38525
    },
    {
      "entropy": 5.700724506378174,
      "epoch": 3.4951015965166907,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000381171616257699,
      "loss": 4.9193,
      "mean_token_accuracy": 0.21582674235105515,
      "num_tokens": 79955072.0,
      "step": 38530
    },
    {
      "entropy": 5.70296630859375,
      "epoch": 3.495555152394775,
      "grad_norm": 1.125,
      "learning_rate": 0.00038114308760228493,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22115064561367034,
      "num_tokens": 79966949.0,
      "step": 38535
    },
    {
      "entropy": 5.631480312347412,
      "epoch": 3.4960087082728593,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003811145567518048,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2172643154859543,
      "num_tokens": 79977109.0,
      "step": 38540
    },
    {
      "entropy": 5.709797334671021,
      "epoch": 3.4964622641509435,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038108602370684854,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2155863016843796,
      "num_tokens": 79987094.0,
      "step": 38545
    },
    {
      "entropy": 5.702955818176269,
      "epoch": 3.4969158200290273,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003810574884680063,
      "loss": 4.995,
      "mean_token_accuracy": 0.21914195418357849,
      "num_tokens": 79997503.0,
      "step": 38550
    },
    {
      "entropy": 5.673231792449951,
      "epoch": 3.4973693759071116,
      "grad_norm": 1.40625,
      "learning_rate": 0.00038102895103586817,
      "loss": 4.9113,
      "mean_token_accuracy": 0.21981395035982132,
      "num_tokens": 80007745.0,
      "step": 38555
    },
    {
      "entropy": 5.719802761077881,
      "epoch": 3.497822931785196,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003810004114110243,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21505439430475234,
      "num_tokens": 80019330.0,
      "step": 38560
    },
    {
      "entropy": 5.706080722808838,
      "epoch": 3.49827648766328,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003809718695940649,
      "loss": 4.9502,
      "mean_token_accuracy": 0.21767064929008484,
      "num_tokens": 80029732.0,
      "step": 38565
    },
    {
      "entropy": 5.66891884803772,
      "epoch": 3.4987300435413644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003809433255855803,
      "loss": 4.8002,
      "mean_token_accuracy": 0.2267008662223816,
      "num_tokens": 80041052.0,
      "step": 38570
    },
    {
      "entropy": 5.686438274383545,
      "epoch": 3.4991835994194487,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038091477938616075,
      "loss": 4.9296,
      "mean_token_accuracy": 0.2250382795929909,
      "num_tokens": 80051929.0,
      "step": 38575
    },
    {
      "entropy": 5.723323011398316,
      "epoch": 3.4996371552975325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038088623099639653,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21618833839893342,
      "num_tokens": 80061153.0,
      "step": 38580
    },
    {
      "entropy": 5.66921215057373,
      "epoch": 3.5000907111756168,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000380857680416878,
      "loss": 4.8902,
      "mean_token_accuracy": 0.2197910487651825,
      "num_tokens": 80070560.0,
      "step": 38585
    },
    {
      "entropy": 5.803826189041137,
      "epoch": 3.500544267053701,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003808291276481957,
      "loss": 4.9414,
      "mean_token_accuracy": 0.21513917446136474,
      "num_tokens": 80081141.0,
      "step": 38590
    },
    {
      "entropy": 5.634916496276856,
      "epoch": 3.5009978229317853,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003808005726909401,
      "loss": 4.8067,
      "mean_token_accuracy": 0.2263699784874916,
      "num_tokens": 80092114.0,
      "step": 38595
    },
    {
      "entropy": 5.621751689910889,
      "epoch": 3.501451378809869,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038077201554570175,
      "loss": 4.8442,
      "mean_token_accuracy": 0.221420356631279,
      "num_tokens": 80101966.0,
      "step": 38600
    },
    {
      "entropy": 5.71883192062378,
      "epoch": 3.5019049346879534,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003807434562130711,
      "loss": 5.0007,
      "mean_token_accuracy": 0.22369088381528854,
      "num_tokens": 80113420.0,
      "step": 38605
    },
    {
      "entropy": 5.660939741134643,
      "epoch": 3.5023584905660377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003807148946936387,
      "loss": 4.9159,
      "mean_token_accuracy": 0.21422133296728135,
      "num_tokens": 80123398.0,
      "step": 38610
    },
    {
      "entropy": 5.690362071990966,
      "epoch": 3.502812046444122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003806863309879956,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22846689671278,
      "num_tokens": 80133885.0,
      "step": 38615
    },
    {
      "entropy": 5.656251716613769,
      "epoch": 3.503265602322206,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003806577650967321,
      "loss": 4.806,
      "mean_token_accuracy": 0.24102170765399933,
      "num_tokens": 80144168.0,
      "step": 38620
    },
    {
      "entropy": 5.607918310165405,
      "epoch": 3.5037191582002905,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003806291970204391,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2215416431427002,
      "num_tokens": 80153956.0,
      "step": 38625
    },
    {
      "entropy": 5.707968378067017,
      "epoch": 3.5041727140783747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00038060062675970743,
      "loss": 4.9339,
      "mean_token_accuracy": 0.21386744678020478,
      "num_tokens": 80163617.0,
      "step": 38630
    },
    {
      "entropy": 5.672884321212768,
      "epoch": 3.5046262699564585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003805720543151278,
      "loss": 4.8323,
      "mean_token_accuracy": 0.22490372210741044,
      "num_tokens": 80174049.0,
      "step": 38635
    },
    {
      "entropy": 5.733846187591553,
      "epoch": 3.505079825834543,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003805434796872912,
      "loss": 4.9292,
      "mean_token_accuracy": 0.2187029480934143,
      "num_tokens": 80185043.0,
      "step": 38640
    },
    {
      "entropy": 5.62802529335022,
      "epoch": 3.505533381712627,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038051490287678856,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22246883660554886,
      "num_tokens": 80194810.0,
      "step": 38645
    },
    {
      "entropy": 5.641334533691406,
      "epoch": 3.5059869375907113,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003804863238842109,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2262333407998085,
      "num_tokens": 80204861.0,
      "step": 38650
    },
    {
      "entropy": 5.672775840759277,
      "epoch": 3.506440493468795,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003804577427101491,
      "loss": 4.8023,
      "mean_token_accuracy": 0.22865131199359895,
      "num_tokens": 80214447.0,
      "step": 38655
    },
    {
      "entropy": 5.643443918228149,
      "epoch": 3.5068940493468794,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003804291593551944,
      "loss": 4.8174,
      "mean_token_accuracy": 0.22229413092136383,
      "num_tokens": 80224213.0,
      "step": 38660
    },
    {
      "entropy": 5.665979814529419,
      "epoch": 3.5073476052249637,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003804005738199377,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22018919736146927,
      "num_tokens": 80234161.0,
      "step": 38665
    },
    {
      "entropy": 5.690096282958985,
      "epoch": 3.507801161103048,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038037198610497037,
      "loss": 4.8705,
      "mean_token_accuracy": 0.2260594755411148,
      "num_tokens": 80243788.0,
      "step": 38670
    },
    {
      "entropy": 5.684653568267822,
      "epoch": 3.5082547169811322,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003803433962108835,
      "loss": 4.893,
      "mean_token_accuracy": 0.22167880535125734,
      "num_tokens": 80254658.0,
      "step": 38675
    },
    {
      "entropy": 5.716538619995117,
      "epoch": 3.5087082728592165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003803148041382683,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22145241796970366,
      "num_tokens": 80264820.0,
      "step": 38680
    },
    {
      "entropy": 5.613696813583374,
      "epoch": 3.5091618287373003,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038028620988771617,
      "loss": 4.8148,
      "mean_token_accuracy": 0.2304685264825821,
      "num_tokens": 80274639.0,
      "step": 38685
    },
    {
      "entropy": 5.631649446487427,
      "epoch": 3.5096153846153846,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003802576134598183,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22631779164075852,
      "num_tokens": 80284542.0,
      "step": 38690
    },
    {
      "entropy": 5.614180374145508,
      "epoch": 3.510068940493469,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00038022901485516625,
      "loss": 4.8129,
      "mean_token_accuracy": 0.22916811406612397,
      "num_tokens": 80294405.0,
      "step": 38695
    },
    {
      "entropy": 5.662906742095947,
      "epoch": 3.510522496371553,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038020041407435133,
      "loss": 4.8527,
      "mean_token_accuracy": 0.21961673498153686,
      "num_tokens": 80305326.0,
      "step": 38700
    },
    {
      "entropy": 5.647483682632446,
      "epoch": 3.510976052249637,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038017181111796506,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22607609182596206,
      "num_tokens": 80315423.0,
      "step": 38705
    },
    {
      "entropy": 5.672399950027466,
      "epoch": 3.511429608127721,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003801432059865989,
      "loss": 4.9268,
      "mean_token_accuracy": 0.21442493051290512,
      "num_tokens": 80325372.0,
      "step": 38710
    },
    {
      "entropy": 5.69990553855896,
      "epoch": 3.5118831640058055,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038011459868084446,
      "loss": 4.8052,
      "mean_token_accuracy": 0.22400895953178407,
      "num_tokens": 80335467.0,
      "step": 38715
    },
    {
      "entropy": 5.6629026412963865,
      "epoch": 3.5123367198838897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003800859892012933,
      "loss": 4.9105,
      "mean_token_accuracy": 0.21953169852495194,
      "num_tokens": 80345420.0,
      "step": 38720
    },
    {
      "entropy": 5.63311996459961,
      "epoch": 3.512790275761974,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003800573775485372,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22780140936374665,
      "num_tokens": 80354369.0,
      "step": 38725
    },
    {
      "entropy": 5.685469150543213,
      "epoch": 3.5132438316400583,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038002876372316774,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22196616232395172,
      "num_tokens": 80365292.0,
      "step": 38730
    },
    {
      "entropy": 5.726894950866699,
      "epoch": 3.513697387518142,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038000014772577657,
      "loss": 4.918,
      "mean_token_accuracy": 0.21927641481161117,
      "num_tokens": 80376340.0,
      "step": 38735
    },
    {
      "entropy": 5.638172245025634,
      "epoch": 3.5141509433962264,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003799715295569557,
      "loss": 4.8575,
      "mean_token_accuracy": 0.21917906701564788,
      "num_tokens": 80386193.0,
      "step": 38740
    },
    {
      "entropy": 5.803388214111328,
      "epoch": 3.5146044992743106,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003799429092172968,
      "loss": 5.052,
      "mean_token_accuracy": 0.21366066634654998,
      "num_tokens": 80397055.0,
      "step": 38745
    },
    {
      "entropy": 5.6963152408599855,
      "epoch": 3.515058055152395,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003799142867073918,
      "loss": 4.9553,
      "mean_token_accuracy": 0.21318659633398057,
      "num_tokens": 80407375.0,
      "step": 38750
    },
    {
      "entropy": 5.695281171798706,
      "epoch": 3.5155116110304787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037988566202783256,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2149920091032982,
      "num_tokens": 80417942.0,
      "step": 38755
    },
    {
      "entropy": 5.6151909828186035,
      "epoch": 3.515965166908563,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003798570351792112,
      "loss": 4.792,
      "mean_token_accuracy": 0.22209582030773162,
      "num_tokens": 80428716.0,
      "step": 38760
    },
    {
      "entropy": 5.665998840332032,
      "epoch": 3.5164187227866472,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037982840616211964,
      "loss": 4.94,
      "mean_token_accuracy": 0.2173798128962517,
      "num_tokens": 80439688.0,
      "step": 38765
    },
    {
      "entropy": 5.7480724334716795,
      "epoch": 3.5168722786647315,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037979977497714984,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21703584641218185,
      "num_tokens": 80450434.0,
      "step": 38770
    },
    {
      "entropy": 5.682918214797974,
      "epoch": 3.5173258345428158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037977114162489404,
      "loss": 4.9679,
      "mean_token_accuracy": 0.21161040663719177,
      "num_tokens": 80462187.0,
      "step": 38775
    },
    {
      "entropy": 5.665341329574585,
      "epoch": 3.5177793904209,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003797425061059443,
      "loss": 4.8507,
      "mean_token_accuracy": 0.21997659802436828,
      "num_tokens": 80473049.0,
      "step": 38780
    },
    {
      "entropy": 5.636114549636841,
      "epoch": 3.5182329462989843,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037971386842089287,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22613928020000457,
      "num_tokens": 80484119.0,
      "step": 38785
    },
    {
      "entropy": 5.631565952301026,
      "epoch": 3.518686502177068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003796852285703319,
      "loss": 4.7827,
      "mean_token_accuracy": 0.23132196813821793,
      "num_tokens": 80494882.0,
      "step": 38790
    },
    {
      "entropy": 5.685667562484741,
      "epoch": 3.5191400580551524,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003796565865548538,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22588509172201157,
      "num_tokens": 80505362.0,
      "step": 38795
    },
    {
      "entropy": 5.672517156600952,
      "epoch": 3.5195936139332367,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003796279423750508,
      "loss": 4.9399,
      "mean_token_accuracy": 0.2148992270231247,
      "num_tokens": 80516027.0,
      "step": 38800
    },
    {
      "entropy": 5.721514797210693,
      "epoch": 3.5200471698113205,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003795992960315153,
      "loss": 4.9437,
      "mean_token_accuracy": 0.21949270963668824,
      "num_tokens": 80527045.0,
      "step": 38805
    },
    {
      "entropy": 5.659644079208374,
      "epoch": 3.5205007256894048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037957064752483966,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23111877888441085,
      "num_tokens": 80537373.0,
      "step": 38810
    },
    {
      "entropy": 5.631891870498658,
      "epoch": 3.520954281567489,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003795419968556164,
      "loss": 4.8972,
      "mean_token_accuracy": 0.21646153330802917,
      "num_tokens": 80546632.0,
      "step": 38815
    },
    {
      "entropy": 5.641005325317383,
      "epoch": 3.5214078374455733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003795133440244381,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22392928302288057,
      "num_tokens": 80556694.0,
      "step": 38820
    },
    {
      "entropy": 5.7098747253417965,
      "epoch": 3.5218613933236576,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003794846890318971,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21654485911130905,
      "num_tokens": 80567834.0,
      "step": 38825
    },
    {
      "entropy": 5.671838569641113,
      "epoch": 3.522314949201742,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003794560318785861,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22473381608724594,
      "num_tokens": 80578299.0,
      "step": 38830
    },
    {
      "entropy": 5.694794654846191,
      "epoch": 3.522768505079826,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037942737256509786,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21924097537994386,
      "num_tokens": 80588620.0,
      "step": 38835
    },
    {
      "entropy": 5.810879468917847,
      "epoch": 3.52322206095791,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003793987110920248,
      "loss": 5.0798,
      "mean_token_accuracy": 0.20884082615375518,
      "num_tokens": 80599076.0,
      "step": 38840
    },
    {
      "entropy": 5.746135425567627,
      "epoch": 3.523675616835994,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037937004745995987,
      "loss": 5.0041,
      "mean_token_accuracy": 0.2127861872315407,
      "num_tokens": 80609440.0,
      "step": 38845
    },
    {
      "entropy": 5.785991191864014,
      "epoch": 3.5241291727140784,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003793413816694957,
      "loss": 4.9733,
      "mean_token_accuracy": 0.2168283700942993,
      "num_tokens": 80619169.0,
      "step": 38850
    },
    {
      "entropy": 5.6547845840454105,
      "epoch": 3.5245827285921627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003793127137212252,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2217233031988144,
      "num_tokens": 80628719.0,
      "step": 38855
    },
    {
      "entropy": 5.662923288345337,
      "epoch": 3.5250362844702465,
      "grad_norm": 1.375,
      "learning_rate": 0.00037928404361574125,
      "loss": 4.8726,
      "mean_token_accuracy": 0.21649819761514663,
      "num_tokens": 80638391.0,
      "step": 38860
    },
    {
      "entropy": 5.6737288475036625,
      "epoch": 3.525489840348331,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003792553713536367,
      "loss": 4.9495,
      "mean_token_accuracy": 0.22030826508998871,
      "num_tokens": 80647720.0,
      "step": 38865
    },
    {
      "entropy": 5.61060266494751,
      "epoch": 3.525943396226415,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003792266969355044,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22741675972938538,
      "num_tokens": 80658841.0,
      "step": 38870
    },
    {
      "entropy": 5.674912166595459,
      "epoch": 3.5263969521044993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037919802036193753,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2227904513478279,
      "num_tokens": 80668735.0,
      "step": 38875
    },
    {
      "entropy": 5.700613927841187,
      "epoch": 3.5268505079825836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037916934163352903,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2249739274382591,
      "num_tokens": 80680757.0,
      "step": 38880
    },
    {
      "entropy": 5.661359643936157,
      "epoch": 3.527304063860668,
      "grad_norm": 1.359375,
      "learning_rate": 0.000379140660750872,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2209099993109703,
      "num_tokens": 80689833.0,
      "step": 38885
    },
    {
      "entropy": 5.6168828964233395,
      "epoch": 3.5277576197387517,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003791119777145595,
      "loss": 4.8267,
      "mean_token_accuracy": 0.22727351784706115,
      "num_tokens": 80700347.0,
      "step": 38890
    },
    {
      "entropy": 5.542741203308106,
      "epoch": 3.528211175616836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003790832925251849,
      "loss": 4.7408,
      "mean_token_accuracy": 0.2413154050707817,
      "num_tokens": 80710559.0,
      "step": 38895
    },
    {
      "entropy": 5.66308560371399,
      "epoch": 3.52866473149492,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003790546051833412,
      "loss": 4.9369,
      "mean_token_accuracy": 0.2140858381986618,
      "num_tokens": 80720805.0,
      "step": 38900
    },
    {
      "entropy": 5.696552324295044,
      "epoch": 3.5291182873730045,
      "grad_norm": 1.25,
      "learning_rate": 0.0003790259156896217,
      "loss": 4.9481,
      "mean_token_accuracy": 0.20508608520030974,
      "num_tokens": 80730937.0,
      "step": 38905
    },
    {
      "entropy": 5.695327806472778,
      "epoch": 3.5295718432510883,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037899722404461983,
      "loss": 4.9083,
      "mean_token_accuracy": 0.22350073456764222,
      "num_tokens": 80741597.0,
      "step": 38910
    },
    {
      "entropy": 5.709546041488648,
      "epoch": 3.5300253991291726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003789685302489289,
      "loss": 4.9044,
      "mean_token_accuracy": 0.21568674594163895,
      "num_tokens": 80752496.0,
      "step": 38915
    },
    {
      "entropy": 5.6948473930358885,
      "epoch": 3.530478955007257,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037893983430314226,
      "loss": 4.8814,
      "mean_token_accuracy": 0.2207223042845726,
      "num_tokens": 80762329.0,
      "step": 38920
    },
    {
      "entropy": 5.666759967803955,
      "epoch": 3.530932510885341,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037891113620785327,
      "loss": 4.9831,
      "mean_token_accuracy": 0.2131723403930664,
      "num_tokens": 80773152.0,
      "step": 38925
    },
    {
      "entropy": 5.665398406982422,
      "epoch": 3.5313860667634254,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037888243596365557,
      "loss": 4.8963,
      "mean_token_accuracy": 0.21958148330450059,
      "num_tokens": 80783502.0,
      "step": 38930
    },
    {
      "entropy": 5.690732192993164,
      "epoch": 3.5318396226415096,
      "grad_norm": 1.25,
      "learning_rate": 0.00037885373357114265,
      "loss": 4.8078,
      "mean_token_accuracy": 0.22268866896629333,
      "num_tokens": 80792644.0,
      "step": 38935
    },
    {
      "entropy": 5.671945571899414,
      "epoch": 3.5322931785195935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000378825029030908,
      "loss": 4.9508,
      "mean_token_accuracy": 0.21928037106990814,
      "num_tokens": 80803184.0,
      "step": 38940
    },
    {
      "entropy": 5.637017059326172,
      "epoch": 3.5327467343976777,
      "grad_norm": 1.25,
      "learning_rate": 0.0003787963223435453,
      "loss": 4.8982,
      "mean_token_accuracy": 0.21587810814380645,
      "num_tokens": 80813353.0,
      "step": 38945
    },
    {
      "entropy": 5.671491622924805,
      "epoch": 3.533200290275762,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003787676135096482,
      "loss": 4.9771,
      "mean_token_accuracy": 0.21644519716501237,
      "num_tokens": 80824663.0,
      "step": 38950
    },
    {
      "entropy": 5.791043376922607,
      "epoch": 3.5336538461538463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003787389025298104,
      "loss": 5.0117,
      "mean_token_accuracy": 0.2102920800447464,
      "num_tokens": 80835507.0,
      "step": 38955
    },
    {
      "entropy": 5.682917308807373,
      "epoch": 3.53410740203193,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003787101894046257,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23192470222711564,
      "num_tokens": 80845035.0,
      "step": 38960
    },
    {
      "entropy": 5.685983037948608,
      "epoch": 3.5345609579100143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037868147413468785,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2210586339235306,
      "num_tokens": 80854992.0,
      "step": 38965
    },
    {
      "entropy": 5.719517230987549,
      "epoch": 3.5350145137880986,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003786527567205907,
      "loss": 4.9943,
      "mean_token_accuracy": 0.21078003644943238,
      "num_tokens": 80864658.0,
      "step": 38970
    },
    {
      "entropy": 5.722558450698853,
      "epoch": 3.535468069666183,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037862403716292805,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22370062321424483,
      "num_tokens": 80873964.0,
      "step": 38975
    },
    {
      "entropy": 5.701463747024536,
      "epoch": 3.535921625544267,
      "grad_norm": 1.25,
      "learning_rate": 0.00037859531546229405,
      "loss": 4.928,
      "mean_token_accuracy": 0.215593583881855,
      "num_tokens": 80883752.0,
      "step": 38980
    },
    {
      "entropy": 5.7043482780456545,
      "epoch": 3.5363751814223514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003785665916192824,
      "loss": 4.9246,
      "mean_token_accuracy": 0.22120550721883775,
      "num_tokens": 80894265.0,
      "step": 38985
    },
    {
      "entropy": 5.609049224853516,
      "epoch": 3.5368287373004357,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037853786563448724,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22402693182229996,
      "num_tokens": 80904246.0,
      "step": 38990
    },
    {
      "entropy": 5.676704692840576,
      "epoch": 3.5372822931785195,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003785091375085027,
      "loss": 4.9054,
      "mean_token_accuracy": 0.21703873723745346,
      "num_tokens": 80913595.0,
      "step": 38995
    },
    {
      "entropy": 5.668815469741821,
      "epoch": 3.5377358490566038,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003784804072419228,
      "loss": 4.9026,
      "mean_token_accuracy": 0.21766487807035445,
      "num_tokens": 80924754.0,
      "step": 39000
    },
    {
      "epoch": 3.5377358490566038,
      "eval_entropy": 5.3975669599766105,
      "eval_loss": 4.972774505615234,
      "eval_mean_token_accuracy": 0.22537762250970392,
      "eval_num_tokens": 80924754.0,
      "eval_runtime": 20.5549,
      "eval_samples_per_second": 1720.268,
      "eval_steps_per_second": 215.033,
      "step": 39000
    },
    {
      "entropy": 5.6199414253234865,
      "epoch": 3.538189404934688,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003784516748353417,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23026662766933442,
      "num_tokens": 80935318.0,
      "step": 39005
    },
    {
      "entropy": 5.645366764068603,
      "epoch": 3.5386429608127723,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003784229402893536,
      "loss": 4.9669,
      "mean_token_accuracy": 0.22032185196876525,
      "num_tokens": 80944886.0,
      "step": 39010
    },
    {
      "entropy": 5.684536933898926,
      "epoch": 3.539096516690856,
      "grad_norm": 1.25,
      "learning_rate": 0.0003783942036045528,
      "loss": 4.9107,
      "mean_token_accuracy": 0.21404988765716554,
      "num_tokens": 80955566.0,
      "step": 39015
    },
    {
      "entropy": 5.739090490341186,
      "epoch": 3.5395500725689404,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037836546478153344,
      "loss": 4.9673,
      "mean_token_accuracy": 0.210165274143219,
      "num_tokens": 80965285.0,
      "step": 39020
    },
    {
      "entropy": 5.623764562606811,
      "epoch": 3.5400036284470247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003783367238208899,
      "loss": 4.8666,
      "mean_token_accuracy": 0.2256423979997635,
      "num_tokens": 80976402.0,
      "step": 39025
    },
    {
      "entropy": 5.693312740325927,
      "epoch": 3.540457184325109,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037830798072321665,
      "loss": 4.9707,
      "mean_token_accuracy": 0.21763391196727752,
      "num_tokens": 80988408.0,
      "step": 39030
    },
    {
      "entropy": 5.678764533996582,
      "epoch": 3.540910740203193,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037827923548910806,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2193120762705803,
      "num_tokens": 80999099.0,
      "step": 39035
    },
    {
      "entropy": 5.631063652038574,
      "epoch": 3.5413642960812775,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037825048811915854,
      "loss": 4.8013,
      "mean_token_accuracy": 0.22957441955804825,
      "num_tokens": 81009350.0,
      "step": 39040
    },
    {
      "entropy": 5.646950626373291,
      "epoch": 3.5418178519593613,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037822173861396256,
      "loss": 4.7988,
      "mean_token_accuracy": 0.22003531754016875,
      "num_tokens": 81019851.0,
      "step": 39045
    },
    {
      "entropy": 5.644749546051026,
      "epoch": 3.5422714078374455,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037819298697411475,
      "loss": 4.917,
      "mean_token_accuracy": 0.217026050388813,
      "num_tokens": 81028987.0,
      "step": 39050
    },
    {
      "entropy": 5.614936876296997,
      "epoch": 3.54272496371553,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037816423320020964,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22462525963783264,
      "num_tokens": 81038081.0,
      "step": 39055
    },
    {
      "entropy": 5.680123949050904,
      "epoch": 3.543178519593614,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037813547729284195,
      "loss": 4.8701,
      "mean_token_accuracy": 0.21908582150936126,
      "num_tokens": 81048066.0,
      "step": 39060
    },
    {
      "entropy": 5.537013101577759,
      "epoch": 3.543632075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037810671925260633,
      "loss": 4.7324,
      "mean_token_accuracy": 0.23865723609924316,
      "num_tokens": 81057421.0,
      "step": 39065
    },
    {
      "entropy": 5.613677549362182,
      "epoch": 3.544085631349782,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037807795908009745,
      "loss": 4.8985,
      "mean_token_accuracy": 0.21641252785921097,
      "num_tokens": 81068435.0,
      "step": 39070
    },
    {
      "entropy": 5.701085805892944,
      "epoch": 3.5445391872278664,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037804919677591006,
      "loss": 4.9344,
      "mean_token_accuracy": 0.21807005405426025,
      "num_tokens": 81079378.0,
      "step": 39075
    },
    {
      "entropy": 5.655683517456055,
      "epoch": 3.5449927431059507,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003780204323406391,
      "loss": 4.9059,
      "mean_token_accuracy": 0.21444810181856155,
      "num_tokens": 81089872.0,
      "step": 39080
    },
    {
      "entropy": 5.627287530899048,
      "epoch": 3.545446298984035,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003779916657748793,
      "loss": 4.827,
      "mean_token_accuracy": 0.22670845091342925,
      "num_tokens": 81099653.0,
      "step": 39085
    },
    {
      "entropy": 5.619866609573364,
      "epoch": 3.5458998548621192,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003779628970792256,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22461666762828827,
      "num_tokens": 81109840.0,
      "step": 39090
    },
    {
      "entropy": 5.6104193210601805,
      "epoch": 3.546353410740203,
      "grad_norm": 1.265625,
      "learning_rate": 0.000377934126254273,
      "loss": 4.8375,
      "mean_token_accuracy": 0.227297081053257,
      "num_tokens": 81120991.0,
      "step": 39095
    },
    {
      "entropy": 5.737735462188721,
      "epoch": 3.5468069666182873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037790535330061635,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22372061908245086,
      "num_tokens": 81129831.0,
      "step": 39100
    },
    {
      "entropy": 5.691895198822022,
      "epoch": 3.5472605224963716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003778765782188508,
      "loss": 4.8767,
      "mean_token_accuracy": 0.22367513328790664,
      "num_tokens": 81140604.0,
      "step": 39105
    },
    {
      "entropy": 5.661174726486206,
      "epoch": 3.547714078374456,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00037784780100957145,
      "loss": 4.893,
      "mean_token_accuracy": 0.22753536850214004,
      "num_tokens": 81150591.0,
      "step": 39110
    },
    {
      "entropy": 5.720527696609497,
      "epoch": 3.5481676342525397,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003778190216733733,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21242575198411942,
      "num_tokens": 81161444.0,
      "step": 39115
    },
    {
      "entropy": 5.692187070846558,
      "epoch": 3.548621190130624,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037779024021085167,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22659531533718108,
      "num_tokens": 81171216.0,
      "step": 39120
    },
    {
      "entropy": 5.681547212600708,
      "epoch": 3.549074746008708,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003777614566226016,
      "loss": 4.9166,
      "mean_token_accuracy": 0.22361256927251816,
      "num_tokens": 81181665.0,
      "step": 39125
    },
    {
      "entropy": 5.566703939437867,
      "epoch": 3.5495283018867925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003777326709092184,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2241305097937584,
      "num_tokens": 81191683.0,
      "step": 39130
    },
    {
      "entropy": 5.648497581481934,
      "epoch": 3.5499818577648767,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037770388307129747,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22878179848194122,
      "num_tokens": 81201211.0,
      "step": 39135
    },
    {
      "entropy": 5.66672420501709,
      "epoch": 3.550435413642961,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037767509310943407,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22501013725996016,
      "num_tokens": 81210668.0,
      "step": 39140
    },
    {
      "entropy": 5.646736192703247,
      "epoch": 3.5508889695210453,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003776463010242235,
      "loss": 4.7538,
      "mean_token_accuracy": 0.22681679874658583,
      "num_tokens": 81221073.0,
      "step": 39145
    },
    {
      "entropy": 5.659432506561279,
      "epoch": 3.551342525399129,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037761750681626134,
      "loss": 4.9262,
      "mean_token_accuracy": 0.216758331656456,
      "num_tokens": 81231687.0,
      "step": 39150
    },
    {
      "entropy": 5.705511951446534,
      "epoch": 3.5517960812772134,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037758871048614295,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21627120077610015,
      "num_tokens": 81242323.0,
      "step": 39155
    },
    {
      "entropy": 5.646678447723389,
      "epoch": 3.5522496371552976,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003775599120344638,
      "loss": 4.7564,
      "mean_token_accuracy": 0.23311761617660523,
      "num_tokens": 81252543.0,
      "step": 39160
    },
    {
      "entropy": 5.661546230316162,
      "epoch": 3.5527031930333814,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037753111146181966,
      "loss": 4.8946,
      "mean_token_accuracy": 0.21643289029598237,
      "num_tokens": 81262180.0,
      "step": 39165
    },
    {
      "entropy": 5.802547550201416,
      "epoch": 3.5531567489114657,
      "grad_norm": 1.28125,
      "learning_rate": 0.000377502308768806,
      "loss": 5.0917,
      "mean_token_accuracy": 0.20186795592308043,
      "num_tokens": 81272088.0,
      "step": 39170
    },
    {
      "entropy": 5.678045606613159,
      "epoch": 3.55361030478955,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003774735039560184,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2288170948624611,
      "num_tokens": 81281723.0,
      "step": 39175
    },
    {
      "entropy": 5.716694307327271,
      "epoch": 3.5540638606676342,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037744469702405273,
      "loss": 4.9471,
      "mean_token_accuracy": 0.220894256234169,
      "num_tokens": 81292030.0,
      "step": 39180
    },
    {
      "entropy": 5.625248384475708,
      "epoch": 3.5545174165457185,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003774158879735045,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22652019709348678,
      "num_tokens": 81302283.0,
      "step": 39185
    },
    {
      "entropy": 5.690362215042114,
      "epoch": 3.5549709724238028,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003773870768049697,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21482795476913452,
      "num_tokens": 81313058.0,
      "step": 39190
    },
    {
      "entropy": 5.669942474365234,
      "epoch": 3.555424528301887,
      "grad_norm": 1.234375,
      "learning_rate": 0.000377358263519044,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22588047236204148,
      "num_tokens": 81323827.0,
      "step": 39195
    },
    {
      "entropy": 5.71724009513855,
      "epoch": 3.555878084179971,
      "grad_norm": 1.375,
      "learning_rate": 0.0003773294481163234,
      "loss": 4.9323,
      "mean_token_accuracy": 0.21463553160429,
      "num_tokens": 81333106.0,
      "step": 39200
    },
    {
      "entropy": 5.664826488494873,
      "epoch": 3.556331640058055,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003773006305974037,
      "loss": 4.8785,
      "mean_token_accuracy": 0.225969335436821,
      "num_tokens": 81343245.0,
      "step": 39205
    },
    {
      "entropy": 5.644611883163452,
      "epoch": 3.5567851959361394,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037727181096288077,
      "loss": 4.8715,
      "mean_token_accuracy": 0.2291637033224106,
      "num_tokens": 81352917.0,
      "step": 39210
    },
    {
      "entropy": 5.608926916122437,
      "epoch": 3.5572387518142237,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003772429892133509,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22133971452713014,
      "num_tokens": 81362838.0,
      "step": 39215
    },
    {
      "entropy": 5.632777261734009,
      "epoch": 3.5576923076923075,
      "grad_norm": 1.25,
      "learning_rate": 0.0003772141653494099,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22803981900215148,
      "num_tokens": 81373395.0,
      "step": 39220
    },
    {
      "entropy": 5.629801082611084,
      "epoch": 3.5581458635703918,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003771853393716538,
      "loss": 4.859,
      "mean_token_accuracy": 0.2212301030755043,
      "num_tokens": 81383976.0,
      "step": 39225
    },
    {
      "entropy": 5.63633394241333,
      "epoch": 3.558599419448476,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000377156511280679,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2125580057501793,
      "num_tokens": 81394328.0,
      "step": 39230
    },
    {
      "entropy": 5.661877155303955,
      "epoch": 3.5590529753265603,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003771276810770815,
      "loss": 4.894,
      "mean_token_accuracy": 0.21903952062129975,
      "num_tokens": 81405451.0,
      "step": 39235
    },
    {
      "entropy": 5.704607629776001,
      "epoch": 3.5595065312046446,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003770988487614575,
      "loss": 4.9857,
      "mean_token_accuracy": 0.21099468618631362,
      "num_tokens": 81416589.0,
      "step": 39240
    },
    {
      "entropy": 5.6146605014801025,
      "epoch": 3.559960087082729,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003770700143344033,
      "loss": 4.8439,
      "mean_token_accuracy": 0.22177279889583587,
      "num_tokens": 81428553.0,
      "step": 39245
    },
    {
      "entropy": 5.8086690425872805,
      "epoch": 3.5604136429608126,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037704117779651514,
      "loss": 4.9953,
      "mean_token_accuracy": 0.21772724539041519,
      "num_tokens": 81438720.0,
      "step": 39250
    },
    {
      "entropy": 5.639038372039795,
      "epoch": 3.560867198838897,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00037701233914838947,
      "loss": 4.9325,
      "mean_token_accuracy": 0.21663607209920882,
      "num_tokens": 81448142.0,
      "step": 39255
    },
    {
      "entropy": 5.670236682891845,
      "epoch": 3.561320754716981,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003769834983906226,
      "loss": 4.954,
      "mean_token_accuracy": 0.21869388818740845,
      "num_tokens": 81459448.0,
      "step": 39260
    },
    {
      "entropy": 5.680305099487304,
      "epoch": 3.5617743105950654,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037695465552381105,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22511059790849686,
      "num_tokens": 81470356.0,
      "step": 39265
    },
    {
      "entropy": 5.647854852676391,
      "epoch": 3.5622278664731493,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037692581054855125,
      "loss": 4.856,
      "mean_token_accuracy": 0.22304714024066924,
      "num_tokens": 81480472.0,
      "step": 39270
    },
    {
      "entropy": 5.6276692867279055,
      "epoch": 3.5626814223512335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037689696346543964,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2254254400730133,
      "num_tokens": 81489949.0,
      "step": 39275
    },
    {
      "entropy": 5.620673370361328,
      "epoch": 3.563134978229318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037686811427507284,
      "loss": 4.7889,
      "mean_token_accuracy": 0.2270620658993721,
      "num_tokens": 81500276.0,
      "step": 39280
    },
    {
      "entropy": 5.657632350921631,
      "epoch": 3.563588534107402,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037683926297804756,
      "loss": 4.7919,
      "mean_token_accuracy": 0.2231597989797592,
      "num_tokens": 81510431.0,
      "step": 39285
    },
    {
      "entropy": 5.621599340438843,
      "epoch": 3.5640420899854863,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003768104095749603,
      "loss": 4.8851,
      "mean_token_accuracy": 0.21941588968038558,
      "num_tokens": 81520718.0,
      "step": 39290
    },
    {
      "entropy": 5.6724639415740965,
      "epoch": 3.5644956458635706,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003767815540664079,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21806146949529648,
      "num_tokens": 81531904.0,
      "step": 39295
    },
    {
      "entropy": 5.666174840927124,
      "epoch": 3.5649492017416544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037675269645298697,
      "loss": 4.9156,
      "mean_token_accuracy": 0.21807790994644166,
      "num_tokens": 81542093.0,
      "step": 39300
    },
    {
      "entropy": 5.646407175064087,
      "epoch": 3.5654027576197387,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003767238367352944,
      "loss": 4.8199,
      "mean_token_accuracy": 0.22963330298662185,
      "num_tokens": 81552020.0,
      "step": 39305
    },
    {
      "entropy": 5.621885681152344,
      "epoch": 3.565856313497823,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037669497491392683,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22844843566417694,
      "num_tokens": 81561644.0,
      "step": 39310
    },
    {
      "entropy": 5.671325778961181,
      "epoch": 3.566309869375907,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037666611098948144,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22219059616327286,
      "num_tokens": 81572158.0,
      "step": 39315
    },
    {
      "entropy": 5.75589804649353,
      "epoch": 3.566763425253991,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003766372449625549,
      "loss": 4.9741,
      "mean_token_accuracy": 0.2117054745554924,
      "num_tokens": 81582335.0,
      "step": 39320
    },
    {
      "entropy": 5.771406555175782,
      "epoch": 3.5672169811320753,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003766083768337442,
      "loss": 4.969,
      "mean_token_accuracy": 0.21337712407112122,
      "num_tokens": 81593004.0,
      "step": 39325
    },
    {
      "entropy": 5.713812160491943,
      "epoch": 3.5676705370101596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003765795066036464,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2206327199935913,
      "num_tokens": 81603656.0,
      "step": 39330
    },
    {
      "entropy": 5.6781110763549805,
      "epoch": 3.568124092888244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037655063427285854,
      "loss": 4.9145,
      "mean_token_accuracy": 0.21984171867370605,
      "num_tokens": 81613039.0,
      "step": 39335
    },
    {
      "entropy": 5.640156936645508,
      "epoch": 3.568577648766328,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037652175984197764,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22307460904121398,
      "num_tokens": 81624410.0,
      "step": 39340
    },
    {
      "entropy": 5.670591068267822,
      "epoch": 3.5690312046444124,
      "grad_norm": 1.25,
      "learning_rate": 0.00037649288331160086,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22253659516572952,
      "num_tokens": 81635160.0,
      "step": 39345
    },
    {
      "entropy": 5.665782165527344,
      "epoch": 3.5694847605224966,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037646400468232546,
      "loss": 4.8654,
      "mean_token_accuracy": 0.21860369741916658,
      "num_tokens": 81645686.0,
      "step": 39350
    },
    {
      "entropy": 5.658524703979492,
      "epoch": 3.5699383164005805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037643512395474853,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22308750003576278,
      "num_tokens": 81656330.0,
      "step": 39355
    },
    {
      "entropy": 5.63554277420044,
      "epoch": 3.5703918722786647,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003764062411294674,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2184840500354767,
      "num_tokens": 81665895.0,
      "step": 39360
    },
    {
      "entropy": 5.743279838562012,
      "epoch": 3.570845428156749,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037637735620707935,
      "loss": 4.924,
      "mean_token_accuracy": 0.21882702708244323,
      "num_tokens": 81676616.0,
      "step": 39365
    },
    {
      "entropy": 5.696913337707519,
      "epoch": 3.5712989840348333,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003763484691881818,
      "loss": 4.8987,
      "mean_token_accuracy": 0.2202715903520584,
      "num_tokens": 81687169.0,
      "step": 39370
    },
    {
      "entropy": 5.718110799789429,
      "epoch": 3.571752539912917,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037631958007337195,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22296208292245864,
      "num_tokens": 81698291.0,
      "step": 39375
    },
    {
      "entropy": 5.690632390975952,
      "epoch": 3.5722060957910013,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003762906888632475,
      "loss": 4.8873,
      "mean_token_accuracy": 0.22014413475990297,
      "num_tokens": 81708750.0,
      "step": 39380
    },
    {
      "entropy": 5.595400857925415,
      "epoch": 3.5726596516690856,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003762617955584058,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22440143972635268,
      "num_tokens": 81719474.0,
      "step": 39385
    },
    {
      "entropy": 5.768841934204102,
      "epoch": 3.57311320754717,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003762329001594443,
      "loss": 4.9514,
      "mean_token_accuracy": 0.22003939896821975,
      "num_tokens": 81730092.0,
      "step": 39390
    },
    {
      "entropy": 5.7015564918518065,
      "epoch": 3.573566763425254,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003762040026669606,
      "loss": 4.8986,
      "mean_token_accuracy": 0.21521924883127214,
      "num_tokens": 81741235.0,
      "step": 39395
    },
    {
      "entropy": 5.623976516723633,
      "epoch": 3.5740203193033384,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003761751030815524,
      "loss": 4.859,
      "mean_token_accuracy": 0.22705985754728317,
      "num_tokens": 81751434.0,
      "step": 39400
    },
    {
      "entropy": 5.663141584396362,
      "epoch": 3.5744738751814222,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003761462014038173,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2141480028629303,
      "num_tokens": 81763935.0,
      "step": 39405
    },
    {
      "entropy": 5.650870323181152,
      "epoch": 3.5749274310595065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003761172976343528,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22021840810775756,
      "num_tokens": 81775007.0,
      "step": 39410
    },
    {
      "entropy": 5.670202732086182,
      "epoch": 3.5753809869375908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037608839177375705,
      "loss": 4.9322,
      "mean_token_accuracy": 0.2241053894162178,
      "num_tokens": 81785690.0,
      "step": 39415
    },
    {
      "entropy": 5.669210004806518,
      "epoch": 3.575834542815675,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003760594838226275,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2243759423494339,
      "num_tokens": 81796143.0,
      "step": 39420
    },
    {
      "entropy": 5.777917098999024,
      "epoch": 3.576288098693759,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003760305737815621,
      "loss": 4.9966,
      "mean_token_accuracy": 0.21600125283002852,
      "num_tokens": 81806607.0,
      "step": 39425
    },
    {
      "entropy": 5.618704128265381,
      "epoch": 3.576741654571843,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003760016616511586,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22732609510421753,
      "num_tokens": 81817216.0,
      "step": 39430
    },
    {
      "entropy": 5.654512882232666,
      "epoch": 3.5771952104499274,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037597274743201507,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22358354032039643,
      "num_tokens": 81827211.0,
      "step": 39435
    },
    {
      "entropy": 5.671589040756226,
      "epoch": 3.5776487663280117,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003759438311247294,
      "loss": 4.8339,
      "mean_token_accuracy": 0.2246164619922638,
      "num_tokens": 81837365.0,
      "step": 39440
    },
    {
      "entropy": 5.655935287475586,
      "epoch": 3.578102322206096,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003759149127298995,
      "loss": 4.854,
      "mean_token_accuracy": 0.22731743454933168,
      "num_tokens": 81847803.0,
      "step": 39445
    },
    {
      "entropy": 5.634450912475586,
      "epoch": 3.57855587808418,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037588599224812363,
      "loss": 4.8304,
      "mean_token_accuracy": 0.22976899296045303,
      "num_tokens": 81858123.0,
      "step": 39450
    },
    {
      "entropy": 5.6110271453857425,
      "epoch": 3.579009433962264,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003758570696799997,
      "loss": 4.8363,
      "mean_token_accuracy": 0.23093357384204866,
      "num_tokens": 81868668.0,
      "step": 39455
    },
    {
      "entropy": 5.643139362335205,
      "epoch": 3.5794629898403483,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00037582814502612575,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23462222069501876,
      "num_tokens": 81877482.0,
      "step": 39460
    },
    {
      "entropy": 5.581671905517578,
      "epoch": 3.5799165457184325,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003757992182871001,
      "loss": 4.7301,
      "mean_token_accuracy": 0.23388663977384566,
      "num_tokens": 81886700.0,
      "step": 39465
    },
    {
      "entropy": 5.665284729003906,
      "epoch": 3.580370101596517,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037577028946352107,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22747864127159118,
      "num_tokens": 81896446.0,
      "step": 39470
    },
    {
      "entropy": 5.696345853805542,
      "epoch": 3.5808236574746006,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003757413585559866,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2305184781551361,
      "num_tokens": 81907418.0,
      "step": 39475
    },
    {
      "entropy": 5.699430274963379,
      "epoch": 3.581277213352685,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037571242556509526,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2273471921682358,
      "num_tokens": 81917940.0,
      "step": 39480
    },
    {
      "entropy": 5.602329921722412,
      "epoch": 3.581730769230769,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003756834904914453,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22235541492700578,
      "num_tokens": 81928158.0,
      "step": 39485
    },
    {
      "entropy": 5.721796226501465,
      "epoch": 3.5821843251088534,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003756545533356351,
      "loss": 4.9096,
      "mean_token_accuracy": 0.2223244458436966,
      "num_tokens": 81938111.0,
      "step": 39490
    },
    {
      "entropy": 5.724743223190307,
      "epoch": 3.5826378809869377,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000375625614098263,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2255377784371376,
      "num_tokens": 81948798.0,
      "step": 39495
    },
    {
      "entropy": 5.682094717025757,
      "epoch": 3.583091436865022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037559667277992765,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22398220449686052,
      "num_tokens": 81959240.0,
      "step": 39500
    },
    {
      "entropy": 5.6626824855804445,
      "epoch": 3.5835449927431062,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003755677293812274,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2253880023956299,
      "num_tokens": 81969929.0,
      "step": 39505
    },
    {
      "entropy": 5.729239511489868,
      "epoch": 3.58399854862119,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003755387839027609,
      "loss": 4.9395,
      "mean_token_accuracy": 0.21058569103479385,
      "num_tokens": 81981007.0,
      "step": 39510
    },
    {
      "entropy": 5.752055692672729,
      "epoch": 3.5844521044992743,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037550983634512683,
      "loss": 5.0032,
      "mean_token_accuracy": 0.21219512969255447,
      "num_tokens": 81991172.0,
      "step": 39515
    },
    {
      "entropy": 5.636320972442627,
      "epoch": 3.5849056603773586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003754808867089236,
      "loss": 4.7856,
      "mean_token_accuracy": 0.2317718744277954,
      "num_tokens": 82001282.0,
      "step": 39520
    },
    {
      "entropy": 5.654307126998901,
      "epoch": 3.5853592162554424,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037545193499475003,
      "loss": 4.8932,
      "mean_token_accuracy": 0.21816196888685227,
      "num_tokens": 82011929.0,
      "step": 39525
    },
    {
      "entropy": 5.754404735565186,
      "epoch": 3.5858127721335267,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003754229812032049,
      "loss": 4.9199,
      "mean_token_accuracy": 0.2293623611330986,
      "num_tokens": 82021663.0,
      "step": 39530
    },
    {
      "entropy": 5.703597974777222,
      "epoch": 3.586266328011611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037539402533488694,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21641401946544647,
      "num_tokens": 82031547.0,
      "step": 39535
    },
    {
      "entropy": 5.7347019672393795,
      "epoch": 3.586719883889695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003753650673903949,
      "loss": 4.928,
      "mean_token_accuracy": 0.2167762354016304,
      "num_tokens": 82041722.0,
      "step": 39540
    },
    {
      "entropy": 5.697149991989136,
      "epoch": 3.5871734397677795,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037533610737032773,
      "loss": 4.9147,
      "mean_token_accuracy": 0.2229364737868309,
      "num_tokens": 82052715.0,
      "step": 39545
    },
    {
      "entropy": 5.665228700637817,
      "epoch": 3.5876269956458637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037530714527528433,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22894608676433564,
      "num_tokens": 82062709.0,
      "step": 39550
    },
    {
      "entropy": 5.71511754989624,
      "epoch": 3.588080551523948,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037527818110586354,
      "loss": 4.9078,
      "mean_token_accuracy": 0.22362364530563356,
      "num_tokens": 82072731.0,
      "step": 39555
    },
    {
      "entropy": 5.706671333312988,
      "epoch": 3.588534107402032,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003752492148626644,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2289661481976509,
      "num_tokens": 82083216.0,
      "step": 39560
    },
    {
      "entropy": 5.699848651885986,
      "epoch": 3.588987663280116,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000375220246546286,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21549673229455948,
      "num_tokens": 82094482.0,
      "step": 39565
    },
    {
      "entropy": 5.699146509170532,
      "epoch": 3.5894412191582004,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003751912761573273,
      "loss": 4.8569,
      "mean_token_accuracy": 0.21918173879384995,
      "num_tokens": 82104988.0,
      "step": 39570
    },
    {
      "entropy": 5.665804767608643,
      "epoch": 3.5898947750362846,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037516230369638755,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21689143031835556,
      "num_tokens": 82114455.0,
      "step": 39575
    },
    {
      "entropy": 5.712041425704956,
      "epoch": 3.5903483309143684,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003751333291640658,
      "loss": 4.8992,
      "mean_token_accuracy": 0.21944786459207535,
      "num_tokens": 82125001.0,
      "step": 39580
    },
    {
      "entropy": 5.697367095947266,
      "epoch": 3.5908018867924527,
      "grad_norm": 1.3125,
      "learning_rate": 0.00037510435256096124,
      "loss": 4.8681,
      "mean_token_accuracy": 0.22204745858907698,
      "num_tokens": 82134586.0,
      "step": 39585
    },
    {
      "entropy": 5.648501205444336,
      "epoch": 3.591255442670537,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037507537388767326,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22423013746738435,
      "num_tokens": 82144612.0,
      "step": 39590
    },
    {
      "entropy": 5.617886161804199,
      "epoch": 3.5917089985486212,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000375046393144801,
      "loss": 4.8592,
      "mean_token_accuracy": 0.23053251057863236,
      "num_tokens": 82154913.0,
      "step": 39595
    },
    {
      "entropy": 5.684206342697143,
      "epoch": 3.5921625544267055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003750174103329438,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2198135659098625,
      "num_tokens": 82164469.0,
      "step": 39600
    },
    {
      "entropy": 5.717775821685791,
      "epoch": 3.59261611030479,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037498842545270105,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22153180837631226,
      "num_tokens": 82174575.0,
      "step": 39605
    },
    {
      "entropy": 5.712337350845337,
      "epoch": 3.5930696661828736,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037495943850467226,
      "loss": 4.8866,
      "mean_token_accuracy": 0.22442503571510314,
      "num_tokens": 82184661.0,
      "step": 39610
    },
    {
      "entropy": 5.735856628417968,
      "epoch": 3.593523222060958,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037493044948945677,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2200234428048134,
      "num_tokens": 82194877.0,
      "step": 39615
    },
    {
      "entropy": 5.688334846496582,
      "epoch": 3.593976777939042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037490145840765414,
      "loss": 4.8906,
      "mean_token_accuracy": 0.2235260307788849,
      "num_tokens": 82206278.0,
      "step": 39620
    },
    {
      "entropy": 5.667785739898681,
      "epoch": 3.5944303338171264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037487246525986384,
      "loss": 4.887,
      "mean_token_accuracy": 0.2203375592827797,
      "num_tokens": 82216140.0,
      "step": 39625
    },
    {
      "entropy": 5.725302410125733,
      "epoch": 3.59488388969521,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003748434700466856,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2271267980337143,
      "num_tokens": 82226239.0,
      "step": 39630
    },
    {
      "entropy": 5.687868595123291,
      "epoch": 3.5953374455732945,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003748144727687188,
      "loss": 4.8725,
      "mean_token_accuracy": 0.2244626373052597,
      "num_tokens": 82235795.0,
      "step": 39635
    },
    {
      "entropy": 5.62402081489563,
      "epoch": 3.5957910014513788,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037478547342656345,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22691652923822403,
      "num_tokens": 82245644.0,
      "step": 39640
    },
    {
      "entropy": 5.660576820373535,
      "epoch": 3.596244557329463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000374756472020819,
      "loss": 4.8614,
      "mean_token_accuracy": 0.22553318589925767,
      "num_tokens": 82255608.0,
      "step": 39645
    },
    {
      "entropy": 5.65521650314331,
      "epoch": 3.5966981132075473,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037472746855208525,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2239788740873337,
      "num_tokens": 82265488.0,
      "step": 39650
    },
    {
      "entropy": 5.656335306167603,
      "epoch": 3.5971516690856316,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003746984630209621,
      "loss": 4.8922,
      "mean_token_accuracy": 0.21513646990060806,
      "num_tokens": 82276866.0,
      "step": 39655
    },
    {
      "entropy": 5.685085153579712,
      "epoch": 3.5976052249637154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00037466945542804933,
      "loss": 4.875,
      "mean_token_accuracy": 0.22881623357534409,
      "num_tokens": 82287125.0,
      "step": 39660
    },
    {
      "entropy": 5.742771196365356,
      "epoch": 3.5980587808417996,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037464044577394676,
      "loss": 4.9272,
      "mean_token_accuracy": 0.2184641495347023,
      "num_tokens": 82296887.0,
      "step": 39665
    },
    {
      "entropy": 5.71889853477478,
      "epoch": 3.598512336719884,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003746114340592544,
      "loss": 4.9784,
      "mean_token_accuracy": 0.20866239964962005,
      "num_tokens": 82306928.0,
      "step": 39670
    },
    {
      "entropy": 5.689286231994629,
      "epoch": 3.598965892597968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037458242028457216,
      "loss": 4.915,
      "mean_token_accuracy": 0.22209928333759307,
      "num_tokens": 82317873.0,
      "step": 39675
    },
    {
      "entropy": 5.589869213104248,
      "epoch": 3.599419448476052,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037455340445050023,
      "loss": 4.7746,
      "mean_token_accuracy": 0.22989826947450637,
      "num_tokens": 82328184.0,
      "step": 39680
    },
    {
      "entropy": 5.634340000152588,
      "epoch": 3.5998730043541363,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037452438655763843,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22481180876493453,
      "num_tokens": 82338191.0,
      "step": 39685
    },
    {
      "entropy": 5.688096237182617,
      "epoch": 3.6003265602322205,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037449536660658696,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21997273564338685,
      "num_tokens": 82348095.0,
      "step": 39690
    },
    {
      "entropy": 5.658121633529663,
      "epoch": 3.600780116110305,
      "grad_norm": 1.3125,
      "learning_rate": 0.000374466344597946,
      "loss": 4.8034,
      "mean_token_accuracy": 0.2255045846104622,
      "num_tokens": 82358216.0,
      "step": 39695
    },
    {
      "entropy": 5.615994167327881,
      "epoch": 3.601233671988389,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003744373205323157,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22161150276660918,
      "num_tokens": 82368581.0,
      "step": 39700
    },
    {
      "entropy": 5.652612686157227,
      "epoch": 3.6016872278664733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003744082944102962,
      "loss": 4.8772,
      "mean_token_accuracy": 0.21787499487400055,
      "num_tokens": 82379609.0,
      "step": 39705
    },
    {
      "entropy": 5.7069737911224365,
      "epoch": 3.6021407837445576,
      "grad_norm": 1.125,
      "learning_rate": 0.0003743792662324878,
      "loss": 4.9139,
      "mean_token_accuracy": 0.22287432551383973,
      "num_tokens": 82390589.0,
      "step": 39710
    },
    {
      "entropy": 5.619561815261841,
      "epoch": 3.6025943396226414,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037435023599949086,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22112976312637328,
      "num_tokens": 82400915.0,
      "step": 39715
    },
    {
      "entropy": 5.654586362838745,
      "epoch": 3.6030478955007257,
      "grad_norm": 1.296875,
      "learning_rate": 0.00037432120371190585,
      "loss": 4.888,
      "mean_token_accuracy": 0.22282626628875732,
      "num_tokens": 82411544.0,
      "step": 39720
    },
    {
      "entropy": 5.622819566726685,
      "epoch": 3.60350145137881,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037429216937033294,
      "loss": 4.8512,
      "mean_token_accuracy": 0.2251242607831955,
      "num_tokens": 82422391.0,
      "step": 39725
    },
    {
      "entropy": 5.620830059051514,
      "epoch": 3.6039550072568938,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037426313297537263,
      "loss": 4.8764,
      "mean_token_accuracy": 0.22143255025148392,
      "num_tokens": 82433564.0,
      "step": 39730
    },
    {
      "entropy": 5.664380407333374,
      "epoch": 3.604408563134978,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003742340945276255,
      "loss": 4.9125,
      "mean_token_accuracy": 0.22205149084329606,
      "num_tokens": 82445005.0,
      "step": 39735
    },
    {
      "entropy": 5.708518838882446,
      "epoch": 3.6048621190130623,
      "grad_norm": 1.25,
      "learning_rate": 0.000374205054027692,
      "loss": 4.947,
      "mean_token_accuracy": 0.21554548889398575,
      "num_tokens": 82455998.0,
      "step": 39740
    },
    {
      "entropy": 5.641297817230225,
      "epoch": 3.6053156748911466,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037417601147617265,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22297605127096176,
      "num_tokens": 82466789.0,
      "step": 39745
    },
    {
      "entropy": 5.682171630859375,
      "epoch": 3.605769230769231,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037414696687366816,
      "loss": 5.0147,
      "mean_token_accuracy": 0.2100578024983406,
      "num_tokens": 82478181.0,
      "step": 39750
    },
    {
      "entropy": 5.667053079605102,
      "epoch": 3.606222786647315,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037411792022077917,
      "loss": 4.8345,
      "mean_token_accuracy": 0.21986059546470643,
      "num_tokens": 82488452.0,
      "step": 39755
    },
    {
      "entropy": 5.712802934646606,
      "epoch": 3.6066763425253994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003740888715181063,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22393935024738312,
      "num_tokens": 82498179.0,
      "step": 39760
    },
    {
      "entropy": 5.638017177581787,
      "epoch": 3.607129898403483,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037405982076625026,
      "loss": 4.9284,
      "mean_token_accuracy": 0.22259201854467392,
      "num_tokens": 82508242.0,
      "step": 39765
    },
    {
      "entropy": 5.69512791633606,
      "epoch": 3.6075834542815675,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037403076796581195,
      "loss": 4.918,
      "mean_token_accuracy": 0.21806730926036835,
      "num_tokens": 82519363.0,
      "step": 39770
    },
    {
      "entropy": 5.73516583442688,
      "epoch": 3.6080370101596517,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003740017131173921,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22069801688194274,
      "num_tokens": 82529745.0,
      "step": 39775
    },
    {
      "entropy": 5.745014190673828,
      "epoch": 3.608490566037736,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003739726562215916,
      "loss": 4.9517,
      "mean_token_accuracy": 0.2167219787836075,
      "num_tokens": 82539573.0,
      "step": 39780
    },
    {
      "entropy": 5.706544971466064,
      "epoch": 3.60894412191582,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037394359727901136,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21726891845464708,
      "num_tokens": 82551313.0,
      "step": 39785
    },
    {
      "entropy": 5.648802328109741,
      "epoch": 3.609397677793904,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003739145362902523,
      "loss": 4.8784,
      "mean_token_accuracy": 0.21816924959421158,
      "num_tokens": 82561283.0,
      "step": 39790
    },
    {
      "entropy": 5.618662929534912,
      "epoch": 3.6098512336719883,
      "grad_norm": 1.375,
      "learning_rate": 0.0003738854732559153,
      "loss": 4.8151,
      "mean_token_accuracy": 0.22747956067323685,
      "num_tokens": 82571118.0,
      "step": 39795
    },
    {
      "entropy": 5.636912202835083,
      "epoch": 3.6103047895500726,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003738564081766017,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22580860704183578,
      "num_tokens": 82581663.0,
      "step": 39800
    },
    {
      "entropy": 5.665033912658691,
      "epoch": 3.610758345428157,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037382734105291234,
      "loss": 4.8503,
      "mean_token_accuracy": 0.2204047977924347,
      "num_tokens": 82592593.0,
      "step": 39805
    },
    {
      "entropy": 5.669158935546875,
      "epoch": 3.611211901306241,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037379827188544835,
      "loss": 4.961,
      "mean_token_accuracy": 0.21837086528539656,
      "num_tokens": 82603047.0,
      "step": 39810
    },
    {
      "entropy": 5.732370233535766,
      "epoch": 3.611665457184325,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037376920067481095,
      "loss": 4.9247,
      "mean_token_accuracy": 0.22156117260456085,
      "num_tokens": 82612789.0,
      "step": 39815
    },
    {
      "entropy": 5.651569080352783,
      "epoch": 3.6121190130624092,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003737401274216012,
      "loss": 4.8717,
      "mean_token_accuracy": 0.21880101710557937,
      "num_tokens": 82622881.0,
      "step": 39820
    },
    {
      "entropy": 5.643942832946777,
      "epoch": 3.6125725689404935,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003737110521264206,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22164962738752364,
      "num_tokens": 82633933.0,
      "step": 39825
    },
    {
      "entropy": 5.6477015018463135,
      "epoch": 3.6130261248185778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003736819747898702,
      "loss": 4.9001,
      "mean_token_accuracy": 0.21923135668039323,
      "num_tokens": 82645069.0,
      "step": 39830
    },
    {
      "entropy": 5.70770492553711,
      "epoch": 3.6134796806966616,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037365289541255154,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2215166762471199,
      "num_tokens": 82655152.0,
      "step": 39835
    },
    {
      "entropy": 5.632300901412964,
      "epoch": 3.613933236574746,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037362381399506575,
      "loss": 4.8554,
      "mean_token_accuracy": 0.2210676446557045,
      "num_tokens": 82665260.0,
      "step": 39840
    },
    {
      "entropy": 5.653606414794922,
      "epoch": 3.61438679245283,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037359473053801435,
      "loss": 4.8746,
      "mean_token_accuracy": 0.2259746313095093,
      "num_tokens": 82676287.0,
      "step": 39845
    },
    {
      "entropy": 5.669798898696899,
      "epoch": 3.6148403483309144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003735656450419988,
      "loss": 4.9861,
      "mean_token_accuracy": 0.20864200741052627,
      "num_tokens": 82686958.0,
      "step": 39850
    },
    {
      "entropy": 5.710234689712524,
      "epoch": 3.6152939042089987,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037353655750762063,
      "loss": 5.0262,
      "mean_token_accuracy": 0.20692770928144455,
      "num_tokens": 82698623.0,
      "step": 39855
    },
    {
      "entropy": 5.761024808883667,
      "epoch": 3.615747460087083,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037350746793548135,
      "loss": 4.9583,
      "mean_token_accuracy": 0.21551277488470078,
      "num_tokens": 82709146.0,
      "step": 39860
    },
    {
      "entropy": 5.688005685806274,
      "epoch": 3.6162010159651667,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037347837632618246,
      "loss": 4.848,
      "mean_token_accuracy": 0.2248758226633072,
      "num_tokens": 82719323.0,
      "step": 39865
    },
    {
      "entropy": 5.698855447769165,
      "epoch": 3.616654571843251,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037344928268032574,
      "loss": 4.9385,
      "mean_token_accuracy": 0.212944395840168,
      "num_tokens": 82730068.0,
      "step": 39870
    },
    {
      "entropy": 5.683442449569702,
      "epoch": 3.6171081277213353,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003734201869985127,
      "loss": 4.9412,
      "mean_token_accuracy": 0.22457023859024047,
      "num_tokens": 82740460.0,
      "step": 39875
    },
    {
      "entropy": 5.670044040679931,
      "epoch": 3.6175616835994195,
      "grad_norm": 1.25,
      "learning_rate": 0.0003733910892813451,
      "loss": 4.8582,
      "mean_token_accuracy": 0.21491052359342575,
      "num_tokens": 82751488.0,
      "step": 39880
    },
    {
      "entropy": 5.583886432647705,
      "epoch": 3.6180152394775034,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037336198952942477,
      "loss": 4.7917,
      "mean_token_accuracy": 0.2368409141898155,
      "num_tokens": 82762061.0,
      "step": 39885
    },
    {
      "entropy": 5.628071308135986,
      "epoch": 3.6184687953555876,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003733328877433534,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23112846463918685,
      "num_tokens": 82773421.0,
      "step": 39890
    },
    {
      "entropy": 5.588958644866944,
      "epoch": 3.618922351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037330378392373285,
      "loss": 4.7432,
      "mean_token_accuracy": 0.2292736440896988,
      "num_tokens": 82783342.0,
      "step": 39895
    },
    {
      "entropy": 5.634418058395386,
      "epoch": 3.619375907111756,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037327467807116507,
      "loss": 4.8954,
      "mean_token_accuracy": 0.21333642303943634,
      "num_tokens": 82793885.0,
      "step": 39900
    },
    {
      "entropy": 5.6729795932769775,
      "epoch": 3.6198294629898404,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037324557018625186,
      "loss": 4.9354,
      "mean_token_accuracy": 0.21814895421266556,
      "num_tokens": 82803746.0,
      "step": 39905
    },
    {
      "entropy": 5.672806596755981,
      "epoch": 3.6202830188679247,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003732164602695952,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22315026819705963,
      "num_tokens": 82813858.0,
      "step": 39910
    },
    {
      "entropy": 5.733905839920044,
      "epoch": 3.620736574746009,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037318734832179705,
      "loss": 4.9328,
      "mean_token_accuracy": 0.22309547513723374,
      "num_tokens": 82824358.0,
      "step": 39915
    },
    {
      "entropy": 5.704417896270752,
      "epoch": 3.621190130624093,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037315823434345963,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2315208524465561,
      "num_tokens": 82835801.0,
      "step": 39920
    },
    {
      "entropy": 5.6033868312835695,
      "epoch": 3.621643686502177,
      "grad_norm": 1.25,
      "learning_rate": 0.0003731291183351848,
      "loss": 4.8153,
      "mean_token_accuracy": 0.23146494925022126,
      "num_tokens": 82845818.0,
      "step": 39925
    },
    {
      "entropy": 5.633762693405151,
      "epoch": 3.6220972423802613,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037310000029757486,
      "loss": 4.865,
      "mean_token_accuracy": 0.22187588661909102,
      "num_tokens": 82856920.0,
      "step": 39930
    },
    {
      "entropy": 5.717894458770752,
      "epoch": 3.6225507982583456,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003730708802312319,
      "loss": 4.9555,
      "mean_token_accuracy": 0.22265180200338364,
      "num_tokens": 82867220.0,
      "step": 39935
    },
    {
      "entropy": 5.641828870773315,
      "epoch": 3.6230043541364294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003730417581367582,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22642106115818023,
      "num_tokens": 82877541.0,
      "step": 39940
    },
    {
      "entropy": 5.705629968643189,
      "epoch": 3.6234579100145137,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003730126340147559,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22633495181798935,
      "num_tokens": 82888346.0,
      "step": 39945
    },
    {
      "entropy": 5.754661512374878,
      "epoch": 3.623911465892598,
      "grad_norm": 1.125,
      "learning_rate": 0.0003729835078658273,
      "loss": 4.9747,
      "mean_token_accuracy": 0.2207999750971794,
      "num_tokens": 82899721.0,
      "step": 39950
    },
    {
      "entropy": 5.687739324569702,
      "epoch": 3.624365021770682,
      "grad_norm": 1.25,
      "learning_rate": 0.00037295437969057493,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22289131879806517,
      "num_tokens": 82910149.0,
      "step": 39955
    },
    {
      "entropy": 5.691064929962158,
      "epoch": 3.6248185776487665,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037292524948960093,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22050891816616058,
      "num_tokens": 82920883.0,
      "step": 39960
    },
    {
      "entropy": 5.695002651214599,
      "epoch": 3.6252721335268507,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003728961172635078,
      "loss": 4.919,
      "mean_token_accuracy": 0.2221357062458992,
      "num_tokens": 82931091.0,
      "step": 39965
    },
    {
      "entropy": 5.68429651260376,
      "epoch": 3.6257256894049346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037286698301289804,
      "loss": 4.8132,
      "mean_token_accuracy": 0.22303204238414764,
      "num_tokens": 82941108.0,
      "step": 39970
    },
    {
      "entropy": 5.6372827053070065,
      "epoch": 3.626179245283019,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003728378467383742,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22563837766647338,
      "num_tokens": 82951943.0,
      "step": 39975
    },
    {
      "entropy": 5.723476600646973,
      "epoch": 3.626632801161103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003728087084405387,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22061725407838823,
      "num_tokens": 82961495.0,
      "step": 39980
    },
    {
      "entropy": 5.725362634658813,
      "epoch": 3.6270863570391874,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037277956811999414,
      "loss": 4.909,
      "mean_token_accuracy": 0.22251984626054763,
      "num_tokens": 82971858.0,
      "step": 39985
    },
    {
      "entropy": 5.681453609466553,
      "epoch": 3.627539912917271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003727504257773433,
      "loss": 4.8783,
      "mean_token_accuracy": 0.222503162920475,
      "num_tokens": 82982805.0,
      "step": 39990
    },
    {
      "entropy": 5.680622720718384,
      "epoch": 3.6279934687953554,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037272128141318866,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2192270427942276,
      "num_tokens": 82993446.0,
      "step": 39995
    },
    {
      "entropy": 5.65215802192688,
      "epoch": 3.6284470246734397,
      "grad_norm": 1.328125,
      "learning_rate": 0.000372692135028133,
      "loss": 4.8085,
      "mean_token_accuracy": 0.22494857907295226,
      "num_tokens": 83003378.0,
      "step": 40000
    },
    {
      "entropy": 5.652303409576416,
      "epoch": 3.628900580551524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037266298662277914,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22131829857826232,
      "num_tokens": 83014223.0,
      "step": 40005
    },
    {
      "entropy": 5.637817287445069,
      "epoch": 3.6293541364296082,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037263383619772986,
      "loss": 4.9273,
      "mean_token_accuracy": 0.2153915971517563,
      "num_tokens": 83024691.0,
      "step": 40010
    },
    {
      "entropy": 5.694369697570801,
      "epoch": 3.6298076923076925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037260468375358794,
      "loss": 4.8961,
      "mean_token_accuracy": 0.217877396941185,
      "num_tokens": 83035704.0,
      "step": 40015
    },
    {
      "entropy": 5.714944744110108,
      "epoch": 3.6302612481857763,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003725755292909562,
      "loss": 4.9915,
      "mean_token_accuracy": 0.21680707782506942,
      "num_tokens": 83047173.0,
      "step": 40020
    },
    {
      "entropy": 5.728207063674927,
      "epoch": 3.6307148040638606,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003725463728104378,
      "loss": 4.9299,
      "mean_token_accuracy": 0.21833959519863128,
      "num_tokens": 83056399.0,
      "step": 40025
    },
    {
      "entropy": 5.6368865966796875,
      "epoch": 3.631168359941945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003725172143126354,
      "loss": 4.8356,
      "mean_token_accuracy": 0.21750013679265975,
      "num_tokens": 83066413.0,
      "step": 40030
    },
    {
      "entropy": 5.644077730178833,
      "epoch": 3.631621915820029,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037248805379815227,
      "loss": 4.9164,
      "mean_token_accuracy": 0.22086847722530364,
      "num_tokens": 83077843.0,
      "step": 40035
    },
    {
      "entropy": 5.727738618850708,
      "epoch": 3.632075471698113,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003724588912675913,
      "loss": 4.9702,
      "mean_token_accuracy": 0.22031510174274443,
      "num_tokens": 83088585.0,
      "step": 40040
    },
    {
      "entropy": 5.7743042469024655,
      "epoch": 3.6325290275761972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003724297267215557,
      "loss": 5.0232,
      "mean_token_accuracy": 0.21739590764045716,
      "num_tokens": 83099295.0,
      "step": 40045
    },
    {
      "entropy": 5.628675889968872,
      "epoch": 3.6329825834542815,
      "grad_norm": 1.375,
      "learning_rate": 0.00037240056016064844,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23047059625387192,
      "num_tokens": 83108530.0,
      "step": 40050
    },
    {
      "entropy": 5.638234186172485,
      "epoch": 3.6334361393323658,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003723713915854728,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22063491642475128,
      "num_tokens": 83119703.0,
      "step": 40055
    },
    {
      "entropy": 5.681389284133911,
      "epoch": 3.63388969521045,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003723422209966319,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22032863050699233,
      "num_tokens": 83129323.0,
      "step": 40060
    },
    {
      "entropy": 5.690658283233643,
      "epoch": 3.6343432510885343,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003723130483947292,
      "loss": 4.8411,
      "mean_token_accuracy": 0.22923820614814758,
      "num_tokens": 83139339.0,
      "step": 40065
    },
    {
      "entropy": 5.650324392318725,
      "epoch": 3.6347968069666186,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003722838737803678,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2249443858861923,
      "num_tokens": 83149845.0,
      "step": 40070
    },
    {
      "entropy": 5.731730508804321,
      "epoch": 3.6352503628447024,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037225469715415106,
      "loss": 4.9125,
      "mean_token_accuracy": 0.21429021656513214,
      "num_tokens": 83160332.0,
      "step": 40075
    },
    {
      "entropy": 5.701664066314697,
      "epoch": 3.6357039187227866,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037222551851668245,
      "loss": 4.9195,
      "mean_token_accuracy": 0.21624959707260133,
      "num_tokens": 83170892.0,
      "step": 40080
    },
    {
      "entropy": 5.637016725540161,
      "epoch": 3.636157474600871,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003721963378685652,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22707261592149736,
      "num_tokens": 83181035.0,
      "step": 40085
    },
    {
      "entropy": 5.618593549728393,
      "epoch": 3.6366110304789547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037216715521040314,
      "loss": 4.874,
      "mean_token_accuracy": 0.22818093299865722,
      "num_tokens": 83191886.0,
      "step": 40090
    },
    {
      "entropy": 5.665148687362671,
      "epoch": 3.637064586357039,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003721379705427994,
      "loss": 4.8736,
      "mean_token_accuracy": 0.22596771121025086,
      "num_tokens": 83202356.0,
      "step": 40095
    },
    {
      "entropy": 5.698509073257446,
      "epoch": 3.6375181422351233,
      "grad_norm": 1.34375,
      "learning_rate": 0.00037210878386635774,
      "loss": 4.9319,
      "mean_token_accuracy": 0.22892556190490723,
      "num_tokens": 83212909.0,
      "step": 40100
    },
    {
      "entropy": 5.7326172351837155,
      "epoch": 3.6379716981132075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037207959518168165,
      "loss": 4.9157,
      "mean_token_accuracy": 0.2213135242462158,
      "num_tokens": 83223523.0,
      "step": 40105
    },
    {
      "entropy": 5.766475057601928,
      "epoch": 3.638425253991292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037205040448937485,
      "loss": 5.013,
      "mean_token_accuracy": 0.21175449341535568,
      "num_tokens": 83233665.0,
      "step": 40110
    },
    {
      "entropy": 5.630761814117432,
      "epoch": 3.638878809869376,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003720212117900409,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22685135900974274,
      "num_tokens": 83243563.0,
      "step": 40115
    },
    {
      "entropy": 5.6522105693817135,
      "epoch": 3.6393323657474603,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037199201708428356,
      "loss": 4.9242,
      "mean_token_accuracy": 0.2208442971110344,
      "num_tokens": 83254246.0,
      "step": 40120
    },
    {
      "entropy": 5.7031529426574705,
      "epoch": 3.639785921625544,
      "grad_norm": 1.25,
      "learning_rate": 0.0003719628203727066,
      "loss": 4.9206,
      "mean_token_accuracy": 0.21978748440742493,
      "num_tokens": 83264361.0,
      "step": 40125
    },
    {
      "entropy": 5.726064014434814,
      "epoch": 3.6402394775036284,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003719336216559138,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2226356416940689,
      "num_tokens": 83274193.0,
      "step": 40130
    },
    {
      "entropy": 5.68732442855835,
      "epoch": 3.6406930333817127,
      "grad_norm": 1.359375,
      "learning_rate": 0.000371904420934509,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22709486335515977,
      "num_tokens": 83284593.0,
      "step": 40135
    },
    {
      "entropy": 5.637549543380738,
      "epoch": 3.641146589259797,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003718752182090961,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22113175839185714,
      "num_tokens": 83294602.0,
      "step": 40140
    },
    {
      "entropy": 5.6569477081298825,
      "epoch": 3.6416001451378808,
      "grad_norm": 1.25,
      "learning_rate": 0.000371846013480279,
      "loss": 4.8766,
      "mean_token_accuracy": 0.2252274438738823,
      "num_tokens": 83305399.0,
      "step": 40145
    },
    {
      "entropy": 5.60464506149292,
      "epoch": 3.642053701015965,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037181680674866165,
      "loss": 4.8391,
      "mean_token_accuracy": 0.22242993265390396,
      "num_tokens": 83315696.0,
      "step": 40150
    },
    {
      "entropy": 5.753710746765137,
      "epoch": 3.6425072568940493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003717875980148481,
      "loss": 5.0312,
      "mean_token_accuracy": 0.2147990271449089,
      "num_tokens": 83326686.0,
      "step": 40155
    },
    {
      "entropy": 5.694650745391845,
      "epoch": 3.6429608127721336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003717583872794423,
      "loss": 4.8588,
      "mean_token_accuracy": 0.2301585152745247,
      "num_tokens": 83336902.0,
      "step": 40160
    },
    {
      "entropy": 5.7124261379241945,
      "epoch": 3.643414368650218,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003717291745430484,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22530333399772645,
      "num_tokens": 83347324.0,
      "step": 40165
    },
    {
      "entropy": 5.60787992477417,
      "epoch": 3.643867924528302,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037169995980627056,
      "loss": 4.8468,
      "mean_token_accuracy": 0.2185913935303688,
      "num_tokens": 83358458.0,
      "step": 40170
    },
    {
      "entropy": 5.586744737625122,
      "epoch": 3.644321480406386,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037167074306971285,
      "loss": 4.9396,
      "mean_token_accuracy": 0.21516866534948348,
      "num_tokens": 83369047.0,
      "step": 40175
    },
    {
      "entropy": 5.700639533996582,
      "epoch": 3.64477503628447,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037164152433397964,
      "loss": 4.8926,
      "mean_token_accuracy": 0.22242873907089233,
      "num_tokens": 83378891.0,
      "step": 40180
    },
    {
      "entropy": 5.683143901824951,
      "epoch": 3.6452285921625545,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000371612303599675,
      "loss": 4.9432,
      "mean_token_accuracy": 0.21869627237319947,
      "num_tokens": 83389289.0,
      "step": 40185
    },
    {
      "entropy": 5.724655628204346,
      "epoch": 3.6456821480406387,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037158308086740336,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2237531915307045,
      "num_tokens": 83399316.0,
      "step": 40190
    },
    {
      "entropy": 5.637756729125977,
      "epoch": 3.6461357039187225,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003715538561377689,
      "loss": 4.9042,
      "mean_token_accuracy": 0.22534003853797913,
      "num_tokens": 83409507.0,
      "step": 40195
    },
    {
      "entropy": 5.733842420578003,
      "epoch": 3.646589259796807,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003715246294113762,
      "loss": 4.9965,
      "mean_token_accuracy": 0.20672013610601425,
      "num_tokens": 83419687.0,
      "step": 40200
    },
    {
      "entropy": 5.767635154724121,
      "epoch": 3.647042815674891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037149540068882945,
      "loss": 4.9975,
      "mean_token_accuracy": 0.21497793346643448,
      "num_tokens": 83429711.0,
      "step": 40205
    },
    {
      "entropy": 5.6566397190094,
      "epoch": 3.6474963715529753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037146616997073327,
      "loss": 4.8541,
      "mean_token_accuracy": 0.21538568139076233,
      "num_tokens": 83440116.0,
      "step": 40210
    },
    {
      "entropy": 5.683903455734253,
      "epoch": 3.6479499274310596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037143693725769214,
      "loss": 4.8742,
      "mean_token_accuracy": 0.21972752809524537,
      "num_tokens": 83450877.0,
      "step": 40215
    },
    {
      "entropy": 5.619827175140381,
      "epoch": 3.648403483309144,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037140770255031066,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22795045971870423,
      "num_tokens": 83460080.0,
      "step": 40220
    },
    {
      "entropy": 5.662614488601685,
      "epoch": 3.6488570391872277,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037137846584919316,
      "loss": 4.9387,
      "mean_token_accuracy": 0.21711855083703996,
      "num_tokens": 83469933.0,
      "step": 40225
    },
    {
      "entropy": 5.622121143341064,
      "epoch": 3.649310595065312,
      "grad_norm": 1.125,
      "learning_rate": 0.0003713492271549445,
      "loss": 4.849,
      "mean_token_accuracy": 0.2308749660849571,
      "num_tokens": 83480477.0,
      "step": 40230
    },
    {
      "entropy": 5.674810552597046,
      "epoch": 3.6497641509433962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003713199864681693,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21794286817312242,
      "num_tokens": 83491022.0,
      "step": 40235
    },
    {
      "entropy": 5.72891674041748,
      "epoch": 3.6502177068214805,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037129074378947216,
      "loss": 4.9449,
      "mean_token_accuracy": 0.22049839049577713,
      "num_tokens": 83501643.0,
      "step": 40240
    },
    {
      "entropy": 5.737355470657349,
      "epoch": 3.6506712626995643,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037126149911945796,
      "loss": 4.9333,
      "mean_token_accuracy": 0.21215626895427703,
      "num_tokens": 83511948.0,
      "step": 40245
    },
    {
      "entropy": 5.697805261611938,
      "epoch": 3.6511248185776486,
      "grad_norm": 1.28125,
      "learning_rate": 0.00037123225245873137,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2223981261253357,
      "num_tokens": 83521939.0,
      "step": 40250
    },
    {
      "entropy": 5.734273862838745,
      "epoch": 3.651578374455733,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037120300380789727,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22584975957870485,
      "num_tokens": 83532098.0,
      "step": 40255
    },
    {
      "entropy": 5.703018665313721,
      "epoch": 3.652031930333817,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037117375316756063,
      "loss": 4.9256,
      "mean_token_accuracy": 0.22374141216278076,
      "num_tokens": 83543238.0,
      "step": 40260
    },
    {
      "entropy": 5.711447668075562,
      "epoch": 3.6524854862119014,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003711445005383263,
      "loss": 4.9242,
      "mean_token_accuracy": 0.21964828968048095,
      "num_tokens": 83552969.0,
      "step": 40265
    },
    {
      "entropy": 5.72319393157959,
      "epoch": 3.6529390420899857,
      "grad_norm": 1.25,
      "learning_rate": 0.00037111524592079896,
      "loss": 4.9598,
      "mean_token_accuracy": 0.21171899884939194,
      "num_tokens": 83564569.0,
      "step": 40270
    },
    {
      "entropy": 5.711685752868652,
      "epoch": 3.65339259796807,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000371085989315584,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2156510755419731,
      "num_tokens": 83575613.0,
      "step": 40275
    },
    {
      "entropy": 5.710333156585693,
      "epoch": 3.6538461538461537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003710567307232864,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21965270936489106,
      "num_tokens": 83586311.0,
      "step": 40280
    },
    {
      "entropy": 5.724339866638184,
      "epoch": 3.654299709724238,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037102747014451095,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21794047951698303,
      "num_tokens": 83596660.0,
      "step": 40285
    },
    {
      "entropy": 5.722314071655274,
      "epoch": 3.6547532656023223,
      "grad_norm": 1.34375,
      "learning_rate": 0.000370998207579863,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2221145063638687,
      "num_tokens": 83606882.0,
      "step": 40290
    },
    {
      "entropy": 5.7511543273925785,
      "epoch": 3.6552068214804065,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003709689430299477,
      "loss": 4.9632,
      "mean_token_accuracy": 0.2198385626077652,
      "num_tokens": 83617380.0,
      "step": 40295
    },
    {
      "entropy": 5.708741331100464,
      "epoch": 3.6556603773584904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003709396764953701,
      "loss": 4.9192,
      "mean_token_accuracy": 0.218975792825222,
      "num_tokens": 83627961.0,
      "step": 40300
    },
    {
      "entropy": 5.57231879234314,
      "epoch": 3.6561139332365746,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003709104079767357,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23020369410514832,
      "num_tokens": 83637646.0,
      "step": 40305
    },
    {
      "entropy": 5.656616449356079,
      "epoch": 3.656567489114659,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037088113747464955,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22575811594724654,
      "num_tokens": 83647807.0,
      "step": 40310
    },
    {
      "entropy": 5.650777816772461,
      "epoch": 3.657021044992743,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000370851864989717,
      "loss": 4.929,
      "mean_token_accuracy": 0.21596544235944748,
      "num_tokens": 83658002.0,
      "step": 40315
    },
    {
      "entropy": 5.661614418029785,
      "epoch": 3.6574746008708274,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003708225905225435,
      "loss": 4.8624,
      "mean_token_accuracy": 0.23165522515773773,
      "num_tokens": 83667875.0,
      "step": 40320
    },
    {
      "entropy": 5.66379599571228,
      "epoch": 3.6579281567489117,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003707933140737345,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22126011848449706,
      "num_tokens": 83678376.0,
      "step": 40325
    },
    {
      "entropy": 5.695123243331909,
      "epoch": 3.6583817126269955,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037076403564389516,
      "loss": 4.9922,
      "mean_token_accuracy": 0.20902622640132903,
      "num_tokens": 83688318.0,
      "step": 40330
    },
    {
      "entropy": 5.679830551147461,
      "epoch": 3.65883526850508,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003707347552336313,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22017918676137924,
      "num_tokens": 83698935.0,
      "step": 40335
    },
    {
      "entropy": 5.712050056457519,
      "epoch": 3.659288824383164,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003707054728435482,
      "loss": 4.971,
      "mean_token_accuracy": 0.22246617823839188,
      "num_tokens": 83709382.0,
      "step": 40340
    },
    {
      "entropy": 5.666761589050293,
      "epoch": 3.6597423802612483,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003706761884742516,
      "loss": 4.8043,
      "mean_token_accuracy": 0.23317637741565705,
      "num_tokens": 83719661.0,
      "step": 40345
    },
    {
      "entropy": 5.770358467102051,
      "epoch": 3.660195936139332,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037064690212634707,
      "loss": 4.9681,
      "mean_token_accuracy": 0.21352480351924896,
      "num_tokens": 83730679.0,
      "step": 40350
    },
    {
      "entropy": 5.602434396743774,
      "epoch": 3.6606494920174164,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003706176138004401,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22150979042053223,
      "num_tokens": 83740595.0,
      "step": 40355
    },
    {
      "entropy": 5.6442187309265135,
      "epoch": 3.6611030478955007,
      "grad_norm": 1.25,
      "learning_rate": 0.00037058832349713667,
      "loss": 4.963,
      "mean_token_accuracy": 0.220208241045475,
      "num_tokens": 83751605.0,
      "step": 40360
    },
    {
      "entropy": 5.780175113677979,
      "epoch": 3.661556603773585,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003705590312170422,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21680912673473357,
      "num_tokens": 83763780.0,
      "step": 40365
    },
    {
      "entropy": 5.688521385192871,
      "epoch": 3.662010159651669,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037052973696076274,
      "loss": 4.8386,
      "mean_token_accuracy": 0.2276061400771141,
      "num_tokens": 83773800.0,
      "step": 40370
    },
    {
      "entropy": 5.739528751373291,
      "epoch": 3.6624637155297535,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037050044072890395,
      "loss": 4.9519,
      "mean_token_accuracy": 0.20574979484081268,
      "num_tokens": 83784402.0,
      "step": 40375
    },
    {
      "entropy": 5.665516662597656,
      "epoch": 3.6629172714078373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003704711425220717,
      "loss": 4.8619,
      "mean_token_accuracy": 0.21817622184753419,
      "num_tokens": 83794852.0,
      "step": 40380
    },
    {
      "entropy": 5.696832084655762,
      "epoch": 3.6633708272859216,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003704418423408718,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2201572149991989,
      "num_tokens": 83805673.0,
      "step": 40385
    },
    {
      "entropy": 5.668293094635009,
      "epoch": 3.663824383164006,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003704125401859104,
      "loss": 4.9049,
      "mean_token_accuracy": 0.2188924565911293,
      "num_tokens": 83815779.0,
      "step": 40390
    },
    {
      "entropy": 5.727199745178223,
      "epoch": 3.66427793904209,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00037038323605779333,
      "loss": 4.8984,
      "mean_token_accuracy": 0.22401462495326996,
      "num_tokens": 83826804.0,
      "step": 40395
    },
    {
      "entropy": 5.7207818031311035,
      "epoch": 3.664731494920174,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003703539299571266,
      "loss": 4.9294,
      "mean_token_accuracy": 0.21933402568101884,
      "num_tokens": 83836500.0,
      "step": 40400
    },
    {
      "entropy": 5.699020671844482,
      "epoch": 3.665185050798258,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003703246218845163,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22574329674243926,
      "num_tokens": 83846647.0,
      "step": 40405
    },
    {
      "entropy": 5.6076253890991214,
      "epoch": 3.6656386066763424,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003702953118405686,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22984277158975602,
      "num_tokens": 83858088.0,
      "step": 40410
    },
    {
      "entropy": 5.682080173492432,
      "epoch": 3.6660921625544267,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003702659998258895,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22277461886405944,
      "num_tokens": 83868936.0,
      "step": 40415
    },
    {
      "entropy": 5.769212436676026,
      "epoch": 3.666545718432511,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037023668584108523,
      "loss": 5.0279,
      "mean_token_accuracy": 0.21163960099220275,
      "num_tokens": 83879813.0,
      "step": 40420
    },
    {
      "entropy": 5.659196376800537,
      "epoch": 3.6669992743105952,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037020736988676206,
      "loss": 4.865,
      "mean_token_accuracy": 0.21778764426708222,
      "num_tokens": 83891619.0,
      "step": 40425
    },
    {
      "entropy": 5.734442949295044,
      "epoch": 3.6674528301886795,
      "grad_norm": 1.328125,
      "learning_rate": 0.00037017805196352623,
      "loss": 4.9992,
      "mean_token_accuracy": 0.22008715867996215,
      "num_tokens": 83901905.0,
      "step": 40430
    },
    {
      "entropy": 5.629171657562256,
      "epoch": 3.6679063860667633,
      "grad_norm": 1.3125,
      "learning_rate": 0.000370148732071984,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22062896192073822,
      "num_tokens": 83912827.0,
      "step": 40435
    },
    {
      "entropy": 5.6329998016357425,
      "epoch": 3.6683599419448476,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003701194102127418,
      "loss": 4.809,
      "mean_token_accuracy": 0.22896229177713395,
      "num_tokens": 83923719.0,
      "step": 40440
    },
    {
      "entropy": 5.582939386367798,
      "epoch": 3.668813497822932,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003700900863864059,
      "loss": 4.8677,
      "mean_token_accuracy": 0.22614769637584686,
      "num_tokens": 83933715.0,
      "step": 40445
    },
    {
      "entropy": 5.689361476898194,
      "epoch": 3.6692670537010157,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003700607605935828,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21600780338048936,
      "num_tokens": 83943946.0,
      "step": 40450
    },
    {
      "entropy": 5.725706768035889,
      "epoch": 3.6697206095791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003700314328348789,
      "loss": 4.9347,
      "mean_token_accuracy": 0.22168777137994766,
      "num_tokens": 83953414.0,
      "step": 40455
    },
    {
      "entropy": 5.654720449447632,
      "epoch": 3.6701741654571842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003700021031109008,
      "loss": 4.8597,
      "mean_token_accuracy": 0.21863938570022584,
      "num_tokens": 83963717.0,
      "step": 40460
    },
    {
      "entropy": 5.659605646133423,
      "epoch": 3.6706277213352685,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036997277142225506,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22110912948846817,
      "num_tokens": 83973467.0,
      "step": 40465
    },
    {
      "entropy": 5.676168394088745,
      "epoch": 3.6710812772133528,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036994343776954806,
      "loss": 4.9178,
      "mean_token_accuracy": 0.2224588766694069,
      "num_tokens": 83984013.0,
      "step": 40470
    },
    {
      "entropy": 5.669651699066162,
      "epoch": 3.671534833091437,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003699141021533867,
      "loss": 4.8082,
      "mean_token_accuracy": 0.22749508768320084,
      "num_tokens": 83994349.0,
      "step": 40475
    },
    {
      "entropy": 5.638855266571045,
      "epoch": 3.6719883889695213,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003698847645743775,
      "loss": 4.807,
      "mean_token_accuracy": 0.2226794794201851,
      "num_tokens": 84004525.0,
      "step": 40480
    },
    {
      "entropy": 5.635551929473877,
      "epoch": 3.672441944847605,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003698554250331271,
      "loss": 4.8904,
      "mean_token_accuracy": 0.21972500085830687,
      "num_tokens": 84014286.0,
      "step": 40485
    },
    {
      "entropy": 5.6224384784698485,
      "epoch": 3.6728955007256894,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003698260835302424,
      "loss": 4.86,
      "mean_token_accuracy": 0.2279515191912651,
      "num_tokens": 84024826.0,
      "step": 40490
    },
    {
      "entropy": 5.712609767913818,
      "epoch": 3.6733490566037736,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003697967400663302,
      "loss": 4.9162,
      "mean_token_accuracy": 0.21949510127305985,
      "num_tokens": 84035726.0,
      "step": 40495
    },
    {
      "entropy": 5.615817356109619,
      "epoch": 3.673802612481858,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036976739464199715,
      "loss": 4.7903,
      "mean_token_accuracy": 0.22861216962337494,
      "num_tokens": 84045573.0,
      "step": 40500
    },
    {
      "entropy": 5.684405088424683,
      "epoch": 3.6742561683599417,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036973804725785024,
      "loss": 4.9097,
      "mean_token_accuracy": 0.2171785742044449,
      "num_tokens": 84055983.0,
      "step": 40505
    },
    {
      "entropy": 5.744190359115601,
      "epoch": 3.674709724238026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036970869791449644,
      "loss": 4.9877,
      "mean_token_accuracy": 0.21835035234689712,
      "num_tokens": 84066905.0,
      "step": 40510
    },
    {
      "entropy": 5.635688638687133,
      "epoch": 3.6751632801161103,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036967934661254263,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23689092546701432,
      "num_tokens": 84076861.0,
      "step": 40515
    },
    {
      "entropy": 5.695151042938233,
      "epoch": 3.6756168359941945,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003696499933525957,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22276612967252732,
      "num_tokens": 84086167.0,
      "step": 40520
    },
    {
      "entropy": 5.615292882919311,
      "epoch": 3.676070391872279,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036962063813526287,
      "loss": 4.861,
      "mean_token_accuracy": 0.22461163401603698,
      "num_tokens": 84095306.0,
      "step": 40525
    },
    {
      "entropy": 5.607744121551514,
      "epoch": 3.676523947750363,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003695912809611511,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22045087665319443,
      "num_tokens": 84106505.0,
      "step": 40530
    },
    {
      "entropy": 5.706215476989746,
      "epoch": 3.676977503628447,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003695619218308676,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2283039793372154,
      "num_tokens": 84116393.0,
      "step": 40535
    },
    {
      "entropy": 5.685896968841552,
      "epoch": 3.677431059506531,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036953256074501937,
      "loss": 4.9213,
      "mean_token_accuracy": 0.22188993245363237,
      "num_tokens": 84125890.0,
      "step": 40540
    },
    {
      "entropy": 5.705290174484253,
      "epoch": 3.6778846153846154,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003695031977042138,
      "loss": 4.8994,
      "mean_token_accuracy": 0.22659522145986558,
      "num_tokens": 84137736.0,
      "step": 40545
    },
    {
      "entropy": 5.678357076644898,
      "epoch": 3.6783381712626997,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036947383270905795,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22135156244039536,
      "num_tokens": 84148974.0,
      "step": 40550
    },
    {
      "entropy": 5.6700136184692385,
      "epoch": 3.6787917271407835,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036944446576015914,
      "loss": 4.9489,
      "mean_token_accuracy": 0.2194286897778511,
      "num_tokens": 84159041.0,
      "step": 40555
    },
    {
      "entropy": 5.627922582626343,
      "epoch": 3.6792452830188678,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036941509685812476,
      "loss": 4.803,
      "mean_token_accuracy": 0.22867572009563447,
      "num_tokens": 84169432.0,
      "step": 40560
    },
    {
      "entropy": 5.640451145172119,
      "epoch": 3.679698838896952,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036938572600356213,
      "loss": 4.856,
      "mean_token_accuracy": 0.22374287545681,
      "num_tokens": 84179964.0,
      "step": 40565
    },
    {
      "entropy": 5.659222555160523,
      "epoch": 3.6801523947750363,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036935635319707856,
      "loss": 4.9244,
      "mean_token_accuracy": 0.21474254578351976,
      "num_tokens": 84189496.0,
      "step": 40570
    },
    {
      "entropy": 5.688618278503418,
      "epoch": 3.6806059506531206,
      "grad_norm": 1.25,
      "learning_rate": 0.0003693269784392816,
      "loss": 4.8749,
      "mean_token_accuracy": 0.23535956740379332,
      "num_tokens": 84200396.0,
      "step": 40575
    },
    {
      "entropy": 5.599977254867554,
      "epoch": 3.681059506531205,
      "grad_norm": 1.484375,
      "learning_rate": 0.00036929760173077863,
      "loss": 4.8306,
      "mean_token_accuracy": 0.2232038050889969,
      "num_tokens": 84210114.0,
      "step": 40580
    },
    {
      "entropy": 5.660508537292481,
      "epoch": 3.6815130624092887,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003692682230721773,
      "loss": 4.872,
      "mean_token_accuracy": 0.2253366470336914,
      "num_tokens": 84220472.0,
      "step": 40585
    },
    {
      "entropy": 5.702978897094726,
      "epoch": 3.681966618287373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003692388424640851,
      "loss": 4.9264,
      "mean_token_accuracy": 0.21808141320943833,
      "num_tokens": 84230718.0,
      "step": 40590
    },
    {
      "entropy": 5.700717449188232,
      "epoch": 3.682420174165457,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036920945990710964,
      "loss": 4.9076,
      "mean_token_accuracy": 0.221554896235466,
      "num_tokens": 84241736.0,
      "step": 40595
    },
    {
      "entropy": 5.581788158416748,
      "epoch": 3.6828737300435415,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036918007540185846,
      "loss": 4.8097,
      "mean_token_accuracy": 0.22812772989273072,
      "num_tokens": 84251770.0,
      "step": 40600
    },
    {
      "entropy": 5.665844535827636,
      "epoch": 3.6833272859216253,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003691506889489394,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22274309396743774,
      "num_tokens": 84263355.0,
      "step": 40605
    },
    {
      "entropy": 5.673920011520385,
      "epoch": 3.6837808417997095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036912130054896005,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21238254755735397,
      "num_tokens": 84274835.0,
      "step": 40610
    },
    {
      "entropy": 5.759057188034058,
      "epoch": 3.684234397677794,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036909191020252825,
      "loss": 5.003,
      "mean_token_accuracy": 0.2085989609360695,
      "num_tokens": 84286099.0,
      "step": 40615
    },
    {
      "entropy": 5.7494548797607425,
      "epoch": 3.684687953555878,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003690625179102517,
      "loss": 4.9608,
      "mean_token_accuracy": 0.21804235130548477,
      "num_tokens": 84296168.0,
      "step": 40620
    },
    {
      "entropy": 5.783842658996582,
      "epoch": 3.6851415094339623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003690331236727384,
      "loss": 5.0239,
      "mean_token_accuracy": 0.20619226843118668,
      "num_tokens": 84307352.0,
      "step": 40625
    },
    {
      "entropy": 5.682150554656983,
      "epoch": 3.6855950653120466,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003690037274905961,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21834431886672973,
      "num_tokens": 84318496.0,
      "step": 40630
    },
    {
      "entropy": 5.618512392044067,
      "epoch": 3.686048621190131,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003689743293644327,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22308747172355653,
      "num_tokens": 84329101.0,
      "step": 40635
    },
    {
      "entropy": 5.649944877624511,
      "epoch": 3.6865021770682147,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003689449292948563,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22666050791740416,
      "num_tokens": 84339143.0,
      "step": 40640
    },
    {
      "entropy": 5.752368974685669,
      "epoch": 3.686955732946299,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036891552728247483,
      "loss": 4.8872,
      "mean_token_accuracy": 0.22207492887973784,
      "num_tokens": 84349693.0,
      "step": 40645
    },
    {
      "entropy": 5.704638719558716,
      "epoch": 3.6874092888243832,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036888612332789636,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21930698901414872,
      "num_tokens": 84359799.0,
      "step": 40650
    },
    {
      "entropy": 5.6655210018157955,
      "epoch": 3.6878628447024675,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036885671743172875,
      "loss": 4.904,
      "mean_token_accuracy": 0.22459696531295775,
      "num_tokens": 84370695.0,
      "step": 40655
    },
    {
      "entropy": 5.673503446578979,
      "epoch": 3.6883164005805513,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003688273095945805,
      "loss": 4.9383,
      "mean_token_accuracy": 0.2119108334183693,
      "num_tokens": 84380205.0,
      "step": 40660
    },
    {
      "entropy": 5.710888051986695,
      "epoch": 3.6887699564586356,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036879789981705954,
      "loss": 4.9052,
      "mean_token_accuracy": 0.22754597216844558,
      "num_tokens": 84391034.0,
      "step": 40665
    },
    {
      "entropy": 5.743909692764282,
      "epoch": 3.68922351233672,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036876848809977404,
      "loss": 4.9765,
      "mean_token_accuracy": 0.21121110022068024,
      "num_tokens": 84401452.0,
      "step": 40670
    },
    {
      "entropy": 5.657085943222046,
      "epoch": 3.689677068214804,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036873907444333234,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2235492616891861,
      "num_tokens": 84411126.0,
      "step": 40675
    },
    {
      "entropy": 5.680313491821289,
      "epoch": 3.6901306240928884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036870965884834274,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22094174474477768,
      "num_tokens": 84422740.0,
      "step": 40680
    },
    {
      "entropy": 5.696887016296387,
      "epoch": 3.6905841799709727,
      "grad_norm": 1.421875,
      "learning_rate": 0.00036868024131541337,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21187540739774705,
      "num_tokens": 84432831.0,
      "step": 40685
    },
    {
      "entropy": 5.734337615966797,
      "epoch": 3.6910377358490565,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036865082184515296,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22372198849916458,
      "num_tokens": 84443176.0,
      "step": 40690
    },
    {
      "entropy": 5.7312757015228275,
      "epoch": 3.6914912917271407,
      "grad_norm": 1.375,
      "learning_rate": 0.00036862140043816957,
      "loss": 4.97,
      "mean_token_accuracy": 0.21707645654678345,
      "num_tokens": 84453185.0,
      "step": 40695
    },
    {
      "entropy": 5.722420024871826,
      "epoch": 3.691944847605225,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003685919770950718,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22058170884847642,
      "num_tokens": 84464303.0,
      "step": 40700
    },
    {
      "entropy": 5.6997528076171875,
      "epoch": 3.6923984034833093,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036856255181646797,
      "loss": 4.9204,
      "mean_token_accuracy": 0.21839746087789536,
      "num_tokens": 84474463.0,
      "step": 40705
    },
    {
      "entropy": 5.660377502441406,
      "epoch": 3.692851959361393,
      "grad_norm": 1.25,
      "learning_rate": 0.0003685331246029668,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22162384986877443,
      "num_tokens": 84484682.0,
      "step": 40710
    },
    {
      "entropy": 5.694066333770752,
      "epoch": 3.6933055152394774,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036850369545517685,
      "loss": 4.8584,
      "mean_token_accuracy": 0.2243731737136841,
      "num_tokens": 84494830.0,
      "step": 40715
    },
    {
      "entropy": 5.654220628738403,
      "epoch": 3.6937590711175616,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003684742643737065,
      "loss": 4.8498,
      "mean_token_accuracy": 0.22292278409004213,
      "num_tokens": 84505449.0,
      "step": 40720
    },
    {
      "entropy": 5.722480249404907,
      "epoch": 3.694212626995646,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036844483135916467,
      "loss": 4.9423,
      "mean_token_accuracy": 0.2236463099718094,
      "num_tokens": 84515833.0,
      "step": 40725
    },
    {
      "entropy": 5.716444110870361,
      "epoch": 3.69466618287373,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003684153964121599,
      "loss": 4.9659,
      "mean_token_accuracy": 0.21427191644906998,
      "num_tokens": 84526677.0,
      "step": 40730
    },
    {
      "entropy": 5.688922500610351,
      "epoch": 3.6951197387518144,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036838595953330095,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22531062364578247,
      "num_tokens": 84536656.0,
      "step": 40735
    },
    {
      "entropy": 5.7193928241729735,
      "epoch": 3.6955732946298983,
      "grad_norm": 1.25,
      "learning_rate": 0.00036835652072319646,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22763002067804336,
      "num_tokens": 84546506.0,
      "step": 40740
    },
    {
      "entropy": 5.700044345855713,
      "epoch": 3.6960268505079825,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003683270799824554,
      "loss": 4.9466,
      "mean_token_accuracy": 0.22135970890522003,
      "num_tokens": 84556921.0,
      "step": 40745
    },
    {
      "entropy": 5.710176610946656,
      "epoch": 3.696480406386067,
      "grad_norm": 1.25,
      "learning_rate": 0.0003682976373116865,
      "loss": 4.8927,
      "mean_token_accuracy": 0.22593822181224824,
      "num_tokens": 84566325.0,
      "step": 40750
    },
    {
      "entropy": 5.619585132598877,
      "epoch": 3.696933962264151,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036826819271149877,
      "loss": 4.8412,
      "mean_token_accuracy": 0.22471843659877777,
      "num_tokens": 84577167.0,
      "step": 40755
    },
    {
      "entropy": 5.652046394348145,
      "epoch": 3.697387518142235,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036823874618250105,
      "loss": 4.8685,
      "mean_token_accuracy": 0.21999358236789704,
      "num_tokens": 84586764.0,
      "step": 40760
    },
    {
      "entropy": 5.639049625396728,
      "epoch": 3.697841074020319,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036820929772530224,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22250303775072097,
      "num_tokens": 84598903.0,
      "step": 40765
    },
    {
      "entropy": 5.635731220245361,
      "epoch": 3.6982946298984034,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003681798473405114,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22968205362558364,
      "num_tokens": 84608695.0,
      "step": 40770
    },
    {
      "entropy": 5.620333003997803,
      "epoch": 3.6987481857764877,
      "grad_norm": 1.25,
      "learning_rate": 0.00036815039502873756,
      "loss": 4.8614,
      "mean_token_accuracy": 0.22320092469453812,
      "num_tokens": 84619377.0,
      "step": 40775
    },
    {
      "entropy": 5.6478018283844,
      "epoch": 3.699201741654572,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036812094079058985,
      "loss": 4.8318,
      "mean_token_accuracy": 0.22340811938047409,
      "num_tokens": 84629716.0,
      "step": 40780
    },
    {
      "entropy": 5.605634641647339,
      "epoch": 3.699655297532656,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003680914846266773,
      "loss": 4.7588,
      "mean_token_accuracy": 0.23379543721675872,
      "num_tokens": 84640214.0,
      "step": 40785
    },
    {
      "entropy": 5.679317951202393,
      "epoch": 3.7001088534107405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003680620265376092,
      "loss": 4.9723,
      "mean_token_accuracy": 0.218359312415123,
      "num_tokens": 84650403.0,
      "step": 40790
    },
    {
      "entropy": 5.679017877578735,
      "epoch": 3.7005624092888243,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036803256652399456,
      "loss": 4.845,
      "mean_token_accuracy": 0.22457738667726518,
      "num_tokens": 84660912.0,
      "step": 40795
    },
    {
      "entropy": 5.6840416431427006,
      "epoch": 3.7010159651669086,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036800310458644285,
      "loss": 4.9313,
      "mean_token_accuracy": 0.2122257396578789,
      "num_tokens": 84672194.0,
      "step": 40800
    },
    {
      "entropy": 5.676962375640869,
      "epoch": 3.701469521044993,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036797364072556316,
      "loss": 4.991,
      "mean_token_accuracy": 0.2174428731203079,
      "num_tokens": 84681850.0,
      "step": 40805
    },
    {
      "entropy": 5.7112311840057375,
      "epoch": 3.7019230769230766,
      "grad_norm": 1.28125,
      "learning_rate": 0.000367944174941965,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22835011929273605,
      "num_tokens": 84691271.0,
      "step": 40810
    },
    {
      "entropy": 5.648849153518677,
      "epoch": 3.702376632801161,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036791470723625746,
      "loss": 4.8822,
      "mean_token_accuracy": 0.21749144792556763,
      "num_tokens": 84702189.0,
      "step": 40815
    },
    {
      "entropy": 5.657384204864502,
      "epoch": 3.702830188679245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003678852376090502,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2315467193722725,
      "num_tokens": 84712399.0,
      "step": 40820
    },
    {
      "entropy": 5.715691375732422,
      "epoch": 3.7032837445573294,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003678557660609525,
      "loss": 5.0199,
      "mean_token_accuracy": 0.21369219571352005,
      "num_tokens": 84723174.0,
      "step": 40825
    },
    {
      "entropy": 5.710756063461304,
      "epoch": 3.7037373004354137,
      "grad_norm": 1.25,
      "learning_rate": 0.000367826292592574,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22941849678754805,
      "num_tokens": 84732498.0,
      "step": 40830
    },
    {
      "entropy": 5.715919733047485,
      "epoch": 3.704190856313498,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003677968172045241,
      "loss": 4.9188,
      "mean_token_accuracy": 0.2171079620718956,
      "num_tokens": 84742362.0,
      "step": 40835
    },
    {
      "entropy": 5.659933280944824,
      "epoch": 3.7046444121915822,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003677673398974122,
      "loss": 4.8953,
      "mean_token_accuracy": 0.21976277977228165,
      "num_tokens": 84753379.0,
      "step": 40840
    },
    {
      "entropy": 5.718342304229736,
      "epoch": 3.705097968069666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003677378606718482,
      "loss": 4.9882,
      "mean_token_accuracy": 0.2170700192451477,
      "num_tokens": 84763245.0,
      "step": 40845
    },
    {
      "entropy": 5.704492902755737,
      "epoch": 3.7055515239477503,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003677083795284416,
      "loss": 4.932,
      "mean_token_accuracy": 0.21407582908868789,
      "num_tokens": 84773929.0,
      "step": 40850
    },
    {
      "entropy": 5.73481330871582,
      "epoch": 3.7060050798258346,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036767889646780204,
      "loss": 4.9315,
      "mean_token_accuracy": 0.2147434875369072,
      "num_tokens": 84784733.0,
      "step": 40855
    },
    {
      "entropy": 5.684029865264892,
      "epoch": 3.706458635703919,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003676494114905394,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2191468134522438,
      "num_tokens": 84794254.0,
      "step": 40860
    },
    {
      "entropy": 5.735726356506348,
      "epoch": 3.7069121915820027,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036761992459726325,
      "loss": 4.9028,
      "mean_token_accuracy": 0.2130702927708626,
      "num_tokens": 84804361.0,
      "step": 40865
    },
    {
      "entropy": 5.6978765487670895,
      "epoch": 3.707365747460087,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036759043578858344,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22717825174331666,
      "num_tokens": 84814497.0,
      "step": 40870
    },
    {
      "entropy": 5.6994226455688475,
      "epoch": 3.7078193033381712,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036756094506510984,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22064704596996307,
      "num_tokens": 84826871.0,
      "step": 40875
    },
    {
      "entropy": 5.667500925064087,
      "epoch": 3.7082728592162555,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00036753145242745226,
      "loss": 4.9044,
      "mean_token_accuracy": 0.2184149831533432,
      "num_tokens": 84837053.0,
      "step": 40880
    },
    {
      "entropy": 5.585552835464478,
      "epoch": 3.7087264150943398,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003675019578762208,
      "loss": 4.843,
      "mean_token_accuracy": 0.22476999014616011,
      "num_tokens": 84846681.0,
      "step": 40885
    },
    {
      "entropy": 5.621462392807007,
      "epoch": 3.709179970972424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003674724614120251,
      "loss": 4.8244,
      "mean_token_accuracy": 0.2224101647734642,
      "num_tokens": 84856651.0,
      "step": 40890
    },
    {
      "entropy": 5.663849830627441,
      "epoch": 3.709633526850508,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036744296303547547,
      "loss": 4.9807,
      "mean_token_accuracy": 0.2159203678369522,
      "num_tokens": 84867531.0,
      "step": 40895
    },
    {
      "entropy": 5.63166298866272,
      "epoch": 3.710087082728592,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003674134627471818,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22616903781890868,
      "num_tokens": 84879132.0,
      "step": 40900
    },
    {
      "entropy": 5.690239477157593,
      "epoch": 3.7105406386066764,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000367383960547754,
      "loss": 4.9055,
      "mean_token_accuracy": 0.23163840770721436,
      "num_tokens": 84888873.0,
      "step": 40905
    },
    {
      "entropy": 5.680754709243774,
      "epoch": 3.7109941944847606,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036735445643780256,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2162686198949814,
      "num_tokens": 84898926.0,
      "step": 40910
    },
    {
      "entropy": 5.684827613830566,
      "epoch": 3.7114477503628445,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003673249504179374,
      "loss": 4.897,
      "mean_token_accuracy": 0.21948649883270263,
      "num_tokens": 84908502.0,
      "step": 40915
    },
    {
      "entropy": 5.677953720092773,
      "epoch": 3.7119013062409287,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003672954424887687,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21775904446840286,
      "num_tokens": 84919681.0,
      "step": 40920
    },
    {
      "entropy": 5.607191514968872,
      "epoch": 3.712354862119013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003672659326509066,
      "loss": 4.8181,
      "mean_token_accuracy": 0.22815295904874802,
      "num_tokens": 84930149.0,
      "step": 40925
    },
    {
      "entropy": 5.61807689666748,
      "epoch": 3.7128084179970973,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036723642090496166,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22373404055833818,
      "num_tokens": 84940086.0,
      "step": 40930
    },
    {
      "entropy": 5.716367244720459,
      "epoch": 3.7132619738751815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036720690725154404,
      "loss": 4.9326,
      "mean_token_accuracy": 0.22301631271839142,
      "num_tokens": 84949248.0,
      "step": 40935
    },
    {
      "entropy": 5.645527648925781,
      "epoch": 3.713715529753266,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036717739169126395,
      "loss": 4.8453,
      "mean_token_accuracy": 0.22551789432764052,
      "num_tokens": 84960892.0,
      "step": 40940
    },
    {
      "entropy": 5.660330677032471,
      "epoch": 3.7141690856313496,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036714787422473206,
      "loss": 4.886,
      "mean_token_accuracy": 0.216607566177845,
      "num_tokens": 84971024.0,
      "step": 40945
    },
    {
      "entropy": 5.608395433425903,
      "epoch": 3.714622641509434,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003671183548525586,
      "loss": 4.8109,
      "mean_token_accuracy": 0.223887699842453,
      "num_tokens": 84980945.0,
      "step": 40950
    },
    {
      "entropy": 5.6810417652130125,
      "epoch": 3.715076197387518,
      "grad_norm": 1.265625,
      "learning_rate": 0.000367088833575354,
      "loss": 4.9483,
      "mean_token_accuracy": 0.2136263832449913,
      "num_tokens": 84992055.0,
      "step": 40955
    },
    {
      "entropy": 5.711512231826783,
      "epoch": 3.7155297532656024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036705931039372904,
      "loss": 4.9237,
      "mean_token_accuracy": 0.22107691317796707,
      "num_tokens": 85002619.0,
      "step": 40960
    },
    {
      "entropy": 5.723572635650635,
      "epoch": 3.7159833091436862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003670297853082939,
      "loss": 4.8948,
      "mean_token_accuracy": 0.2250138908624649,
      "num_tokens": 85013317.0,
      "step": 40965
    },
    {
      "entropy": 5.694820976257324,
      "epoch": 3.7164368650217705,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036700025831965937,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22417686581611634,
      "num_tokens": 85023457.0,
      "step": 40970
    },
    {
      "entropy": 5.661444234848022,
      "epoch": 3.7168904208998548,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036697072942843616,
      "loss": 4.879,
      "mean_token_accuracy": 0.2200169399380684,
      "num_tokens": 85033498.0,
      "step": 40975
    },
    {
      "entropy": 5.822324514389038,
      "epoch": 3.717343976777939,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003669411986352348,
      "loss": 5.0912,
      "mean_token_accuracy": 0.21519254744052888,
      "num_tokens": 85043687.0,
      "step": 40980
    },
    {
      "entropy": 5.7554267883300785,
      "epoch": 3.7177975326560233,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000366911665940666,
      "loss": 4.9392,
      "mean_token_accuracy": 0.21988731622695923,
      "num_tokens": 85055045.0,
      "step": 40985
    },
    {
      "entropy": 5.671528148651123,
      "epoch": 3.7182510885341076,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036688213134534044,
      "loss": 4.8449,
      "mean_token_accuracy": 0.2341105580329895,
      "num_tokens": 85064625.0,
      "step": 40990
    },
    {
      "entropy": 5.759242391586303,
      "epoch": 3.718704644412192,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003668525948498691,
      "loss": 5.049,
      "mean_token_accuracy": 0.20288423597812652,
      "num_tokens": 85075249.0,
      "step": 40995
    },
    {
      "entropy": 5.673050832748413,
      "epoch": 3.7191582002902757,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036682305645486266,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22368443757295609,
      "num_tokens": 85085900.0,
      "step": 41000
    },
    {
      "entropy": 5.680192708969116,
      "epoch": 3.71961175616836,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036679351616093194,
      "loss": 4.9337,
      "mean_token_accuracy": 0.21972575038671494,
      "num_tokens": 85095772.0,
      "step": 41005
    },
    {
      "entropy": 5.668636465072632,
      "epoch": 3.720065312046444,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036676397396868797,
      "loss": 4.8485,
      "mean_token_accuracy": 0.2213755413889885,
      "num_tokens": 85105653.0,
      "step": 41010
    },
    {
      "entropy": 5.714299488067627,
      "epoch": 3.7205188679245285,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003667344298787417,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2167710095643997,
      "num_tokens": 85115432.0,
      "step": 41015
    },
    {
      "entropy": 5.761515140533447,
      "epoch": 3.7209724238026123,
      "grad_norm": 1.25,
      "learning_rate": 0.00036670488389170395,
      "loss": 5.075,
      "mean_token_accuracy": 0.20838387459516525,
      "num_tokens": 85126572.0,
      "step": 41020
    },
    {
      "entropy": 5.620250558853149,
      "epoch": 3.7214259796806965,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036667533600818586,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2284772589802742,
      "num_tokens": 85136809.0,
      "step": 41025
    },
    {
      "entropy": 5.871505498886108,
      "epoch": 3.721879535558781,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003666457862287985,
      "loss": 5.0646,
      "mean_token_accuracy": 0.20963408797979355,
      "num_tokens": 85148346.0,
      "step": 41030
    },
    {
      "entropy": 5.684710168838501,
      "epoch": 3.722333091436865,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036661623455415296,
      "loss": 4.9033,
      "mean_token_accuracy": 0.21899901777505876,
      "num_tokens": 85158807.0,
      "step": 41035
    },
    {
      "entropy": 5.702105474472046,
      "epoch": 3.7227866473149493,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003665866809848602,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2097207114100456,
      "num_tokens": 85170752.0,
      "step": 41040
    },
    {
      "entropy": 5.664738607406616,
      "epoch": 3.7232402031930336,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036655712552153164,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2209311008453369,
      "num_tokens": 85180779.0,
      "step": 41045
    },
    {
      "entropy": 5.7262864112854,
      "epoch": 3.7236937590711174,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003665275681647784,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22774630635976792,
      "num_tokens": 85191391.0,
      "step": 41050
    },
    {
      "entropy": 5.716647291183472,
      "epoch": 3.7241473149492017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003664980089152116,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22189580500125886,
      "num_tokens": 85201357.0,
      "step": 41055
    },
    {
      "entropy": 5.691839218139648,
      "epoch": 3.724600870827286,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00036646844777344284,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22176516503095628,
      "num_tokens": 85212193.0,
      "step": 41060
    },
    {
      "entropy": 5.701785135269165,
      "epoch": 3.7250544267053702,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036643888474008313,
      "loss": 4.9312,
      "mean_token_accuracy": 0.22183466255664824,
      "num_tokens": 85223809.0,
      "step": 41065
    },
    {
      "entropy": 5.686916828155518,
      "epoch": 3.725507982583454,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003664093198157441,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22361328303813935,
      "num_tokens": 85234992.0,
      "step": 41070
    },
    {
      "entropy": 5.765382957458496,
      "epoch": 3.7259615384615383,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003663797530010369,
      "loss": 4.9468,
      "mean_token_accuracy": 0.21633796244859696,
      "num_tokens": 85244157.0,
      "step": 41075
    },
    {
      "entropy": 5.653511190414429,
      "epoch": 3.7264150943396226,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003663501842965732,
      "loss": 4.8734,
      "mean_token_accuracy": 0.23080657720565795,
      "num_tokens": 85255946.0,
      "step": 41080
    },
    {
      "entropy": 5.648684883117676,
      "epoch": 3.726868650217707,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003663206137029645,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22713794857263564,
      "num_tokens": 85265273.0,
      "step": 41085
    },
    {
      "entropy": 5.6115447044372555,
      "epoch": 3.727322206095791,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036629104122082215,
      "loss": 4.7281,
      "mean_token_accuracy": 0.231294147670269,
      "num_tokens": 85275794.0,
      "step": 41090
    },
    {
      "entropy": 5.68388295173645,
      "epoch": 3.7277757619738754,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036626146685075775,
      "loss": 4.9923,
      "mean_token_accuracy": 0.21017343550920486,
      "num_tokens": 85287334.0,
      "step": 41095
    },
    {
      "entropy": 5.703068017959595,
      "epoch": 3.728229317851959,
      "grad_norm": 1.296875,
      "learning_rate": 0.000366231890593383,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2216135859489441,
      "num_tokens": 85297320.0,
      "step": 41100
    },
    {
      "entropy": 5.663260173797608,
      "epoch": 3.7286828737300435,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003662023124493095,
      "loss": 4.8158,
      "mean_token_accuracy": 0.2258962407708168,
      "num_tokens": 85308071.0,
      "step": 41105
    },
    {
      "entropy": 5.72218017578125,
      "epoch": 3.7291364296081277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000366172732419149,
      "loss": 4.9209,
      "mean_token_accuracy": 0.2156672239303589,
      "num_tokens": 85317790.0,
      "step": 41110
    },
    {
      "entropy": 5.71658844947815,
      "epoch": 3.729589985486212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036614315050351317,
      "loss": 4.9877,
      "mean_token_accuracy": 0.21400294303894044,
      "num_tokens": 85329008.0,
      "step": 41115
    },
    {
      "entropy": 5.796836280822754,
      "epoch": 3.730043541364296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003661135667030137,
      "loss": 4.9993,
      "mean_token_accuracy": 0.21302020251750947,
      "num_tokens": 85339095.0,
      "step": 41120
    },
    {
      "entropy": 5.726266384124756,
      "epoch": 3.73049709724238,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003660839810182624,
      "loss": 4.9357,
      "mean_token_accuracy": 0.21982518285512925,
      "num_tokens": 85349833.0,
      "step": 41125
    },
    {
      "entropy": 5.7357179641723635,
      "epoch": 3.7309506531204644,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036605439344987125,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21451949179172516,
      "num_tokens": 85359512.0,
      "step": 41130
    },
    {
      "entropy": 5.720212841033936,
      "epoch": 3.7314042089985486,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000366024803998452,
      "loss": 4.9437,
      "mean_token_accuracy": 0.22229668498039246,
      "num_tokens": 85370053.0,
      "step": 41135
    },
    {
      "entropy": 5.692185544967652,
      "epoch": 3.731857764876633,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036599521266461653,
      "loss": 4.9067,
      "mean_token_accuracy": 0.222061362862587,
      "num_tokens": 85380020.0,
      "step": 41140
    },
    {
      "entropy": 5.691695976257324,
      "epoch": 3.732311320754717,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036596561944897703,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22381123453378676,
      "num_tokens": 85390232.0,
      "step": 41145
    },
    {
      "entropy": 5.605344629287719,
      "epoch": 3.7327648766328014,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003659360243521453,
      "loss": 4.8211,
      "mean_token_accuracy": 0.23350778818130494,
      "num_tokens": 85400401.0,
      "step": 41150
    },
    {
      "entropy": 5.673161602020263,
      "epoch": 3.7332184325108853,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003659064273747334,
      "loss": 4.9364,
      "mean_token_accuracy": 0.21452258229255677,
      "num_tokens": 85411749.0,
      "step": 41155
    },
    {
      "entropy": 5.74624662399292,
      "epoch": 3.7336719883889695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036587682851735336,
      "loss": 4.9966,
      "mean_token_accuracy": 0.21385712623596193,
      "num_tokens": 85424065.0,
      "step": 41160
    },
    {
      "entropy": 5.7279259204864506,
      "epoch": 3.734125544267054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036584722778061743,
      "loss": 4.9358,
      "mean_token_accuracy": 0.22008656710386276,
      "num_tokens": 85434412.0,
      "step": 41165
    },
    {
      "entropy": 5.695285511016846,
      "epoch": 3.7345791001451376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003658176251651376,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22418590635061264,
      "num_tokens": 85444489.0,
      "step": 41170
    },
    {
      "entropy": 5.721828556060791,
      "epoch": 3.735032656023222,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036578802067152615,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22025670409202575,
      "num_tokens": 85454556.0,
      "step": 41175
    },
    {
      "entropy": 5.63813099861145,
      "epoch": 3.735486211901306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003657584143003954,
      "loss": 4.8329,
      "mean_token_accuracy": 0.22858185470104217,
      "num_tokens": 85464701.0,
      "step": 41180
    },
    {
      "entropy": 5.69109616279602,
      "epoch": 3.7359397677793904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003657288060523574,
      "loss": 4.9638,
      "mean_token_accuracy": 0.22157543599605561,
      "num_tokens": 85475657.0,
      "step": 41185
    },
    {
      "entropy": 5.7124816417694095,
      "epoch": 3.7363933236574747,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036569919592802466,
      "loss": 4.8898,
      "mean_token_accuracy": 0.22905835807323455,
      "num_tokens": 85486358.0,
      "step": 41190
    },
    {
      "entropy": 5.7485902309417725,
      "epoch": 3.736846879535559,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003656695839280093,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21779648065567017,
      "num_tokens": 85497913.0,
      "step": 41195
    },
    {
      "entropy": 5.687849473953247,
      "epoch": 3.737300435413643,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036563997005292396,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2272608682513237,
      "num_tokens": 85508314.0,
      "step": 41200
    },
    {
      "entropy": 5.631940793991089,
      "epoch": 3.737753991291727,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036561035430338084,
      "loss": 4.8943,
      "mean_token_accuracy": 0.21926860511302948,
      "num_tokens": 85518142.0,
      "step": 41205
    },
    {
      "entropy": 5.663657379150391,
      "epoch": 3.7382075471698113,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036558073667999257,
      "loss": 4.7836,
      "mean_token_accuracy": 0.22842101156711578,
      "num_tokens": 85526854.0,
      "step": 41210
    },
    {
      "entropy": 5.583002662658691,
      "epoch": 3.7386611030478956,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00036555111718337166,
      "loss": 4.7935,
      "mean_token_accuracy": 0.2333999067544937,
      "num_tokens": 85537153.0,
      "step": 41215
    },
    {
      "entropy": 5.7152691841125485,
      "epoch": 3.73911465892598,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003655214958141305,
      "loss": 4.9984,
      "mean_token_accuracy": 0.20878121703863145,
      "num_tokens": 85548583.0,
      "step": 41220
    },
    {
      "entropy": 5.6988016128540036,
      "epoch": 3.7395682148040637,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00036549187257288164,
      "loss": 4.9449,
      "mean_token_accuracy": 0.22795640975236892,
      "num_tokens": 85559228.0,
      "step": 41225
    },
    {
      "entropy": 5.7186614036560055,
      "epoch": 3.740021770682148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036546224746023795,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2229028105735779,
      "num_tokens": 85569757.0,
      "step": 41230
    },
    {
      "entropy": 5.661580610275268,
      "epoch": 3.740475326560232,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036543262047681186,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22795261293649674,
      "num_tokens": 85582013.0,
      "step": 41235
    },
    {
      "entropy": 5.699161767959595,
      "epoch": 3.7409288824383164,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003654029916232161,
      "loss": 4.8824,
      "mean_token_accuracy": 0.23454479575157167,
      "num_tokens": 85594049.0,
      "step": 41240
    },
    {
      "entropy": 5.706962203979492,
      "epoch": 3.7413824383164007,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036537336090006347,
      "loss": 4.9311,
      "mean_token_accuracy": 0.21831998080015183,
      "num_tokens": 85605253.0,
      "step": 41245
    },
    {
      "entropy": 5.6473023891448975,
      "epoch": 3.741835994194485,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036534372830796674,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22834509164094924,
      "num_tokens": 85615032.0,
      "step": 41250
    },
    {
      "entropy": 5.651205444335938,
      "epoch": 3.742289550072569,
      "grad_norm": 1.25,
      "learning_rate": 0.0003653140938475386,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2306559979915619,
      "num_tokens": 85625497.0,
      "step": 41255
    },
    {
      "entropy": 5.652278804779053,
      "epoch": 3.742743105950653,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036528445751939206,
      "loss": 4.9276,
      "mean_token_accuracy": 0.2206938534975052,
      "num_tokens": 85636058.0,
      "step": 41260
    },
    {
      "entropy": 5.679962825775147,
      "epoch": 3.7431966618287373,
      "grad_norm": 1.359375,
      "learning_rate": 0.00036525481932413997,
      "loss": 4.9117,
      "mean_token_accuracy": 0.21783520728349687,
      "num_tokens": 85646505.0,
      "step": 41265
    },
    {
      "entropy": 5.738751363754273,
      "epoch": 3.7436502177068216,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036522517926239513,
      "loss": 4.8983,
      "mean_token_accuracy": 0.21850587874650956,
      "num_tokens": 85656864.0,
      "step": 41270
    },
    {
      "entropy": 5.623754978179932,
      "epoch": 3.7441037735849054,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036519553733477055,
      "loss": 4.825,
      "mean_token_accuracy": 0.22952149510383607,
      "num_tokens": 85666783.0,
      "step": 41275
    },
    {
      "entropy": 5.593802976608276,
      "epoch": 3.7445573294629897,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003651658935418794,
      "loss": 4.785,
      "mean_token_accuracy": 0.23101411610841752,
      "num_tokens": 85677102.0,
      "step": 41280
    },
    {
      "entropy": 5.620873308181762,
      "epoch": 3.745010885341074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036513624788433454,
      "loss": 4.8332,
      "mean_token_accuracy": 0.23130874186754227,
      "num_tokens": 85687048.0,
      "step": 41285
    },
    {
      "entropy": 5.629025173187256,
      "epoch": 3.7454644412191582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003651066003627491,
      "loss": 4.8907,
      "mean_token_accuracy": 0.21603457629680634,
      "num_tokens": 85698314.0,
      "step": 41290
    },
    {
      "entropy": 5.730273962020874,
      "epoch": 3.7459179970972425,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003650769509777362,
      "loss": 4.9614,
      "mean_token_accuracy": 0.21609768271446228,
      "num_tokens": 85708039.0,
      "step": 41295
    },
    {
      "entropy": 5.65498948097229,
      "epoch": 3.7463715529753268,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00036504729972990904,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2350754976272583,
      "num_tokens": 85717741.0,
      "step": 41300
    },
    {
      "entropy": 5.701639795303345,
      "epoch": 3.7468251088534106,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003650176466198807,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22574660629034043,
      "num_tokens": 85728286.0,
      "step": 41305
    },
    {
      "entropy": 5.653979253768921,
      "epoch": 3.747278664731495,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00036498799164826453,
      "loss": 4.873,
      "mean_token_accuracy": 0.22077085822820663,
      "num_tokens": 85738970.0,
      "step": 41310
    },
    {
      "entropy": 5.6406408786773685,
      "epoch": 3.747732220609579,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036495833481567375,
      "loss": 4.8409,
      "mean_token_accuracy": 0.22405920177698135,
      "num_tokens": 85749284.0,
      "step": 41315
    },
    {
      "entropy": 5.725311708450318,
      "epoch": 3.7481857764876634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036492867612272176,
      "loss": 4.9513,
      "mean_token_accuracy": 0.213972070813179,
      "num_tokens": 85759595.0,
      "step": 41320
    },
    {
      "entropy": 5.763889741897583,
      "epoch": 3.748639332365747,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003648990155700217,
      "loss": 4.9934,
      "mean_token_accuracy": 0.21423787921667098,
      "num_tokens": 85771049.0,
      "step": 41325
    },
    {
      "entropy": 5.682758378982544,
      "epoch": 3.7490928882438315,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036486935315818713,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2193435698747635,
      "num_tokens": 85781656.0,
      "step": 41330
    },
    {
      "entropy": 5.58990216255188,
      "epoch": 3.7495464441219157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036483968888783154,
      "loss": 4.753,
      "mean_token_accuracy": 0.23176947385072708,
      "num_tokens": 85790922.0,
      "step": 41335
    },
    {
      "entropy": 5.651162242889404,
      "epoch": 3.75,
      "grad_norm": 1.390625,
      "learning_rate": 0.00036481002275956813,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22995253652334213,
      "num_tokens": 85799928.0,
      "step": 41340
    },
    {
      "entropy": 5.68808422088623,
      "epoch": 3.7504535558780843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003647803547740106,
      "loss": 4.941,
      "mean_token_accuracy": 0.2140167772769928,
      "num_tokens": 85810594.0,
      "step": 41345
    },
    {
      "entropy": 5.6785407066345215,
      "epoch": 3.7509071117561685,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036475068493177254,
      "loss": 4.9148,
      "mean_token_accuracy": 0.21832190901041032,
      "num_tokens": 85820690.0,
      "step": 41350
    },
    {
      "entropy": 5.671871900558472,
      "epoch": 3.751360667634253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003647210132334674,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22500260323286056,
      "num_tokens": 85830314.0,
      "step": 41355
    },
    {
      "entropy": 5.662873458862305,
      "epoch": 3.7518142235123366,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003646913396797088,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2084158331155777,
      "num_tokens": 85841099.0,
      "step": 41360
    },
    {
      "entropy": 5.718393850326538,
      "epoch": 3.752267779390421,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036466166427111046,
      "loss": 4.8969,
      "mean_token_accuracy": 0.22112439572811127,
      "num_tokens": 85852813.0,
      "step": 41365
    },
    {
      "entropy": 5.690115118026734,
      "epoch": 3.752721335268505,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036463198700828616,
      "loss": 4.8366,
      "mean_token_accuracy": 0.22139388024806977,
      "num_tokens": 85862778.0,
      "step": 41370
    },
    {
      "entropy": 5.634028911590576,
      "epoch": 3.7531748911465894,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003646023078918494,
      "loss": 4.9132,
      "mean_token_accuracy": 0.22033293545246124,
      "num_tokens": 85872729.0,
      "step": 41375
    },
    {
      "entropy": 5.676275730133057,
      "epoch": 3.7536284470246732,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003645726269224141,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2259535610675812,
      "num_tokens": 85883289.0,
      "step": 41380
    },
    {
      "entropy": 5.618963766098022,
      "epoch": 3.7540820029027575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036454294410059407,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23423250168561935,
      "num_tokens": 85892318.0,
      "step": 41385
    },
    {
      "entropy": 5.631635332107544,
      "epoch": 3.7545355587808418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003645132594270032,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22386680245399476,
      "num_tokens": 85902719.0,
      "step": 41390
    },
    {
      "entropy": 5.652317333221435,
      "epoch": 3.754989114658926,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036448357290225523,
      "loss": 4.8621,
      "mean_token_accuracy": 0.22026864141225816,
      "num_tokens": 85913271.0,
      "step": 41395
    },
    {
      "entropy": 5.627652645111084,
      "epoch": 3.7554426705370103,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003644538845269642,
      "loss": 4.7985,
      "mean_token_accuracy": 0.2271735891699791,
      "num_tokens": 85922397.0,
      "step": 41400
    },
    {
      "entropy": 5.723268651962281,
      "epoch": 3.7558962264150946,
      "grad_norm": 1.203125,
      "learning_rate": 0.000364424194301744,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21691277176141738,
      "num_tokens": 85933261.0,
      "step": 41405
    },
    {
      "entropy": 5.76209454536438,
      "epoch": 3.7563497822931784,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003643945022272087,
      "loss": 4.967,
      "mean_token_accuracy": 0.22054621130228041,
      "num_tokens": 85943860.0,
      "step": 41410
    },
    {
      "entropy": 5.736132717132568,
      "epoch": 3.7568033381712627,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003643648083039724,
      "loss": 4.9663,
      "mean_token_accuracy": 0.21328079104423522,
      "num_tokens": 85953984.0,
      "step": 41415
    },
    {
      "entropy": 5.699458408355713,
      "epoch": 3.757256894049347,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000364335112532649,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22114909142255784,
      "num_tokens": 85964771.0,
      "step": 41420
    },
    {
      "entropy": 5.672175359725952,
      "epoch": 3.757710449927431,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003643054149138527,
      "loss": 4.9438,
      "mean_token_accuracy": 0.21471927911043168,
      "num_tokens": 85975859.0,
      "step": 41425
    },
    {
      "entropy": 5.763905477523804,
      "epoch": 3.758164005805515,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036427571544819765,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21731624901294708,
      "num_tokens": 85987022.0,
      "step": 41430
    },
    {
      "entropy": 5.656785535812378,
      "epoch": 3.7586175616835993,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003642460141362981,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22487738132476806,
      "num_tokens": 85997056.0,
      "step": 41435
    },
    {
      "entropy": 5.7189067840576175,
      "epoch": 3.7590711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.00036421631097876823,
      "loss": 4.968,
      "mean_token_accuracy": 0.21532460451126098,
      "num_tokens": 86008853.0,
      "step": 41440
    },
    {
      "entropy": 5.6411303043365475,
      "epoch": 3.759524673439768,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003641866059762223,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22100775092840194,
      "num_tokens": 86020358.0,
      "step": 41445
    },
    {
      "entropy": 5.689276313781738,
      "epoch": 3.759978229317852,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003641568991292747,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21461450159549714,
      "num_tokens": 86030706.0,
      "step": 41450
    },
    {
      "entropy": 5.780449151992798,
      "epoch": 3.7604317851959363,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036412719043853956,
      "loss": 4.9724,
      "mean_token_accuracy": 0.21685199439525604,
      "num_tokens": 86041964.0,
      "step": 41455
    },
    {
      "entropy": 5.734532451629638,
      "epoch": 3.76088534107402,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036409747990463146,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22409389764070511,
      "num_tokens": 86052213.0,
      "step": 41460
    },
    {
      "entropy": 5.621679830551147,
      "epoch": 3.7613388969521044,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003640677675281649,
      "loss": 4.9,
      "mean_token_accuracy": 0.21544238775968552,
      "num_tokens": 86061419.0,
      "step": 41465
    },
    {
      "entropy": 5.713098001480103,
      "epoch": 3.7617924528301887,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003640380533097541,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22301627248525618,
      "num_tokens": 86071585.0,
      "step": 41470
    },
    {
      "entropy": 5.687799978256225,
      "epoch": 3.762246008708273,
      "grad_norm": 1.328125,
      "learning_rate": 0.00036400833725001364,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2190204456448555,
      "num_tokens": 86081175.0,
      "step": 41475
    },
    {
      "entropy": 5.711560440063477,
      "epoch": 3.762699564586357,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003639786193495581,
      "loss": 4.952,
      "mean_token_accuracy": 0.21608454138040542,
      "num_tokens": 86092407.0,
      "step": 41480
    },
    {
      "entropy": 5.772978162765503,
      "epoch": 3.763153120464441,
      "grad_norm": 1.25,
      "learning_rate": 0.00036394889960900215,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21905975490808488,
      "num_tokens": 86102218.0,
      "step": 41485
    },
    {
      "entropy": 5.774706268310547,
      "epoch": 3.7636066763425253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003639191780289602,
      "loss": 4.9351,
      "mean_token_accuracy": 0.22194332182407378,
      "num_tokens": 86112631.0,
      "step": 41490
    },
    {
      "entropy": 5.687812232971192,
      "epoch": 3.7640602322206096,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003638894546100469,
      "loss": 4.953,
      "mean_token_accuracy": 0.20991104245185851,
      "num_tokens": 86123675.0,
      "step": 41495
    },
    {
      "entropy": 5.605954170227051,
      "epoch": 3.764513788098694,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036385972935287715,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22531697452068328,
      "num_tokens": 86133909.0,
      "step": 41500
    },
    {
      "entropy": 5.699400901794434,
      "epoch": 3.764967343976778,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003638300022580656,
      "loss": 4.916,
      "mean_token_accuracy": 0.2203238859772682,
      "num_tokens": 86144560.0,
      "step": 41505
    },
    {
      "entropy": 5.777115058898926,
      "epoch": 3.7654208998548624,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036380027332622684,
      "loss": 4.9622,
      "mean_token_accuracy": 0.21474549174308777,
      "num_tokens": 86155140.0,
      "step": 41510
    },
    {
      "entropy": 5.729059076309204,
      "epoch": 3.765874455732946,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00036377054255797584,
      "loss": 4.9312,
      "mean_token_accuracy": 0.21773631125688553,
      "num_tokens": 86165555.0,
      "step": 41515
    },
    {
      "entropy": 5.693825340270996,
      "epoch": 3.7663280116110305,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036374080995392737,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22147663682699203,
      "num_tokens": 86176204.0,
      "step": 41520
    },
    {
      "entropy": 5.72369532585144,
      "epoch": 3.7667815674891147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003637110755146964,
      "loss": 4.9726,
      "mean_token_accuracy": 0.21922099441289902,
      "num_tokens": 86186888.0,
      "step": 41525
    },
    {
      "entropy": 5.698044061660767,
      "epoch": 3.7672351233671986,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036368133924089766,
      "loss": 4.8586,
      "mean_token_accuracy": 0.2270179271697998,
      "num_tokens": 86196654.0,
      "step": 41530
    },
    {
      "entropy": 5.688607978820801,
      "epoch": 3.767688679245283,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003636516011331464,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22412585020065307,
      "num_tokens": 86207045.0,
      "step": 41535
    },
    {
      "entropy": 5.724132394790649,
      "epoch": 3.768142235123367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036362186119205734,
      "loss": 4.9511,
      "mean_token_accuracy": 0.21064476370811464,
      "num_tokens": 86216906.0,
      "step": 41540
    },
    {
      "entropy": 5.642863512039185,
      "epoch": 3.7685957910014514,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003635921194182457,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2209583804011345,
      "num_tokens": 86227380.0,
      "step": 41545
    },
    {
      "entropy": 5.714413022994995,
      "epoch": 3.7690493468795356,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003635623758123263,
      "loss": 5.022,
      "mean_token_accuracy": 0.22172670513391496,
      "num_tokens": 86236948.0,
      "step": 41550
    },
    {
      "entropy": 5.730819749832153,
      "epoch": 3.76950290275762,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036353263037491445,
      "loss": 4.9735,
      "mean_token_accuracy": 0.2147914484143257,
      "num_tokens": 86246784.0,
      "step": 41555
    },
    {
      "entropy": 5.668892478942871,
      "epoch": 3.769956458635704,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003635028831066253,
      "loss": 4.8381,
      "mean_token_accuracy": 0.2246150404214859,
      "num_tokens": 86257381.0,
      "step": 41560
    },
    {
      "entropy": 5.647143936157226,
      "epoch": 3.770410014513788,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003634731340080739,
      "loss": 4.8548,
      "mean_token_accuracy": 0.22447453886270524,
      "num_tokens": 86268117.0,
      "step": 41565
    },
    {
      "entropy": 5.813835477828979,
      "epoch": 3.7708635703918723,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003634433830798756,
      "loss": 5.0426,
      "mean_token_accuracy": 0.21462751030921937,
      "num_tokens": 86279195.0,
      "step": 41570
    },
    {
      "entropy": 5.607689666748047,
      "epoch": 3.7713171262699565,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003634136303226457,
      "loss": 4.7537,
      "mean_token_accuracy": 0.23417772054672242,
      "num_tokens": 86288154.0,
      "step": 41575
    },
    {
      "entropy": 5.671529626846313,
      "epoch": 3.771770682148041,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003633838757369992,
      "loss": 4.9122,
      "mean_token_accuracy": 0.21788786649703978,
      "num_tokens": 86297524.0,
      "step": 41580
    },
    {
      "entropy": 5.661235666275024,
      "epoch": 3.7722242380261246,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003633541193235517,
      "loss": 4.9792,
      "mean_token_accuracy": 0.21313594281673431,
      "num_tokens": 86307031.0,
      "step": 41585
    },
    {
      "entropy": 5.676587343215942,
      "epoch": 3.772677793904209,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003633243610829185,
      "loss": 4.939,
      "mean_token_accuracy": 0.22437526434659957,
      "num_tokens": 86317629.0,
      "step": 41590
    },
    {
      "entropy": 5.7200634479522705,
      "epoch": 3.773131349782293,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036329460101571496,
      "loss": 4.9784,
      "mean_token_accuracy": 0.2188064217567444,
      "num_tokens": 86328243.0,
      "step": 41595
    },
    {
      "entropy": 5.787331533432007,
      "epoch": 3.7735849056603774,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003632648391225568,
      "loss": 5.0046,
      "mean_token_accuracy": 0.21249126344919206,
      "num_tokens": 86338905.0,
      "step": 41600
    },
    {
      "entropy": 5.712730550765992,
      "epoch": 3.7740384615384617,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036323507540405903,
      "loss": 4.9449,
      "mean_token_accuracy": 0.22006287723779677,
      "num_tokens": 86350433.0,
      "step": 41605
    },
    {
      "entropy": 5.7437280178070065,
      "epoch": 3.774492017416546,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003632053098608374,
      "loss": 4.9376,
      "mean_token_accuracy": 0.2149263620376587,
      "num_tokens": 86360664.0,
      "step": 41610
    },
    {
      "entropy": 5.72393388748169,
      "epoch": 3.7749455732946298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036317554249350767,
      "loss": 4.9712,
      "mean_token_accuracy": 0.22229407727718353,
      "num_tokens": 86372101.0,
      "step": 41615
    },
    {
      "entropy": 5.671252346038818,
      "epoch": 3.775399129172714,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003631457733026851,
      "loss": 4.8218,
      "mean_token_accuracy": 0.22755935937166213,
      "num_tokens": 86382494.0,
      "step": 41620
    },
    {
      "entropy": 5.785129022598267,
      "epoch": 3.7758526850507983,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036311600228898554,
      "loss": 5.0595,
      "mean_token_accuracy": 0.20868812650442123,
      "num_tokens": 86393292.0,
      "step": 41625
    },
    {
      "entropy": 5.636383533477783,
      "epoch": 3.7763062409288826,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036308622945302455,
      "loss": 4.8412,
      "mean_token_accuracy": 0.223045015335083,
      "num_tokens": 86402940.0,
      "step": 41630
    },
    {
      "entropy": 5.63857889175415,
      "epoch": 3.7767597968069664,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000363056454795418,
      "loss": 4.879,
      "mean_token_accuracy": 0.21326947808265687,
      "num_tokens": 86413382.0,
      "step": 41635
    },
    {
      "entropy": 5.7477428913116455,
      "epoch": 3.7772133526850507,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003630266783167814,
      "loss": 4.9476,
      "mean_token_accuracy": 0.2155992403626442,
      "num_tokens": 86423866.0,
      "step": 41640
    },
    {
      "entropy": 5.787046909332275,
      "epoch": 3.777666908563135,
      "grad_norm": 1.640625,
      "learning_rate": 0.0003629969000177307,
      "loss": 5.0175,
      "mean_token_accuracy": 0.20921713262796401,
      "num_tokens": 86434908.0,
      "step": 41645
    },
    {
      "entropy": 5.615132856369018,
      "epoch": 3.778120464441219,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003629671198988817,
      "loss": 4.8212,
      "mean_token_accuracy": 0.23117538988590242,
      "num_tokens": 86445604.0,
      "step": 41650
    },
    {
      "entropy": 5.616502666473389,
      "epoch": 3.7785740203193035,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003629373379608502,
      "loss": 4.774,
      "mean_token_accuracy": 0.2339630350470543,
      "num_tokens": 86455813.0,
      "step": 41655
    },
    {
      "entropy": 5.597483396530151,
      "epoch": 3.7790275761973877,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036290755420425214,
      "loss": 4.7628,
      "mean_token_accuracy": 0.2384694755077362,
      "num_tokens": 86465256.0,
      "step": 41660
    },
    {
      "entropy": 5.6814440250396725,
      "epoch": 3.7794811320754715,
      "grad_norm": 1.3125,
      "learning_rate": 0.00036287776862970356,
      "loss": 4.89,
      "mean_token_accuracy": 0.22276489287614823,
      "num_tokens": 86474605.0,
      "step": 41665
    },
    {
      "entropy": 5.691185140609742,
      "epoch": 3.779934687953556,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036284798123782024,
      "loss": 4.8951,
      "mean_token_accuracy": 0.2197098597884178,
      "num_tokens": 86484886.0,
      "step": 41670
    },
    {
      "entropy": 5.6543690204620365,
      "epoch": 3.78038824383164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003628181920292182,
      "loss": 4.9703,
      "mean_token_accuracy": 0.21535971611738206,
      "num_tokens": 86495816.0,
      "step": 41675
    },
    {
      "entropy": 5.6754837989807125,
      "epoch": 3.7808417997097243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003627884010045138,
      "loss": 4.8656,
      "mean_token_accuracy": 0.21969487965106965,
      "num_tokens": 86506319.0,
      "step": 41680
    },
    {
      "entropy": 5.7648265838623045,
      "epoch": 3.781295355587808,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003627586081643227,
      "loss": 4.905,
      "mean_token_accuracy": 0.2221885934472084,
      "num_tokens": 86516952.0,
      "step": 41685
    },
    {
      "entropy": 5.760353088378906,
      "epoch": 3.7817489114658924,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036272881350926135,
      "loss": 5.0171,
      "mean_token_accuracy": 0.2143901988863945,
      "num_tokens": 86527327.0,
      "step": 41690
    },
    {
      "entropy": 5.698188400268554,
      "epoch": 3.7822024673439767,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003626990170399457,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22474419325590134,
      "num_tokens": 86538281.0,
      "step": 41695
    },
    {
      "entropy": 5.7246919631958,
      "epoch": 3.782656023222061,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036266921875699205,
      "loss": 4.9654,
      "mean_token_accuracy": 0.21591188460588456,
      "num_tokens": 86549034.0,
      "step": 41700
    },
    {
      "entropy": 5.690959739685058,
      "epoch": 3.7831095791001452,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003626394186610166,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21811742335557938,
      "num_tokens": 86559306.0,
      "step": 41705
    },
    {
      "entropy": 5.7528785228729244,
      "epoch": 3.7835631349782295,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003626096167526356,
      "loss": 4.9288,
      "mean_token_accuracy": 0.2223096564412117,
      "num_tokens": 86569414.0,
      "step": 41710
    },
    {
      "entropy": 5.582431507110596,
      "epoch": 3.7840166908563138,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003625798130324655,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2268217995762825,
      "num_tokens": 86578554.0,
      "step": 41715
    },
    {
      "entropy": 5.683692693710327,
      "epoch": 3.7844702467343976,
      "grad_norm": 1.453125,
      "learning_rate": 0.00036255000750112256,
      "loss": 4.9002,
      "mean_token_accuracy": 0.22122927010059357,
      "num_tokens": 86588624.0,
      "step": 41720
    },
    {
      "entropy": 5.713313055038452,
      "epoch": 3.784923802612482,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036252020015922304,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21660754531621934,
      "num_tokens": 86599266.0,
      "step": 41725
    },
    {
      "entropy": 5.751156520843506,
      "epoch": 3.785377358490566,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036249039100738357,
      "loss": 4.9595,
      "mean_token_accuracy": 0.21281423419713974,
      "num_tokens": 86609853.0,
      "step": 41730
    },
    {
      "entropy": 5.614102220535278,
      "epoch": 3.7858309143686504,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003624605800462205,
      "loss": 4.8562,
      "mean_token_accuracy": 0.22482120096683503,
      "num_tokens": 86619927.0,
      "step": 41735
    },
    {
      "entropy": 5.748379373550415,
      "epoch": 3.786284470246734,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036243076727635027,
      "loss": 4.9256,
      "mean_token_accuracy": 0.21848769932985307,
      "num_tokens": 86630079.0,
      "step": 41740
    },
    {
      "entropy": 5.7045533657073975,
      "epoch": 3.7867380261248185,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036240095269838955,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22210616767406463,
      "num_tokens": 86639744.0,
      "step": 41745
    },
    {
      "entropy": 5.663864803314209,
      "epoch": 3.7871915820029027,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036237113631295485,
      "loss": 4.9516,
      "mean_token_accuracy": 0.21452151536941527,
      "num_tokens": 86650866.0,
      "step": 41750
    },
    {
      "entropy": 5.665346717834472,
      "epoch": 3.787645137880987,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036234131812066286,
      "loss": 4.8112,
      "mean_token_accuracy": 0.229076486825943,
      "num_tokens": 86659864.0,
      "step": 41755
    },
    {
      "entropy": 5.618901348114013,
      "epoch": 3.7880986937590713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036231149812213006,
      "loss": 4.8211,
      "mean_token_accuracy": 0.22624040395021439,
      "num_tokens": 86670039.0,
      "step": 41760
    },
    {
      "entropy": 5.628277015686035,
      "epoch": 3.7885522496371555,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003622816763179732,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2267481192946434,
      "num_tokens": 86680434.0,
      "step": 41765
    },
    {
      "entropy": 5.709825420379639,
      "epoch": 3.7890058055152394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003622518527088091,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2131723165512085,
      "num_tokens": 86691026.0,
      "step": 41770
    },
    {
      "entropy": 5.738569593429565,
      "epoch": 3.7894593613933236,
      "grad_norm": 1.34375,
      "learning_rate": 0.00036222202729525453,
      "loss": 4.974,
      "mean_token_accuracy": 0.21591583043336868,
      "num_tokens": 86701323.0,
      "step": 41775
    },
    {
      "entropy": 5.684671306610108,
      "epoch": 3.789912917271408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000362192200077926,
      "loss": 4.853,
      "mean_token_accuracy": 0.23281288295984268,
      "num_tokens": 86711152.0,
      "step": 41780
    },
    {
      "entropy": 5.746676445007324,
      "epoch": 3.790366473149492,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036216237105744076,
      "loss": 5.0124,
      "mean_token_accuracy": 0.22284946143627166,
      "num_tokens": 86720814.0,
      "step": 41785
    },
    {
      "entropy": 5.7027716636657715,
      "epoch": 3.790820029027576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003621325402344154,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22832615822553634,
      "num_tokens": 86731229.0,
      "step": 41790
    },
    {
      "entropy": 5.654819297790527,
      "epoch": 3.7912735849056602,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036210270760946684,
      "loss": 4.8763,
      "mean_token_accuracy": 0.21928852200508117,
      "num_tokens": 86741440.0,
      "step": 41795
    },
    {
      "entropy": 5.749005651473999,
      "epoch": 3.7917271407837445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003620728731832122,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21414629817008973,
      "num_tokens": 86752022.0,
      "step": 41800
    },
    {
      "entropy": 5.719813394546509,
      "epoch": 3.7921806966618288,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003620430369562683,
      "loss": 4.884,
      "mean_token_accuracy": 0.2204085946083069,
      "num_tokens": 86763090.0,
      "step": 41805
    },
    {
      "entropy": 5.688506460189819,
      "epoch": 3.792634252539913,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036201319892925214,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22363327741622924,
      "num_tokens": 86773357.0,
      "step": 41810
    },
    {
      "entropy": 5.643548917770386,
      "epoch": 3.7930878084179973,
      "grad_norm": 1.359375,
      "learning_rate": 0.000361983359102781,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22414054870605468,
      "num_tokens": 86784402.0,
      "step": 41815
    },
    {
      "entropy": 5.697834539413452,
      "epoch": 3.793541364296081,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036195351747747177,
      "loss": 4.939,
      "mean_token_accuracy": 0.21922467052936553,
      "num_tokens": 86794945.0,
      "step": 41820
    },
    {
      "entropy": 5.685494899749756,
      "epoch": 3.7939949201741654,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036192367405394157,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2223448783159256,
      "num_tokens": 86806936.0,
      "step": 41825
    },
    {
      "entropy": 5.760114526748657,
      "epoch": 3.7944484760522497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003618938288328077,
      "loss": 5.0082,
      "mean_token_accuracy": 0.214934866130352,
      "num_tokens": 86817776.0,
      "step": 41830
    },
    {
      "entropy": 5.650275373458863,
      "epoch": 3.794902031930334,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003618639818146873,
      "loss": 4.7893,
      "mean_token_accuracy": 0.22985864132642747,
      "num_tokens": 86828054.0,
      "step": 41835
    },
    {
      "entropy": 5.689925909042358,
      "epoch": 3.7953555878084178,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036183413300019767,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22443366497755052,
      "num_tokens": 86838132.0,
      "step": 41840
    },
    {
      "entropy": 5.669130468368531,
      "epoch": 3.795809143686502,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036180428238995595,
      "loss": 4.9081,
      "mean_token_accuracy": 0.2204018324613571,
      "num_tokens": 86848219.0,
      "step": 41845
    },
    {
      "entropy": 5.627791452407837,
      "epoch": 3.7962626995645863,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003617744299845796,
      "loss": 4.8266,
      "mean_token_accuracy": 0.22443221211433412,
      "num_tokens": 86858559.0,
      "step": 41850
    },
    {
      "entropy": 5.651916646957398,
      "epoch": 3.7967162554426706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003617445757846859,
      "loss": 4.9303,
      "mean_token_accuracy": 0.22195221483707428,
      "num_tokens": 86868573.0,
      "step": 41855
    },
    {
      "entropy": 5.765279626846313,
      "epoch": 3.797169811320755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036171471979089227,
      "loss": 4.9069,
      "mean_token_accuracy": 0.2244272381067276,
      "num_tokens": 86878641.0,
      "step": 41860
    },
    {
      "entropy": 5.743291187286377,
      "epoch": 3.797623367198839,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003616848620038162,
      "loss": 4.9525,
      "mean_token_accuracy": 0.21600497364997864,
      "num_tokens": 86888644.0,
      "step": 41865
    },
    {
      "entropy": 5.778665351867676,
      "epoch": 3.7980769230769234,
      "grad_norm": 1.234375,
      "learning_rate": 0.000361655002424075,
      "loss": 5.0467,
      "mean_token_accuracy": 0.2046968922019005,
      "num_tokens": 86899428.0,
      "step": 41870
    },
    {
      "entropy": 5.734430170059204,
      "epoch": 3.798530478955007,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036162514105228635,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22590023577213286,
      "num_tokens": 86910155.0,
      "step": 41875
    },
    {
      "entropy": 5.683951950073242,
      "epoch": 3.7989840348330914,
      "grad_norm": 1.375,
      "learning_rate": 0.0003615952778890676,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2244248390197754,
      "num_tokens": 86919783.0,
      "step": 41880
    },
    {
      "entropy": 5.681609296798706,
      "epoch": 3.7994375907111757,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036156541293503653,
      "loss": 4.9222,
      "mean_token_accuracy": 0.2182750806212425,
      "num_tokens": 86929647.0,
      "step": 41885
    },
    {
      "entropy": 5.703765773773194,
      "epoch": 3.7998911465892595,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036153554619081064,
      "loss": 4.8594,
      "mean_token_accuracy": 0.2197545975446701,
      "num_tokens": 86940113.0,
      "step": 41890
    },
    {
      "entropy": 5.733808994293213,
      "epoch": 3.800344702467344,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003615056776570076,
      "loss": 4.9662,
      "mean_token_accuracy": 0.21871128678321838,
      "num_tokens": 86949950.0,
      "step": 41895
    },
    {
      "entropy": 5.668528270721436,
      "epoch": 3.800798258345428,
      "grad_norm": 1.375,
      "learning_rate": 0.00036147580733424513,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23255461305379868,
      "num_tokens": 86959286.0,
      "step": 41900
    },
    {
      "entropy": 5.642542028427124,
      "epoch": 3.8012518142235123,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003614459352231409,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22376033365726472,
      "num_tokens": 86970165.0,
      "step": 41905
    },
    {
      "entropy": 5.66293773651123,
      "epoch": 3.8017053701015966,
      "grad_norm": 1.25,
      "learning_rate": 0.0003614160613243127,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2197091355919838,
      "num_tokens": 86980331.0,
      "step": 41910
    },
    {
      "entropy": 5.795016098022461,
      "epoch": 3.802158925979681,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003613861856383783,
      "loss": 4.9653,
      "mean_token_accuracy": 0.21410493850708007,
      "num_tokens": 86991790.0,
      "step": 41915
    },
    {
      "entropy": 5.7666949272155765,
      "epoch": 3.802612481857765,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003613563081659556,
      "loss": 5.0248,
      "mean_token_accuracy": 0.2211061254143715,
      "num_tokens": 87001220.0,
      "step": 41920
    },
    {
      "entropy": 5.728192710876465,
      "epoch": 3.803066037735849,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003613264289076625,
      "loss": 4.8784,
      "mean_token_accuracy": 0.2229156583547592,
      "num_tokens": 87012720.0,
      "step": 41925
    },
    {
      "entropy": 5.717841482162475,
      "epoch": 3.803519593613933,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003612965478641167,
      "loss": 4.9409,
      "mean_token_accuracy": 0.21893847733736038,
      "num_tokens": 87021891.0,
      "step": 41930
    },
    {
      "entropy": 5.786494874954224,
      "epoch": 3.8039731494920175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036126666503593643,
      "loss": 5.0374,
      "mean_token_accuracy": 0.21520183682441713,
      "num_tokens": 87032546.0,
      "step": 41935
    },
    {
      "entropy": 5.720000219345093,
      "epoch": 3.8044267053701017,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003612367804237395,
      "loss": 4.8628,
      "mean_token_accuracy": 0.21866170465946197,
      "num_tokens": 87042173.0,
      "step": 41940
    },
    {
      "entropy": 5.739330816268921,
      "epoch": 3.8048802612481856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003612068940281439,
      "loss": 4.9699,
      "mean_token_accuracy": 0.2201453223824501,
      "num_tokens": 87052475.0,
      "step": 41945
    },
    {
      "entropy": 5.702307844161988,
      "epoch": 3.80533381712627,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003611770058497679,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21731145530939103,
      "num_tokens": 87062698.0,
      "step": 41950
    },
    {
      "entropy": 5.673658752441407,
      "epoch": 3.805787373004354,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036114711588922945,
      "loss": 4.8861,
      "mean_token_accuracy": 0.21768244504928588,
      "num_tokens": 87073168.0,
      "step": 41955
    },
    {
      "entropy": 5.756040859222412,
      "epoch": 3.8062409288824384,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003611172241471466,
      "loss": 4.9849,
      "mean_token_accuracy": 0.21372864693403243,
      "num_tokens": 87084270.0,
      "step": 41960
    },
    {
      "entropy": 5.679685783386231,
      "epoch": 3.8066944847605226,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003610873306241376,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2166223183274269,
      "num_tokens": 87093869.0,
      "step": 41965
    },
    {
      "entropy": 5.685136127471924,
      "epoch": 3.807148040638607,
      "grad_norm": 1.265625,
      "learning_rate": 0.00036105743532082075,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22219808846712114,
      "num_tokens": 87103371.0,
      "step": 41970
    },
    {
      "entropy": 5.710441255569458,
      "epoch": 3.8076015965166907,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003610275382378142,
      "loss": 4.8924,
      "mean_token_accuracy": 0.21857076287269592,
      "num_tokens": 87113384.0,
      "step": 41975
    },
    {
      "entropy": 5.769006395339966,
      "epoch": 3.808055152394775,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003609976393757362,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21955859661102295,
      "num_tokens": 87123925.0,
      "step": 41980
    },
    {
      "entropy": 5.7010733127594,
      "epoch": 3.8085087082728593,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036096773873520513,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21950035840272902,
      "num_tokens": 87134608.0,
      "step": 41985
    },
    {
      "entropy": 5.699103975296021,
      "epoch": 3.8089622641509435,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00036093783631683927,
      "loss": 4.895,
      "mean_token_accuracy": 0.2193385660648346,
      "num_tokens": 87145184.0,
      "step": 41990
    },
    {
      "entropy": 5.657801532745362,
      "epoch": 3.8094158200290273,
      "grad_norm": 1.25,
      "learning_rate": 0.0003609079321212571,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2289069652557373,
      "num_tokens": 87155483.0,
      "step": 41995
    },
    {
      "entropy": 5.679376029968262,
      "epoch": 3.8098693759071116,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036087802614907696,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21563355326652528,
      "num_tokens": 87166615.0,
      "step": 42000
    },
    {
      "epoch": 3.8098693759071116,
      "eval_entropy": 5.396120522896089,
      "eval_loss": 4.969227313995361,
      "eval_mean_token_accuracy": 0.22692512883488677,
      "eval_num_tokens": 87166615.0,
      "eval_runtime": 20.509,
      "eval_samples_per_second": 1724.121,
      "eval_steps_per_second": 215.515,
      "step": 42000
    },
    {
      "entropy": 5.698106050491333,
      "epoch": 3.810322931785196,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036084811840091736,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22113138437271118,
      "num_tokens": 87177151.0,
      "step": 42005
    },
    {
      "entropy": 5.656077527999878,
      "epoch": 3.81077648766328,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003608182088773969,
      "loss": 4.8953,
      "mean_token_accuracy": 0.22082448601722718,
      "num_tokens": 87188085.0,
      "step": 42010
    },
    {
      "entropy": 5.673961067199707,
      "epoch": 3.8112300435413644,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003607882975791338,
      "loss": 4.9471,
      "mean_token_accuracy": 0.21804846078157425,
      "num_tokens": 87197240.0,
      "step": 42015
    },
    {
      "entropy": 5.695687198638916,
      "epoch": 3.8116835994194487,
      "grad_norm": 1.171875,
      "learning_rate": 0.000360758384506747,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2309401139616966,
      "num_tokens": 87208064.0,
      "step": 42020
    },
    {
      "entropy": 5.690528869628906,
      "epoch": 3.8121371552975325,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036072846966085496,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22233093678951263,
      "num_tokens": 87218386.0,
      "step": 42025
    },
    {
      "entropy": 5.636750268936157,
      "epoch": 3.8125907111756168,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00036069855304207634,
      "loss": 4.8056,
      "mean_token_accuracy": 0.22786920964717866,
      "num_tokens": 87228459.0,
      "step": 42030
    },
    {
      "entropy": 5.706151819229126,
      "epoch": 3.813044267053701,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003606686346510297,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21173974722623826,
      "num_tokens": 87240317.0,
      "step": 42035
    },
    {
      "entropy": 5.6186200141906735,
      "epoch": 3.8134978229317853,
      "grad_norm": 1.234375,
      "learning_rate": 0.000360638714488334,
      "loss": 4.8021,
      "mean_token_accuracy": 0.22624580711126327,
      "num_tokens": 87250197.0,
      "step": 42040
    },
    {
      "entropy": 5.64288067817688,
      "epoch": 3.813951378809869,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036060879255460766,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2216366097331047,
      "num_tokens": 87261266.0,
      "step": 42045
    },
    {
      "entropy": 5.6509119987487795,
      "epoch": 3.8144049346879534,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003605788688504699,
      "loss": 4.7818,
      "mean_token_accuracy": 0.2338416948914528,
      "num_tokens": 87270490.0,
      "step": 42050
    },
    {
      "entropy": 5.73028039932251,
      "epoch": 3.8148584905660377,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003605489433765392,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22237619310617446,
      "num_tokens": 87281260.0,
      "step": 42055
    },
    {
      "entropy": 5.655322885513305,
      "epoch": 3.815312046444122,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003605190161334345,
      "loss": 4.8198,
      "mean_token_accuracy": 0.22955294996500014,
      "num_tokens": 87290577.0,
      "step": 42060
    },
    {
      "entropy": 5.611524295806885,
      "epoch": 3.815765602322206,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003604890871217748,
      "loss": 4.7528,
      "mean_token_accuracy": 0.2314566895365715,
      "num_tokens": 87300185.0,
      "step": 42065
    },
    {
      "entropy": 5.695080709457398,
      "epoch": 3.8162191582002905,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003604591563421791,
      "loss": 4.9938,
      "mean_token_accuracy": 0.21752720326185226,
      "num_tokens": 87310088.0,
      "step": 42070
    },
    {
      "entropy": 5.714201879501343,
      "epoch": 3.8166727140783747,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003604292237952662,
      "loss": 4.9153,
      "mean_token_accuracy": 0.21619799137115478,
      "num_tokens": 87319942.0,
      "step": 42075
    },
    {
      "entropy": 5.7396574974060055,
      "epoch": 3.8171262699564585,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003603992894816551,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2247095912694931,
      "num_tokens": 87329175.0,
      "step": 42080
    },
    {
      "entropy": 5.595544481277466,
      "epoch": 3.817579825834543,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000360369353401965,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22752521336078643,
      "num_tokens": 87338555.0,
      "step": 42085
    },
    {
      "entropy": 5.652551317214966,
      "epoch": 3.818033381712627,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00036033941555681486,
      "loss": 4.8674,
      "mean_token_accuracy": 0.23163282126188278,
      "num_tokens": 87348424.0,
      "step": 42090
    },
    {
      "entropy": 5.657241535186768,
      "epoch": 3.8184869375907113,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00036030947594682395,
      "loss": 4.813,
      "mean_token_accuracy": 0.22819059193134308,
      "num_tokens": 87359724.0,
      "step": 42095
    },
    {
      "entropy": 5.663462924957275,
      "epoch": 3.818940493468795,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003602795345726113,
      "loss": 4.9707,
      "mean_token_accuracy": 0.219003364443779,
      "num_tokens": 87371324.0,
      "step": 42100
    },
    {
      "entropy": 5.7177215099334715,
      "epoch": 3.8193940493468794,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003602495914347962,
      "loss": 4.8195,
      "mean_token_accuracy": 0.2313385412096977,
      "num_tokens": 87380770.0,
      "step": 42105
    },
    {
      "entropy": 5.6727618217468265,
      "epoch": 3.8198476052249637,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036021964653399773,
      "loss": 4.9221,
      "mean_token_accuracy": 0.22347851544618608,
      "num_tokens": 87391123.0,
      "step": 42110
    },
    {
      "entropy": 5.697835922241211,
      "epoch": 3.820301161103048,
      "grad_norm": 1.296875,
      "learning_rate": 0.00036018969987083524,
      "loss": 4.932,
      "mean_token_accuracy": 0.21795317977666856,
      "num_tokens": 87401357.0,
      "step": 42115
    },
    {
      "entropy": 5.689523077011108,
      "epoch": 3.8207547169811322,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003601597514459281,
      "loss": 4.8787,
      "mean_token_accuracy": 0.22052175551652908,
      "num_tokens": 87411373.0,
      "step": 42120
    },
    {
      "entropy": 5.6626784324646,
      "epoch": 3.8212082728592165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00036012980125989567,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22817035764455795,
      "num_tokens": 87421915.0,
      "step": 42125
    },
    {
      "entropy": 5.70165786743164,
      "epoch": 3.8216618287373003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003600998493133571,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22132140547037124,
      "num_tokens": 87432148.0,
      "step": 42130
    },
    {
      "entropy": 5.693114328384399,
      "epoch": 3.8221153846153846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003600698956069321,
      "loss": 4.932,
      "mean_token_accuracy": 0.22332126051187515,
      "num_tokens": 87442605.0,
      "step": 42135
    },
    {
      "entropy": 5.70950288772583,
      "epoch": 3.822568940493469,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003600399401412399,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22317626029253007,
      "num_tokens": 87452482.0,
      "step": 42140
    },
    {
      "entropy": 5.582072830200195,
      "epoch": 3.823022496371553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036000998291690005,
      "loss": 4.7582,
      "mean_token_accuracy": 0.22759491354227065,
      "num_tokens": 87462074.0,
      "step": 42145
    },
    {
      "entropy": 5.693901395797729,
      "epoch": 3.823476052249637,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003599800239345321,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2206367149949074,
      "num_tokens": 87472377.0,
      "step": 42150
    },
    {
      "entropy": 5.663262462615966,
      "epoch": 3.823929608127721,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003599500631947557,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22198552191257476,
      "num_tokens": 87483436.0,
      "step": 42155
    },
    {
      "entropy": 5.75718412399292,
      "epoch": 3.8243831640058055,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035992010069819025,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22456623017787933,
      "num_tokens": 87492231.0,
      "step": 42160
    },
    {
      "entropy": 5.821539831161499,
      "epoch": 3.8248367198838897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035989013644545555,
      "loss": 4.9752,
      "mean_token_accuracy": 0.2229054868221283,
      "num_tokens": 87503416.0,
      "step": 42165
    },
    {
      "entropy": 5.691490840911865,
      "epoch": 3.825290275761974,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035986017043717114,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22416868954896926,
      "num_tokens": 87513737.0,
      "step": 42170
    },
    {
      "entropy": 5.689023780822754,
      "epoch": 3.8257438316400583,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003598302026739568,
      "loss": 4.9353,
      "mean_token_accuracy": 0.21543175876140594,
      "num_tokens": 87523532.0,
      "step": 42175
    },
    {
      "entropy": 5.6363301277160645,
      "epoch": 3.826197387518142,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035980023315643217,
      "loss": 4.782,
      "mean_token_accuracy": 0.22993355095386506,
      "num_tokens": 87534066.0,
      "step": 42180
    },
    {
      "entropy": 5.767252445220947,
      "epoch": 3.8266509433962264,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003597702618852172,
      "loss": 5.0159,
      "mean_token_accuracy": 0.21259225457906722,
      "num_tokens": 87544520.0,
      "step": 42185
    },
    {
      "entropy": 5.7044915676116945,
      "epoch": 3.8271044992743106,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003597402888609316,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22322292774915695,
      "num_tokens": 87553913.0,
      "step": 42190
    },
    {
      "entropy": 5.71387128829956,
      "epoch": 3.827558055152395,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035971031408419524,
      "loss": 4.898,
      "mean_token_accuracy": 0.21989507377147674,
      "num_tokens": 87563434.0,
      "step": 42195
    },
    {
      "entropy": 5.6991057872772215,
      "epoch": 3.8280116110304787,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035968033755562795,
      "loss": 4.9247,
      "mean_token_accuracy": 0.21967702507972717,
      "num_tokens": 87573137.0,
      "step": 42200
    },
    {
      "entropy": 5.757615089416504,
      "epoch": 3.828465166908563,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003596503592758497,
      "loss": 4.8915,
      "mean_token_accuracy": 0.22377588003873825,
      "num_tokens": 87582610.0,
      "step": 42205
    },
    {
      "entropy": 5.683595180511475,
      "epoch": 3.8289187227866472,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003596203792454805,
      "loss": 4.876,
      "mean_token_accuracy": 0.22166688442230226,
      "num_tokens": 87593062.0,
      "step": 42210
    },
    {
      "entropy": 5.706789207458496,
      "epoch": 3.8293722786647315,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003595903974651403,
      "loss": 4.9639,
      "mean_token_accuracy": 0.22055468410253526,
      "num_tokens": 87604250.0,
      "step": 42215
    },
    {
      "entropy": 5.7752673625946045,
      "epoch": 3.8298258345428158,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003595604139354491,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2148015096783638,
      "num_tokens": 87615434.0,
      "step": 42220
    },
    {
      "entropy": 5.728222274780274,
      "epoch": 3.8302793904209,
      "grad_norm": 1.3125,
      "learning_rate": 0.000359530428657027,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22806479930877685,
      "num_tokens": 87625005.0,
      "step": 42225
    },
    {
      "entropy": 5.653257226943969,
      "epoch": 3.8307329462989843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003595004416304941,
      "loss": 4.8525,
      "mean_token_accuracy": 0.2238890677690506,
      "num_tokens": 87635486.0,
      "step": 42230
    },
    {
      "entropy": 5.68800458908081,
      "epoch": 3.831186502177068,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035947045285647045,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22609685361385345,
      "num_tokens": 87645771.0,
      "step": 42235
    },
    {
      "entropy": 5.7767256736755375,
      "epoch": 3.8316400580551524,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003594404623355765,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22015759348869324,
      "num_tokens": 87656116.0,
      "step": 42240
    },
    {
      "entropy": 5.8501979351043705,
      "epoch": 3.8320936139332367,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003594104700684321,
      "loss": 5.0993,
      "mean_token_accuracy": 0.2104796662926674,
      "num_tokens": 87666476.0,
      "step": 42245
    },
    {
      "entropy": 5.709457111358643,
      "epoch": 3.8325471698113205,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035938047605565764,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22960105389356614,
      "num_tokens": 87677406.0,
      "step": 42250
    },
    {
      "entropy": 5.67279281616211,
      "epoch": 3.8330007256894048,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035935048029787356,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2233297899365425,
      "num_tokens": 87687461.0,
      "step": 42255
    },
    {
      "entropy": 5.6823495388031,
      "epoch": 3.833454281567489,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003593204827957,
      "loss": 4.8471,
      "mean_token_accuracy": 0.21706174015998841,
      "num_tokens": 87696470.0,
      "step": 42260
    },
    {
      "entropy": 5.677671098709107,
      "epoch": 3.8339078374455733,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003592904835497574,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21666138023138046,
      "num_tokens": 87706598.0,
      "step": 42265
    },
    {
      "entropy": 5.782623529434204,
      "epoch": 3.8343613933236576,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003592604825606661,
      "loss": 5.006,
      "mean_token_accuracy": 0.21567748188972474,
      "num_tokens": 87717749.0,
      "step": 42270
    },
    {
      "entropy": 5.713708162307739,
      "epoch": 3.834814949201742,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003592304798290465,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22860334515571595,
      "num_tokens": 87727893.0,
      "step": 42275
    },
    {
      "entropy": 5.664063549041748,
      "epoch": 3.835268505079826,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003592004753555191,
      "loss": 4.9192,
      "mean_token_accuracy": 0.21941585689783097,
      "num_tokens": 87737980.0,
      "step": 42280
    },
    {
      "entropy": 5.701055908203125,
      "epoch": 3.83572206095791,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003591704691407045,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2218065083026886,
      "num_tokens": 87749271.0,
      "step": 42285
    },
    {
      "entropy": 5.675433540344239,
      "epoch": 3.836175616835994,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035914046118522313,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22552073001861572,
      "num_tokens": 87760523.0,
      "step": 42290
    },
    {
      "entropy": 5.6721498489379885,
      "epoch": 3.8366291727140784,
      "grad_norm": 1.375,
      "learning_rate": 0.00035911045148969556,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2223702549934387,
      "num_tokens": 87770084.0,
      "step": 42295
    },
    {
      "entropy": 5.78171672821045,
      "epoch": 3.8370827285921627,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035908044005474244,
      "loss": 4.9611,
      "mean_token_accuracy": 0.21497900038957596,
      "num_tokens": 87780163.0,
      "step": 42300
    },
    {
      "entropy": 5.635128211975098,
      "epoch": 3.8375362844702465,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003590504268809843,
      "loss": 4.8456,
      "mean_token_accuracy": 0.22408992946147918,
      "num_tokens": 87791382.0,
      "step": 42305
    },
    {
      "entropy": 5.580571794509888,
      "epoch": 3.837989840348331,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035902041196904204,
      "loss": 4.856,
      "mean_token_accuracy": 0.2211557701230049,
      "num_tokens": 87802143.0,
      "step": 42310
    },
    {
      "entropy": 5.678957557678222,
      "epoch": 3.838443396226415,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003589903953195362,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2224273920059204,
      "num_tokens": 87812255.0,
      "step": 42315
    },
    {
      "entropy": 5.736659145355224,
      "epoch": 3.8388969521044993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035896037693308757,
      "loss": 4.9151,
      "mean_token_accuracy": 0.21901753395795823,
      "num_tokens": 87822744.0,
      "step": 42320
    },
    {
      "entropy": 5.803732585906983,
      "epoch": 3.8393505079825836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000358930356810317,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21910140365362168,
      "num_tokens": 87833701.0,
      "step": 42325
    },
    {
      "entropy": 5.736588478088379,
      "epoch": 3.839804063860668,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003589003349518452,
      "loss": 4.9513,
      "mean_token_accuracy": 0.2113957241177559,
      "num_tokens": 87844215.0,
      "step": 42330
    },
    {
      "entropy": 5.751780462265015,
      "epoch": 3.8402576197387517,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035887031135829305,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2166260600090027,
      "num_tokens": 87854281.0,
      "step": 42335
    },
    {
      "entropy": 5.764206838607788,
      "epoch": 3.840711175616836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003588402860302816,
      "loss": 4.979,
      "mean_token_accuracy": 0.21392544358968735,
      "num_tokens": 87866003.0,
      "step": 42340
    },
    {
      "entropy": 5.709864377975464,
      "epoch": 3.84116473149492,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035881025896843165,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22359587550163268,
      "num_tokens": 87875854.0,
      "step": 42345
    },
    {
      "entropy": 5.7785240650177006,
      "epoch": 3.8416182873730045,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035878023017336415,
      "loss": 5.0171,
      "mean_token_accuracy": 0.21475472301244736,
      "num_tokens": 87887242.0,
      "step": 42350
    },
    {
      "entropy": 5.733989953994751,
      "epoch": 3.8420718432510883,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003587501996457001,
      "loss": 4.9117,
      "mean_token_accuracy": 0.2214742809534073,
      "num_tokens": 87897859.0,
      "step": 42355
    },
    {
      "entropy": 5.690939950942993,
      "epoch": 3.8425253991291726,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035872016738606067,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22491035610437393,
      "num_tokens": 87908945.0,
      "step": 42360
    },
    {
      "entropy": 5.707577228546143,
      "epoch": 3.842978955007257,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035869013339506674,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22741124480962754,
      "num_tokens": 87919572.0,
      "step": 42365
    },
    {
      "entropy": 5.756832885742187,
      "epoch": 3.843432510885341,
      "grad_norm": 1.375,
      "learning_rate": 0.00035866009767333964,
      "loss": 4.9538,
      "mean_token_accuracy": 0.21872526854276658,
      "num_tokens": 87929337.0,
      "step": 42370
    },
    {
      "entropy": 5.7840502738952635,
      "epoch": 3.8438860667634254,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035863006022150034,
      "loss": 4.9935,
      "mean_token_accuracy": 0.2153836891055107,
      "num_tokens": 87940513.0,
      "step": 42375
    },
    {
      "entropy": 5.677812719345093,
      "epoch": 3.8443396226415096,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035860002104017014,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23151499778032303,
      "num_tokens": 87951181.0,
      "step": 42380
    },
    {
      "entropy": 5.687026739120483,
      "epoch": 3.8447931785195935,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003585699801299701,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22194211184978485,
      "num_tokens": 87961925.0,
      "step": 42385
    },
    {
      "entropy": 5.705641365051269,
      "epoch": 3.8452467343976777,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035853993749152165,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22954976707696914,
      "num_tokens": 87972233.0,
      "step": 42390
    },
    {
      "entropy": 5.728885459899902,
      "epoch": 3.845700290275762,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035850989312544596,
      "loss": 4.9477,
      "mean_token_accuracy": 0.21732770949602126,
      "num_tokens": 87983131.0,
      "step": 42395
    },
    {
      "entropy": 5.683565378189087,
      "epoch": 3.8461538461538463,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003584798470323644,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2223536491394043,
      "num_tokens": 87993926.0,
      "step": 42400
    },
    {
      "entropy": 5.778710842132568,
      "epoch": 3.84660740203193,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003584497992128984,
      "loss": 5.0297,
      "mean_token_accuracy": 0.21180527955293654,
      "num_tokens": 88004403.0,
      "step": 42405
    },
    {
      "entropy": 5.679070949554443,
      "epoch": 3.8470609579100143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003584197496676692,
      "loss": 4.861,
      "mean_token_accuracy": 0.22809899747371673,
      "num_tokens": 88015289.0,
      "step": 42410
    },
    {
      "entropy": 5.640101766586303,
      "epoch": 3.8475145137880986,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035838969839729846,
      "loss": 4.8494,
      "mean_token_accuracy": 0.22137422263622283,
      "num_tokens": 88025724.0,
      "step": 42415
    },
    {
      "entropy": 5.6739335536956785,
      "epoch": 3.847968069666183,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003583596454024073,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2179590493440628,
      "num_tokens": 88036444.0,
      "step": 42420
    },
    {
      "entropy": 5.775604295730591,
      "epoch": 3.848421625544267,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003583295906836176,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22423885613679886,
      "num_tokens": 88047334.0,
      "step": 42425
    },
    {
      "entropy": 5.721081256866455,
      "epoch": 3.8488751814223514,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035829953424155065,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22154112607240678,
      "num_tokens": 88057268.0,
      "step": 42430
    },
    {
      "entropy": 5.658592224121094,
      "epoch": 3.8493287373004357,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003582694760768281,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22194052785634993,
      "num_tokens": 88067615.0,
      "step": 42435
    },
    {
      "entropy": 5.762260818481446,
      "epoch": 3.8497822931785195,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035823941619007155,
      "loss": 4.987,
      "mean_token_accuracy": 0.22107081413269042,
      "num_tokens": 88077754.0,
      "step": 42440
    },
    {
      "entropy": 5.6961202144622805,
      "epoch": 3.8502358490566038,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035820935458190275,
      "loss": 4.8762,
      "mean_token_accuracy": 0.2195508137345314,
      "num_tokens": 88088346.0,
      "step": 42445
    },
    {
      "entropy": 5.631023120880127,
      "epoch": 3.850689404934688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035817929125294323,
      "loss": 4.8507,
      "mean_token_accuracy": 0.22588455229997634,
      "num_tokens": 88098534.0,
      "step": 42450
    },
    {
      "entropy": 5.650130987167358,
      "epoch": 3.8511429608127723,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003581492262038148,
      "loss": 4.9293,
      "mean_token_accuracy": 0.21836246103048323,
      "num_tokens": 88109264.0,
      "step": 42455
    },
    {
      "entropy": 5.72651047706604,
      "epoch": 3.851596516690856,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003581191594351392,
      "loss": 4.934,
      "mean_token_accuracy": 0.22214488089084625,
      "num_tokens": 88120407.0,
      "step": 42460
    },
    {
      "entropy": 5.744232225418091,
      "epoch": 3.8520500725689404,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003580890909475381,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2188191071152687,
      "num_tokens": 88131908.0,
      "step": 42465
    },
    {
      "entropy": 5.660446357727051,
      "epoch": 3.8525036284470247,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035805902074163354,
      "loss": 4.8686,
      "mean_token_accuracy": 0.21799683719873428,
      "num_tokens": 88142559.0,
      "step": 42470
    },
    {
      "entropy": 5.670582246780396,
      "epoch": 3.852957184325109,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003580289488180472,
      "loss": 4.9269,
      "mean_token_accuracy": 0.227896811068058,
      "num_tokens": 88153441.0,
      "step": 42475
    },
    {
      "entropy": 5.690734148025513,
      "epoch": 3.853410740203193,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00035799887517740103,
      "loss": 4.899,
      "mean_token_accuracy": 0.2248982548713684,
      "num_tokens": 88164153.0,
      "step": 42480
    },
    {
      "entropy": 5.728591823577881,
      "epoch": 3.8538642960812775,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000357968799820317,
      "loss": 4.9114,
      "mean_token_accuracy": 0.21746076196432113,
      "num_tokens": 88174138.0,
      "step": 42485
    },
    {
      "entropy": 5.6981734275817875,
      "epoch": 3.8543178519593613,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003579387227474171,
      "loss": 4.9178,
      "mean_token_accuracy": 0.22188902795314788,
      "num_tokens": 88184704.0,
      "step": 42490
    },
    {
      "entropy": 5.797828912734985,
      "epoch": 3.8547714078374455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003579086439593232,
      "loss": 4.9411,
      "mean_token_accuracy": 0.2166929617524147,
      "num_tokens": 88194225.0,
      "step": 42495
    },
    {
      "entropy": 5.751197004318238,
      "epoch": 3.85522496371553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003578785634566574,
      "loss": 4.9265,
      "mean_token_accuracy": 0.21983438432216645,
      "num_tokens": 88206125.0,
      "step": 42500
    },
    {
      "entropy": 5.648555421829224,
      "epoch": 3.855678519593614,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035784848124004187,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23218234926462172,
      "num_tokens": 88217085.0,
      "step": 42505
    },
    {
      "entropy": 5.666171312332153,
      "epoch": 3.856132075471698,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003578183973100986,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21562377661466597,
      "num_tokens": 88228126.0,
      "step": 42510
    },
    {
      "entropy": 5.69417667388916,
      "epoch": 3.856585631349782,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035778831166744966,
      "loss": 4.9995,
      "mean_token_accuracy": 0.21322721242904663,
      "num_tokens": 88238829.0,
      "step": 42515
    },
    {
      "entropy": 5.718562316894531,
      "epoch": 3.8570391872278664,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003577582243127175,
      "loss": 4.9169,
      "mean_token_accuracy": 0.21971588879823684,
      "num_tokens": 88249821.0,
      "step": 42520
    },
    {
      "entropy": 5.70019679069519,
      "epoch": 3.8574927431059507,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003577281352465241,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22574142515659332,
      "num_tokens": 88260126.0,
      "step": 42525
    },
    {
      "entropy": 5.688124370574951,
      "epoch": 3.857946298984035,
      "grad_norm": 1.25,
      "learning_rate": 0.00035769804446949177,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22495007514953613,
      "num_tokens": 88271492.0,
      "step": 42530
    },
    {
      "entropy": 5.7549505710601805,
      "epoch": 3.8583998548621192,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003576679519822428,
      "loss": 4.9865,
      "mean_token_accuracy": 0.21016664654016495,
      "num_tokens": 88281966.0,
      "step": 42535
    },
    {
      "entropy": 5.617628717422486,
      "epoch": 3.858853410740203,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035763785778539963,
      "loss": 4.8038,
      "mean_token_accuracy": 0.2375423088669777,
      "num_tokens": 88292805.0,
      "step": 42540
    },
    {
      "entropy": 5.674291515350342,
      "epoch": 3.8593069666182873,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003576077618795844,
      "loss": 4.8452,
      "mean_token_accuracy": 0.2271577760577202,
      "num_tokens": 88302727.0,
      "step": 42545
    },
    {
      "entropy": 5.589399671554565,
      "epoch": 3.8597605224963716,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003575776642654196,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22513065189123155,
      "num_tokens": 88313150.0,
      "step": 42550
    },
    {
      "entropy": 5.678622913360596,
      "epoch": 3.860214078374456,
      "grad_norm": 1.421875,
      "learning_rate": 0.00035754756494352767,
      "loss": 4.835,
      "mean_token_accuracy": 0.22992342114448547,
      "num_tokens": 88323060.0,
      "step": 42555
    },
    {
      "entropy": 5.778040361404419,
      "epoch": 3.8606676342525397,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003575174639145311,
      "loss": 4.8812,
      "mean_token_accuracy": 0.22726214230060576,
      "num_tokens": 88332641.0,
      "step": 42560
    },
    {
      "entropy": 5.635117101669311,
      "epoch": 3.861121190130624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003574873611790523,
      "loss": 4.899,
      "mean_token_accuracy": 0.22802033573389052,
      "num_tokens": 88342807.0,
      "step": 42565
    },
    {
      "entropy": 5.596829557418824,
      "epoch": 3.861574746008708,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035745725673771387,
      "loss": 4.791,
      "mean_token_accuracy": 0.23186014145612716,
      "num_tokens": 88352958.0,
      "step": 42570
    },
    {
      "entropy": 5.717193460464477,
      "epoch": 3.8620283018867925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035742715059113836,
      "loss": 4.9398,
      "mean_token_accuracy": 0.21541982144117355,
      "num_tokens": 88363184.0,
      "step": 42575
    },
    {
      "entropy": 5.624860334396362,
      "epoch": 3.8624818577648767,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003573970427399484,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22888078689575195,
      "num_tokens": 88373991.0,
      "step": 42580
    },
    {
      "entropy": 5.731839323043824,
      "epoch": 3.862935413642961,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035736693318476646,
      "loss": 4.91,
      "mean_token_accuracy": 0.21535104662179946,
      "num_tokens": 88384518.0,
      "step": 42585
    },
    {
      "entropy": 5.60913667678833,
      "epoch": 3.8633889695210453,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035733682192621555,
      "loss": 4.7858,
      "mean_token_accuracy": 0.22804877758026124,
      "num_tokens": 88394506.0,
      "step": 42590
    },
    {
      "entropy": 5.704238510131836,
      "epoch": 3.863842525399129,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003573067089649181,
      "loss": 4.8899,
      "mean_token_accuracy": 0.21950516998767852,
      "num_tokens": 88404625.0,
      "step": 42595
    },
    {
      "entropy": 5.705760526657104,
      "epoch": 3.8642960812772134,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00035727659430149686,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22199369966983795,
      "num_tokens": 88413754.0,
      "step": 42600
    },
    {
      "entropy": 5.741912031173706,
      "epoch": 3.8647496371552976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003572464779365748,
      "loss": 4.9929,
      "mean_token_accuracy": 0.21422540843486787,
      "num_tokens": 88424613.0,
      "step": 42605
    },
    {
      "entropy": 5.717949485778808,
      "epoch": 3.8652031930333814,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035721635987077455,
      "loss": 4.895,
      "mean_token_accuracy": 0.22636578232049942,
      "num_tokens": 88435248.0,
      "step": 42610
    },
    {
      "entropy": 5.689569187164307,
      "epoch": 3.8656567489114657,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035718624010471904,
      "loss": 4.7766,
      "mean_token_accuracy": 0.23485510498285295,
      "num_tokens": 88444802.0,
      "step": 42615
    },
    {
      "entropy": 5.703863286972046,
      "epoch": 3.86611030478955,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003571561186390311,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21708950102329255,
      "num_tokens": 88455335.0,
      "step": 42620
    },
    {
      "entropy": 5.717738294601441,
      "epoch": 3.8665638606676342,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003571259954743338,
      "loss": 4.903,
      "mean_token_accuracy": 0.22895789593458177,
      "num_tokens": 88465857.0,
      "step": 42625
    },
    {
      "entropy": 5.689590883255005,
      "epoch": 3.8670174165457185,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035709587061124985,
      "loss": 4.9523,
      "mean_token_accuracy": 0.21596571505069734,
      "num_tokens": 88475930.0,
      "step": 42630
    },
    {
      "entropy": 5.730015420913697,
      "epoch": 3.8674709724238028,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035706574405040245,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22516955733299254,
      "num_tokens": 88486732.0,
      "step": 42635
    },
    {
      "entropy": 5.745752906799316,
      "epoch": 3.867924528301887,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003570356157924145,
      "loss": 4.907,
      "mean_token_accuracy": 0.22421524226665496,
      "num_tokens": 88497017.0,
      "step": 42640
    },
    {
      "entropy": 5.681430816650391,
      "epoch": 3.868378084179971,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003570054858379092,
      "loss": 4.802,
      "mean_token_accuracy": 0.238104048371315,
      "num_tokens": 88507423.0,
      "step": 42645
    },
    {
      "entropy": 5.682418727874756,
      "epoch": 3.868831640058055,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035697535418750955,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22642449140548707,
      "num_tokens": 88517795.0,
      "step": 42650
    },
    {
      "entropy": 5.614846181869507,
      "epoch": 3.8692851959361394,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035694522084183864,
      "loss": 4.7996,
      "mean_token_accuracy": 0.22631308883428575,
      "num_tokens": 88527641.0,
      "step": 42655
    },
    {
      "entropy": 5.7047354221344,
      "epoch": 3.8697387518142237,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035691508580151966,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2138011321425438,
      "num_tokens": 88537874.0,
      "step": 42660
    },
    {
      "entropy": 5.708426141738892,
      "epoch": 3.8701923076923075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003568849490671759,
      "loss": 4.9171,
      "mean_token_accuracy": 0.22096857726573943,
      "num_tokens": 88548108.0,
      "step": 42665
    },
    {
      "entropy": 5.645083856582642,
      "epoch": 3.8706458635703918,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003568548106394305,
      "loss": 4.7961,
      "mean_token_accuracy": 0.2289535880088806,
      "num_tokens": 88558284.0,
      "step": 42670
    },
    {
      "entropy": 5.704737377166748,
      "epoch": 3.871099419448476,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003568246705189067,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2223971515893936,
      "num_tokens": 88568635.0,
      "step": 42675
    },
    {
      "entropy": 5.687330913543701,
      "epoch": 3.8715529753265603,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003567945287062279,
      "loss": 4.9088,
      "mean_token_accuracy": 0.21853030771017073,
      "num_tokens": 88577968.0,
      "step": 42680
    },
    {
      "entropy": 5.7637317180633545,
      "epoch": 3.8720065312046446,
      "grad_norm": 1.125,
      "learning_rate": 0.00035676438520201745,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22073307633399963,
      "num_tokens": 88590562.0,
      "step": 42685
    },
    {
      "entropy": 5.706578016281128,
      "epoch": 3.872460087082729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035673424000689857,
      "loss": 4.92,
      "mean_token_accuracy": 0.21736774742603301,
      "num_tokens": 88601187.0,
      "step": 42690
    },
    {
      "entropy": 5.636082601547241,
      "epoch": 3.8729136429608126,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035670409312149495,
      "loss": 4.8195,
      "mean_token_accuracy": 0.22202365696430207,
      "num_tokens": 88611459.0,
      "step": 42695
    },
    {
      "entropy": 5.7014635562896725,
      "epoch": 3.873367198838897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035667394454642963,
      "loss": 4.918,
      "mean_token_accuracy": 0.22797491252422333,
      "num_tokens": 88621652.0,
      "step": 42700
    },
    {
      "entropy": 5.691184711456299,
      "epoch": 3.873820754716981,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003566437942823265,
      "loss": 4.7958,
      "mean_token_accuracy": 0.22297163009643556,
      "num_tokens": 88632425.0,
      "step": 42705
    },
    {
      "entropy": 5.645804500579834,
      "epoch": 3.8742743105950654,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035661364232980884,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23363922983407975,
      "num_tokens": 88641263.0,
      "step": 42710
    },
    {
      "entropy": 5.6501070022583,
      "epoch": 3.8747278664731493,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035658348868950027,
      "loss": 4.8936,
      "mean_token_accuracy": 0.21811120510101317,
      "num_tokens": 88651665.0,
      "step": 42715
    },
    {
      "entropy": 5.675134086608887,
      "epoch": 3.8751814223512335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035655333336202436,
      "loss": 4.8297,
      "mean_token_accuracy": 0.2271896556019783,
      "num_tokens": 88660791.0,
      "step": 42720
    },
    {
      "entropy": 5.628390550613403,
      "epoch": 3.875634978229318,
      "grad_norm": 1.375,
      "learning_rate": 0.0003565231763480047,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23422186225652694,
      "num_tokens": 88671341.0,
      "step": 42725
    },
    {
      "entropy": 5.665875387191773,
      "epoch": 3.876088534107402,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003564930176480651,
      "loss": 4.9065,
      "mean_token_accuracy": 0.21552653163671492,
      "num_tokens": 88682500.0,
      "step": 42730
    },
    {
      "entropy": 5.72534294128418,
      "epoch": 3.8765420899854863,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035646285726282896,
      "loss": 4.8885,
      "mean_token_accuracy": 0.21584065705537797,
      "num_tokens": 88693513.0,
      "step": 42735
    },
    {
      "entropy": 5.758978652954101,
      "epoch": 3.8769956458635706,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035643269519292027,
      "loss": 4.951,
      "mean_token_accuracy": 0.2183902904391289,
      "num_tokens": 88703498.0,
      "step": 42740
    },
    {
      "entropy": 5.668791723251343,
      "epoch": 3.8774492017416544,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003564025314389627,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22215707451105118,
      "num_tokens": 88714217.0,
      "step": 42745
    },
    {
      "entropy": 5.699587011337281,
      "epoch": 3.8779027576197387,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035637236600157994,
      "loss": 4.9238,
      "mean_token_accuracy": 0.2165780782699585,
      "num_tokens": 88725006.0,
      "step": 42750
    },
    {
      "entropy": 5.666853046417236,
      "epoch": 3.878356313497823,
      "grad_norm": 1.296875,
      "learning_rate": 0.000356342198881396,
      "loss": 4.8353,
      "mean_token_accuracy": 0.22679433077573777,
      "num_tokens": 88735253.0,
      "step": 42755
    },
    {
      "entropy": 5.7381609916687015,
      "epoch": 3.878809869375907,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035631203007903465,
      "loss": 4.9464,
      "mean_token_accuracy": 0.21591888070106507,
      "num_tokens": 88746313.0,
      "step": 42760
    },
    {
      "entropy": 5.663065195083618,
      "epoch": 3.879263425253991,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003562818595951198,
      "loss": 4.8507,
      "mean_token_accuracy": 0.22631865739822388,
      "num_tokens": 88757448.0,
      "step": 42765
    },
    {
      "entropy": 5.702502346038818,
      "epoch": 3.8797169811320753,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003562516874302754,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22603631019592285,
      "num_tokens": 88767426.0,
      "step": 42770
    },
    {
      "entropy": 5.704104471206665,
      "epoch": 3.8801705370101596,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003562215135851253,
      "loss": 4.9196,
      "mean_token_accuracy": 0.22068741768598557,
      "num_tokens": 88778188.0,
      "step": 42775
    },
    {
      "entropy": 5.7872625350952145,
      "epoch": 3.880624092888244,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035619133806029374,
      "loss": 4.9635,
      "mean_token_accuracy": 0.2255626380443573,
      "num_tokens": 88788927.0,
      "step": 42780
    },
    {
      "entropy": 5.749645614624024,
      "epoch": 3.881077648766328,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035616116085640455,
      "loss": 4.9506,
      "mean_token_accuracy": 0.21690801978111268,
      "num_tokens": 88798809.0,
      "step": 42785
    },
    {
      "entropy": 5.618165969848633,
      "epoch": 3.8815312046444124,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003561309819740819,
      "loss": 4.7858,
      "mean_token_accuracy": 0.23431729674339294,
      "num_tokens": 88808688.0,
      "step": 42790
    },
    {
      "entropy": 5.676987504959106,
      "epoch": 3.8819847605224966,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035610080141394983,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2243338868021965,
      "num_tokens": 88818400.0,
      "step": 42795
    },
    {
      "entropy": 5.654087495803833,
      "epoch": 3.8824383164005805,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035607061917663254,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22215137034654617,
      "num_tokens": 88828032.0,
      "step": 42800
    },
    {
      "entropy": 5.685729503631592,
      "epoch": 3.8828918722786647,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035604043526275413,
      "loss": 4.9826,
      "mean_token_accuracy": 0.2220853701233864,
      "num_tokens": 88839856.0,
      "step": 42805
    },
    {
      "entropy": 5.702771282196045,
      "epoch": 3.883345428156749,
      "grad_norm": 1.265625,
      "learning_rate": 0.000356010249672939,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22573107928037645,
      "num_tokens": 88851016.0,
      "step": 42810
    },
    {
      "entropy": 5.705296039581299,
      "epoch": 3.8837989840348333,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003559800624078111,
      "loss": 4.914,
      "mean_token_accuracy": 0.21998802870512008,
      "num_tokens": 88859795.0,
      "step": 42815
    },
    {
      "entropy": 5.6938769817352295,
      "epoch": 3.884252539912917,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003559498734679949,
      "loss": 4.9294,
      "mean_token_accuracy": 0.22071518301963805,
      "num_tokens": 88869595.0,
      "step": 42820
    },
    {
      "entropy": 5.7423947811126705,
      "epoch": 3.8847060957910013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003559196828541148,
      "loss": 4.9404,
      "mean_token_accuracy": 0.21705970913171768,
      "num_tokens": 88880857.0,
      "step": 42825
    },
    {
      "entropy": 5.665018749237061,
      "epoch": 3.8851596516690856,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000355889490566795,
      "loss": 4.8265,
      "mean_token_accuracy": 0.23481018841266632,
      "num_tokens": 88890687.0,
      "step": 42830
    },
    {
      "entropy": 5.676608037948609,
      "epoch": 3.88561320754717,
      "grad_norm": 1.25,
      "learning_rate": 0.00035585929660665984,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22297827601432801,
      "num_tokens": 88899600.0,
      "step": 42835
    },
    {
      "entropy": 5.705259466171265,
      "epoch": 3.886066763425254,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003558291009743339,
      "loss": 4.8692,
      "mean_token_accuracy": 0.2267654001712799,
      "num_tokens": 88909521.0,
      "step": 42840
    },
    {
      "entropy": 5.698551845550537,
      "epoch": 3.8865203193033384,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00035579890367044153,
      "loss": 4.9417,
      "mean_token_accuracy": 0.2210657551884651,
      "num_tokens": 88920061.0,
      "step": 42845
    },
    {
      "entropy": 5.631320810317993,
      "epoch": 3.8869738751814222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003557687046956073,
      "loss": 4.8004,
      "mean_token_accuracy": 0.22743285149335862,
      "num_tokens": 88929504.0,
      "step": 42850
    },
    {
      "entropy": 5.773502922058105,
      "epoch": 3.8874274310595065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035573850405045554,
      "loss": 5.0192,
      "mean_token_accuracy": 0.20954971760511398,
      "num_tokens": 88940653.0,
      "step": 42855
    },
    {
      "entropy": 5.749763822555542,
      "epoch": 3.8878809869375908,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003557083017356111,
      "loss": 4.9593,
      "mean_token_accuracy": 0.22401134371757508,
      "num_tokens": 88952133.0,
      "step": 42860
    },
    {
      "entropy": 5.662125968933106,
      "epoch": 3.888334542815675,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003556780977516983,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22554275691509246,
      "num_tokens": 88963190.0,
      "step": 42865
    },
    {
      "entropy": 5.610604381561279,
      "epoch": 3.888788098693759,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035564789209934183,
      "loss": 4.8008,
      "mean_token_accuracy": 0.2290722146630287,
      "num_tokens": 88973034.0,
      "step": 42870
    },
    {
      "entropy": 5.688286209106446,
      "epoch": 3.889241654571843,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035561768477916655,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23403785079717637,
      "num_tokens": 88982809.0,
      "step": 42875
    },
    {
      "entropy": 5.688271427154541,
      "epoch": 3.8896952104499274,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000355587475791797,
      "loss": 4.8857,
      "mean_token_accuracy": 0.21751513481140136,
      "num_tokens": 88993563.0,
      "step": 42880
    },
    {
      "entropy": 5.586067247390747,
      "epoch": 3.8901487663280117,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00035555726513785785,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2350510761141777,
      "num_tokens": 89002408.0,
      "step": 42885
    },
    {
      "entropy": 5.631853914260864,
      "epoch": 3.890602322206096,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003555270528179739,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22491317987442017,
      "num_tokens": 89013060.0,
      "step": 42890
    },
    {
      "entropy": 5.750309038162231,
      "epoch": 3.89105587808418,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035549683883277,
      "loss": 4.9329,
      "mean_token_accuracy": 0.22158871591091156,
      "num_tokens": 89023406.0,
      "step": 42895
    },
    {
      "entropy": 5.732694149017334,
      "epoch": 3.891509433962264,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000355466623182871,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2266659766435623,
      "num_tokens": 89033800.0,
      "step": 42900
    },
    {
      "entropy": 5.766987466812134,
      "epoch": 3.8919629898403483,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003554364058689017,
      "loss": 4.9133,
      "mean_token_accuracy": 0.22028793692588805,
      "num_tokens": 89044601.0,
      "step": 42905
    },
    {
      "entropy": 5.74994387626648,
      "epoch": 3.8924165457184325,
      "grad_norm": 1.171875,
      "learning_rate": 0.000355406186891487,
      "loss": 4.9441,
      "mean_token_accuracy": 0.21885457783937454,
      "num_tokens": 89056138.0,
      "step": 42910
    },
    {
      "entropy": 5.7116485118865965,
      "epoch": 3.892870101596517,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003553759662512519,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2295670211315155,
      "num_tokens": 89065776.0,
      "step": 42915
    },
    {
      "entropy": 5.826280212402343,
      "epoch": 3.8933236574746006,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035534574394882135,
      "loss": 5.0025,
      "mean_token_accuracy": 0.20831212401390076,
      "num_tokens": 89076926.0,
      "step": 42920
    },
    {
      "entropy": 5.707504796981811,
      "epoch": 3.893777213352685,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003553155199848203,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21342908591032028,
      "num_tokens": 89088036.0,
      "step": 42925
    },
    {
      "entropy": 5.678588581085205,
      "epoch": 3.894230769230769,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003552852943598739,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22523169368505477,
      "num_tokens": 89098606.0,
      "step": 42930
    },
    {
      "entropy": 5.671547365188599,
      "epoch": 3.8946843251088534,
      "grad_norm": 1.296875,
      "learning_rate": 0.000355255067074607,
      "loss": 4.8202,
      "mean_token_accuracy": 0.23018813133239746,
      "num_tokens": 89107733.0,
      "step": 42935
    },
    {
      "entropy": 5.676202917098999,
      "epoch": 3.8951378809869377,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035522483812964496,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22751456648111343,
      "num_tokens": 89119094.0,
      "step": 42940
    },
    {
      "entropy": 5.664641857147217,
      "epoch": 3.895591436865022,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035519460752561285,
      "loss": 4.7688,
      "mean_token_accuracy": 0.2281271055340767,
      "num_tokens": 89129254.0,
      "step": 42945
    },
    {
      "entropy": 5.69221510887146,
      "epoch": 3.8960449927431062,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003551643752631358,
      "loss": 4.8155,
      "mean_token_accuracy": 0.23000796139240265,
      "num_tokens": 89139609.0,
      "step": 42950
    },
    {
      "entropy": 5.700359630584717,
      "epoch": 3.89649854862119,
      "grad_norm": 1.25,
      "learning_rate": 0.0003551341413428391,
      "loss": 4.8542,
      "mean_token_accuracy": 0.21780824214220046,
      "num_tokens": 89149504.0,
      "step": 42955
    },
    {
      "entropy": 5.70113844871521,
      "epoch": 3.8969521044992743,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003551039057653478,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22681287825107574,
      "num_tokens": 89160036.0,
      "step": 42960
    },
    {
      "entropy": 5.700443601608276,
      "epoch": 3.8974056603773586,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003550736685312874,
      "loss": 4.9216,
      "mean_token_accuracy": 0.21788099855184556,
      "num_tokens": 89170960.0,
      "step": 42965
    },
    {
      "entropy": 5.671843671798706,
      "epoch": 3.8978592162554424,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003550434296412831,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23031731098890304,
      "num_tokens": 89179630.0,
      "step": 42970
    },
    {
      "entropy": 5.693530750274658,
      "epoch": 3.8983127721335267,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003550131890959603,
      "loss": 4.8905,
      "mean_token_accuracy": 0.22811541259288787,
      "num_tokens": 89188880.0,
      "step": 42975
    },
    {
      "entropy": 5.733471250534057,
      "epoch": 3.898766328011611,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035498294689594435,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22238024622201918,
      "num_tokens": 89199585.0,
      "step": 42980
    },
    {
      "entropy": 5.698893594741821,
      "epoch": 3.899219883889695,
      "grad_norm": 1.25,
      "learning_rate": 0.00035495270304186074,
      "loss": 4.8709,
      "mean_token_accuracy": 0.21533643156290055,
      "num_tokens": 89209812.0,
      "step": 42985
    },
    {
      "entropy": 5.668211603164673,
      "epoch": 3.8996734397677795,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003549224575343347,
      "loss": 4.8709,
      "mean_token_accuracy": 0.2247721254825592,
      "num_tokens": 89220244.0,
      "step": 42990
    },
    {
      "entropy": 5.6706677913665775,
      "epoch": 3.9001269956458637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035489221037399203,
      "loss": 4.9281,
      "mean_token_accuracy": 0.21916999965906142,
      "num_tokens": 89230618.0,
      "step": 42995
    },
    {
      "entropy": 5.693241310119629,
      "epoch": 3.900580551523948,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035486196156145805,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22346450239419938,
      "num_tokens": 89242041.0,
      "step": 43000
    },
    {
      "entropy": 5.716432428359985,
      "epoch": 3.901034107402032,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003548317110973583,
      "loss": 4.9137,
      "mean_token_accuracy": 0.22300271838903427,
      "num_tokens": 89252817.0,
      "step": 43005
    },
    {
      "entropy": 5.704679489135742,
      "epoch": 3.901487663280116,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003548014589823185,
      "loss": 4.9311,
      "mean_token_accuracy": 0.22356182634830474,
      "num_tokens": 89263702.0,
      "step": 43010
    },
    {
      "entropy": 5.674681806564331,
      "epoch": 3.9019412191582004,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035477120521696417,
      "loss": 4.8423,
      "mean_token_accuracy": 0.2320714548230171,
      "num_tokens": 89273822.0,
      "step": 43015
    },
    {
      "entropy": 5.753475284576416,
      "epoch": 3.9023947750362846,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003547409498019209,
      "loss": 4.9971,
      "mean_token_accuracy": 0.21308946758508682,
      "num_tokens": 89283973.0,
      "step": 43020
    },
    {
      "entropy": 5.723907423019409,
      "epoch": 3.9028483309143684,
      "grad_norm": 1.40625,
      "learning_rate": 0.00035471069273781454,
      "loss": 4.93,
      "mean_token_accuracy": 0.22746729850769043,
      "num_tokens": 89294059.0,
      "step": 43025
    },
    {
      "entropy": 5.712174606323242,
      "epoch": 3.9033018867924527,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003546804340252707,
      "loss": 4.8511,
      "mean_token_accuracy": 0.223424531519413,
      "num_tokens": 89304049.0,
      "step": 43030
    },
    {
      "entropy": 5.64229736328125,
      "epoch": 3.903755442670537,
      "grad_norm": 1.34375,
      "learning_rate": 0.00035465017366491527,
      "loss": 4.7826,
      "mean_token_accuracy": 0.236358243227005,
      "num_tokens": 89314468.0,
      "step": 43035
    },
    {
      "entropy": 5.617630529403686,
      "epoch": 3.9042089985486212,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035461991165737386,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2247973695397377,
      "num_tokens": 89324580.0,
      "step": 43040
    },
    {
      "entropy": 5.752272844314575,
      "epoch": 3.9046625544267055,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035458964800327245,
      "loss": 4.949,
      "mean_token_accuracy": 0.2141755297780037,
      "num_tokens": 89334976.0,
      "step": 43045
    },
    {
      "entropy": 5.720316219329834,
      "epoch": 3.90511611030479,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035455938270323684,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2208763465285301,
      "num_tokens": 89345210.0,
      "step": 43050
    },
    {
      "entropy": 5.668814849853516,
      "epoch": 3.9055696661828736,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003545291157578928,
      "loss": 4.7773,
      "mean_token_accuracy": 0.232504503428936,
      "num_tokens": 89354623.0,
      "step": 43055
    },
    {
      "entropy": 5.65915732383728,
      "epoch": 3.906023222060958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035449884716786653,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22308213263750076,
      "num_tokens": 89364684.0,
      "step": 43060
    },
    {
      "entropy": 5.679146337509155,
      "epoch": 3.906476777939042,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035446857693378377,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22045473456382753,
      "num_tokens": 89375415.0,
      "step": 43065
    },
    {
      "entropy": 5.742120599746704,
      "epoch": 3.9069303338171264,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003544383050562706,
      "loss": 4.9738,
      "mean_token_accuracy": 0.22321866899728776,
      "num_tokens": 89385791.0,
      "step": 43070
    },
    {
      "entropy": 5.730193710327148,
      "epoch": 3.90738388969521,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000354408031535953,
      "loss": 5.0109,
      "mean_token_accuracy": 0.2195617824792862,
      "num_tokens": 89395345.0,
      "step": 43075
    },
    {
      "entropy": 5.675777816772461,
      "epoch": 3.9078374455732945,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035437775637345715,
      "loss": 4.8498,
      "mean_token_accuracy": 0.22920776754617692,
      "num_tokens": 89405413.0,
      "step": 43080
    },
    {
      "entropy": 5.698183965682984,
      "epoch": 3.9082910014513788,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035434747956940906,
      "loss": 4.877,
      "mean_token_accuracy": 0.22580647021532058,
      "num_tokens": 89415168.0,
      "step": 43085
    },
    {
      "entropy": 5.631013965606689,
      "epoch": 3.908744557329463,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035431720112443475,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2268574446439743,
      "num_tokens": 89426438.0,
      "step": 43090
    },
    {
      "entropy": 5.6644776344299315,
      "epoch": 3.9091981132075473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035428692103916067,
      "loss": 4.8222,
      "mean_token_accuracy": 0.22299847453832627,
      "num_tokens": 89437454.0,
      "step": 43095
    },
    {
      "entropy": 5.734905624389649,
      "epoch": 3.9096516690856316,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003542566393142128,
      "loss": 4.8517,
      "mean_token_accuracy": 0.22591111809015274,
      "num_tokens": 89448082.0,
      "step": 43100
    },
    {
      "entropy": 5.643299007415772,
      "epoch": 3.9101052249637154,
      "grad_norm": 1.328125,
      "learning_rate": 0.00035422635595021745,
      "loss": 4.8262,
      "mean_token_accuracy": 0.22751634269952775,
      "num_tokens": 89458227.0,
      "step": 43105
    },
    {
      "entropy": 5.644638538360596,
      "epoch": 3.9105587808417996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035419607094780084,
      "loss": 4.8853,
      "mean_token_accuracy": 0.22369194477796556,
      "num_tokens": 89468374.0,
      "step": 43110
    },
    {
      "entropy": 5.687273693084717,
      "epoch": 3.911012336719884,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035416578430758934,
      "loss": 4.9247,
      "mean_token_accuracy": 0.223680017888546,
      "num_tokens": 89477728.0,
      "step": 43115
    },
    {
      "entropy": 5.747713947296143,
      "epoch": 3.911465892597968,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035413549603020915,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2223873794078827,
      "num_tokens": 89487107.0,
      "step": 43120
    },
    {
      "entropy": 5.7543418407440186,
      "epoch": 3.911919448476052,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035410520611628676,
      "loss": 4.9176,
      "mean_token_accuracy": 0.23255050778388978,
      "num_tokens": 89497075.0,
      "step": 43125
    },
    {
      "entropy": 5.579173994064331,
      "epoch": 3.9123730043541363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003540749145664486,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22416334152221679,
      "num_tokens": 89508068.0,
      "step": 43130
    },
    {
      "entropy": 5.68187985420227,
      "epoch": 3.9128265602322205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003540446213813211,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23354481756687165,
      "num_tokens": 89518371.0,
      "step": 43135
    },
    {
      "entropy": 5.703180170059204,
      "epoch": 3.913280116110305,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035401432656153057,
      "loss": 4.8932,
      "mean_token_accuracy": 0.22183385938405992,
      "num_tokens": 89529024.0,
      "step": 43140
    },
    {
      "entropy": 5.644742202758789,
      "epoch": 3.913733671988389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035398403010770377,
      "loss": 4.7807,
      "mean_token_accuracy": 0.22984619438648224,
      "num_tokens": 89540331.0,
      "step": 43145
    },
    {
      "entropy": 5.589857196807861,
      "epoch": 3.9141872278664733,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000353953732020467,
      "loss": 4.7956,
      "mean_token_accuracy": 0.2315685287117958,
      "num_tokens": 89550385.0,
      "step": 43150
    },
    {
      "entropy": 5.69700026512146,
      "epoch": 3.9146407837445576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035392343230044687,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21501801759004593,
      "num_tokens": 89561480.0,
      "step": 43155
    },
    {
      "entropy": 5.73602442741394,
      "epoch": 3.9150943396226414,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035389313094827017,
      "loss": 4.8758,
      "mean_token_accuracy": 0.21795661449432374,
      "num_tokens": 89571897.0,
      "step": 43160
    },
    {
      "entropy": 5.732106685638428,
      "epoch": 3.9155478955007257,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035386282796456343,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22340949177742003,
      "num_tokens": 89582019.0,
      "step": 43165
    },
    {
      "entropy": 5.6827805519104,
      "epoch": 3.91600145137881,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003538325233499532,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2228444665670395,
      "num_tokens": 89591806.0,
      "step": 43170
    },
    {
      "entropy": 5.625305557250977,
      "epoch": 3.9164550072568938,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035380221710506636,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22374015152454377,
      "num_tokens": 89601845.0,
      "step": 43175
    },
    {
      "entropy": 5.712595796585083,
      "epoch": 3.916908563134978,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035377190923052964,
      "loss": 4.9661,
      "mean_token_accuracy": 0.22097087949514388,
      "num_tokens": 89612331.0,
      "step": 43180
    },
    {
      "entropy": 5.730075073242188,
      "epoch": 3.9173621190130623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035374159972696966,
      "loss": 4.9452,
      "mean_token_accuracy": 0.22049085199832916,
      "num_tokens": 89622631.0,
      "step": 43185
    },
    {
      "entropy": 5.744501209259033,
      "epoch": 3.9178156748911466,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003537112885950133,
      "loss": 4.9434,
      "mean_token_accuracy": 0.22112187296152114,
      "num_tokens": 89633065.0,
      "step": 43190
    },
    {
      "entropy": 5.743510484695435,
      "epoch": 3.918269230769231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035368097583528756,
      "loss": 4.9266,
      "mean_token_accuracy": 0.22347161471843718,
      "num_tokens": 89643451.0,
      "step": 43195
    },
    {
      "entropy": 5.784735202789307,
      "epoch": 3.918722786647315,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003536506614484192,
      "loss": 4.9361,
      "mean_token_accuracy": 0.21853674799203873,
      "num_tokens": 89654786.0,
      "step": 43200
    },
    {
      "entropy": 5.747192239761352,
      "epoch": 3.9191763425253994,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003536203454350351,
      "loss": 4.9447,
      "mean_token_accuracy": 0.21793559938669205,
      "num_tokens": 89665459.0,
      "step": 43205
    },
    {
      "entropy": 5.677040529251099,
      "epoch": 3.919629898403483,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003535900277957621,
      "loss": 4.8796,
      "mean_token_accuracy": 0.22407102584838867,
      "num_tokens": 89675149.0,
      "step": 43210
    },
    {
      "entropy": 5.693553829193116,
      "epoch": 3.9200834542815675,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035355970853122743,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2289108157157898,
      "num_tokens": 89684994.0,
      "step": 43215
    },
    {
      "entropy": 5.758742332458496,
      "epoch": 3.9205370101596517,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035352938764205793,
      "loss": 5.0411,
      "mean_token_accuracy": 0.20810346752405168,
      "num_tokens": 89694751.0,
      "step": 43220
    },
    {
      "entropy": 5.7102463722229,
      "epoch": 3.920990566037736,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003534990651288807,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2195572406053543,
      "num_tokens": 89704976.0,
      "step": 43225
    },
    {
      "entropy": 5.717789936065674,
      "epoch": 3.92144412191582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035346874099232274,
      "loss": 4.9676,
      "mean_token_accuracy": 0.21705664098262786,
      "num_tokens": 89715863.0,
      "step": 43230
    },
    {
      "entropy": 5.6769129753112795,
      "epoch": 3.921897677793904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035343841523301125,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22075092196464538,
      "num_tokens": 89725856.0,
      "step": 43235
    },
    {
      "entropy": 5.702417278289795,
      "epoch": 3.9223512336719883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003534080878515733,
      "loss": 4.9383,
      "mean_token_accuracy": 0.21830409467220308,
      "num_tokens": 89736345.0,
      "step": 43240
    },
    {
      "entropy": 5.757352590560913,
      "epoch": 3.9228047895500726,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003533777588486362,
      "loss": 4.938,
      "mean_token_accuracy": 0.22455221712589263,
      "num_tokens": 89746960.0,
      "step": 43245
    },
    {
      "entropy": 5.7087586402893065,
      "epoch": 3.923258345428157,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000353347428224827,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21986302584409714,
      "num_tokens": 89757373.0,
      "step": 43250
    },
    {
      "entropy": 5.735062551498413,
      "epoch": 3.923711901306241,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035331709598077304,
      "loss": 4.912,
      "mean_token_accuracy": 0.2200719818472862,
      "num_tokens": 89767895.0,
      "step": 43255
    },
    {
      "entropy": 5.711250162124633,
      "epoch": 3.924165457184325,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003532867621171016,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22296363413333892,
      "num_tokens": 89777512.0,
      "step": 43260
    },
    {
      "entropy": 5.710016775131225,
      "epoch": 3.9246190130624092,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035325642663443994,
      "loss": 4.907,
      "mean_token_accuracy": 0.22619131207466125,
      "num_tokens": 89787308.0,
      "step": 43265
    },
    {
      "entropy": 5.727097415924073,
      "epoch": 3.9250725689404935,
      "grad_norm": 1.375,
      "learning_rate": 0.0003532260895334155,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22859703451395036,
      "num_tokens": 89798384.0,
      "step": 43270
    },
    {
      "entropy": 5.731836175918579,
      "epoch": 3.9255261248185778,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035319575081465556,
      "loss": 4.9473,
      "mean_token_accuracy": 0.21806685775518417,
      "num_tokens": 89808757.0,
      "step": 43275
    },
    {
      "entropy": 5.725045871734619,
      "epoch": 3.9259796806966616,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035316541047878753,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21657539159059525,
      "num_tokens": 89818925.0,
      "step": 43280
    },
    {
      "entropy": 5.704287052154541,
      "epoch": 3.926433236574746,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003531350685264389,
      "loss": 4.9473,
      "mean_token_accuracy": 0.2222840279340744,
      "num_tokens": 89828647.0,
      "step": 43285
    },
    {
      "entropy": 5.7898218631744385,
      "epoch": 3.92688679245283,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035310472495823713,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2144137755036354,
      "num_tokens": 89838784.0,
      "step": 43290
    },
    {
      "entropy": 5.692929983139038,
      "epoch": 3.9273403483309144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003530743797748098,
      "loss": 4.8902,
      "mean_token_accuracy": 0.22276646494865418,
      "num_tokens": 89848464.0,
      "step": 43295
    },
    {
      "entropy": 5.626317501068115,
      "epoch": 3.9277939042089987,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003530440329767844,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22068150639533995,
      "num_tokens": 89859594.0,
      "step": 43300
    },
    {
      "entropy": 5.674175786972046,
      "epoch": 3.928247460087083,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003530136845647885,
      "loss": 4.8448,
      "mean_token_accuracy": 0.22197034060955048,
      "num_tokens": 89869869.0,
      "step": 43305
    },
    {
      "entropy": 5.7484660148620605,
      "epoch": 3.9287010159651667,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035298333453944976,
      "loss": 4.9392,
      "mean_token_accuracy": 0.224872425198555,
      "num_tokens": 89879604.0,
      "step": 43310
    },
    {
      "entropy": 5.601915788650513,
      "epoch": 3.929154571843251,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003529529829013957,
      "loss": 4.766,
      "mean_token_accuracy": 0.2280060663819313,
      "num_tokens": 89889200.0,
      "step": 43315
    },
    {
      "entropy": 5.696368074417114,
      "epoch": 3.9296081277213353,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003529226296512541,
      "loss": 4.9094,
      "mean_token_accuracy": 0.228561869263649,
      "num_tokens": 89898525.0,
      "step": 43320
    },
    {
      "entropy": 5.662487411499024,
      "epoch": 3.9300616835994195,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00035289227478965267,
      "loss": 4.9365,
      "mean_token_accuracy": 0.2179530844092369,
      "num_tokens": 89909053.0,
      "step": 43325
    },
    {
      "entropy": 5.651578760147094,
      "epoch": 3.9305152394775034,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003528619183172192,
      "loss": 4.8253,
      "mean_token_accuracy": 0.2301553875207901,
      "num_tokens": 89918375.0,
      "step": 43330
    },
    {
      "entropy": 5.772030687332153,
      "epoch": 3.9309687953555876,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003528315602345814,
      "loss": 4.9616,
      "mean_token_accuracy": 0.22663764506578446,
      "num_tokens": 89928252.0,
      "step": 43335
    },
    {
      "entropy": 5.688143873214722,
      "epoch": 3.931422351233672,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003528012005423671,
      "loss": 4.8952,
      "mean_token_accuracy": 0.2179613396525383,
      "num_tokens": 89938613.0,
      "step": 43340
    },
    {
      "entropy": 5.742148923873901,
      "epoch": 3.931875907111756,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035277083924120406,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22446449100971222,
      "num_tokens": 89948781.0,
      "step": 43345
    },
    {
      "entropy": 5.692275047302246,
      "epoch": 3.9323294629898404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035274047633172037,
      "loss": 4.9166,
      "mean_token_accuracy": 0.21644906997680663,
      "num_tokens": 89959928.0,
      "step": 43350
    },
    {
      "entropy": 5.762853288650513,
      "epoch": 3.9327830188679247,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003527101118145437,
      "loss": 4.9465,
      "mean_token_accuracy": 0.2154739663004875,
      "num_tokens": 89970756.0,
      "step": 43355
    },
    {
      "entropy": 5.7425648212432865,
      "epoch": 3.933236574746009,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003526797456903022,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2225078597664833,
      "num_tokens": 89981332.0,
      "step": 43360
    },
    {
      "entropy": 5.68741021156311,
      "epoch": 3.933690130624093,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035264937795962374,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22295965105295182,
      "num_tokens": 89991427.0,
      "step": 43365
    },
    {
      "entropy": 5.713253450393677,
      "epoch": 3.934143686502177,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003526190086231364,
      "loss": 4.8656,
      "mean_token_accuracy": 0.23191914856433868,
      "num_tokens": 90001599.0,
      "step": 43370
    },
    {
      "entropy": 5.704966592788696,
      "epoch": 3.9345972423802613,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035258863768146816,
      "loss": 4.9037,
      "mean_token_accuracy": 0.224869741499424,
      "num_tokens": 90011932.0,
      "step": 43375
    },
    {
      "entropy": 5.7174766063690186,
      "epoch": 3.9350507982583456,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003525582651352471,
      "loss": 4.9981,
      "mean_token_accuracy": 0.22066253125667573,
      "num_tokens": 90022123.0,
      "step": 43380
    },
    {
      "entropy": 5.65465784072876,
      "epoch": 3.9355043541364294,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003525278909851013,
      "loss": 4.8338,
      "mean_token_accuracy": 0.2317067265510559,
      "num_tokens": 90033024.0,
      "step": 43385
    },
    {
      "entropy": 5.717942094802856,
      "epoch": 3.9359579100145137,
      "grad_norm": 1.234375,
      "learning_rate": 0.000352497515231659,
      "loss": 4.878,
      "mean_token_accuracy": 0.2273194044828415,
      "num_tokens": 90043742.0,
      "step": 43390
    },
    {
      "entropy": 5.674232101440429,
      "epoch": 3.936411465892598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003524671378755484,
      "loss": 4.86,
      "mean_token_accuracy": 0.22144481241703035,
      "num_tokens": 90054508.0,
      "step": 43395
    },
    {
      "entropy": 5.688033485412598,
      "epoch": 3.936865021770682,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003524367589173976,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2241895779967308,
      "num_tokens": 90065058.0,
      "step": 43400
    },
    {
      "entropy": 5.695210886001587,
      "epoch": 3.9373185776487665,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00035240637835783485,
      "loss": 4.9136,
      "mean_token_accuracy": 0.223476280272007,
      "num_tokens": 90075813.0,
      "step": 43405
    },
    {
      "entropy": 5.75432939529419,
      "epoch": 3.9377721335268507,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035237599619748853,
      "loss": 4.9572,
      "mean_token_accuracy": 0.21534956842660904,
      "num_tokens": 90087064.0,
      "step": 43410
    },
    {
      "entropy": 5.686218214035034,
      "epoch": 3.9382256894049346,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003523456124369868,
      "loss": 4.8368,
      "mean_token_accuracy": 0.22935550957918166,
      "num_tokens": 90096668.0,
      "step": 43415
    },
    {
      "entropy": 5.638683557510376,
      "epoch": 3.938679245283019,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003523152270769582,
      "loss": 4.7697,
      "mean_token_accuracy": 0.23469507694244385,
      "num_tokens": 90105569.0,
      "step": 43420
    },
    {
      "entropy": 5.697639894485474,
      "epoch": 3.939132801161103,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000352284840118031,
      "loss": 4.941,
      "mean_token_accuracy": 0.21961626708507537,
      "num_tokens": 90116040.0,
      "step": 43425
    },
    {
      "entropy": 5.733775806427002,
      "epoch": 3.9395863570391874,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003522544515608335,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22025926411151886,
      "num_tokens": 90125929.0,
      "step": 43430
    },
    {
      "entropy": 5.814044666290283,
      "epoch": 3.940039912917271,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003522240614059944,
      "loss": 4.9815,
      "mean_token_accuracy": 0.2191663533449173,
      "num_tokens": 90136251.0,
      "step": 43435
    },
    {
      "entropy": 5.7245259284973145,
      "epoch": 3.9404934687953554,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000352193669654142,
      "loss": 4.851,
      "mean_token_accuracy": 0.23167499601840974,
      "num_tokens": 90146480.0,
      "step": 43440
    },
    {
      "entropy": 5.726921892166137,
      "epoch": 3.9409470246734397,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035216327630590476,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2199931338429451,
      "num_tokens": 90157192.0,
      "step": 43445
    },
    {
      "entropy": 5.711971426010132,
      "epoch": 3.941400580551524,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035213288136191136,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22148375064134598,
      "num_tokens": 90167120.0,
      "step": 43450
    },
    {
      "entropy": 5.6866796016693115,
      "epoch": 3.9418541364296082,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035210248482279034,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22462537586688996,
      "num_tokens": 90176276.0,
      "step": 43455
    },
    {
      "entropy": 5.655860090255738,
      "epoch": 3.9423076923076925,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035207208668917026,
      "loss": 4.8971,
      "mean_token_accuracy": 0.2214522734284401,
      "num_tokens": 90187150.0,
      "step": 43460
    },
    {
      "entropy": 5.640939140319825,
      "epoch": 3.9427612481857763,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035204168696167975,
      "loss": 4.8281,
      "mean_token_accuracy": 0.21816634386777878,
      "num_tokens": 90198112.0,
      "step": 43465
    },
    {
      "entropy": 5.675923681259155,
      "epoch": 3.9432148040638606,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035201128564094753,
      "loss": 4.943,
      "mean_token_accuracy": 0.22197229266166688,
      "num_tokens": 90208940.0,
      "step": 43470
    },
    {
      "entropy": 5.709405136108399,
      "epoch": 3.943668359941945,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003519808827276023,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22684278935194016,
      "num_tokens": 90220073.0,
      "step": 43475
    },
    {
      "entropy": 5.6813537120819095,
      "epoch": 3.944121915820029,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035195047822227283,
      "loss": 4.9315,
      "mean_token_accuracy": 0.21667447686195374,
      "num_tokens": 90230059.0,
      "step": 43480
    },
    {
      "entropy": 5.618416213989258,
      "epoch": 3.944575471698113,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00035192007212558787,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22674121409654618,
      "num_tokens": 90240486.0,
      "step": 43485
    },
    {
      "entropy": 5.758249187469483,
      "epoch": 3.9450290275761972,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003518896644381762,
      "loss": 4.9641,
      "mean_token_accuracy": 0.21738604009151458,
      "num_tokens": 90250014.0,
      "step": 43490
    },
    {
      "entropy": 5.7031261920928955,
      "epoch": 3.9454825834542815,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003518592551606666,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2150621995329857,
      "num_tokens": 90261040.0,
      "step": 43495
    },
    {
      "entropy": 5.743045043945313,
      "epoch": 3.9459361393323658,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003518288442936881,
      "loss": 4.9346,
      "mean_token_accuracy": 0.2263345956802368,
      "num_tokens": 90271243.0,
      "step": 43500
    },
    {
      "entropy": 5.694262599945068,
      "epoch": 3.94638969521045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003517984318378695,
      "loss": 4.8944,
      "mean_token_accuracy": 0.22142966091632843,
      "num_tokens": 90281176.0,
      "step": 43505
    },
    {
      "entropy": 5.794441366195679,
      "epoch": 3.9468432510885343,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003517680177938398,
      "loss": 4.9721,
      "mean_token_accuracy": 0.22160424292087555,
      "num_tokens": 90292465.0,
      "step": 43510
    },
    {
      "entropy": 5.734121561050415,
      "epoch": 3.9472968069666186,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003517376021622279,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2288681074976921,
      "num_tokens": 90303482.0,
      "step": 43515
    },
    {
      "entropy": 5.69947509765625,
      "epoch": 3.9477503628447024,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003517071849436629,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22025609165430068,
      "num_tokens": 90313696.0,
      "step": 43520
    },
    {
      "entropy": 5.6887610912322994,
      "epoch": 3.9482039187227866,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035167676613877366,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21415794491767884,
      "num_tokens": 90324564.0,
      "step": 43525
    },
    {
      "entropy": 5.661167573928833,
      "epoch": 3.948657474600871,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003516463457481894,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23634923994541168,
      "num_tokens": 90334440.0,
      "step": 43530
    },
    {
      "entropy": 5.726218891143799,
      "epoch": 3.9491110304789547,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003516159237725392,
      "loss": 4.9251,
      "mean_token_accuracy": 0.2209845080971718,
      "num_tokens": 90344029.0,
      "step": 43535
    },
    {
      "entropy": 5.7383016586303714,
      "epoch": 3.949564586357039,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003515855002124521,
      "loss": 4.9425,
      "mean_token_accuracy": 0.21595233231782912,
      "num_tokens": 90354628.0,
      "step": 43540
    },
    {
      "entropy": 5.716387176513672,
      "epoch": 3.9500181422351233,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003515550750685573,
      "loss": 4.8136,
      "mean_token_accuracy": 0.23017212599515915,
      "num_tokens": 90365091.0,
      "step": 43545
    },
    {
      "entropy": 5.700933599472046,
      "epoch": 3.9504716981132075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003515246483414842,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22156794220209122,
      "num_tokens": 90374197.0,
      "step": 43550
    },
    {
      "entropy": 5.731234407424926,
      "epoch": 3.950925253991292,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035149422003186176,
      "loss": 4.9404,
      "mean_token_accuracy": 0.22234025448560715,
      "num_tokens": 90384686.0,
      "step": 43555
    },
    {
      "entropy": 5.770050477981568,
      "epoch": 3.951378809869376,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035146379014031943,
      "loss": 4.9725,
      "mean_token_accuracy": 0.22053298056125642,
      "num_tokens": 90396329.0,
      "step": 43560
    },
    {
      "entropy": 5.749848079681397,
      "epoch": 3.9518323657474603,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003514333586674864,
      "loss": 4.9689,
      "mean_token_accuracy": 0.21848632544279098,
      "num_tokens": 90408957.0,
      "step": 43565
    },
    {
      "entropy": 5.640656518936157,
      "epoch": 3.952285921625544,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000351402925613992,
      "loss": 4.8529,
      "mean_token_accuracy": 0.22737266570329667,
      "num_tokens": 90420194.0,
      "step": 43570
    },
    {
      "entropy": 5.669390535354614,
      "epoch": 3.9527394775036284,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003513724909804656,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2255084693431854,
      "num_tokens": 90430881.0,
      "step": 43575
    },
    {
      "entropy": 5.6828831195831295,
      "epoch": 3.9531930333817127,
      "grad_norm": 1.390625,
      "learning_rate": 0.00035134205476753674,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2154932513833046,
      "num_tokens": 90440713.0,
      "step": 43580
    },
    {
      "entropy": 5.715508079528808,
      "epoch": 3.953646589259797,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003513116169758347,
      "loss": 4.9589,
      "mean_token_accuracy": 0.22007109820842743,
      "num_tokens": 90451274.0,
      "step": 43585
    },
    {
      "entropy": 5.75844898223877,
      "epoch": 3.9541001451378808,
      "grad_norm": 1.25,
      "learning_rate": 0.0003512811776059889,
      "loss": 4.9918,
      "mean_token_accuracy": 0.21818943023681642,
      "num_tokens": 90461503.0,
      "step": 43590
    },
    {
      "entropy": 5.696061420440674,
      "epoch": 3.954553701015965,
      "grad_norm": 1.34375,
      "learning_rate": 0.000351250736658629,
      "loss": 4.8214,
      "mean_token_accuracy": 0.23215697854757308,
      "num_tokens": 90471178.0,
      "step": 43595
    },
    {
      "entropy": 5.677223825454712,
      "epoch": 3.9550072568940493,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003512202941343844,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22452804446220398,
      "num_tokens": 90481573.0,
      "step": 43600
    },
    {
      "entropy": 5.721448087692261,
      "epoch": 3.9554608127721336,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003511898500338847,
      "loss": 4.9547,
      "mean_token_accuracy": 0.2184355691075325,
      "num_tokens": 90491853.0,
      "step": 43605
    },
    {
      "entropy": 5.715265369415283,
      "epoch": 3.955914368650218,
      "grad_norm": 1.25,
      "learning_rate": 0.0003511594043577594,
      "loss": 4.9359,
      "mean_token_accuracy": 0.21365946680307388,
      "num_tokens": 90502958.0,
      "step": 43610
    },
    {
      "entropy": 5.649737882614136,
      "epoch": 3.956367924528302,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035112895710663833,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22374788224697112,
      "num_tokens": 90512555.0,
      "step": 43615
    },
    {
      "entropy": 5.726055860519409,
      "epoch": 3.956821480406386,
      "grad_norm": 1.25,
      "learning_rate": 0.00035109850828115105,
      "loss": 4.9277,
      "mean_token_accuracy": 0.22756720930337906,
      "num_tokens": 90522722.0,
      "step": 43620
    },
    {
      "entropy": 5.7051492691040036,
      "epoch": 3.95727503628447,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035106805788192707,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22285272926092148,
      "num_tokens": 90533182.0,
      "step": 43625
    },
    {
      "entropy": 5.761256361007691,
      "epoch": 3.9577285921625545,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003510376059095963,
      "loss": 4.9539,
      "mean_token_accuracy": 0.21384112536907196,
      "num_tokens": 90543495.0,
      "step": 43630
    },
    {
      "entropy": 5.653493595123291,
      "epoch": 3.9581821480406387,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003510071523647886,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2256562039256096,
      "num_tokens": 90553680.0,
      "step": 43635
    },
    {
      "entropy": 5.681036710739136,
      "epoch": 3.9586357039187225,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003509766972481336,
      "loss": 4.8224,
      "mean_token_accuracy": 0.22594478726387024,
      "num_tokens": 90563513.0,
      "step": 43640
    },
    {
      "entropy": 5.800935220718384,
      "epoch": 3.959089259796807,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000350946240560261,
      "loss": 4.9537,
      "mean_token_accuracy": 0.21822860091924667,
      "num_tokens": 90573337.0,
      "step": 43645
    },
    {
      "entropy": 5.685687685012818,
      "epoch": 3.959542815674891,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003509157823018009,
      "loss": 4.8816,
      "mean_token_accuracy": 0.22647543400526046,
      "num_tokens": 90582979.0,
      "step": 43650
    },
    {
      "entropy": 5.769347238540649,
      "epoch": 3.9599963715529753,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035088532247338306,
      "loss": 4.9928,
      "mean_token_accuracy": 0.21049410998821258,
      "num_tokens": 90593552.0,
      "step": 43655
    },
    {
      "entropy": 5.759906530380249,
      "epoch": 3.9604499274310596,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035085486107563733,
      "loss": 4.9165,
      "mean_token_accuracy": 0.22160508930683137,
      "num_tokens": 90603813.0,
      "step": 43660
    },
    {
      "entropy": 5.7381353855133055,
      "epoch": 3.960903483309144,
      "grad_norm": 1.265625,
      "learning_rate": 0.00035082439810919385,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2267778441309929,
      "num_tokens": 90613885.0,
      "step": 43665
    },
    {
      "entropy": 5.694240236282349,
      "epoch": 3.9613570391872277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003507939335746824,
      "loss": 4.8046,
      "mean_token_accuracy": 0.22059134691953658,
      "num_tokens": 90624665.0,
      "step": 43670
    },
    {
      "entropy": 5.681277275085449,
      "epoch": 3.961810595065312,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003507634674727332,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22494624555110931,
      "num_tokens": 90635496.0,
      "step": 43675
    },
    {
      "entropy": 5.719098329544067,
      "epoch": 3.9622641509433962,
      "grad_norm": 1.375,
      "learning_rate": 0.0003507329998039762,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22567182034254074,
      "num_tokens": 90645502.0,
      "step": 43680
    },
    {
      "entropy": 5.61095118522644,
      "epoch": 3.9627177068214805,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035070253056904133,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23037520796060562,
      "num_tokens": 90654724.0,
      "step": 43685
    },
    {
      "entropy": 5.668018817901611,
      "epoch": 3.9631712626995643,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003506720597685589,
      "loss": 4.8688,
      "mean_token_accuracy": 0.22323452383279802,
      "num_tokens": 90665280.0,
      "step": 43690
    },
    {
      "entropy": 5.704755687713623,
      "epoch": 3.9636248185776486,
      "grad_norm": 1.28125,
      "learning_rate": 0.00035064158740315894,
      "loss": 4.9331,
      "mean_token_accuracy": 0.2212210476398468,
      "num_tokens": 90675804.0,
      "step": 43695
    },
    {
      "entropy": 5.695136642456054,
      "epoch": 3.964078374455733,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035061111347347174,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22346337735652924,
      "num_tokens": 90686636.0,
      "step": 43700
    },
    {
      "entropy": 5.660781574249268,
      "epoch": 3.964531930333817,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035058063798012746,
      "loss": 4.8881,
      "mean_token_accuracy": 0.22585732042789458,
      "num_tokens": 90697666.0,
      "step": 43705
    },
    {
      "entropy": 5.741635322570801,
      "epoch": 3.9649854862119014,
      "grad_norm": 1.765625,
      "learning_rate": 0.0003505501609237563,
      "loss": 4.9799,
      "mean_token_accuracy": 0.22086500823497773,
      "num_tokens": 90708111.0,
      "step": 43710
    },
    {
      "entropy": 5.6993889808654785,
      "epoch": 3.9654390420899857,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035051968230498853,
      "loss": 4.8455,
      "mean_token_accuracy": 0.22709561586380006,
      "num_tokens": 90718711.0,
      "step": 43715
    },
    {
      "entropy": 5.648063802719117,
      "epoch": 3.96589259796807,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035048920212445454,
      "loss": 4.841,
      "mean_token_accuracy": 0.2289152055978775,
      "num_tokens": 90729585.0,
      "step": 43720
    },
    {
      "entropy": 5.629609155654907,
      "epoch": 3.9663461538461537,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035045872038278467,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23136287182569504,
      "num_tokens": 90739654.0,
      "step": 43725
    },
    {
      "entropy": 5.694218730926513,
      "epoch": 3.966799709724238,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003504282370806091,
      "loss": 4.8449,
      "mean_token_accuracy": 0.22711179703474044,
      "num_tokens": 90749456.0,
      "step": 43730
    },
    {
      "entropy": 5.670402574539184,
      "epoch": 3.9672532656023223,
      "grad_norm": 1.359375,
      "learning_rate": 0.00035039775221855845,
      "loss": 4.8459,
      "mean_token_accuracy": 0.231731815636158,
      "num_tokens": 90760051.0,
      "step": 43735
    },
    {
      "entropy": 5.632954168319702,
      "epoch": 3.9677068214804065,
      "grad_norm": 1.25,
      "learning_rate": 0.000350367265797263,
      "loss": 4.8213,
      "mean_token_accuracy": 0.22650574445724486,
      "num_tokens": 90770452.0,
      "step": 43740
    },
    {
      "entropy": 5.6726155281066895,
      "epoch": 3.9681603773584904,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035033677781735336,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22172708660364152,
      "num_tokens": 90780150.0,
      "step": 43745
    },
    {
      "entropy": 5.672044324874878,
      "epoch": 3.9686139332365746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003503062882794599,
      "loss": 4.8168,
      "mean_token_accuracy": 0.23037809729576111,
      "num_tokens": 90790338.0,
      "step": 43750
    },
    {
      "entropy": 5.669913911819458,
      "epoch": 3.969067489114659,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00035027579718421327,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22585249096155166,
      "num_tokens": 90800195.0,
      "step": 43755
    },
    {
      "entropy": 5.709565448760986,
      "epoch": 3.969521044992743,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003502453045322439,
      "loss": 4.961,
      "mean_token_accuracy": 0.21745812594890596,
      "num_tokens": 90810872.0,
      "step": 43760
    },
    {
      "entropy": 5.716914939880371,
      "epoch": 3.9699746008708274,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00035021481032418254,
      "loss": 4.8736,
      "mean_token_accuracy": 0.22125812470912934,
      "num_tokens": 90821018.0,
      "step": 43765
    },
    {
      "entropy": 5.721039438247681,
      "epoch": 3.9704281567489117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003501843145606597,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22068131268024443,
      "num_tokens": 90830274.0,
      "step": 43770
    },
    {
      "entropy": 5.639467334747314,
      "epoch": 3.9708817126269955,
      "grad_norm": 1.28125,
      "learning_rate": 0.000350153817242306,
      "loss": 4.7942,
      "mean_token_accuracy": 0.2361720994114876,
      "num_tokens": 90839926.0,
      "step": 43775
    },
    {
      "entropy": 5.601562547683716,
      "epoch": 3.97133526850508,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003501233183697523,
      "loss": 4.795,
      "mean_token_accuracy": 0.22289225608110427,
      "num_tokens": 90850263.0,
      "step": 43780
    },
    {
      "entropy": 5.698162603378296,
      "epoch": 3.971788824383164,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00035009281794362926,
      "loss": 4.9078,
      "mean_token_accuracy": 0.21344294846057893,
      "num_tokens": 90861747.0,
      "step": 43785
    },
    {
      "entropy": 5.724799585342407,
      "epoch": 3.9722423802612483,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035006231596456755,
      "loss": 4.9693,
      "mean_token_accuracy": 0.21479806154966355,
      "num_tokens": 90872186.0,
      "step": 43790
    },
    {
      "entropy": 5.745276021957397,
      "epoch": 3.972695936139332,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00035003181243319797,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22638722509145737,
      "num_tokens": 90881088.0,
      "step": 43795
    },
    {
      "entropy": 5.635633134841919,
      "epoch": 3.9731494920174164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003500013073501514,
      "loss": 4.8432,
      "mean_token_accuracy": 0.2237011596560478,
      "num_tokens": 90890589.0,
      "step": 43800
    },
    {
      "entropy": 5.684895849227905,
      "epoch": 3.9736030478955007,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003499708007160588,
      "loss": 4.94,
      "mean_token_accuracy": 0.22109041661024093,
      "num_tokens": 90901692.0,
      "step": 43805
    },
    {
      "entropy": 5.750027847290039,
      "epoch": 3.974056603773585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003499402925315509,
      "loss": 4.9631,
      "mean_token_accuracy": 0.2198677808046341,
      "num_tokens": 90911811.0,
      "step": 43810
    },
    {
      "entropy": 5.729765748977661,
      "epoch": 3.974510159651669,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034990978279725846,
      "loss": 4.8039,
      "mean_token_accuracy": 0.22888574302196502,
      "num_tokens": 90922323.0,
      "step": 43815
    },
    {
      "entropy": 5.795409488677978,
      "epoch": 3.9749637155297535,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034987927151381283,
      "loss": 5.0442,
      "mean_token_accuracy": 0.2101174548268318,
      "num_tokens": 90932834.0,
      "step": 43820
    },
    {
      "entropy": 5.728570747375488,
      "epoch": 3.9754172714078373,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034984875868184466,
      "loss": 4.8886,
      "mean_token_accuracy": 0.22751642018556595,
      "num_tokens": 90943616.0,
      "step": 43825
    },
    {
      "entropy": 5.733536243438721,
      "epoch": 3.9758708272859216,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034981824430198517,
      "loss": 4.9762,
      "mean_token_accuracy": 0.22046145647764206,
      "num_tokens": 90954468.0,
      "step": 43830
    },
    {
      "entropy": 5.696765756607055,
      "epoch": 3.976324383164006,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034978772837486536,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2285262793302536,
      "num_tokens": 90965109.0,
      "step": 43835
    },
    {
      "entropy": 5.719534206390381,
      "epoch": 3.97677793904209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034975721090111625,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22307549715042113,
      "num_tokens": 90974789.0,
      "step": 43840
    },
    {
      "entropy": 5.720041131973266,
      "epoch": 3.977231494920174,
      "grad_norm": 1.25,
      "learning_rate": 0.0003497266918813689,
      "loss": 4.9757,
      "mean_token_accuracy": 0.21816346049308777,
      "num_tokens": 90985154.0,
      "step": 43845
    },
    {
      "entropy": 5.715408182144165,
      "epoch": 3.977685050798258,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003496961713162545,
      "loss": 4.9371,
      "mean_token_accuracy": 0.22252209931612016,
      "num_tokens": 90996007.0,
      "step": 43850
    },
    {
      "entropy": 5.776365613937378,
      "epoch": 3.9781386066763424,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003496656492064043,
      "loss": 4.9629,
      "mean_token_accuracy": 0.21786343008279802,
      "num_tokens": 91006253.0,
      "step": 43855
    },
    {
      "entropy": 5.771911144256592,
      "epoch": 3.9785921625544267,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003496351255524494,
      "loss": 5.0281,
      "mean_token_accuracy": 0.21573567837476731,
      "num_tokens": 91017169.0,
      "step": 43860
    },
    {
      "entropy": 5.6889454364776615,
      "epoch": 3.979045718432511,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034960460035502117,
      "loss": 4.892,
      "mean_token_accuracy": 0.2200805813074112,
      "num_tokens": 91026684.0,
      "step": 43865
    },
    {
      "entropy": 5.671290111541748,
      "epoch": 3.9794992743105952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003495740736147508,
      "loss": 4.8309,
      "mean_token_accuracy": 0.22544642984867097,
      "num_tokens": 91037462.0,
      "step": 43870
    },
    {
      "entropy": 5.763996171951294,
      "epoch": 3.9799528301886795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034954354533226953,
      "loss": 4.982,
      "mean_token_accuracy": 0.21669506430625915,
      "num_tokens": 91049418.0,
      "step": 43875
    },
    {
      "entropy": 5.678443765640258,
      "epoch": 3.9804063860667633,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003495130155082087,
      "loss": 4.8453,
      "mean_token_accuracy": 0.2220543846487999,
      "num_tokens": 91061013.0,
      "step": 43880
    },
    {
      "entropy": 5.6805847644805905,
      "epoch": 3.9808599419448476,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003494824841431998,
      "loss": 4.8781,
      "mean_token_accuracy": 0.2251404568552971,
      "num_tokens": 91070974.0,
      "step": 43885
    },
    {
      "entropy": 5.711058950424194,
      "epoch": 3.981313497822932,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003494519512378741,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2247988447546959,
      "num_tokens": 91082681.0,
      "step": 43890
    },
    {
      "entropy": 5.718933010101319,
      "epoch": 3.9817670537010157,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034942141679286307,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2295877978205681,
      "num_tokens": 91093552.0,
      "step": 43895
    },
    {
      "entropy": 5.725760078430175,
      "epoch": 3.9822206095791,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003493908808087982,
      "loss": 4.93,
      "mean_token_accuracy": 0.21561136841773987,
      "num_tokens": 91105470.0,
      "step": 43900
    },
    {
      "entropy": 5.716452789306641,
      "epoch": 3.9826741654571842,
      "grad_norm": 1.09375,
      "learning_rate": 0.000349360343286311,
      "loss": 4.8685,
      "mean_token_accuracy": 0.22770961970090867,
      "num_tokens": 91115822.0,
      "step": 43905
    },
    {
      "entropy": 5.692395210266113,
      "epoch": 3.9831277213352685,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00034932980422603283,
      "loss": 4.8466,
      "mean_token_accuracy": 0.2298087179660797,
      "num_tokens": 91125722.0,
      "step": 43910
    },
    {
      "entropy": 5.802732944488525,
      "epoch": 3.9835812772133528,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003492992636285955,
      "loss": 5.0214,
      "mean_token_accuracy": 0.2256566360592842,
      "num_tokens": 91135783.0,
      "step": 43915
    },
    {
      "entropy": 5.7217858791351315,
      "epoch": 3.984034833091437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003492687214946304,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2282187208533287,
      "num_tokens": 91146151.0,
      "step": 43920
    },
    {
      "entropy": 5.68949522972107,
      "epoch": 3.9844883889695213,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034923817782476927,
      "loss": 4.907,
      "mean_token_accuracy": 0.21798891574144363,
      "num_tokens": 91156711.0,
      "step": 43925
    },
    {
      "entropy": 5.679375410079956,
      "epoch": 3.984941944847605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003492076326196436,
      "loss": 4.9351,
      "mean_token_accuracy": 0.21624661087989808,
      "num_tokens": 91166601.0,
      "step": 43930
    },
    {
      "entropy": 5.765782070159912,
      "epoch": 3.9853955007256894,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034917708587988523,
      "loss": 4.9562,
      "mean_token_accuracy": 0.2093972608447075,
      "num_tokens": 91176556.0,
      "step": 43935
    },
    {
      "entropy": 5.714406538009643,
      "epoch": 3.9858490566037736,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034914653760612584,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22366479337215422,
      "num_tokens": 91186499.0,
      "step": 43940
    },
    {
      "entropy": 5.682843828201294,
      "epoch": 3.986302612481858,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003491159877989971,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2170662209391594,
      "num_tokens": 91197088.0,
      "step": 43945
    },
    {
      "entropy": 5.680780220031738,
      "epoch": 3.9867561683599417,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000349085436459131,
      "loss": 4.9038,
      "mean_token_accuracy": 0.22644558995962144,
      "num_tokens": 91207188.0,
      "step": 43950
    },
    {
      "entropy": 5.673380327224732,
      "epoch": 3.987209724238026,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034905488358715907,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22777435630559922,
      "num_tokens": 91217349.0,
      "step": 43955
    },
    {
      "entropy": 5.728506660461425,
      "epoch": 3.9876632801161103,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003490243291837133,
      "loss": 4.9358,
      "mean_token_accuracy": 0.2204625725746155,
      "num_tokens": 91227775.0,
      "step": 43960
    },
    {
      "entropy": 5.699440288543701,
      "epoch": 3.9881168359941945,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034899377324942556,
      "loss": 4.794,
      "mean_token_accuracy": 0.23354199826717376,
      "num_tokens": 91238315.0,
      "step": 43965
    },
    {
      "entropy": 5.697731161117554,
      "epoch": 3.988570391872279,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034896321578492767,
      "loss": 5.0061,
      "mean_token_accuracy": 0.21632297337055206,
      "num_tokens": 91249178.0,
      "step": 43970
    },
    {
      "entropy": 5.659305477142334,
      "epoch": 3.989023947750363,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003489326567908516,
      "loss": 4.8704,
      "mean_token_accuracy": 0.22513043731451035,
      "num_tokens": 91257911.0,
      "step": 43975
    },
    {
      "entropy": 5.704394721984864,
      "epoch": 3.989477503628447,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034890209626782943,
      "loss": 4.908,
      "mean_token_accuracy": 0.21791397780179977,
      "num_tokens": 91267491.0,
      "step": 43980
    },
    {
      "entropy": 5.677386951446533,
      "epoch": 3.989931059506531,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003488715342164931,
      "loss": 4.8937,
      "mean_token_accuracy": 0.2273339182138443,
      "num_tokens": 91278292.0,
      "step": 43985
    },
    {
      "entropy": 5.731640481948853,
      "epoch": 3.9903846153846154,
      "grad_norm": 1.46875,
      "learning_rate": 0.00034884097063747453,
      "loss": 4.9895,
      "mean_token_accuracy": 0.21107571125030516,
      "num_tokens": 91287871.0,
      "step": 43990
    },
    {
      "entropy": 5.6711469173431395,
      "epoch": 3.9908381712626997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003488104055314059,
      "loss": 4.8297,
      "mean_token_accuracy": 0.22915210127830504,
      "num_tokens": 91297201.0,
      "step": 43995
    },
    {
      "entropy": 5.6121907234191895,
      "epoch": 3.9912917271407835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003487798388989193,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23009294420480728,
      "num_tokens": 91306449.0,
      "step": 44000
    },
    {
      "entropy": 5.617125511169434,
      "epoch": 3.9917452830188678,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034874927074064684,
      "loss": 4.8016,
      "mean_token_accuracy": 0.22320926189422607,
      "num_tokens": 91317516.0,
      "step": 44005
    },
    {
      "entropy": 5.689169120788574,
      "epoch": 3.992198838896952,
      "grad_norm": 1.375,
      "learning_rate": 0.0003487187010572206,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21937841325998306,
      "num_tokens": 91326850.0,
      "step": 44010
    },
    {
      "entropy": 5.6886955261230465,
      "epoch": 3.9926523947750363,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003486881298492729,
      "loss": 4.9152,
      "mean_token_accuracy": 0.2206160306930542,
      "num_tokens": 91338068.0,
      "step": 44015
    },
    {
      "entropy": 5.776484775543213,
      "epoch": 3.9931059506531206,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003486575571174359,
      "loss": 4.8972,
      "mean_token_accuracy": 0.22222470492124557,
      "num_tokens": 91349627.0,
      "step": 44020
    },
    {
      "entropy": 5.73153862953186,
      "epoch": 3.993559506531205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034862698286234185,
      "loss": 4.9652,
      "mean_token_accuracy": 0.21464336663484573,
      "num_tokens": 91361229.0,
      "step": 44025
    },
    {
      "entropy": 5.736625385284424,
      "epoch": 3.9940130624092887,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003485964070846229,
      "loss": 4.9874,
      "mean_token_accuracy": 0.21108559966087342,
      "num_tokens": 91372348.0,
      "step": 44030
    },
    {
      "entropy": 5.675780391693115,
      "epoch": 3.994466618287373,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003485658297849117,
      "loss": 4.839,
      "mean_token_accuracy": 0.22279536575078965,
      "num_tokens": 91383028.0,
      "step": 44035
    },
    {
      "entropy": 5.725448846817017,
      "epoch": 3.994920174165457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034853525096384034,
      "loss": 4.9578,
      "mean_token_accuracy": 0.21627615243196488,
      "num_tokens": 91393263.0,
      "step": 44040
    },
    {
      "entropy": 5.756396722793579,
      "epoch": 3.9953737300435415,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003485046706220411,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22138756215572358,
      "num_tokens": 91403881.0,
      "step": 44045
    },
    {
      "entropy": 5.72415509223938,
      "epoch": 3.9958272859216253,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003484740887601467,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22164282947778702,
      "num_tokens": 91413723.0,
      "step": 44050
    },
    {
      "entropy": 5.712735366821289,
      "epoch": 3.9962808417997095,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034844350537878937,
      "loss": 4.88,
      "mean_token_accuracy": 0.22282869070768357,
      "num_tokens": 91424168.0,
      "step": 44055
    },
    {
      "entropy": 5.79491810798645,
      "epoch": 3.996734397677794,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034841292047860164,
      "loss": 4.9922,
      "mean_token_accuracy": 0.2106110855937004,
      "num_tokens": 91433133.0,
      "step": 44060
    },
    {
      "entropy": 5.737838792800903,
      "epoch": 3.997187953555878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000348382334060216,
      "loss": 4.9836,
      "mean_token_accuracy": 0.21028833091259003,
      "num_tokens": 91444521.0,
      "step": 44065
    },
    {
      "entropy": 5.656456232070923,
      "epoch": 3.9976415094339623,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000348351746124265,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2283462330698967,
      "num_tokens": 91454870.0,
      "step": 44070
    },
    {
      "entropy": 5.684705066680908,
      "epoch": 3.9980950653120466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003483211566713812,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2325435072183609,
      "num_tokens": 91465472.0,
      "step": 44075
    },
    {
      "entropy": 5.7271263122558596,
      "epoch": 3.998548621190131,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003482905657021971,
      "loss": 4.901,
      "mean_token_accuracy": 0.22087326049804687,
      "num_tokens": 91475467.0,
      "step": 44080
    },
    {
      "entropy": 5.742320871353149,
      "epoch": 3.9990021770682147,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003482599732173456,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2240229293704033,
      "num_tokens": 91485615.0,
      "step": 44085
    },
    {
      "entropy": 5.6885459423065186,
      "epoch": 3.999455732946299,
      "grad_norm": 1.25,
      "learning_rate": 0.0003482293792174591,
      "loss": 4.8727,
      "mean_token_accuracy": 0.22675559371709825,
      "num_tokens": 91495790.0,
      "step": 44090
    },
    {
      "entropy": 5.688253927230835,
      "epoch": 3.9999092888243832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003481987837031703,
      "loss": 4.9513,
      "mean_token_accuracy": 0.21600220650434493,
      "num_tokens": 91507164.0,
      "step": 44095
    },
    {
      "entropy": 5.65847749710083,
      "epoch": 4.000362844702467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003481681866751121,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22942764163017274,
      "num_tokens": 91516732.0,
      "step": 44100
    },
    {
      "entropy": 5.643223667144776,
      "epoch": 4.000816400580551,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034813758813391716,
      "loss": 4.7813,
      "mean_token_accuracy": 0.23443893790245057,
      "num_tokens": 91527073.0,
      "step": 44105
    },
    {
      "entropy": 5.662564039230347,
      "epoch": 4.001269956458636,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003481069880802183,
      "loss": 4.8383,
      "mean_token_accuracy": 0.22924644351005555,
      "num_tokens": 91538314.0,
      "step": 44110
    },
    {
      "entropy": 5.690345621109008,
      "epoch": 4.00172351233672,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034807638651464815,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2279155284166336,
      "num_tokens": 91549159.0,
      "step": 44115
    },
    {
      "entropy": 5.6798787117004395,
      "epoch": 4.002177068214804,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003480457834378398,
      "loss": 4.8736,
      "mean_token_accuracy": 0.2212012991309166,
      "num_tokens": 91560901.0,
      "step": 44120
    },
    {
      "entropy": 5.693486166000366,
      "epoch": 4.002630624092888,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034801517885042597,
      "loss": 4.838,
      "mean_token_accuracy": 0.2236404076218605,
      "num_tokens": 91570797.0,
      "step": 44125
    },
    {
      "entropy": 5.800421953201294,
      "epoch": 4.003084179970973,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003479845727530397,
      "loss": 4.9124,
      "mean_token_accuracy": 0.22272665649652482,
      "num_tokens": 91580971.0,
      "step": 44130
    },
    {
      "entropy": 5.699944925308228,
      "epoch": 4.003537735849057,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034795396514631386,
      "loss": 4.8654,
      "mean_token_accuracy": 0.23142243325710296,
      "num_tokens": 91590834.0,
      "step": 44135
    },
    {
      "entropy": 5.681314373016358,
      "epoch": 4.003991291727141,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034792335603088135,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2216480478644371,
      "num_tokens": 91602099.0,
      "step": 44140
    },
    {
      "entropy": 5.639006900787353,
      "epoch": 4.004444847605225,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034789274540737535,
      "loss": 4.7974,
      "mean_token_accuracy": 0.23417962491512298,
      "num_tokens": 91612269.0,
      "step": 44145
    },
    {
      "entropy": 5.7166983127594,
      "epoch": 4.004898403483309,
      "grad_norm": 1.25,
      "learning_rate": 0.00034786213327642874,
      "loss": 4.8501,
      "mean_token_accuracy": 0.22549672573804855,
      "num_tokens": 91623644.0,
      "step": 44150
    },
    {
      "entropy": 5.7069989204406735,
      "epoch": 4.005351959361393,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003478315196386747,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23420860767364501,
      "num_tokens": 91633563.0,
      "step": 44155
    },
    {
      "entropy": 5.695914316177368,
      "epoch": 4.005805515239477,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003478009044947463,
      "loss": 4.815,
      "mean_token_accuracy": 0.23067686408758165,
      "num_tokens": 91644648.0,
      "step": 44160
    },
    {
      "entropy": 5.667443180084229,
      "epoch": 4.006259071117562,
      "grad_norm": 1.25,
      "learning_rate": 0.0003477702878452766,
      "loss": 4.6997,
      "mean_token_accuracy": 0.25245318114757537,
      "num_tokens": 91655366.0,
      "step": 44165
    },
    {
      "entropy": 5.639540910720825,
      "epoch": 4.006712626995646,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003477396696908988,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22800018787384033,
      "num_tokens": 91665474.0,
      "step": 44170
    },
    {
      "entropy": 5.677517318725586,
      "epoch": 4.00716618287373,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034770905003224606,
      "loss": 4.8511,
      "mean_token_accuracy": 0.22021735608577728,
      "num_tokens": 91675291.0,
      "step": 44175
    },
    {
      "entropy": 5.598275518417358,
      "epoch": 4.007619738751814,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003476784288699517,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23097187429666519,
      "num_tokens": 91686472.0,
      "step": 44180
    },
    {
      "entropy": 5.704984140396118,
      "epoch": 4.008073294629899,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034764780620464886,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22616575360298158,
      "num_tokens": 91696347.0,
      "step": 44185
    },
    {
      "entropy": 5.701453495025635,
      "epoch": 4.008526850507983,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003476171820369709,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2200729638338089,
      "num_tokens": 91706504.0,
      "step": 44190
    },
    {
      "entropy": 5.667878246307373,
      "epoch": 4.008980406386066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034758655636755103,
      "loss": 4.8029,
      "mean_token_accuracy": 0.23962463587522506,
      "num_tokens": 91716559.0,
      "step": 44195
    },
    {
      "entropy": 5.7046562194824215,
      "epoch": 4.009433962264151,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00034755592919702275,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2201984390616417,
      "num_tokens": 91726312.0,
      "step": 44200
    },
    {
      "entropy": 5.6582612037658695,
      "epoch": 4.009887518142235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034752530052601944,
      "loss": 4.7958,
      "mean_token_accuracy": 0.22858455181121826,
      "num_tokens": 91736882.0,
      "step": 44205
    },
    {
      "entropy": 5.705489778518677,
      "epoch": 4.010341074020319,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034749467035517437,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22234297692775726,
      "num_tokens": 91746905.0,
      "step": 44210
    },
    {
      "entropy": 5.676013994216919,
      "epoch": 4.010794629898403,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034746403868512094,
      "loss": 4.8983,
      "mean_token_accuracy": 0.22211127281188964,
      "num_tokens": 91757609.0,
      "step": 44215
    },
    {
      "entropy": 5.67447190284729,
      "epoch": 4.011248185776488,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034743340551649285,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23089018911123277,
      "num_tokens": 91768017.0,
      "step": 44220
    },
    {
      "entropy": 5.611434364318848,
      "epoch": 4.011701741654572,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034740277084992346,
      "loss": 4.7348,
      "mean_token_accuracy": 0.24121248722076416,
      "num_tokens": 91777820.0,
      "step": 44225
    },
    {
      "entropy": 5.613953018188477,
      "epoch": 4.012155297532656,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034737213468604626,
      "loss": 4.846,
      "mean_token_accuracy": 0.22106989175081254,
      "num_tokens": 91789011.0,
      "step": 44230
    },
    {
      "entropy": 5.729775238037109,
      "epoch": 4.0126088534107405,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003473414970254949,
      "loss": 4.9126,
      "mean_token_accuracy": 0.2268722102046013,
      "num_tokens": 91799165.0,
      "step": 44235
    },
    {
      "entropy": 5.683861827850341,
      "epoch": 4.013062409288825,
      "grad_norm": 1.375,
      "learning_rate": 0.00034731085786890306,
      "loss": 4.8628,
      "mean_token_accuracy": 0.2309691697359085,
      "num_tokens": 91809602.0,
      "step": 44240
    },
    {
      "entropy": 5.714441537857056,
      "epoch": 4.013515965166909,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034728021721690407,
      "loss": 4.9216,
      "mean_token_accuracy": 0.22430762201547622,
      "num_tokens": 91820092.0,
      "step": 44245
    },
    {
      "entropy": 5.690649127960205,
      "epoch": 4.013969521044992,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003472495750701318,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2304888501763344,
      "num_tokens": 91829730.0,
      "step": 44250
    },
    {
      "entropy": 5.70622673034668,
      "epoch": 4.014423076923077,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034721893142922,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22210321128368377,
      "num_tokens": 91840697.0,
      "step": 44255
    },
    {
      "entropy": 5.681812858581543,
      "epoch": 4.014876632801161,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003471882862948023,
      "loss": 4.8781,
      "mean_token_accuracy": 0.22331944704055787,
      "num_tokens": 91852145.0,
      "step": 44260
    },
    {
      "entropy": 5.695208358764648,
      "epoch": 4.015330188679245,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034715763966751233,
      "loss": 4.8344,
      "mean_token_accuracy": 0.22121974378824233,
      "num_tokens": 91861940.0,
      "step": 44265
    },
    {
      "entropy": 5.752874660491943,
      "epoch": 4.0157837445573294,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034712699154798403,
      "loss": 4.867,
      "mean_token_accuracy": 0.22496105283498763,
      "num_tokens": 91872179.0,
      "step": 44270
    },
    {
      "entropy": 5.706880331039429,
      "epoch": 4.016237300435414,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034709634193685126,
      "loss": 4.8125,
      "mean_token_accuracy": 0.2301259458065033,
      "num_tokens": 91882051.0,
      "step": 44275
    },
    {
      "entropy": 5.652614307403565,
      "epoch": 4.016690856313498,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003470656908347476,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22296436876058578,
      "num_tokens": 91892672.0,
      "step": 44280
    },
    {
      "entropy": 5.597233533859253,
      "epoch": 4.017144412191582,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003470350382423072,
      "loss": 4.7521,
      "mean_token_accuracy": 0.235336971282959,
      "num_tokens": 91903418.0,
      "step": 44285
    },
    {
      "entropy": 5.674207305908203,
      "epoch": 4.0175979680696665,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003470043841601639,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23122962117195128,
      "num_tokens": 91913076.0,
      "step": 44290
    },
    {
      "entropy": 5.667631769180298,
      "epoch": 4.018051523947751,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034697372858895147,
      "loss": 4.8298,
      "mean_token_accuracy": 0.23289192020893096,
      "num_tokens": 91923464.0,
      "step": 44295
    },
    {
      "entropy": 5.713213109970093,
      "epoch": 4.018505079825834,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000346943071529304,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2212130159139633,
      "num_tokens": 91934399.0,
      "step": 44300
    },
    {
      "entropy": 5.732719612121582,
      "epoch": 4.018958635703918,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034691241298185553,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22816334515810013,
      "num_tokens": 91945330.0,
      "step": 44305
    },
    {
      "entropy": 5.696770191192627,
      "epoch": 4.019412191582003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034688175294724004,
      "loss": 4.8862,
      "mean_token_accuracy": 0.22006458789110184,
      "num_tokens": 91956379.0,
      "step": 44310
    },
    {
      "entropy": 5.563412046432495,
      "epoch": 4.019865747460087,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034685109142609154,
      "loss": 4.7187,
      "mean_token_accuracy": 0.23325911462306975,
      "num_tokens": 91965969.0,
      "step": 44315
    },
    {
      "entropy": 5.7503965377807615,
      "epoch": 4.020319303338171,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003468204284190442,
      "loss": 4.983,
      "mean_token_accuracy": 0.21282863318920137,
      "num_tokens": 91977005.0,
      "step": 44320
    },
    {
      "entropy": 5.710402107238769,
      "epoch": 4.0207728592162555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034678976392673206,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2271944209933281,
      "num_tokens": 91988735.0,
      "step": 44325
    },
    {
      "entropy": 5.733931303024292,
      "epoch": 4.02122641509434,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00034675909794978926,
      "loss": 4.859,
      "mean_token_accuracy": 0.22972788512706757,
      "num_tokens": 91998743.0,
      "step": 44330
    },
    {
      "entropy": 5.719684314727783,
      "epoch": 4.021679970972424,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034672843048885017,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23061641603708266,
      "num_tokens": 92009465.0,
      "step": 44335
    },
    {
      "entropy": 5.619619941711425,
      "epoch": 4.022133526850508,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034669776154454875,
      "loss": 4.8174,
      "mean_token_accuracy": 0.23445021361112595,
      "num_tokens": 92020818.0,
      "step": 44340
    },
    {
      "entropy": 5.692957544326783,
      "epoch": 4.0225870827285926,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034666709111751945,
      "loss": 4.923,
      "mean_token_accuracy": 0.22366418838500976,
      "num_tokens": 92031528.0,
      "step": 44345
    },
    {
      "entropy": 5.723623895645142,
      "epoch": 4.023040638606676,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003466364192083963,
      "loss": 4.9073,
      "mean_token_accuracy": 0.22181422859430314,
      "num_tokens": 92041856.0,
      "step": 44350
    },
    {
      "entropy": 5.700039625167847,
      "epoch": 4.02349419448476,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034660574581781386,
      "loss": 4.8353,
      "mean_token_accuracy": 0.23066414445638656,
      "num_tokens": 92052117.0,
      "step": 44355
    },
    {
      "entropy": 5.724466133117676,
      "epoch": 4.0239477503628445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003465750709464063,
      "loss": 4.887,
      "mean_token_accuracy": 0.22509430348873138,
      "num_tokens": 92062405.0,
      "step": 44360
    },
    {
      "entropy": 5.718185138702393,
      "epoch": 4.024401306240929,
      "grad_norm": 1.25,
      "learning_rate": 0.000346544394594808,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22802191823720933,
      "num_tokens": 92073269.0,
      "step": 44365
    },
    {
      "entropy": 5.6868323802948,
      "epoch": 4.024854862119013,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034651371676365343,
      "loss": 4.8074,
      "mean_token_accuracy": 0.22835942208766938,
      "num_tokens": 92083645.0,
      "step": 44370
    },
    {
      "entropy": 5.806156826019287,
      "epoch": 4.025308417997097,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034648303745357705,
      "loss": 4.8716,
      "mean_token_accuracy": 0.22640362977981568,
      "num_tokens": 92093696.0,
      "step": 44375
    },
    {
      "entropy": 5.694458150863648,
      "epoch": 4.0257619738751815,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003464523566652131,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21972382813692093,
      "num_tokens": 92103217.0,
      "step": 44380
    },
    {
      "entropy": 5.621622085571289,
      "epoch": 4.026215529753266,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034642167439919634,
      "loss": 4.7898,
      "mean_token_accuracy": 0.22937562465667724,
      "num_tokens": 92112934.0,
      "step": 44385
    },
    {
      "entropy": 5.651074028015136,
      "epoch": 4.02666908563135,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003463909906561611,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2269120067358017,
      "num_tokens": 92123066.0,
      "step": 44390
    },
    {
      "entropy": 5.7112147331237795,
      "epoch": 4.027122641509434,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003463603054367419,
      "loss": 4.8407,
      "mean_token_accuracy": 0.2310007855296135,
      "num_tokens": 92133911.0,
      "step": 44395
    },
    {
      "entropy": 5.637371587753296,
      "epoch": 4.027576197387518,
      "grad_norm": 1.25,
      "learning_rate": 0.0003463296187415735,
      "loss": 4.781,
      "mean_token_accuracy": 0.22730980813503265,
      "num_tokens": 92143941.0,
      "step": 44400
    },
    {
      "entropy": 5.658279132843018,
      "epoch": 4.028029753265602,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003462989305712904,
      "loss": 4.8001,
      "mean_token_accuracy": 0.22805669903755188,
      "num_tokens": 92153934.0,
      "step": 44405
    },
    {
      "entropy": 5.631411838531494,
      "epoch": 4.028483309143686,
      "grad_norm": 1.6875,
      "learning_rate": 0.0003462682409265272,
      "loss": 4.796,
      "mean_token_accuracy": 0.23362559527158738,
      "num_tokens": 92163642.0,
      "step": 44410
    },
    {
      "entropy": 5.65312933921814,
      "epoch": 4.0289368650217705,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003462375498079186,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22468093186616897,
      "num_tokens": 92174034.0,
      "step": 44415
    },
    {
      "entropy": 5.717608213424683,
      "epoch": 4.029390420899855,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034620685721609937,
      "loss": 4.8546,
      "mean_token_accuracy": 0.2285498633980751,
      "num_tokens": 92183714.0,
      "step": 44420
    },
    {
      "entropy": 5.716265678405762,
      "epoch": 4.029843976777939,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034617616315170413,
      "loss": 4.9137,
      "mean_token_accuracy": 0.2200876295566559,
      "num_tokens": 92195291.0,
      "step": 44425
    },
    {
      "entropy": 5.696355199813842,
      "epoch": 4.030297532656023,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003461454676153678,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2268739551305771,
      "num_tokens": 92206159.0,
      "step": 44430
    },
    {
      "entropy": 5.704968500137329,
      "epoch": 4.030751088534108,
      "grad_norm": 1.28125,
      "learning_rate": 0.000346114770607725,
      "loss": 4.888,
      "mean_token_accuracy": 0.2233716920018196,
      "num_tokens": 92216928.0,
      "step": 44435
    },
    {
      "entropy": 5.636195945739746,
      "epoch": 4.031204644412192,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003460840721294106,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23043448776006697,
      "num_tokens": 92227029.0,
      "step": 44440
    },
    {
      "entropy": 5.6818702697753904,
      "epoch": 4.031658200290276,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034605337218105955,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22880153506994247,
      "num_tokens": 92237310.0,
      "step": 44445
    },
    {
      "entropy": 5.638952827453613,
      "epoch": 4.03211175616836,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003460226707633066,
      "loss": 4.8286,
      "mean_token_accuracy": 0.22679798007011415,
      "num_tokens": 92247841.0,
      "step": 44450
    },
    {
      "entropy": 5.715380096435547,
      "epoch": 4.032565312046444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034599196787678667,
      "loss": 4.8487,
      "mean_token_accuracy": 0.21976906508207322,
      "num_tokens": 92258513.0,
      "step": 44455
    },
    {
      "entropy": 5.689907550811768,
      "epoch": 4.033018867924528,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003459612635221349,
      "loss": 4.865,
      "mean_token_accuracy": 0.22422935366630553,
      "num_tokens": 92268659.0,
      "step": 44460
    },
    {
      "entropy": 5.663132238388061,
      "epoch": 4.033472423802612,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034593055769998595,
      "loss": 4.8573,
      "mean_token_accuracy": 0.22282302975654603,
      "num_tokens": 92279810.0,
      "step": 44465
    },
    {
      "entropy": 5.6584522247314455,
      "epoch": 4.0339259796806965,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003458998504109751,
      "loss": 4.8631,
      "mean_token_accuracy": 0.2285860687494278,
      "num_tokens": 92289625.0,
      "step": 44470
    },
    {
      "entropy": 5.659290170669555,
      "epoch": 4.034379535558781,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003458691416557372,
      "loss": 4.8209,
      "mean_token_accuracy": 0.2303316354751587,
      "num_tokens": 92300669.0,
      "step": 44475
    },
    {
      "entropy": 5.682563018798828,
      "epoch": 4.034833091436865,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034583843143490744,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22958121448755264,
      "num_tokens": 92312160.0,
      "step": 44480
    },
    {
      "entropy": 5.725495767593384,
      "epoch": 4.035286647314949,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003458077197491208,
      "loss": 4.9472,
      "mean_token_accuracy": 0.2178875833749771,
      "num_tokens": 92322677.0,
      "step": 44485
    },
    {
      "entropy": 5.673046207427978,
      "epoch": 4.035740203193034,
      "grad_norm": 1.25,
      "learning_rate": 0.0003457770065990126,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23379191607236863,
      "num_tokens": 92333148.0,
      "step": 44490
    },
    {
      "entropy": 5.725253677368164,
      "epoch": 4.036193759071118,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003457462919852178,
      "loss": 4.892,
      "mean_token_accuracy": 0.22464536428451537,
      "num_tokens": 92344189.0,
      "step": 44495
    },
    {
      "entropy": 5.745694637298584,
      "epoch": 4.036647314949202,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003457155759083717,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2316122680902481,
      "num_tokens": 92356136.0,
      "step": 44500
    },
    {
      "entropy": 5.66837100982666,
      "epoch": 4.0371008708272855,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003456848583691094,
      "loss": 4.7472,
      "mean_token_accuracy": 0.2287842944264412,
      "num_tokens": 92366393.0,
      "step": 44505
    },
    {
      "entropy": 5.6672460556030275,
      "epoch": 4.03755442670537,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003456541393680662,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2320243552327156,
      "num_tokens": 92376819.0,
      "step": 44510
    },
    {
      "entropy": 5.6751988410949705,
      "epoch": 4.038007982583454,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034562341890587747,
      "loss": 4.885,
      "mean_token_accuracy": 0.2208233430981636,
      "num_tokens": 92386704.0,
      "step": 44515
    },
    {
      "entropy": 5.681497526168823,
      "epoch": 4.038461538461538,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003455926969831784,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22001320719718934,
      "num_tokens": 92397412.0,
      "step": 44520
    },
    {
      "entropy": 5.771698236465454,
      "epoch": 4.038915094339623,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034556197360060436,
      "loss": 4.9402,
      "mean_token_accuracy": 0.2198114961385727,
      "num_tokens": 92408641.0,
      "step": 44525
    },
    {
      "entropy": 5.670127201080322,
      "epoch": 4.039368650217707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003455312487587908,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2263902097940445,
      "num_tokens": 92419341.0,
      "step": 44530
    },
    {
      "entropy": 5.692534685134888,
      "epoch": 4.039822206095791,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003455005224583729,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22528087645769118,
      "num_tokens": 92429297.0,
      "step": 44535
    },
    {
      "entropy": 5.75702977180481,
      "epoch": 4.040275761973875,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034546979469998637,
      "loss": 4.9279,
      "mean_token_accuracy": 0.22372381687164306,
      "num_tokens": 92439851.0,
      "step": 44540
    },
    {
      "entropy": 5.623942422866821,
      "epoch": 4.04072931785196,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034543906548426646,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23412166088819503,
      "num_tokens": 92451096.0,
      "step": 44545
    },
    {
      "entropy": 5.686883401870728,
      "epoch": 4.041182873730044,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034540833481184867,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22943923324346543,
      "num_tokens": 92460516.0,
      "step": 44550
    },
    {
      "entropy": 5.756282997131348,
      "epoch": 4.041636429608127,
      "grad_norm": 1.125,
      "learning_rate": 0.00034537760268336866,
      "loss": 4.9096,
      "mean_token_accuracy": 0.21721821427345275,
      "num_tokens": 92471429.0,
      "step": 44555
    },
    {
      "entropy": 5.719521427154541,
      "epoch": 4.042089985486212,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003453468690994619,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22551417499780654,
      "num_tokens": 92482047.0,
      "step": 44560
    },
    {
      "entropy": 5.643525218963623,
      "epoch": 4.042543541364296,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034531613406076387,
      "loss": 4.8063,
      "mean_token_accuracy": 0.2332908406853676,
      "num_tokens": 92492206.0,
      "step": 44565
    },
    {
      "entropy": 5.646417236328125,
      "epoch": 4.04299709724238,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003452853975679103,
      "loss": 4.7721,
      "mean_token_accuracy": 0.23105165362358093,
      "num_tokens": 92501979.0,
      "step": 44570
    },
    {
      "entropy": 5.822420501708985,
      "epoch": 4.043450653120464,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034525465962153683,
      "loss": 4.9351,
      "mean_token_accuracy": 0.2212986946105957,
      "num_tokens": 92512388.0,
      "step": 44575
    },
    {
      "entropy": 5.763976764678955,
      "epoch": 4.043904208998549,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034522392022227915,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2223334327340126,
      "num_tokens": 92522805.0,
      "step": 44580
    },
    {
      "entropy": 5.655717849731445,
      "epoch": 4.044357764876633,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034519317937077283,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22229886353015899,
      "num_tokens": 92532957.0,
      "step": 44585
    },
    {
      "entropy": 5.675611114501953,
      "epoch": 4.044811320754717,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003451624370676537,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22181422114372254,
      "num_tokens": 92543451.0,
      "step": 44590
    },
    {
      "entropy": 5.773923015594482,
      "epoch": 4.045264876632801,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034513169331355754,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2234429508447647,
      "num_tokens": 92553856.0,
      "step": 44595
    },
    {
      "entropy": 5.655421924591065,
      "epoch": 4.045718432510886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034510094810911995,
      "loss": 4.7871,
      "mean_token_accuracy": 0.23022069185972213,
      "num_tokens": 92563816.0,
      "step": 44600
    },
    {
      "entropy": 5.660015535354614,
      "epoch": 4.04617198838897,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034507020145497705,
      "loss": 4.8728,
      "mean_token_accuracy": 0.23081972897052766,
      "num_tokens": 92574062.0,
      "step": 44605
    },
    {
      "entropy": 5.629234266281128,
      "epoch": 4.046625544267053,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034503945335176444,
      "loss": 4.8057,
      "mean_token_accuracy": 0.23437357544898987,
      "num_tokens": 92583935.0,
      "step": 44610
    },
    {
      "entropy": 5.744720983505249,
      "epoch": 4.047079100145138,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034500870380011807,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2270135834813118,
      "num_tokens": 92594037.0,
      "step": 44615
    },
    {
      "entropy": 5.692263746261597,
      "epoch": 4.047532656023222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034497795280067386,
      "loss": 4.8359,
      "mean_token_accuracy": 0.22331465035676956,
      "num_tokens": 92603336.0,
      "step": 44620
    },
    {
      "entropy": 5.624852895736694,
      "epoch": 4.047986211901306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034494720035406777,
      "loss": 4.7687,
      "mean_token_accuracy": 0.23151130676269532,
      "num_tokens": 92614210.0,
      "step": 44625
    },
    {
      "entropy": 5.72175440788269,
      "epoch": 4.04843976777939,
      "grad_norm": 1.25,
      "learning_rate": 0.0003449164464609358,
      "loss": 4.8798,
      "mean_token_accuracy": 0.22379563748836517,
      "num_tokens": 92624036.0,
      "step": 44630
    },
    {
      "entropy": 5.696424150466919,
      "epoch": 4.048893323657475,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003448856911219139,
      "loss": 4.9255,
      "mean_token_accuracy": 0.2186195284128189,
      "num_tokens": 92634588.0,
      "step": 44635
    },
    {
      "entropy": 5.645076036453247,
      "epoch": 4.049346879535559,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034485493433763804,
      "loss": 4.8099,
      "mean_token_accuracy": 0.2305596426129341,
      "num_tokens": 92645027.0,
      "step": 44640
    },
    {
      "entropy": 5.807842540740967,
      "epoch": 4.049800435413643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034482417610874437,
      "loss": 4.901,
      "mean_token_accuracy": 0.22165869027376175,
      "num_tokens": 92656669.0,
      "step": 44645
    },
    {
      "entropy": 5.709304761886597,
      "epoch": 4.0502539912917275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003447934164358689,
      "loss": 4.9436,
      "mean_token_accuracy": 0.2168800413608551,
      "num_tokens": 92666558.0,
      "step": 44650
    },
    {
      "entropy": 5.6990845680236815,
      "epoch": 4.050707547169812,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034476265531964786,
      "loss": 4.877,
      "mean_token_accuracy": 0.23064761310815812,
      "num_tokens": 92677384.0,
      "step": 44655
    },
    {
      "entropy": 5.659431409835816,
      "epoch": 4.051161103047895,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003447318927607173,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23241167068481444,
      "num_tokens": 92687619.0,
      "step": 44660
    },
    {
      "entropy": 5.628617668151856,
      "epoch": 4.051614658925979,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003447011287597134,
      "loss": 4.7376,
      "mean_token_accuracy": 0.23461686968803405,
      "num_tokens": 92697784.0,
      "step": 44665
    },
    {
      "entropy": 5.697090101242066,
      "epoch": 4.052068214804064,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003446703633172724,
      "loss": 4.9261,
      "mean_token_accuracy": 0.21679323315620422,
      "num_tokens": 92709255.0,
      "step": 44670
    },
    {
      "entropy": 5.644973707199097,
      "epoch": 4.052521770682148,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003446395964340305,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23222300559282302,
      "num_tokens": 92720714.0,
      "step": 44675
    },
    {
      "entropy": 5.691215753555298,
      "epoch": 4.052975326560232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034460882811062417,
      "loss": 4.8272,
      "mean_token_accuracy": 0.22296714037656784,
      "num_tokens": 92730783.0,
      "step": 44680
    },
    {
      "entropy": 5.751193618774414,
      "epoch": 4.0534288824383164,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034457805834768937,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22103992551565171,
      "num_tokens": 92741943.0,
      "step": 44685
    },
    {
      "entropy": 5.661552476882934,
      "epoch": 4.053882438316401,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003445472871458627,
      "loss": 4.8535,
      "mean_token_accuracy": 0.23114657253026963,
      "num_tokens": 92753510.0,
      "step": 44690
    },
    {
      "entropy": 5.650116205215454,
      "epoch": 4.054335994194485,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003445165145057803,
      "loss": 4.8025,
      "mean_token_accuracy": 0.23432120829820632,
      "num_tokens": 92763655.0,
      "step": 44695
    },
    {
      "entropy": 5.711693906784058,
      "epoch": 4.054789550072569,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034448574042807866,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22356671541929246,
      "num_tokens": 92773732.0,
      "step": 44700
    },
    {
      "entropy": 5.737141370773315,
      "epoch": 4.0552431059506535,
      "grad_norm": 1.25,
      "learning_rate": 0.0003444549649133943,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22262146472930908,
      "num_tokens": 92784822.0,
      "step": 44705
    },
    {
      "entropy": 5.760669708251953,
      "epoch": 4.055696661828737,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003444241879623636,
      "loss": 4.9545,
      "mean_token_accuracy": 0.21099607646465302,
      "num_tokens": 92795270.0,
      "step": 44710
    },
    {
      "entropy": 5.730932283401489,
      "epoch": 4.056150217706821,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003443934095756229,
      "loss": 4.8442,
      "mean_token_accuracy": 0.22749514132738113,
      "num_tokens": 92806578.0,
      "step": 44715
    },
    {
      "entropy": 5.721300554275513,
      "epoch": 4.056603773584905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003443626297538088,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22396479696035385,
      "num_tokens": 92815932.0,
      "step": 44720
    },
    {
      "entropy": 5.671530771255493,
      "epoch": 4.05705732946299,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003443318484975579,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23082755953073503,
      "num_tokens": 92825536.0,
      "step": 44725
    },
    {
      "entropy": 5.669141149520874,
      "epoch": 4.057510885341074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003443010658075066,
      "loss": 4.8147,
      "mean_token_accuracy": 0.22423868626356125,
      "num_tokens": 92836451.0,
      "step": 44730
    },
    {
      "entropy": 5.711064672470092,
      "epoch": 4.057964441219158,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003442702816842917,
      "loss": 4.9487,
      "mean_token_accuracy": 0.22057753652334214,
      "num_tokens": 92846923.0,
      "step": 44735
    },
    {
      "entropy": 5.675275897979736,
      "epoch": 4.0584179970972425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034423949612854974,
      "loss": 4.7925,
      "mean_token_accuracy": 0.2319028601050377,
      "num_tokens": 92857549.0,
      "step": 44740
    },
    {
      "entropy": 5.74143533706665,
      "epoch": 4.058871552975327,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003442087091409173,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22399177700281142,
      "num_tokens": 92867059.0,
      "step": 44745
    },
    {
      "entropy": 5.701948118209839,
      "epoch": 4.059325108853411,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003441779207220311,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22026100754737854,
      "num_tokens": 92878138.0,
      "step": 44750
    },
    {
      "entropy": 5.686127662658691,
      "epoch": 4.059778664731495,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003441471308725279,
      "loss": 4.8719,
      "mean_token_accuracy": 0.22804326117038726,
      "num_tokens": 92888931.0,
      "step": 44755
    },
    {
      "entropy": 5.720755672454834,
      "epoch": 4.060232220609579,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034411633959304444,
      "loss": 4.9238,
      "mean_token_accuracy": 0.21542762070894242,
      "num_tokens": 92899718.0,
      "step": 44760
    },
    {
      "entropy": 5.702804374694824,
      "epoch": 4.060685776487663,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034408554688421743,
      "loss": 4.8263,
      "mean_token_accuracy": 0.23294223099946976,
      "num_tokens": 92910381.0,
      "step": 44765
    },
    {
      "entropy": 5.650000524520874,
      "epoch": 4.061139332365747,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003440547527466837,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23375887274742127,
      "num_tokens": 92919905.0,
      "step": 44770
    },
    {
      "entropy": 5.636110401153564,
      "epoch": 4.0615928882438315,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034402395718108016,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23154082298278808,
      "num_tokens": 92930210.0,
      "step": 44775
    },
    {
      "entropy": 5.694787931442261,
      "epoch": 4.062046444121916,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034399316018804356,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22769869416952132,
      "num_tokens": 92940832.0,
      "step": 44780
    },
    {
      "entropy": 5.775491619110108,
      "epoch": 4.0625,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003439623617682107,
      "loss": 4.9563,
      "mean_token_accuracy": 0.22008833289146423,
      "num_tokens": 92951349.0,
      "step": 44785
    },
    {
      "entropy": 5.694003295898438,
      "epoch": 4.062953555878084,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003439315619222188,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22863494753837585,
      "num_tokens": 92961642.0,
      "step": 44790
    },
    {
      "entropy": 5.672139644622803,
      "epoch": 4.0634071117561685,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003439007606507046,
      "loss": 4.8779,
      "mean_token_accuracy": 0.2302296668291092,
      "num_tokens": 92971061.0,
      "step": 44795
    },
    {
      "entropy": 5.753585338592529,
      "epoch": 4.063860667634253,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003438699579543051,
      "loss": 4.9518,
      "mean_token_accuracy": 0.219168458878994,
      "num_tokens": 92981732.0,
      "step": 44800
    },
    {
      "entropy": 5.7241175174713135,
      "epoch": 4.064314223512337,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034383915383365726,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22117528915405274,
      "num_tokens": 92991570.0,
      "step": 44805
    },
    {
      "entropy": 5.699480438232422,
      "epoch": 4.064767779390421,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003438083482893982,
      "loss": 4.8617,
      "mean_token_accuracy": 0.22717937380075454,
      "num_tokens": 93001623.0,
      "step": 44810
    },
    {
      "entropy": 5.725891256332398,
      "epoch": 4.065221335268505,
      "grad_norm": 1.34375,
      "learning_rate": 0.00034377754132216496,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2259368821978569,
      "num_tokens": 93012167.0,
      "step": 44815
    },
    {
      "entropy": 5.663627195358276,
      "epoch": 4.065674891146589,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003437467329325947,
      "loss": 4.8028,
      "mean_token_accuracy": 0.23507469445466994,
      "num_tokens": 93022090.0,
      "step": 44820
    },
    {
      "entropy": 5.659423303604126,
      "epoch": 4.066128447024673,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003437159231213244,
      "loss": 4.7306,
      "mean_token_accuracy": 0.23610540181398393,
      "num_tokens": 93030917.0,
      "step": 44825
    },
    {
      "entropy": 5.665352392196655,
      "epoch": 4.0665820029027575,
      "grad_norm": 1.375,
      "learning_rate": 0.0003436851118889914,
      "loss": 4.8239,
      "mean_token_accuracy": 0.22218020856380463,
      "num_tokens": 93040113.0,
      "step": 44830
    },
    {
      "entropy": 5.705097484588623,
      "epoch": 4.067035558780842,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003436542992362326,
      "loss": 4.8656,
      "mean_token_accuracy": 0.2272750288248062,
      "num_tokens": 93050053.0,
      "step": 44835
    },
    {
      "entropy": 5.764237785339356,
      "epoch": 4.067489114658926,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003436234851636856,
      "loss": 4.9208,
      "mean_token_accuracy": 0.21356096416711806,
      "num_tokens": 93060581.0,
      "step": 44840
    },
    {
      "entropy": 5.758456563949585,
      "epoch": 4.06794267053701,
      "grad_norm": 1.4375,
      "learning_rate": 0.00034359266967198734,
      "loss": 4.9164,
      "mean_token_accuracy": 0.22472504228353501,
      "num_tokens": 93070158.0,
      "step": 44845
    },
    {
      "entropy": 5.624843788146973,
      "epoch": 4.068396226415095,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003435618527617752,
      "loss": 4.8084,
      "mean_token_accuracy": 0.23151520043611526,
      "num_tokens": 93080804.0,
      "step": 44850
    },
    {
      "entropy": 5.710074090957642,
      "epoch": 4.068849782293179,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003435310344336865,
      "loss": 4.8454,
      "mean_token_accuracy": 0.22514020055532455,
      "num_tokens": 93091346.0,
      "step": 44855
    },
    {
      "entropy": 5.654680061340332,
      "epoch": 4.069303338171263,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003435002146883585,
      "loss": 4.8252,
      "mean_token_accuracy": 0.2282971516251564,
      "num_tokens": 93100931.0,
      "step": 44860
    },
    {
      "entropy": 5.679315042495728,
      "epoch": 4.0697568940493465,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003434693935264286,
      "loss": 4.8331,
      "mean_token_accuracy": 0.2334408476948738,
      "num_tokens": 93110181.0,
      "step": 44865
    },
    {
      "entropy": 5.723607206344605,
      "epoch": 4.070210449927431,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003434385709485343,
      "loss": 4.8279,
      "mean_token_accuracy": 0.23044648468494416,
      "num_tokens": 93120140.0,
      "step": 44870
    },
    {
      "entropy": 5.753378105163574,
      "epoch": 4.070664005805515,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003434077469553128,
      "loss": 4.9079,
      "mean_token_accuracy": 0.21870438903570175,
      "num_tokens": 93130853.0,
      "step": 44875
    },
    {
      "entropy": 5.688275575637817,
      "epoch": 4.071117561683599,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003433769215474017,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2291127324104309,
      "num_tokens": 93141430.0,
      "step": 44880
    },
    {
      "entropy": 5.768914747238159,
      "epoch": 4.0715711175616836,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034334609472543845,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21876697689294816,
      "num_tokens": 93152270.0,
      "step": 44885
    },
    {
      "entropy": 5.745680618286133,
      "epoch": 4.072024673439768,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003433152664900605,
      "loss": 4.9177,
      "mean_token_accuracy": 0.21661687940359114,
      "num_tokens": 93162850.0,
      "step": 44890
    },
    {
      "entropy": 5.6924022197723385,
      "epoch": 4.072478229317852,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003432844368419055,
      "loss": 4.7475,
      "mean_token_accuracy": 0.236563578248024,
      "num_tokens": 93172265.0,
      "step": 44895
    },
    {
      "entropy": 5.655765438079834,
      "epoch": 4.072931785195936,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003432536057816108,
      "loss": 4.8051,
      "mean_token_accuracy": 0.23508071899414062,
      "num_tokens": 93182756.0,
      "step": 44900
    },
    {
      "entropy": 5.671857690811157,
      "epoch": 4.073385341074021,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034322277330981426,
      "loss": 4.8735,
      "mean_token_accuracy": 0.21983840614557265,
      "num_tokens": 93192710.0,
      "step": 44905
    },
    {
      "entropy": 5.6819655418396,
      "epoch": 4.073838896952105,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034319193942715343,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2326256439089775,
      "num_tokens": 93202531.0,
      "step": 44910
    },
    {
      "entropy": 5.742272090911865,
      "epoch": 4.074292452830188,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034316110413426586,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22216078042984008,
      "num_tokens": 93212208.0,
      "step": 44915
    },
    {
      "entropy": 5.782126092910767,
      "epoch": 4.0747460087082725,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003431302674317892,
      "loss": 4.9271,
      "mean_token_accuracy": 0.22087455540895462,
      "num_tokens": 93222932.0,
      "step": 44920
    },
    {
      "entropy": 5.717576742172241,
      "epoch": 4.075199564586357,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034309942932036136,
      "loss": 4.7659,
      "mean_token_accuracy": 0.2357520654797554,
      "num_tokens": 93232049.0,
      "step": 44925
    },
    {
      "entropy": 5.6888158321380615,
      "epoch": 4.075653120464441,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034306858980061994,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22663516253232957,
      "num_tokens": 93244230.0,
      "step": 44930
    },
    {
      "entropy": 5.678946495056152,
      "epoch": 4.076106676342525,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034303774887320274,
      "loss": 4.87,
      "mean_token_accuracy": 0.2333603322505951,
      "num_tokens": 93254159.0,
      "step": 44935
    },
    {
      "entropy": 5.695543670654297,
      "epoch": 4.07656023222061,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034300690653874753,
      "loss": 4.7876,
      "mean_token_accuracy": 0.23109705299139022,
      "num_tokens": 93263742.0,
      "step": 44940
    },
    {
      "entropy": 5.69886736869812,
      "epoch": 4.077013788098694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034297606279789215,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21355538070201874,
      "num_tokens": 93274446.0,
      "step": 44945
    },
    {
      "entropy": 5.703141403198242,
      "epoch": 4.077467343976778,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003429452176512745,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2192397892475128,
      "num_tokens": 93284102.0,
      "step": 44950
    },
    {
      "entropy": 5.696621656417847,
      "epoch": 4.077920899854862,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034291437109953244,
      "loss": 4.8054,
      "mean_token_accuracy": 0.23829540163278579,
      "num_tokens": 93295096.0,
      "step": 44955
    },
    {
      "entropy": 5.67803635597229,
      "epoch": 4.078374455732947,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003428835231433039,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22570423036813736,
      "num_tokens": 93305377.0,
      "step": 44960
    },
    {
      "entropy": 5.787143135070801,
      "epoch": 4.07882801161103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034285267378322674,
      "loss": 4.932,
      "mean_token_accuracy": 0.22447891533374786,
      "num_tokens": 93315859.0,
      "step": 44965
    },
    {
      "entropy": 5.702143383026123,
      "epoch": 4.079281567489114,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00034282182301993903,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22847175896167754,
      "num_tokens": 93326519.0,
      "step": 44970
    },
    {
      "entropy": 5.714928102493286,
      "epoch": 4.079735123367199,
      "grad_norm": 1.25,
      "learning_rate": 0.0003427909708540787,
      "loss": 4.8377,
      "mean_token_accuracy": 0.22240885347127914,
      "num_tokens": 93337055.0,
      "step": 44975
    },
    {
      "entropy": 5.7040696144104,
      "epoch": 4.080188679245283,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034276011728628393,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22343938499689103,
      "num_tokens": 93346565.0,
      "step": 44980
    },
    {
      "entropy": 5.712832307815551,
      "epoch": 4.080642235123367,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00034272926231719255,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2302722230553627,
      "num_tokens": 93356366.0,
      "step": 44985
    },
    {
      "entropy": 5.7517012596130375,
      "epoch": 4.081095791001451,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003426984059474428,
      "loss": 4.8662,
      "mean_token_accuracy": 0.21985929310321808,
      "num_tokens": 93367317.0,
      "step": 44990
    },
    {
      "entropy": 5.698300313949585,
      "epoch": 4.081549346879536,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003426675481776728,
      "loss": 4.878,
      "mean_token_accuracy": 0.2266569748520851,
      "num_tokens": 93379601.0,
      "step": 44995
    },
    {
      "entropy": 5.698724269866943,
      "epoch": 4.08200290275762,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034263668900852065,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23339014500379562,
      "num_tokens": 93389740.0,
      "step": 45000
    },
    {
      "epoch": 4.08200290275762,
      "eval_entropy": 5.436114283600544,
      "eval_loss": 4.968533992767334,
      "eval_mean_token_accuracy": 0.22858855953161938,
      "eval_num_tokens": 93389740.0,
      "eval_runtime": 20.5485,
      "eval_samples_per_second": 1720.81,
      "eval_steps_per_second": 215.101,
      "step": 45000
    },
    {
      "entropy": 5.780312728881836,
      "epoch": 4.082456458635704,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003426058284406244,
      "loss": 4.9789,
      "mean_token_accuracy": 0.22023508697748184,
      "num_tokens": 93400232.0,
      "step": 45005
    },
    {
      "entropy": 5.759642887115478,
      "epoch": 4.082910014513788,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003425749664746226,
      "loss": 4.9048,
      "mean_token_accuracy": 0.22537297308444976,
      "num_tokens": 93410662.0,
      "step": 45010
    },
    {
      "entropy": 5.695563459396363,
      "epoch": 4.083363570391873,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003425441031111532,
      "loss": 4.7499,
      "mean_token_accuracy": 0.23547904938459396,
      "num_tokens": 93420328.0,
      "step": 45015
    },
    {
      "entropy": 5.732776117324829,
      "epoch": 4.083817126269956,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034251323835085453,
      "loss": 4.9126,
      "mean_token_accuracy": 0.21588945090770723,
      "num_tokens": 93430878.0,
      "step": 45020
    },
    {
      "entropy": 5.791356706619263,
      "epoch": 4.08427068214804,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003424823721943649,
      "loss": 4.9837,
      "mean_token_accuracy": 0.21977088004350662,
      "num_tokens": 93441784.0,
      "step": 45025
    },
    {
      "entropy": 5.687831115722656,
      "epoch": 4.084724238026125,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003424515046423226,
      "loss": 4.8952,
      "mean_token_accuracy": 0.22451057583093642,
      "num_tokens": 93453058.0,
      "step": 45030
    },
    {
      "entropy": 5.718330574035645,
      "epoch": 4.085177793904209,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003424206356953659,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22315017133951187,
      "num_tokens": 93463089.0,
      "step": 45035
    },
    {
      "entropy": 5.747386884689331,
      "epoch": 4.085631349782293,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034238976535413333,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22693369537591934,
      "num_tokens": 93473315.0,
      "step": 45040
    },
    {
      "entropy": 5.817064905166626,
      "epoch": 4.086084905660377,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003423588936192633,
      "loss": 4.9722,
      "mean_token_accuracy": 0.21400206387043,
      "num_tokens": 93485113.0,
      "step": 45045
    },
    {
      "entropy": 5.678099775314331,
      "epoch": 4.086538461538462,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003423280204913941,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23074843287467955,
      "num_tokens": 93494269.0,
      "step": 45050
    },
    {
      "entropy": 5.610371780395508,
      "epoch": 4.086992017416546,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003422971459711644,
      "loss": 4.8016,
      "mean_token_accuracy": 0.23330307602882386,
      "num_tokens": 93505052.0,
      "step": 45055
    },
    {
      "entropy": 5.697642850875854,
      "epoch": 4.08744557329463,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034226627005921244,
      "loss": 4.8513,
      "mean_token_accuracy": 0.22637408822774888,
      "num_tokens": 93514986.0,
      "step": 45060
    },
    {
      "entropy": 5.728595733642578,
      "epoch": 4.0878991291727145,
      "grad_norm": 1.34375,
      "learning_rate": 0.000342235392756177,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22495073229074478,
      "num_tokens": 93525286.0,
      "step": 45065
    },
    {
      "entropy": 5.763792181015015,
      "epoch": 4.088352685050798,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034220451406269636,
      "loss": 4.9093,
      "mean_token_accuracy": 0.22106239944696426,
      "num_tokens": 93536034.0,
      "step": 45070
    },
    {
      "entropy": 5.648965120315552,
      "epoch": 4.088806240928882,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00034217363397940937,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22953555434942247,
      "num_tokens": 93546283.0,
      "step": 45075
    },
    {
      "entropy": 5.658751392364502,
      "epoch": 4.089259796806966,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034214275250695446,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2243357256054878,
      "num_tokens": 93555811.0,
      "step": 45080
    },
    {
      "entropy": 5.742631292343139,
      "epoch": 4.089713352685051,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003421118696459703,
      "loss": 4.9346,
      "mean_token_accuracy": 0.22053607255220414,
      "num_tokens": 93567154.0,
      "step": 45085
    },
    {
      "entropy": 5.692592811584473,
      "epoch": 4.090166908563135,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034208098539709566,
      "loss": 4.823,
      "mean_token_accuracy": 0.22961331754922867,
      "num_tokens": 93577654.0,
      "step": 45090
    },
    {
      "entropy": 5.681293630599976,
      "epoch": 4.090620464441219,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034205009976096906,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22204892635345458,
      "num_tokens": 93588712.0,
      "step": 45095
    },
    {
      "entropy": 5.701315212249756,
      "epoch": 4.0910740203193035,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003420192127382293,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22460637092590333,
      "num_tokens": 93599186.0,
      "step": 45100
    },
    {
      "entropy": 5.696014642715454,
      "epoch": 4.091527576197388,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034198832432951517,
      "loss": 4.7756,
      "mean_token_accuracy": 0.2382916823029518,
      "num_tokens": 93609439.0,
      "step": 45105
    },
    {
      "entropy": 5.710198450088501,
      "epoch": 4.091981132075472,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003419574345354654,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22245459258556366,
      "num_tokens": 93619903.0,
      "step": 45110
    },
    {
      "entropy": 5.682673025131225,
      "epoch": 4.092434687953556,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034192654335671893,
      "loss": 4.8102,
      "mean_token_accuracy": 0.23425752967596053,
      "num_tokens": 93629895.0,
      "step": 45115
    },
    {
      "entropy": 5.710788202285767,
      "epoch": 4.09288824383164,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034189565079391436,
      "loss": 4.9386,
      "mean_token_accuracy": 0.2164367377758026,
      "num_tokens": 93640677.0,
      "step": 45120
    },
    {
      "entropy": 5.730302524566651,
      "epoch": 4.093341799709724,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003418647568476907,
      "loss": 4.9499,
      "mean_token_accuracy": 0.22796961218118666,
      "num_tokens": 93651541.0,
      "step": 45125
    },
    {
      "entropy": 5.739527988433838,
      "epoch": 4.093795355587808,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003418338615186869,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2223595827817917,
      "num_tokens": 93661901.0,
      "step": 45130
    },
    {
      "entropy": 5.721985101699829,
      "epoch": 4.094248911465892,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034180296480754174,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22661933451890945,
      "num_tokens": 93672307.0,
      "step": 45135
    },
    {
      "entropy": 5.701646327972412,
      "epoch": 4.094702467343977,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003417720667148943,
      "loss": 4.8233,
      "mean_token_accuracy": 0.21985965520143508,
      "num_tokens": 93683802.0,
      "step": 45140
    },
    {
      "entropy": 5.756990385055542,
      "epoch": 4.095156023222061,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003417411672413835,
      "loss": 4.9352,
      "mean_token_accuracy": 0.2118327498435974,
      "num_tokens": 93693467.0,
      "step": 45145
    },
    {
      "entropy": 5.706813764572144,
      "epoch": 4.095609579100145,
      "grad_norm": 1.390625,
      "learning_rate": 0.00034171026638764833,
      "loss": 4.8492,
      "mean_token_accuracy": 0.23244535624980928,
      "num_tokens": 93703858.0,
      "step": 45150
    },
    {
      "entropy": 5.69429292678833,
      "epoch": 4.0960631349782295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034167936415432785,
      "loss": 4.8539,
      "mean_token_accuracy": 0.2317863404750824,
      "num_tokens": 93714457.0,
      "step": 45155
    },
    {
      "entropy": 5.705235862731934,
      "epoch": 4.096516690856314,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003416484605420611,
      "loss": 4.8819,
      "mean_token_accuracy": 0.23097188919782638,
      "num_tokens": 93725009.0,
      "step": 45160
    },
    {
      "entropy": 5.768914842605591,
      "epoch": 4.096970246734398,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034161755555148725,
      "loss": 4.9088,
      "mean_token_accuracy": 0.2175307974219322,
      "num_tokens": 93736357.0,
      "step": 45165
    },
    {
      "entropy": 5.686219120025635,
      "epoch": 4.097423802612482,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034158664918324536,
      "loss": 4.8157,
      "mean_token_accuracy": 0.2259742334485054,
      "num_tokens": 93746711.0,
      "step": 45170
    },
    {
      "entropy": 5.728860521316529,
      "epoch": 4.097877358490566,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034155574143797455,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22190282940864564,
      "num_tokens": 93757732.0,
      "step": 45175
    },
    {
      "entropy": 5.70110125541687,
      "epoch": 4.09833091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00034152483231631413,
      "loss": 4.8368,
      "mean_token_accuracy": 0.2282794490456581,
      "num_tokens": 93767048.0,
      "step": 45180
    },
    {
      "entropy": 5.84728970527649,
      "epoch": 4.098784470246734,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003414939218189032,
      "loss": 5.0552,
      "mean_token_accuracy": 0.21597282588481903,
      "num_tokens": 93778283.0,
      "step": 45185
    },
    {
      "entropy": 5.769228935241699,
      "epoch": 4.0992380261248185,
      "grad_norm": 1.28125,
      "learning_rate": 0.000341463009946381,
      "loss": 4.9635,
      "mean_token_accuracy": 0.21797900795936584,
      "num_tokens": 93788776.0,
      "step": 45190
    },
    {
      "entropy": 5.748241758346557,
      "epoch": 4.099691582002903,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034143209669938685,
      "loss": 4.8789,
      "mean_token_accuracy": 0.21802350133657455,
      "num_tokens": 93798695.0,
      "step": 45195
    },
    {
      "entropy": 5.793102169036866,
      "epoch": 4.100145137880987,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034140118207856,
      "loss": 4.975,
      "mean_token_accuracy": 0.22397787123918533,
      "num_tokens": 93808993.0,
      "step": 45200
    },
    {
      "entropy": 5.792188024520874,
      "epoch": 4.100598693759071,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034137026608453973,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21846818923950195,
      "num_tokens": 93818361.0,
      "step": 45205
    },
    {
      "entropy": 5.798019361495972,
      "epoch": 4.1010522496371555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003413393487179655,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2342015877366066,
      "num_tokens": 93828988.0,
      "step": 45210
    },
    {
      "entropy": 5.812813329696655,
      "epoch": 4.10150580551524,
      "grad_norm": 1.25,
      "learning_rate": 0.0003413084299794767,
      "loss": 4.9456,
      "mean_token_accuracy": 0.2176952600479126,
      "num_tokens": 93839350.0,
      "step": 45215
    },
    {
      "entropy": 5.728455400466919,
      "epoch": 4.101959361393324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034127750986971255,
      "loss": 4.83,
      "mean_token_accuracy": 0.23052688986063002,
      "num_tokens": 93849939.0,
      "step": 45220
    },
    {
      "entropy": 5.7156750679016115,
      "epoch": 4.1024129172714074,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003412465883893126,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22105774879455567,
      "num_tokens": 93861278.0,
      "step": 45225
    },
    {
      "entropy": 5.760678672790528,
      "epoch": 4.102866473149492,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003412156655389164,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22260782718658448,
      "num_tokens": 93871693.0,
      "step": 45230
    },
    {
      "entropy": 5.725436353683472,
      "epoch": 4.103320029027576,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003411847413191633,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22489041090011597,
      "num_tokens": 93883098.0,
      "step": 45235
    },
    {
      "entropy": 5.675696992874146,
      "epoch": 4.10377358490566,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034115381573069297,
      "loss": 4.7956,
      "mean_token_accuracy": 0.2291572645306587,
      "num_tokens": 93893347.0,
      "step": 45240
    },
    {
      "entropy": 5.743026876449585,
      "epoch": 4.1042271407837445,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034112288877414485,
      "loss": 4.9347,
      "mean_token_accuracy": 0.21608586758375167,
      "num_tokens": 93903964.0,
      "step": 45245
    },
    {
      "entropy": 5.728967189788818,
      "epoch": 4.104680696661829,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003410919604501585,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22395798116922377,
      "num_tokens": 93914526.0,
      "step": 45250
    },
    {
      "entropy": 5.782363033294677,
      "epoch": 4.105134252539913,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003410610307593735,
      "loss": 4.9704,
      "mean_token_accuracy": 0.22771930694580078,
      "num_tokens": 93924430.0,
      "step": 45255
    },
    {
      "entropy": 5.74453444480896,
      "epoch": 4.105587808417997,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003410300997024297,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22877823412418366,
      "num_tokens": 93935150.0,
      "step": 45260
    },
    {
      "entropy": 5.662942123413086,
      "epoch": 4.106041364296082,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003409991672799665,
      "loss": 4.8108,
      "mean_token_accuracy": 0.23596953302621843,
      "num_tokens": 93944496.0,
      "step": 45265
    },
    {
      "entropy": 5.730890846252441,
      "epoch": 4.106494920174166,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034096823349262376,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22035014778375625,
      "num_tokens": 93955745.0,
      "step": 45270
    },
    {
      "entropy": 5.658197546005249,
      "epoch": 4.106948476052249,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003409372983410411,
      "loss": 4.8022,
      "mean_token_accuracy": 0.22870007902383804,
      "num_tokens": 93965930.0,
      "step": 45275
    },
    {
      "entropy": 5.699591159820557,
      "epoch": 4.1074020319303335,
      "grad_norm": 1.375,
      "learning_rate": 0.0003409063618258583,
      "loss": 4.7913,
      "mean_token_accuracy": 0.2310325399041176,
      "num_tokens": 93976435.0,
      "step": 45280
    },
    {
      "entropy": 5.585588502883911,
      "epoch": 4.107855587808418,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00034087542394771524,
      "loss": 4.7239,
      "mean_token_accuracy": 0.23684583604335785,
      "num_tokens": 93987294.0,
      "step": 45285
    },
    {
      "entropy": 5.696119070053101,
      "epoch": 4.108309143686502,
      "grad_norm": 1.359375,
      "learning_rate": 0.00034084448470725147,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22486038506031036,
      "num_tokens": 93997177.0,
      "step": 45290
    },
    {
      "entropy": 5.777114915847778,
      "epoch": 4.108762699564586,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003408135441051071,
      "loss": 4.9694,
      "mean_token_accuracy": 0.21778407394886018,
      "num_tokens": 94007663.0,
      "step": 45295
    },
    {
      "entropy": 5.706534099578858,
      "epoch": 4.1092162554426706,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003407826021419218,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2283296212553978,
      "num_tokens": 94016899.0,
      "step": 45300
    },
    {
      "entropy": 5.710009384155273,
      "epoch": 4.109669811320755,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003407516588183356,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22124049812555313,
      "num_tokens": 94027696.0,
      "step": 45305
    },
    {
      "entropy": 5.664752006530762,
      "epoch": 4.110123367198839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003407207141349882,
      "loss": 4.888,
      "mean_token_accuracy": 0.222280590236187,
      "num_tokens": 94038075.0,
      "step": 45310
    },
    {
      "entropy": 5.7483924388885494,
      "epoch": 4.110576923076923,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003406897680925199,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22109536826610565,
      "num_tokens": 94049082.0,
      "step": 45315
    },
    {
      "entropy": 5.754404830932617,
      "epoch": 4.111030478955008,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034065882069157023,
      "loss": 4.8888,
      "mean_token_accuracy": 0.21813881397247314,
      "num_tokens": 94060367.0,
      "step": 45320
    },
    {
      "entropy": 5.686347484588623,
      "epoch": 4.111484034833092,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034062787193277954,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22974654287099838,
      "num_tokens": 94069415.0,
      "step": 45325
    },
    {
      "entropy": 5.724197912216186,
      "epoch": 4.111937590711175,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034059692181678775,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2210339143872261,
      "num_tokens": 94079047.0,
      "step": 45330
    },
    {
      "entropy": 5.69263014793396,
      "epoch": 4.1123911465892595,
      "grad_norm": 1.3125,
      "learning_rate": 0.00034056597034423484,
      "loss": 4.8319,
      "mean_token_accuracy": 0.22634677737951278,
      "num_tokens": 94089631.0,
      "step": 45335
    },
    {
      "entropy": 5.699172115325927,
      "epoch": 4.112844702467344,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00034053501751576097,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22968704402446746,
      "num_tokens": 94099265.0,
      "step": 45340
    },
    {
      "entropy": 5.668064069747925,
      "epoch": 4.113298258345428,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003405040633320063,
      "loss": 4.8604,
      "mean_token_accuracy": 0.21694139987230301,
      "num_tokens": 94109673.0,
      "step": 45345
    },
    {
      "entropy": 5.673715972900391,
      "epoch": 4.113751814223512,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003404731077936108,
      "loss": 4.7755,
      "mean_token_accuracy": 0.2319294810295105,
      "num_tokens": 94120004.0,
      "step": 45350
    },
    {
      "entropy": 5.648002338409424,
      "epoch": 4.114205370101597,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003404421509012148,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2237185060977936,
      "num_tokens": 94129935.0,
      "step": 45355
    },
    {
      "entropy": 5.7071535110473635,
      "epoch": 4.114658925979681,
      "grad_norm": 1.453125,
      "learning_rate": 0.00034041119265545847,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2196139171719551,
      "num_tokens": 94139035.0,
      "step": 45360
    },
    {
      "entropy": 5.762556648254394,
      "epoch": 4.115112481857765,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034038023305698196,
      "loss": 4.8418,
      "mean_token_accuracy": 0.22550002485513687,
      "num_tokens": 94149281.0,
      "step": 45365
    },
    {
      "entropy": 5.708716678619385,
      "epoch": 4.115566037735849,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003403492721064256,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22718836069107057,
      "num_tokens": 94160455.0,
      "step": 45370
    },
    {
      "entropy": 5.744315528869629,
      "epoch": 4.116019593613934,
      "grad_norm": 1.375,
      "learning_rate": 0.00034031830980442955,
      "loss": 4.8933,
      "mean_token_accuracy": 0.2224472716450691,
      "num_tokens": 94170334.0,
      "step": 45375
    },
    {
      "entropy": 5.672414636611938,
      "epoch": 4.116473149492017,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034028734615163436,
      "loss": 4.806,
      "mean_token_accuracy": 0.2272252306342125,
      "num_tokens": 94181055.0,
      "step": 45380
    },
    {
      "entropy": 5.689899110794068,
      "epoch": 4.116926705370101,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003402563811486802,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2315835937857628,
      "num_tokens": 94191182.0,
      "step": 45385
    },
    {
      "entropy": 5.725643825531006,
      "epoch": 4.117380261248186,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003402254147962073,
      "loss": 4.9094,
      "mean_token_accuracy": 0.21733620911836624,
      "num_tokens": 94201229.0,
      "step": 45390
    },
    {
      "entropy": 5.65181794166565,
      "epoch": 4.11783381712627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003401944470948563,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22866207957267762,
      "num_tokens": 94210956.0,
      "step": 45395
    },
    {
      "entropy": 5.695469665527344,
      "epoch": 4.118287373004354,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034016347804526753,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22111136317253113,
      "num_tokens": 94220740.0,
      "step": 45400
    },
    {
      "entropy": 5.628972387313842,
      "epoch": 4.118740928882438,
      "grad_norm": 1.125,
      "learning_rate": 0.00034013250764808133,
      "loss": 4.7609,
      "mean_token_accuracy": 0.2412831813097,
      "num_tokens": 94231497.0,
      "step": 45405
    },
    {
      "entropy": 5.703517150878906,
      "epoch": 4.119194484760523,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003401015359039384,
      "loss": 4.8842,
      "mean_token_accuracy": 0.22846257984638213,
      "num_tokens": 94240492.0,
      "step": 45410
    },
    {
      "entropy": 5.66874361038208,
      "epoch": 4.119648040638607,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003400705628134791,
      "loss": 4.839,
      "mean_token_accuracy": 0.23066378235816956,
      "num_tokens": 94251586.0,
      "step": 45415
    },
    {
      "entropy": 5.690919494628906,
      "epoch": 4.120101596516691,
      "grad_norm": 1.328125,
      "learning_rate": 0.000340039588377344,
      "loss": 4.8671,
      "mean_token_accuracy": 0.230966517329216,
      "num_tokens": 94261635.0,
      "step": 45420
    },
    {
      "entropy": 5.73991961479187,
      "epoch": 4.120555152394775,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003400086125961735,
      "loss": 4.8962,
      "mean_token_accuracy": 0.2195743978023529,
      "num_tokens": 94269881.0,
      "step": 45425
    },
    {
      "entropy": 5.706535959243775,
      "epoch": 4.121008708272859,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00033997763547060847,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22952235639095306,
      "num_tokens": 94279412.0,
      "step": 45430
    },
    {
      "entropy": 5.686516427993775,
      "epoch": 4.121462264150943,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003399466570012893,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2199428379535675,
      "num_tokens": 94290157.0,
      "step": 45435
    },
    {
      "entropy": 5.723792934417725,
      "epoch": 4.121915820029027,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033991567718885674,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22251606434583665,
      "num_tokens": 94299693.0,
      "step": 45440
    },
    {
      "entropy": 5.802095556259156,
      "epoch": 4.122369375907112,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003398846960339515,
      "loss": 4.9937,
      "mean_token_accuracy": 0.22011375278234482,
      "num_tokens": 94310245.0,
      "step": 45445
    },
    {
      "entropy": 5.715678548812866,
      "epoch": 4.122822931785196,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003398537135372141,
      "loss": 4.85,
      "mean_token_accuracy": 0.22355229109525682,
      "num_tokens": 94321190.0,
      "step": 45450
    },
    {
      "entropy": 5.751244735717774,
      "epoch": 4.12327648766328,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003398227296992854,
      "loss": 4.9334,
      "mean_token_accuracy": 0.2285672679543495,
      "num_tokens": 94330273.0,
      "step": 45455
    },
    {
      "entropy": 5.766734600067139,
      "epoch": 4.123730043541364,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033979174452080616,
      "loss": 4.9709,
      "mean_token_accuracy": 0.21296848654747008,
      "num_tokens": 94341097.0,
      "step": 45460
    },
    {
      "entropy": 5.7117030143737795,
      "epoch": 4.124183599419449,
      "grad_norm": 1.25,
      "learning_rate": 0.0003397607580024171,
      "loss": 4.8679,
      "mean_token_accuracy": 0.22234456092119217,
      "num_tokens": 94352329.0,
      "step": 45465
    },
    {
      "entropy": 5.721599912643432,
      "epoch": 4.124637155297533,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003397297701447591,
      "loss": 4.8212,
      "mean_token_accuracy": 0.2279725581407547,
      "num_tokens": 94362551.0,
      "step": 45470
    },
    {
      "entropy": 5.695858716964722,
      "epoch": 4.125090711175617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000339698780948473,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22182660400867463,
      "num_tokens": 94372405.0,
      "step": 45475
    },
    {
      "entropy": 5.757310962677002,
      "epoch": 4.125544267053701,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003396677904141995,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22956510484218598,
      "num_tokens": 94382347.0,
      "step": 45480
    },
    {
      "entropy": 5.674115705490112,
      "epoch": 4.125997822931785,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033963679854257965,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22320032119750977,
      "num_tokens": 94393428.0,
      "step": 45485
    },
    {
      "entropy": 5.725510358810425,
      "epoch": 4.126451378809869,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003396058053342544,
      "loss": 4.8914,
      "mean_token_accuracy": 0.22735315859317778,
      "num_tokens": 94403941.0,
      "step": 45490
    },
    {
      "entropy": 5.728762483596801,
      "epoch": 4.126904934687953,
      "grad_norm": 1.25,
      "learning_rate": 0.00033957481078986465,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22155938297510147,
      "num_tokens": 94414206.0,
      "step": 45495
    },
    {
      "entropy": 5.75198802947998,
      "epoch": 4.127358490566038,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033954381491005126,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23031072467565536,
      "num_tokens": 94424481.0,
      "step": 45500
    },
    {
      "entropy": 5.69146728515625,
      "epoch": 4.127812046444122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003395128176954554,
      "loss": 4.8294,
      "mean_token_accuracy": 0.23091253489255906,
      "num_tokens": 94434714.0,
      "step": 45505
    },
    {
      "entropy": 5.752295637130738,
      "epoch": 4.128265602322206,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000339481819146718,
      "loss": 4.9104,
      "mean_token_accuracy": 0.22142142355442046,
      "num_tokens": 94445267.0,
      "step": 45510
    },
    {
      "entropy": 5.716652965545654,
      "epoch": 4.1287191582002905,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033945081926448013,
      "loss": 4.8712,
      "mean_token_accuracy": 0.2247446745634079,
      "num_tokens": 94455724.0,
      "step": 45515
    },
    {
      "entropy": 5.74789605140686,
      "epoch": 4.129172714078375,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033941981804938294,
      "loss": 4.9279,
      "mean_token_accuracy": 0.21574294120073317,
      "num_tokens": 94466655.0,
      "step": 45520
    },
    {
      "entropy": 5.667703533172608,
      "epoch": 4.129626269956459,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033938881550206744,
      "loss": 4.8151,
      "mean_token_accuracy": 0.22987173944711686,
      "num_tokens": 94477699.0,
      "step": 45525
    },
    {
      "entropy": 5.731016683578491,
      "epoch": 4.130079825834542,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033935781162317485,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22422097772359847,
      "num_tokens": 94487639.0,
      "step": 45530
    },
    {
      "entropy": 5.818817567825318,
      "epoch": 4.130533381712627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033932680641334635,
      "loss": 5.0053,
      "mean_token_accuracy": 0.21628451347351074,
      "num_tokens": 94498687.0,
      "step": 45535
    },
    {
      "entropy": 5.760129308700561,
      "epoch": 4.130986937590711,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003392957998732231,
      "loss": 4.8409,
      "mean_token_accuracy": 0.225234317779541,
      "num_tokens": 94508706.0,
      "step": 45540
    },
    {
      "entropy": 5.676712608337402,
      "epoch": 4.131440493468795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033926479200344626,
      "loss": 4.7722,
      "mean_token_accuracy": 0.2326131746172905,
      "num_tokens": 94517246.0,
      "step": 45545
    },
    {
      "entropy": 5.675050210952759,
      "epoch": 4.131894049346879,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003392337828046572,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23133792728185654,
      "num_tokens": 94527956.0,
      "step": 45550
    },
    {
      "entropy": 5.759914302825928,
      "epoch": 4.132347605224964,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003392027722774971,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21918095797300338,
      "num_tokens": 94539101.0,
      "step": 45555
    },
    {
      "entropy": 5.695048713684082,
      "epoch": 4.132801161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033917176042260735,
      "loss": 4.8148,
      "mean_token_accuracy": 0.22576915621757507,
      "num_tokens": 94549597.0,
      "step": 45560
    },
    {
      "entropy": 5.7599684715271,
      "epoch": 4.133254716981132,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003391407472406292,
      "loss": 4.9388,
      "mean_token_accuracy": 0.22078569680452348,
      "num_tokens": 94559911.0,
      "step": 45565
    },
    {
      "entropy": 5.710391283035278,
      "epoch": 4.1337082728592165,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033910973273220416,
      "loss": 4.8071,
      "mean_token_accuracy": 0.2232665613293648,
      "num_tokens": 94569396.0,
      "step": 45570
    },
    {
      "entropy": 5.713934230804443,
      "epoch": 4.134161828737301,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033907871689797337,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2281242862343788,
      "num_tokens": 94579817.0,
      "step": 45575
    },
    {
      "entropy": 5.632025718688965,
      "epoch": 4.134615384615385,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003390476997385785,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22509572207927703,
      "num_tokens": 94590398.0,
      "step": 45580
    },
    {
      "entropy": 5.651340627670288,
      "epoch": 4.135068940493468,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003390166812546608,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22372719794511794,
      "num_tokens": 94600528.0,
      "step": 45585
    },
    {
      "entropy": 5.712470436096192,
      "epoch": 4.135522496371553,
      "grad_norm": 1.25,
      "learning_rate": 0.00033898566144686184,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21940924376249313,
      "num_tokens": 94611107.0,
      "step": 45590
    },
    {
      "entropy": 5.774176549911499,
      "epoch": 4.135976052249637,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003389546403158231,
      "loss": 4.944,
      "mean_token_accuracy": 0.21889222115278245,
      "num_tokens": 94620932.0,
      "step": 45595
    },
    {
      "entropy": 5.774168825149536,
      "epoch": 4.136429608127721,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033892361786218604,
      "loss": 4.9405,
      "mean_token_accuracy": 0.21746836602687836,
      "num_tokens": 94630476.0,
      "step": 45600
    },
    {
      "entropy": 5.67464656829834,
      "epoch": 4.1368831640058055,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003388925940865924,
      "loss": 4.8675,
      "mean_token_accuracy": 0.2214506521821022,
      "num_tokens": 94641487.0,
      "step": 45605
    },
    {
      "entropy": 5.728150224685669,
      "epoch": 4.13733671988389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033886156898968353,
      "loss": 4.7911,
      "mean_token_accuracy": 0.23146069198846816,
      "num_tokens": 94651668.0,
      "step": 45610
    },
    {
      "entropy": 5.663774251937866,
      "epoch": 4.137790275761974,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033883054257210116,
      "loss": 4.8792,
      "mean_token_accuracy": 0.2185288205742836,
      "num_tokens": 94661218.0,
      "step": 45615
    },
    {
      "entropy": 5.6626220703125,
      "epoch": 4.138243831640058,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000338799514834487,
      "loss": 4.822,
      "mean_token_accuracy": 0.22418735027313233,
      "num_tokens": 94672412.0,
      "step": 45620
    },
    {
      "entropy": 5.721090412139892,
      "epoch": 4.1386973875181425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003387684857774824,
      "loss": 4.9352,
      "mean_token_accuracy": 0.21356236934661865,
      "num_tokens": 94683102.0,
      "step": 45625
    },
    {
      "entropy": 5.728157567977905,
      "epoch": 4.139150943396227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003387374554017294,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22535331249237062,
      "num_tokens": 94692869.0,
      "step": 45630
    },
    {
      "entropy": 5.711226272583008,
      "epoch": 4.13960449927431,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033870642370786966,
      "loss": 4.821,
      "mean_token_accuracy": 0.2282906472682953,
      "num_tokens": 94702437.0,
      "step": 45635
    },
    {
      "entropy": 5.67783555984497,
      "epoch": 4.1400580551523944,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033867539069654476,
      "loss": 4.9077,
      "mean_token_accuracy": 0.22388543784618378,
      "num_tokens": 94713130.0,
      "step": 45640
    },
    {
      "entropy": 5.764367580413818,
      "epoch": 4.140511611030479,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033864435636839647,
      "loss": 4.9684,
      "mean_token_accuracy": 0.21833201497793198,
      "num_tokens": 94723251.0,
      "step": 45645
    },
    {
      "entropy": 5.730973720550537,
      "epoch": 4.140965166908563,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003386133207240668,
      "loss": 4.8767,
      "mean_token_accuracy": 0.23355486690998079,
      "num_tokens": 94733805.0,
      "step": 45650
    },
    {
      "entropy": 5.707059526443482,
      "epoch": 4.141418722786647,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003385822837641974,
      "loss": 4.8128,
      "mean_token_accuracy": 0.22314165234565736,
      "num_tokens": 94743654.0,
      "step": 45655
    },
    {
      "entropy": 5.659441757202148,
      "epoch": 4.1418722786647315,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033855124548943025,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23096459060907365,
      "num_tokens": 94754635.0,
      "step": 45660
    },
    {
      "entropy": 5.700114965438843,
      "epoch": 4.142325834542816,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033852020590040705,
      "loss": 4.8993,
      "mean_token_accuracy": 0.22502158731222152,
      "num_tokens": 94765371.0,
      "step": 45665
    },
    {
      "entropy": 5.69606032371521,
      "epoch": 4.1427793904209,
      "grad_norm": 1.375,
      "learning_rate": 0.0003384891649977698,
      "loss": 4.8057,
      "mean_token_accuracy": 0.2330516830086708,
      "num_tokens": 94775780.0,
      "step": 45670
    },
    {
      "entropy": 5.755156993865967,
      "epoch": 4.143232946298984,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033845812278216045,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21843330562114716,
      "num_tokens": 94785345.0,
      "step": 45675
    },
    {
      "entropy": 5.68817253112793,
      "epoch": 4.143686502177069,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033842707925422095,
      "loss": 4.8223,
      "mean_token_accuracy": 0.2274296075105667,
      "num_tokens": 94796059.0,
      "step": 45680
    },
    {
      "entropy": 5.658831310272217,
      "epoch": 4.144140058055152,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033839603441459333,
      "loss": 4.8537,
      "mean_token_accuracy": 0.22837753891944884,
      "num_tokens": 94805867.0,
      "step": 45685
    },
    {
      "entropy": 5.682936286926269,
      "epoch": 4.144593613933236,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003383649882639195,
      "loss": 4.8512,
      "mean_token_accuracy": 0.22595291137695311,
      "num_tokens": 94817172.0,
      "step": 45690
    },
    {
      "entropy": 5.707564449310302,
      "epoch": 4.1450471698113205,
      "grad_norm": 1.25,
      "learning_rate": 0.0003383339408028416,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22444123029708862,
      "num_tokens": 94828748.0,
      "step": 45695
    },
    {
      "entropy": 5.674473237991333,
      "epoch": 4.145500725689405,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003383028920320017,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2288126066327095,
      "num_tokens": 94838704.0,
      "step": 45700
    },
    {
      "entropy": 5.734043025970459,
      "epoch": 4.145954281567489,
      "grad_norm": 1.25,
      "learning_rate": 0.00033827184195204174,
      "loss": 4.9446,
      "mean_token_accuracy": 0.21601210236549379,
      "num_tokens": 94849555.0,
      "step": 45705
    },
    {
      "entropy": 5.787162113189697,
      "epoch": 4.146407837445573,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003382407905636041,
      "loss": 4.9615,
      "mean_token_accuracy": 0.21975075155496598,
      "num_tokens": 94862238.0,
      "step": 45710
    },
    {
      "entropy": 5.625086641311645,
      "epoch": 4.1468613933236576,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003382097378673308,
      "loss": 4.7476,
      "mean_token_accuracy": 0.23923512995243074,
      "num_tokens": 94872960.0,
      "step": 45715
    },
    {
      "entropy": 5.6462969303131105,
      "epoch": 4.147314949201742,
      "grad_norm": 1.234375,
      "learning_rate": 0.000338178683863864,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23086034059524535,
      "num_tokens": 94883612.0,
      "step": 45720
    },
    {
      "entropy": 5.637822246551513,
      "epoch": 4.147768505079826,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003381476285538458,
      "loss": 4.806,
      "mean_token_accuracy": 0.23581933826208115,
      "num_tokens": 94893176.0,
      "step": 45725
    },
    {
      "entropy": 5.673295164108277,
      "epoch": 4.14822206095791,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003381165719379187,
      "loss": 4.8126,
      "mean_token_accuracy": 0.223284712433815,
      "num_tokens": 94903860.0,
      "step": 45730
    },
    {
      "entropy": 5.6618884086608885,
      "epoch": 4.148675616835995,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003380855140167247,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23273450434207915,
      "num_tokens": 94915681.0,
      "step": 45735
    },
    {
      "entropy": 5.6879514217376705,
      "epoch": 4.149129172714078,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003380544547909061,
      "loss": 4.8474,
      "mean_token_accuracy": 0.22431155443191528,
      "num_tokens": 94927062.0,
      "step": 45740
    },
    {
      "entropy": 5.691237068176269,
      "epoch": 4.149582728592162,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003380233942611055,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22869154661893845,
      "num_tokens": 94938838.0,
      "step": 45745
    },
    {
      "entropy": 5.658385515213013,
      "epoch": 4.1500362844702465,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337992332427965,
      "loss": 4.808,
      "mean_token_accuracy": 0.22599396854639053,
      "num_tokens": 94949678.0,
      "step": 45750
    },
    {
      "entropy": 5.716036319732666,
      "epoch": 4.150489840348331,
      "grad_norm": 1.296875,
      "learning_rate": 0.000337961269292127,
      "loss": 4.9073,
      "mean_token_accuracy": 0.22498934715986252,
      "num_tokens": 94960314.0,
      "step": 45755
    },
    {
      "entropy": 5.717112064361572,
      "epoch": 4.150943396226415,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033793020485423385,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22490934282541275,
      "num_tokens": 94969831.0,
      "step": 45760
    },
    {
      "entropy": 5.711872100830078,
      "epoch": 4.151396952104499,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033789913911492814,
      "loss": 4.8478,
      "mean_token_accuracy": 0.223559732735157,
      "num_tokens": 94980046.0,
      "step": 45765
    },
    {
      "entropy": 5.690056896209716,
      "epoch": 4.151850507982584,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003378680720748522,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22624509781599045,
      "num_tokens": 94990850.0,
      "step": 45770
    },
    {
      "entropy": 5.75922737121582,
      "epoch": 4.152304063860668,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003378370037346484,
      "loss": 4.9134,
      "mean_token_accuracy": 0.22130387127399445,
      "num_tokens": 95001658.0,
      "step": 45775
    },
    {
      "entropy": 5.710968685150147,
      "epoch": 4.152757619738752,
      "grad_norm": 1.46875,
      "learning_rate": 0.00033780593409495936,
      "loss": 4.8302,
      "mean_token_accuracy": 0.22911507189273833,
      "num_tokens": 95011912.0,
      "step": 45780
    },
    {
      "entropy": 5.696898746490478,
      "epoch": 4.153211175616836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003377748631564276,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22236702740192413,
      "num_tokens": 95021299.0,
      "step": 45785
    },
    {
      "entropy": 5.693111515045166,
      "epoch": 4.15366473149492,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033774379091969583,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23339128941297532,
      "num_tokens": 95031520.0,
      "step": 45790
    },
    {
      "entropy": 5.615486097335816,
      "epoch": 4.154118287373004,
      "grad_norm": 1.25,
      "learning_rate": 0.00033771271738540634,
      "loss": 4.7779,
      "mean_token_accuracy": 0.23967468440532685,
      "num_tokens": 95042127.0,
      "step": 45795
    },
    {
      "entropy": 5.693194150924683,
      "epoch": 4.154571843251088,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033768164255420184,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2226952061057091,
      "num_tokens": 95052181.0,
      "step": 45800
    },
    {
      "entropy": 5.780291318893433,
      "epoch": 4.155025399129173,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033765056642672507,
      "loss": 4.9035,
      "mean_token_accuracy": 0.22503660917282103,
      "num_tokens": 95062583.0,
      "step": 45805
    },
    {
      "entropy": 5.737106275558472,
      "epoch": 4.155478955007257,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033761948900361857,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22698320895433427,
      "num_tokens": 95073250.0,
      "step": 45810
    },
    {
      "entropy": 5.726348876953125,
      "epoch": 4.155932510885341,
      "grad_norm": 1.25,
      "learning_rate": 0.0003375884102855251,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2274572655558586,
      "num_tokens": 95083544.0,
      "step": 45815
    },
    {
      "entropy": 5.643235063552856,
      "epoch": 4.156386066763425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033755733027308746,
      "loss": 4.851,
      "mean_token_accuracy": 0.22937243580818176,
      "num_tokens": 95093933.0,
      "step": 45820
    },
    {
      "entropy": 5.653103542327881,
      "epoch": 4.15683962264151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003375262489669481,
      "loss": 4.8945,
      "mean_token_accuracy": 0.2277132272720337,
      "num_tokens": 95104468.0,
      "step": 45825
    },
    {
      "entropy": 5.77847843170166,
      "epoch": 4.157293178519594,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033749516636775014,
      "loss": 4.8903,
      "mean_token_accuracy": 0.2219273954629898,
      "num_tokens": 95114285.0,
      "step": 45830
    },
    {
      "entropy": 5.777374744415283,
      "epoch": 4.157746734397678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033746408247613616,
      "loss": 4.9245,
      "mean_token_accuracy": 0.22104430198669434,
      "num_tokens": 95123984.0,
      "step": 45835
    },
    {
      "entropy": 5.767383241653443,
      "epoch": 4.1582002902757615,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000337432997292749,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2236990362405777,
      "num_tokens": 95135466.0,
      "step": 45840
    },
    {
      "entropy": 5.648140478134155,
      "epoch": 4.158653846153846,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003374019108182315,
      "loss": 4.7764,
      "mean_token_accuracy": 0.2305083841085434,
      "num_tokens": 95145223.0,
      "step": 45845
    },
    {
      "entropy": 5.7665407180786135,
      "epoch": 4.15910740203193,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003373708230532267,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22452492117881775,
      "num_tokens": 95154916.0,
      "step": 45850
    },
    {
      "entropy": 5.697826290130616,
      "epoch": 4.159560957910014,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033733973399837735,
      "loss": 4.81,
      "mean_token_accuracy": 0.22206875830888748,
      "num_tokens": 95164717.0,
      "step": 45855
    },
    {
      "entropy": 5.80001220703125,
      "epoch": 4.160014513788099,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003373086436543263,
      "loss": 4.9132,
      "mean_token_accuracy": 0.21781759113073348,
      "num_tokens": 95175544.0,
      "step": 45860
    },
    {
      "entropy": 5.712191104888916,
      "epoch": 4.160468069666183,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003372775520217168,
      "loss": 4.9243,
      "mean_token_accuracy": 0.2246760442852974,
      "num_tokens": 95185997.0,
      "step": 45865
    },
    {
      "entropy": 5.641669225692749,
      "epoch": 4.160921625544267,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003372464591011916,
      "loss": 4.8725,
      "mean_token_accuracy": 0.21997740268707275,
      "num_tokens": 95196967.0,
      "step": 45870
    },
    {
      "entropy": 5.66511287689209,
      "epoch": 4.161375181422351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033721536489339374,
      "loss": 4.8114,
      "mean_token_accuracy": 0.22963499873876572,
      "num_tokens": 95206788.0,
      "step": 45875
    },
    {
      "entropy": 5.796599102020264,
      "epoch": 4.161828737300436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033718426939896627,
      "loss": 4.8914,
      "mean_token_accuracy": 0.2283964768052101,
      "num_tokens": 95217806.0,
      "step": 45880
    },
    {
      "entropy": 5.704882860183716,
      "epoch": 4.16228229317852,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033715317261855235,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23746591806411743,
      "num_tokens": 95228311.0,
      "step": 45885
    },
    {
      "entropy": 5.666493988037109,
      "epoch": 4.162735849056604,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033712207455279485,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22192852348089218,
      "num_tokens": 95238872.0,
      "step": 45890
    },
    {
      "entropy": 5.624309873580932,
      "epoch": 4.163189404934688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003370909752023371,
      "loss": 4.8081,
      "mean_token_accuracy": 0.23804305642843246,
      "num_tokens": 95248575.0,
      "step": 45895
    },
    {
      "entropy": 5.7143903255462645,
      "epoch": 4.163642960812772,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003370598745678221,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22153713405132294,
      "num_tokens": 95259323.0,
      "step": 45900
    },
    {
      "entropy": 5.743830347061158,
      "epoch": 4.164096516690856,
      "grad_norm": 1.25,
      "learning_rate": 0.0003370287726498932,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23166890144348146,
      "num_tokens": 95270018.0,
      "step": 45905
    },
    {
      "entropy": 5.706224727630615,
      "epoch": 4.16455007256894,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033699766944919325,
      "loss": 4.8293,
      "mean_token_accuracy": 0.2287753090262413,
      "num_tokens": 95279783.0,
      "step": 45910
    },
    {
      "entropy": 5.723867893218994,
      "epoch": 4.165003628447025,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033696656496636586,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22477851063013077,
      "num_tokens": 95291200.0,
      "step": 45915
    },
    {
      "entropy": 5.704744148254394,
      "epoch": 4.165457184325109,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033693545920205416,
      "loss": 4.9072,
      "mean_token_accuracy": 0.21702313423156738,
      "num_tokens": 95301641.0,
      "step": 45920
    },
    {
      "entropy": 5.634801149368286,
      "epoch": 4.165910740203193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033690435215690125,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23206814527511596,
      "num_tokens": 95312636.0,
      "step": 45925
    },
    {
      "entropy": 5.658076572418213,
      "epoch": 4.1663642960812775,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003368732438315506,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23245650082826613,
      "num_tokens": 95322168.0,
      "step": 45930
    },
    {
      "entropy": 5.669683933258057,
      "epoch": 4.166817851959362,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033684213422664554,
      "loss": 4.8032,
      "mean_token_accuracy": 0.22894673496484758,
      "num_tokens": 95333162.0,
      "step": 45935
    },
    {
      "entropy": 5.664115047454834,
      "epoch": 4.167271407837446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003368110233428294,
      "loss": 4.8334,
      "mean_token_accuracy": 0.22892070263624192,
      "num_tokens": 95343806.0,
      "step": 45940
    },
    {
      "entropy": 5.696900463104248,
      "epoch": 4.167724963715529,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033677991118074554,
      "loss": 4.7807,
      "mean_token_accuracy": 0.23741073161363602,
      "num_tokens": 95353906.0,
      "step": 45945
    },
    {
      "entropy": 5.781115341186523,
      "epoch": 4.168178519593614,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003367487977410373,
      "loss": 4.9765,
      "mean_token_accuracy": 0.2147670716047287,
      "num_tokens": 95363801.0,
      "step": 45950
    },
    {
      "entropy": 5.7500204086303714,
      "epoch": 4.168632075471698,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033671768302434823,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2219286233186722,
      "num_tokens": 95375012.0,
      "step": 45955
    },
    {
      "entropy": 5.758822870254517,
      "epoch": 4.169085631349782,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033668656703132174,
      "loss": 4.8845,
      "mean_token_accuracy": 0.2265584349632263,
      "num_tokens": 95386429.0,
      "step": 45960
    },
    {
      "entropy": 5.653383874893189,
      "epoch": 4.169539187227866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003366554497626014,
      "loss": 4.8681,
      "mean_token_accuracy": 0.21877045184373856,
      "num_tokens": 95396534.0,
      "step": 45965
    },
    {
      "entropy": 5.681945562362671,
      "epoch": 4.169992743105951,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033662433121883057,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2274239182472229,
      "num_tokens": 95406621.0,
      "step": 45970
    },
    {
      "entropy": 5.707672834396362,
      "epoch": 4.170446298984035,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003365932114006528,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22473012804985046,
      "num_tokens": 95417587.0,
      "step": 45975
    },
    {
      "entropy": 5.625006198883057,
      "epoch": 4.170899854862119,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033656209030871184,
      "loss": 4.7452,
      "mean_token_accuracy": 0.236981625854969,
      "num_tokens": 95428288.0,
      "step": 45980
    },
    {
      "entropy": 5.6600322246551515,
      "epoch": 4.1713534107402035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033653096794365115,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22147093564271927,
      "num_tokens": 95438093.0,
      "step": 45985
    },
    {
      "entropy": 5.605151224136352,
      "epoch": 4.171806966618288,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033649984430611437,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23228452801704408,
      "num_tokens": 95448397.0,
      "step": 45990
    },
    {
      "entropy": 5.79120626449585,
      "epoch": 4.172260522496371,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033646871939674506,
      "loss": 4.9986,
      "mean_token_accuracy": 0.20981747657060623,
      "num_tokens": 95459533.0,
      "step": 45995
    },
    {
      "entropy": 5.703748607635498,
      "epoch": 4.172714078374455,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000336437593216187,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22629403024911882,
      "num_tokens": 95469507.0,
      "step": 46000
    },
    {
      "entropy": 5.796309328079223,
      "epoch": 4.17316763425254,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003364064657650839,
      "loss": 4.9333,
      "mean_token_accuracy": 0.2175938069820404,
      "num_tokens": 95479742.0,
      "step": 46005
    },
    {
      "entropy": 5.662481594085693,
      "epoch": 4.173621190130624,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033637533704407936,
      "loss": 4.8024,
      "mean_token_accuracy": 0.23629496693611146,
      "num_tokens": 95489751.0,
      "step": 46010
    },
    {
      "entropy": 5.7347430229187015,
      "epoch": 4.174074746008708,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003363442070538173,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2171835258603096,
      "num_tokens": 95500580.0,
      "step": 46015
    },
    {
      "entropy": 5.752248620986938,
      "epoch": 4.1745283018867925,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033631307579494137,
      "loss": 4.9508,
      "mean_token_accuracy": 0.22439915239810942,
      "num_tokens": 95511018.0,
      "step": 46020
    },
    {
      "entropy": 5.789962339401245,
      "epoch": 4.174981857764877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033628194326809543,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22505989372730256,
      "num_tokens": 95521220.0,
      "step": 46025
    },
    {
      "entropy": 5.706304502487183,
      "epoch": 4.175435413642961,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033625080947392324,
      "loss": 4.8599,
      "mean_token_accuracy": 0.22181616127490997,
      "num_tokens": 95531426.0,
      "step": 46030
    },
    {
      "entropy": 5.735952758789063,
      "epoch": 4.175888969521045,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003362196744130687,
      "loss": 4.8871,
      "mean_token_accuracy": 0.2333260655403137,
      "num_tokens": 95542698.0,
      "step": 46035
    },
    {
      "entropy": 5.759538412094116,
      "epoch": 4.1763425253991295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033618853808617573,
      "loss": 4.813,
      "mean_token_accuracy": 0.23738054186105728,
      "num_tokens": 95553545.0,
      "step": 46040
    },
    {
      "entropy": 5.782966184616089,
      "epoch": 4.176796081277214,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003361574004938882,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22445379495620726,
      "num_tokens": 95564452.0,
      "step": 46045
    },
    {
      "entropy": 5.695712947845459,
      "epoch": 4.177249637155297,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033612626163685,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2347927749156952,
      "num_tokens": 95574756.0,
      "step": 46050
    },
    {
      "entropy": 5.7958262920379635,
      "epoch": 4.1777031930333814,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033609512151570517,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21641970574855804,
      "num_tokens": 95585147.0,
      "step": 46055
    },
    {
      "entropy": 5.734976005554199,
      "epoch": 4.178156748911466,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033606398013109765,
      "loss": 4.9734,
      "mean_token_accuracy": 0.22271275222301484,
      "num_tokens": 95596025.0,
      "step": 46060
    },
    {
      "entropy": 5.679508304595947,
      "epoch": 4.17861030478955,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003360328374836715,
      "loss": 4.8609,
      "mean_token_accuracy": 0.21967616379261018,
      "num_tokens": 95607191.0,
      "step": 46065
    },
    {
      "entropy": 5.722225713729858,
      "epoch": 4.179063860667634,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003360016935740706,
      "loss": 4.884,
      "mean_token_accuracy": 0.22075208723545076,
      "num_tokens": 95617256.0,
      "step": 46070
    },
    {
      "entropy": 5.76864972114563,
      "epoch": 4.1795174165457185,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033597054840293927,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22344349473714828,
      "num_tokens": 95627638.0,
      "step": 46075
    },
    {
      "entropy": 5.729780149459839,
      "epoch": 4.179970972423803,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003359394019709214,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2265940472483635,
      "num_tokens": 95637662.0,
      "step": 46080
    },
    {
      "entropy": 5.700526189804077,
      "epoch": 4.180424528301887,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033590825427866117,
      "loss": 4.8375,
      "mean_token_accuracy": 0.2230824425816536,
      "num_tokens": 95647505.0,
      "step": 46085
    },
    {
      "entropy": 5.664188480377197,
      "epoch": 4.180878084179971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003358771053268027,
      "loss": 4.8016,
      "mean_token_accuracy": 0.22308114022016526,
      "num_tokens": 95657810.0,
      "step": 46090
    },
    {
      "entropy": 5.669896268844605,
      "epoch": 4.181331640058056,
      "grad_norm": 1.375,
      "learning_rate": 0.00033584595511599024,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2240704596042633,
      "num_tokens": 95667838.0,
      "step": 46095
    },
    {
      "entropy": 5.735329341888428,
      "epoch": 4.181785195936139,
      "grad_norm": 1.25,
      "learning_rate": 0.0003358148036468679,
      "loss": 4.8594,
      "mean_token_accuracy": 0.23159938156604767,
      "num_tokens": 95678497.0,
      "step": 46100
    },
    {
      "entropy": 5.69722695350647,
      "epoch": 4.182238751814223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033578365092007986,
      "loss": 4.765,
      "mean_token_accuracy": 0.2286176159977913,
      "num_tokens": 95688656.0,
      "step": 46105
    },
    {
      "entropy": 5.757472848892212,
      "epoch": 4.1826923076923075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003357524969362705,
      "loss": 4.952,
      "mean_token_accuracy": 0.21739222556352616,
      "num_tokens": 95699156.0,
      "step": 46110
    },
    {
      "entropy": 5.721827650070191,
      "epoch": 4.183145863570392,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033572134169608395,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2294031336903572,
      "num_tokens": 95709680.0,
      "step": 46115
    },
    {
      "entropy": 5.79875078201294,
      "epoch": 4.183599419448476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033569018520016465,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2267226755619049,
      "num_tokens": 95719142.0,
      "step": 46120
    },
    {
      "entropy": 5.719665050506592,
      "epoch": 4.18405297532656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003356590274491568,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2177251771092415,
      "num_tokens": 95729415.0,
      "step": 46125
    },
    {
      "entropy": 5.648221874237061,
      "epoch": 4.1845065312046446,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003356278684437049,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22332355380058289,
      "num_tokens": 95739775.0,
      "step": 46130
    },
    {
      "entropy": 5.7452627658844,
      "epoch": 4.184960087082729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033559670818445315,
      "loss": 4.9327,
      "mean_token_accuracy": 0.21870820522308348,
      "num_tokens": 95751540.0,
      "step": 46135
    },
    {
      "entropy": 5.833878040313721,
      "epoch": 4.185413642960813,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000335565546672046,
      "loss": 5.0534,
      "mean_token_accuracy": 0.2146403521299362,
      "num_tokens": 95761885.0,
      "step": 46140
    },
    {
      "entropy": 5.7896318435668945,
      "epoch": 4.185867198838897,
      "grad_norm": 1.203125,
      "learning_rate": 0.000335534383907128,
      "loss": 4.8823,
      "mean_token_accuracy": 0.2274986371397972,
      "num_tokens": 95772284.0,
      "step": 46145
    },
    {
      "entropy": 5.6864134788513185,
      "epoch": 4.186320754716981,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003355032198903434,
      "loss": 4.8184,
      "mean_token_accuracy": 0.2249468296766281,
      "num_tokens": 95783373.0,
      "step": 46150
    },
    {
      "entropy": 5.61769208908081,
      "epoch": 4.186774310595065,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033547205462233694,
      "loss": 4.7569,
      "mean_token_accuracy": 0.2358310803771019,
      "num_tokens": 95793226.0,
      "step": 46155
    },
    {
      "entropy": 5.5834746837615965,
      "epoch": 4.187227866473149,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033544088810375287,
      "loss": 4.7713,
      "mean_token_accuracy": 0.2294733077287674,
      "num_tokens": 95803559.0,
      "step": 46160
    },
    {
      "entropy": 5.682831478118897,
      "epoch": 4.1876814223512335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003354097203352358,
      "loss": 4.8494,
      "mean_token_accuracy": 0.2257275938987732,
      "num_tokens": 95814620.0,
      "step": 46165
    },
    {
      "entropy": 5.676716947555542,
      "epoch": 4.188134978229318,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003353785513174305,
      "loss": 4.8444,
      "mean_token_accuracy": 0.23089781999588013,
      "num_tokens": 95825024.0,
      "step": 46170
    },
    {
      "entropy": 5.729108810424805,
      "epoch": 4.188588534107402,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033534738105098124,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22997616976499557,
      "num_tokens": 95837373.0,
      "step": 46175
    },
    {
      "entropy": 5.707740211486817,
      "epoch": 4.189042089985486,
      "grad_norm": 1.40625,
      "learning_rate": 0.00033531620953653276,
      "loss": 4.8752,
      "mean_token_accuracy": 0.22428961098194122,
      "num_tokens": 95846962.0,
      "step": 46180
    },
    {
      "entropy": 5.700722122192383,
      "epoch": 4.189495645863571,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003352850367747297,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22837209105491638,
      "num_tokens": 95857492.0,
      "step": 46185
    },
    {
      "entropy": 5.683300447463989,
      "epoch": 4.189949201741655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003352538627662168,
      "loss": 4.8166,
      "mean_token_accuracy": 0.22403760105371476,
      "num_tokens": 95867903.0,
      "step": 46190
    },
    {
      "entropy": 5.804043292999268,
      "epoch": 4.190402757619739,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003352226875116386,
      "loss": 4.9876,
      "mean_token_accuracy": 0.21725146621465682,
      "num_tokens": 95878720.0,
      "step": 46195
    },
    {
      "entropy": 5.704366636276245,
      "epoch": 4.1908563134978225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003351915110116399,
      "loss": 4.8957,
      "mean_token_accuracy": 0.2181372493505478,
      "num_tokens": 95888406.0,
      "step": 46200
    },
    {
      "entropy": 5.709855651855468,
      "epoch": 4.191309869375907,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033516033326686537,
      "loss": 4.8947,
      "mean_token_accuracy": 0.21616960018873216,
      "num_tokens": 95898431.0,
      "step": 46205
    },
    {
      "entropy": 5.736049938201904,
      "epoch": 4.191763425253991,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033512915427795975,
      "loss": 4.859,
      "mean_token_accuracy": 0.22646784037351608,
      "num_tokens": 95909089.0,
      "step": 46210
    },
    {
      "entropy": 5.665524530410766,
      "epoch": 4.192216981132075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000335097974045568,
      "loss": 4.838,
      "mean_token_accuracy": 0.22766564786434174,
      "num_tokens": 95919787.0,
      "step": 46215
    },
    {
      "entropy": 5.722571849822998,
      "epoch": 4.19267053701016,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033506679257033485,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2262008309364319,
      "num_tokens": 95930620.0,
      "step": 46220
    },
    {
      "entropy": 5.7456090450286865,
      "epoch": 4.193124092888244,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003350356098529051,
      "loss": 4.8367,
      "mean_token_accuracy": 0.22530510574579238,
      "num_tokens": 95940258.0,
      "step": 46225
    },
    {
      "entropy": 5.752669906616211,
      "epoch": 4.193577648766328,
      "grad_norm": 1.375,
      "learning_rate": 0.0003350044258939236,
      "loss": 4.9326,
      "mean_token_accuracy": 0.2259291023015976,
      "num_tokens": 95950062.0,
      "step": 46230
    },
    {
      "entropy": 5.681144523620605,
      "epoch": 4.194031204644412,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033497324069403536,
      "loss": 4.7759,
      "mean_token_accuracy": 0.22978834956884384,
      "num_tokens": 95960658.0,
      "step": 46235
    },
    {
      "entropy": 5.717984962463379,
      "epoch": 4.194484760522497,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003349420542538851,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22571849077939987,
      "num_tokens": 95970497.0,
      "step": 46240
    },
    {
      "entropy": 5.747794342041016,
      "epoch": 4.194938316400581,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033491086657411797,
      "loss": 4.9534,
      "mean_token_accuracy": 0.2149714484810829,
      "num_tokens": 95980405.0,
      "step": 46245
    },
    {
      "entropy": 5.772362804412841,
      "epoch": 4.195391872278664,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003348796776553788,
      "loss": 4.97,
      "mean_token_accuracy": 0.21731164753437043,
      "num_tokens": 95991111.0,
      "step": 46250
    },
    {
      "entropy": 5.777197599411011,
      "epoch": 4.1958454281567485,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003348484874983127,
      "loss": 4.897,
      "mean_token_accuracy": 0.2274266764521599,
      "num_tokens": 96001962.0,
      "step": 46255
    },
    {
      "entropy": 5.720953798294067,
      "epoch": 4.196298984034833,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003348172961035646,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22518610209226608,
      "num_tokens": 96011774.0,
      "step": 46260
    },
    {
      "entropy": 5.708581113815308,
      "epoch": 4.196752539912917,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003347861034717795,
      "loss": 4.8901,
      "mean_token_accuracy": 0.22112664878368377,
      "num_tokens": 96021915.0,
      "step": 46265
    },
    {
      "entropy": 5.695303535461425,
      "epoch": 4.197206095791001,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033475490960360266,
      "loss": 4.8696,
      "mean_token_accuracy": 0.2237190917134285,
      "num_tokens": 96032851.0,
      "step": 46270
    },
    {
      "entropy": 5.670752954483032,
      "epoch": 4.197659651669086,
      "grad_norm": 1.328125,
      "learning_rate": 0.000334723714499679,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23046067655086516,
      "num_tokens": 96042630.0,
      "step": 46275
    },
    {
      "entropy": 5.677845811843872,
      "epoch": 4.19811320754717,
      "grad_norm": 1.515625,
      "learning_rate": 0.00033469251816065375,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22659361064434053,
      "num_tokens": 96052154.0,
      "step": 46280
    },
    {
      "entropy": 5.7193992137908936,
      "epoch": 4.198566763425254,
      "grad_norm": 1.390625,
      "learning_rate": 0.000334661320587172,
      "loss": 4.8607,
      "mean_token_accuracy": 0.22604342997074128,
      "num_tokens": 96062301.0,
      "step": 46285
    },
    {
      "entropy": 5.739561271667481,
      "epoch": 4.199020319303338,
      "grad_norm": 1.375,
      "learning_rate": 0.000334630121779879,
      "loss": 4.9263,
      "mean_token_accuracy": 0.22259617894887923,
      "num_tokens": 96072792.0,
      "step": 46290
    },
    {
      "entropy": 5.762626838684082,
      "epoch": 4.199473875181423,
      "grad_norm": 1.375,
      "learning_rate": 0.00033459892173941973,
      "loss": 4.931,
      "mean_token_accuracy": 0.21947454214096068,
      "num_tokens": 96082610.0,
      "step": 46295
    },
    {
      "entropy": 5.7934534549713135,
      "epoch": 4.199927431059507,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003345677204664397,
      "loss": 4.9501,
      "mean_token_accuracy": 0.2225771278142929,
      "num_tokens": 96093095.0,
      "step": 46300
    },
    {
      "entropy": 5.710545492172241,
      "epoch": 4.20038098693759,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033453651796158397,
      "loss": 4.8509,
      "mean_token_accuracy": 0.22926794290542601,
      "num_tokens": 96103218.0,
      "step": 46305
    },
    {
      "entropy": 5.691026592254639,
      "epoch": 4.200834542815675,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033450531422549786,
      "loss": 4.845,
      "mean_token_accuracy": 0.22923587560653685,
      "num_tokens": 96113937.0,
      "step": 46310
    },
    {
      "entropy": 5.798322772979736,
      "epoch": 4.201288098693759,
      "grad_norm": 1.640625,
      "learning_rate": 0.00033447410925882673,
      "loss": 5.0062,
      "mean_token_accuracy": 0.22200922220945357,
      "num_tokens": 96124378.0,
      "step": 46315
    },
    {
      "entropy": 5.632367038726807,
      "epoch": 4.201741654571843,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003344429030622159,
      "loss": 4.7528,
      "mean_token_accuracy": 0.23358624279499055,
      "num_tokens": 96133657.0,
      "step": 46320
    },
    {
      "entropy": 5.687938404083252,
      "epoch": 4.202195210449927,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033441169563631055,
      "loss": 4.8701,
      "mean_token_accuracy": 0.2288980334997177,
      "num_tokens": 96143669.0,
      "step": 46325
    },
    {
      "entropy": 5.715309906005859,
      "epoch": 4.202648766328012,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033438048698175625,
      "loss": 4.848,
      "mean_token_accuracy": 0.2246783509850502,
      "num_tokens": 96154782.0,
      "step": 46330
    },
    {
      "entropy": 5.7240838527679445,
      "epoch": 4.203102322206096,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003343492770991984,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22767011672258378,
      "num_tokens": 96165128.0,
      "step": 46335
    },
    {
      "entropy": 5.683759641647339,
      "epoch": 4.20355587808418,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003343180659892824,
      "loss": 4.884,
      "mean_token_accuracy": 0.22308689057826997,
      "num_tokens": 96176909.0,
      "step": 46340
    },
    {
      "entropy": 5.710403156280518,
      "epoch": 4.2040094339622645,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003342868536526536,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23232154548168182,
      "num_tokens": 96188287.0,
      "step": 46345
    },
    {
      "entropy": 5.6582419872283936,
      "epoch": 4.204462989840349,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033425564008995764,
      "loss": 4.834,
      "mean_token_accuracy": 0.22887646406888962,
      "num_tokens": 96198852.0,
      "step": 46350
    },
    {
      "entropy": 5.686162090301513,
      "epoch": 4.204916545718432,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003342244253018399,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22499673217535018,
      "num_tokens": 96208111.0,
      "step": 46355
    },
    {
      "entropy": 5.705805683135987,
      "epoch": 4.205370101596516,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003341932092889459,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23097601234912873,
      "num_tokens": 96218925.0,
      "step": 46360
    },
    {
      "entropy": 5.734135150909424,
      "epoch": 4.205823657474601,
      "grad_norm": 1.375,
      "learning_rate": 0.0003341619920519213,
      "loss": 4.9094,
      "mean_token_accuracy": 0.23252851963043214,
      "num_tokens": 96228573.0,
      "step": 46365
    },
    {
      "entropy": 5.732025575637818,
      "epoch": 4.206277213352685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033413077359141166,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22934682667255402,
      "num_tokens": 96239109.0,
      "step": 46370
    },
    {
      "entropy": 5.738963031768799,
      "epoch": 4.206730769230769,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003340995539080625,
      "loss": 4.9183,
      "mean_token_accuracy": 0.2209576666355133,
      "num_tokens": 96249286.0,
      "step": 46375
    },
    {
      "entropy": 5.6607061386108395,
      "epoch": 4.207184325108853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033406833300251947,
      "loss": 4.8405,
      "mean_token_accuracy": 0.228090737760067,
      "num_tokens": 96259515.0,
      "step": 46380
    },
    {
      "entropy": 5.687766361236572,
      "epoch": 4.207637880986938,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003340371108754284,
      "loss": 4.807,
      "mean_token_accuracy": 0.22759128212928773,
      "num_tokens": 96269136.0,
      "step": 46385
    },
    {
      "entropy": 5.742825078964233,
      "epoch": 4.208091436865022,
      "grad_norm": 1.25,
      "learning_rate": 0.0003340058875274347,
      "loss": 4.9015,
      "mean_token_accuracy": 0.22750115245580674,
      "num_tokens": 96279792.0,
      "step": 46390
    },
    {
      "entropy": 5.720653343200683,
      "epoch": 4.208544992743106,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003339746629591842,
      "loss": 4.9201,
      "mean_token_accuracy": 0.2221170961856842,
      "num_tokens": 96289606.0,
      "step": 46395
    },
    {
      "entropy": 5.693995094299316,
      "epoch": 4.2089985486211905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033394343717132275,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2167604848742485,
      "num_tokens": 96299903.0,
      "step": 46400
    },
    {
      "entropy": 5.788566398620605,
      "epoch": 4.209452104499274,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033391221016449595,
      "loss": 4.9474,
      "mean_token_accuracy": 0.21528271585702896,
      "num_tokens": 96310179.0,
      "step": 46405
    },
    {
      "entropy": 5.748233652114868,
      "epoch": 4.209905660377358,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033388098193934965,
      "loss": 4.9033,
      "mean_token_accuracy": 0.22187597453594207,
      "num_tokens": 96320948.0,
      "step": 46410
    },
    {
      "entropy": 5.685335397720337,
      "epoch": 4.210359216255442,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003338497524965296,
      "loss": 4.8286,
      "mean_token_accuracy": 0.2266976460814476,
      "num_tokens": 96332046.0,
      "step": 46415
    },
    {
      "entropy": 5.70167384147644,
      "epoch": 4.210812772133527,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00033381852183668166,
      "loss": 4.8178,
      "mean_token_accuracy": 0.22920230627059937,
      "num_tokens": 96342327.0,
      "step": 46420
    },
    {
      "entropy": 5.686598443984986,
      "epoch": 4.211266328011611,
      "grad_norm": 1.5,
      "learning_rate": 0.0003337872899604517,
      "loss": 4.8632,
      "mean_token_accuracy": 0.21791262477636336,
      "num_tokens": 96352048.0,
      "step": 46425
    },
    {
      "entropy": 5.678032112121582,
      "epoch": 4.211719883889695,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033375605686848566,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22906270027160644,
      "num_tokens": 96362917.0,
      "step": 46430
    },
    {
      "entropy": 5.7292357921600345,
      "epoch": 4.2121734397677795,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003337248225614293,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2276833787560463,
      "num_tokens": 96373261.0,
      "step": 46435
    },
    {
      "entropy": 5.749915885925293,
      "epoch": 4.212626995645864,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003336935870399288,
      "loss": 4.9008,
      "mean_token_accuracy": 0.21564942002296447,
      "num_tokens": 96384279.0,
      "step": 46440
    },
    {
      "entropy": 5.754499292373657,
      "epoch": 4.213080551523948,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003336623503046299,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22264978736639024,
      "num_tokens": 96394071.0,
      "step": 46445
    },
    {
      "entropy": 5.685178518295288,
      "epoch": 4.213534107402032,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033363111235617854,
      "loss": 4.8718,
      "mean_token_accuracy": 0.23106679469347,
      "num_tokens": 96404233.0,
      "step": 46450
    },
    {
      "entropy": 5.747089910507202,
      "epoch": 4.2139876632801165,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033359987319522095,
      "loss": 4.8919,
      "mean_token_accuracy": 0.2276756003499031,
      "num_tokens": 96415388.0,
      "step": 46455
    },
    {
      "entropy": 5.7812446594238285,
      "epoch": 4.2144412191582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033356863282240295,
      "loss": 4.8792,
      "mean_token_accuracy": 0.22296981364488602,
      "num_tokens": 96425744.0,
      "step": 46460
    },
    {
      "entropy": 5.737930536270142,
      "epoch": 4.214894775036284,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003335373912383707,
      "loss": 4.8721,
      "mean_token_accuracy": 0.23375994265079497,
      "num_tokens": 96436340.0,
      "step": 46465
    },
    {
      "entropy": 5.783969497680664,
      "epoch": 4.2153483309143684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003335061484437704,
      "loss": 4.9804,
      "mean_token_accuracy": 0.21307456493377686,
      "num_tokens": 96447384.0,
      "step": 46470
    },
    {
      "entropy": 5.672529792785644,
      "epoch": 4.215801886792453,
      "grad_norm": 1.3125,
      "learning_rate": 0.000333474904439248,
      "loss": 4.8471,
      "mean_token_accuracy": 0.227647602558136,
      "num_tokens": 96456938.0,
      "step": 46475
    },
    {
      "entropy": 5.666154098510742,
      "epoch": 4.216255442670537,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003334436592254496,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22397653758525848,
      "num_tokens": 96467343.0,
      "step": 46480
    },
    {
      "entropy": 5.714778184890747,
      "epoch": 4.216708998548621,
      "grad_norm": 1.390625,
      "learning_rate": 0.00033341241280302153,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22790463864803315,
      "num_tokens": 96477443.0,
      "step": 46485
    },
    {
      "entropy": 5.712932395935058,
      "epoch": 4.2171625544267055,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033338116517260985,
      "loss": 4.9138,
      "mean_token_accuracy": 0.21801262646913527,
      "num_tokens": 96487711.0,
      "step": 46490
    },
    {
      "entropy": 5.8109753131866455,
      "epoch": 4.21761611030479,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003333499163348608,
      "loss": 5.0074,
      "mean_token_accuracy": 0.21426323056221008,
      "num_tokens": 96497839.0,
      "step": 46495
    },
    {
      "entropy": 5.737284469604492,
      "epoch": 4.218069666182874,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033331866629042055,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22088176310062407,
      "num_tokens": 96508237.0,
      "step": 46500
    },
    {
      "entropy": 5.656514883041382,
      "epoch": 4.218523222060958,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003332874150399355,
      "loss": 4.7675,
      "mean_token_accuracy": 0.23072372525930404,
      "num_tokens": 96518877.0,
      "step": 46505
    },
    {
      "entropy": 5.7005260467529295,
      "epoch": 4.218976777939042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003332561625840519,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2251722112298012,
      "num_tokens": 96530734.0,
      "step": 46510
    },
    {
      "entropy": 5.765560102462769,
      "epoch": 4.219430333817126,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003332249089234159,
      "loss": 4.9432,
      "mean_token_accuracy": 0.22299870550632478,
      "num_tokens": 96540983.0,
      "step": 46515
    },
    {
      "entropy": 5.771080064773559,
      "epoch": 4.21988388969521,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033319365405867394,
      "loss": 4.9498,
      "mean_token_accuracy": 0.22328597903251649,
      "num_tokens": 96551866.0,
      "step": 46520
    },
    {
      "entropy": 5.722241353988648,
      "epoch": 4.2203374455732945,
      "grad_norm": 1.375,
      "learning_rate": 0.00033316239799047247,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22483631521463393,
      "num_tokens": 96562515.0,
      "step": 46525
    },
    {
      "entropy": 5.672820711135865,
      "epoch": 4.220791001451379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033313114071945764,
      "loss": 4.8606,
      "mean_token_accuracy": 0.22442122399806977,
      "num_tokens": 96573266.0,
      "step": 46530
    },
    {
      "entropy": 5.6991236209869385,
      "epoch": 4.221244557329463,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003330998822462762,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23384528756141662,
      "num_tokens": 96583070.0,
      "step": 46535
    },
    {
      "entropy": 5.705707693099976,
      "epoch": 4.221698113207547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033306862257157427,
      "loss": 4.8095,
      "mean_token_accuracy": 0.2297367885708809,
      "num_tokens": 96593582.0,
      "step": 46540
    },
    {
      "entropy": 5.688095378875732,
      "epoch": 4.2221516690856316,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003330373616959984,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23001430332660674,
      "num_tokens": 96603233.0,
      "step": 46545
    },
    {
      "entropy": 5.690533876419067,
      "epoch": 4.222605224963716,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033300609962019523,
      "loss": 4.8321,
      "mean_token_accuracy": 0.22947851121425628,
      "num_tokens": 96614361.0,
      "step": 46550
    },
    {
      "entropy": 5.742845773696899,
      "epoch": 4.2230587808418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000332974836344811,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22380124181509017,
      "num_tokens": 96624161.0,
      "step": 46555
    },
    {
      "entropy": 5.735103559494019,
      "epoch": 4.2235123367198835,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033294357187049246,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22143205255270004,
      "num_tokens": 96634329.0,
      "step": 46560
    },
    {
      "entropy": 5.765730094909668,
      "epoch": 4.223965892597968,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033291230619788604,
      "loss": 4.9478,
      "mean_token_accuracy": 0.22629334032535553,
      "num_tokens": 96645575.0,
      "step": 46565
    },
    {
      "entropy": 5.737837314605713,
      "epoch": 4.224419448476052,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033288103932763846,
      "loss": 4.8863,
      "mean_token_accuracy": 0.21793497651815413,
      "num_tokens": 96656224.0,
      "step": 46570
    },
    {
      "entropy": 5.784933996200562,
      "epoch": 4.224873004354136,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003328497712603962,
      "loss": 4.869,
      "mean_token_accuracy": 0.23093735426664352,
      "num_tokens": 96665950.0,
      "step": 46575
    },
    {
      "entropy": 5.656629848480224,
      "epoch": 4.2253265602322205,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033281850199680586,
      "loss": 4.7611,
      "mean_token_accuracy": 0.2320149138569832,
      "num_tokens": 96676916.0,
      "step": 46580
    },
    {
      "entropy": 5.71456561088562,
      "epoch": 4.225780116110305,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003327872315375142,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2166626825928688,
      "num_tokens": 96687289.0,
      "step": 46585
    },
    {
      "entropy": 5.739950227737427,
      "epoch": 4.226233671988389,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003327559598831679,
      "loss": 4.893,
      "mean_token_accuracy": 0.22803975045681,
      "num_tokens": 96698149.0,
      "step": 46590
    },
    {
      "entropy": 5.72047643661499,
      "epoch": 4.226687227866473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033272468703441363,
      "loss": 4.8601,
      "mean_token_accuracy": 0.225819955766201,
      "num_tokens": 96709466.0,
      "step": 46595
    },
    {
      "entropy": 5.752383756637573,
      "epoch": 4.227140783744558,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000332693412991898,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22248856127262115,
      "num_tokens": 96720179.0,
      "step": 46600
    },
    {
      "entropy": 5.785198593139649,
      "epoch": 4.227594339622642,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003326621377562681,
      "loss": 4.9182,
      "mean_token_accuracy": 0.21766040921211244,
      "num_tokens": 96730783.0,
      "step": 46605
    },
    {
      "entropy": 5.775998783111572,
      "epoch": 4.228047895500726,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033263086132817035,
      "loss": 4.9799,
      "mean_token_accuracy": 0.2185059443116188,
      "num_tokens": 96740903.0,
      "step": 46610
    },
    {
      "entropy": 5.713264465332031,
      "epoch": 4.2285014513788095,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033259958370825165,
      "loss": 4.8593,
      "mean_token_accuracy": 0.23127478808164598,
      "num_tokens": 96750600.0,
      "step": 46615
    },
    {
      "entropy": 5.6771289825439455,
      "epoch": 4.228955007256894,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033256830489715905,
      "loss": 4.8194,
      "mean_token_accuracy": 0.22842043936252593,
      "num_tokens": 96760550.0,
      "step": 46620
    },
    {
      "entropy": 5.718864154815674,
      "epoch": 4.229408563134978,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003325370248955392,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22591228634119034,
      "num_tokens": 96770784.0,
      "step": 46625
    },
    {
      "entropy": 5.6901750564575195,
      "epoch": 4.229862119013062,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003325057437040389,
      "loss": 4.7923,
      "mean_token_accuracy": 0.2324146404862404,
      "num_tokens": 96780572.0,
      "step": 46630
    },
    {
      "entropy": 5.693010854721069,
      "epoch": 4.230315674891147,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003324744613233052,
      "loss": 4.8057,
      "mean_token_accuracy": 0.22388217151165007,
      "num_tokens": 96790504.0,
      "step": 46635
    },
    {
      "entropy": 5.733713245391845,
      "epoch": 4.230769230769231,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033244317775398506,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2239360421895981,
      "num_tokens": 96800856.0,
      "step": 46640
    },
    {
      "entropy": 5.711424493789673,
      "epoch": 4.231222786647315,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003324118929967253,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22201080173254012,
      "num_tokens": 96811293.0,
      "step": 46645
    },
    {
      "entropy": 5.627492189407349,
      "epoch": 4.231676342525399,
      "grad_norm": 1.28125,
      "learning_rate": 0.000332380607052173,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23413191139698028,
      "num_tokens": 96821844.0,
      "step": 46650
    },
    {
      "entropy": 5.719759941101074,
      "epoch": 4.232129898403484,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003323493199209751,
      "loss": 4.9458,
      "mean_token_accuracy": 0.22005916237831116,
      "num_tokens": 96833438.0,
      "step": 46655
    },
    {
      "entropy": 5.656706380844116,
      "epoch": 4.232583454281568,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003323180316037787,
      "loss": 4.8091,
      "mean_token_accuracy": 0.2285059466958046,
      "num_tokens": 96843086.0,
      "step": 46660
    },
    {
      "entropy": 5.713295793533325,
      "epoch": 4.233037010159651,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003322867421012308,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22624849230051042,
      "num_tokens": 96853064.0,
      "step": 46665
    },
    {
      "entropy": 5.791063976287842,
      "epoch": 4.2334905660377355,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003322554514139784,
      "loss": 4.9458,
      "mean_token_accuracy": 0.21970123648643494,
      "num_tokens": 96864213.0,
      "step": 46670
    },
    {
      "entropy": 5.814552068710327,
      "epoch": 4.23394412191582,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033222415954266873,
      "loss": 5.01,
      "mean_token_accuracy": 0.21283529996871947,
      "num_tokens": 96875344.0,
      "step": 46675
    },
    {
      "entropy": 5.783625268936158,
      "epoch": 4.234397677793904,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033219286648794886,
      "loss": 4.9444,
      "mean_token_accuracy": 0.2257297784090042,
      "num_tokens": 96885985.0,
      "step": 46680
    },
    {
      "entropy": 5.686002969741821,
      "epoch": 4.234851233671988,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003321615722504659,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23789321929216384,
      "num_tokens": 96897512.0,
      "step": 46685
    },
    {
      "entropy": 5.692858505249023,
      "epoch": 4.235304789550073,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003321302768308671,
      "loss": 4.9527,
      "mean_token_accuracy": 0.21218942552804948,
      "num_tokens": 96907630.0,
      "step": 46690
    },
    {
      "entropy": 5.647123193740844,
      "epoch": 4.235758345428157,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033209898022979964,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23648007959127426,
      "num_tokens": 96917400.0,
      "step": 46695
    },
    {
      "entropy": 5.790036392211914,
      "epoch": 4.236211901306241,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003320676824479106,
      "loss": 4.9787,
      "mean_token_accuracy": 0.22079810351133347,
      "num_tokens": 96927700.0,
      "step": 46700
    },
    {
      "entropy": 5.776599597930908,
      "epoch": 4.236665457184325,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003320363834858474,
      "loss": 4.831,
      "mean_token_accuracy": 0.2237682819366455,
      "num_tokens": 96938925.0,
      "step": 46705
    },
    {
      "entropy": 5.677910804748535,
      "epoch": 4.23711901306241,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003320050833442573,
      "loss": 4.8104,
      "mean_token_accuracy": 0.22634731978178024,
      "num_tokens": 96948956.0,
      "step": 46710
    },
    {
      "entropy": 5.6868452548980715,
      "epoch": 4.237572568940493,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003319737820237875,
      "loss": 4.9116,
      "mean_token_accuracy": 0.22509346157312393,
      "num_tokens": 96958565.0,
      "step": 46715
    },
    {
      "entropy": 5.690572214126587,
      "epoch": 4.238026124818577,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033194247952508544,
      "loss": 4.8807,
      "mean_token_accuracy": 0.22418861091136932,
      "num_tokens": 96968261.0,
      "step": 46720
    },
    {
      "entropy": 5.771971035003662,
      "epoch": 4.238479680696662,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00033191117584879836,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22177067399024963,
      "num_tokens": 96978621.0,
      "step": 46725
    },
    {
      "entropy": 5.702822065353393,
      "epoch": 4.238933236574746,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003318798709955736,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22159192115068435,
      "num_tokens": 96989081.0,
      "step": 46730
    },
    {
      "entropy": 5.783305883407593,
      "epoch": 4.23938679245283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003318485649660586,
      "loss": 4.99,
      "mean_token_accuracy": 0.21565618813037873,
      "num_tokens": 97000518.0,
      "step": 46735
    },
    {
      "entropy": 5.710417556762695,
      "epoch": 4.239840348330914,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033181725776090093,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22827820032835006,
      "num_tokens": 97012038.0,
      "step": 46740
    },
    {
      "entropy": 5.71870436668396,
      "epoch": 4.240293904208999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033178594938074787,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2232270359992981,
      "num_tokens": 97022256.0,
      "step": 46745
    },
    {
      "entropy": 5.7228563785552975,
      "epoch": 4.240747460087083,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003317546398262468,
      "loss": 4.9573,
      "mean_token_accuracy": 0.22339436709880828,
      "num_tokens": 97033063.0,
      "step": 46750
    },
    {
      "entropy": 5.703003740310669,
      "epoch": 4.241201015965167,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003317233290980454,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22746617794036866,
      "num_tokens": 97043711.0,
      "step": 46755
    },
    {
      "entropy": 5.717815923690796,
      "epoch": 4.2416545718432515,
      "grad_norm": 1.4375,
      "learning_rate": 0.00033169201719679105,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23250004947185515,
      "num_tokens": 97053582.0,
      "step": 46760
    },
    {
      "entropy": 5.7520787715911865,
      "epoch": 4.242108127721336,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033166070412313133,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22584278136491776,
      "num_tokens": 97064535.0,
      "step": 46765
    },
    {
      "entropy": 5.696024227142334,
      "epoch": 4.242561683599419,
      "grad_norm": 1.375,
      "learning_rate": 0.0003316293898777138,
      "loss": 4.9174,
      "mean_token_accuracy": 0.21856116950511933,
      "num_tokens": 97074295.0,
      "step": 46770
    },
    {
      "entropy": 5.684445190429687,
      "epoch": 4.243015239477503,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033159807446118614,
      "loss": 4.8941,
      "mean_token_accuracy": 0.22101756185293198,
      "num_tokens": 97084781.0,
      "step": 46775
    },
    {
      "entropy": 5.806559944152832,
      "epoch": 4.243468795355588,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003315667578741958,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21913011819124223,
      "num_tokens": 97094848.0,
      "step": 46780
    },
    {
      "entropy": 5.76139349937439,
      "epoch": 4.243922351233672,
      "grad_norm": 1.25,
      "learning_rate": 0.0003315354401173905,
      "loss": 4.8567,
      "mean_token_accuracy": 0.22550999075174333,
      "num_tokens": 97104747.0,
      "step": 46785
    },
    {
      "entropy": 5.684535360336303,
      "epoch": 4.244375907111756,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003315041211914179,
      "loss": 4.8271,
      "mean_token_accuracy": 0.22302540093660356,
      "num_tokens": 97115599.0,
      "step": 46790
    },
    {
      "entropy": 5.706315898895264,
      "epoch": 4.24482946298984,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003314728010969257,
      "loss": 4.9843,
      "mean_token_accuracy": 0.22323342859745027,
      "num_tokens": 97127492.0,
      "step": 46795
    },
    {
      "entropy": 5.735858535766601,
      "epoch": 4.245283018867925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033144147983456154,
      "loss": 4.8937,
      "mean_token_accuracy": 0.21981571316719056,
      "num_tokens": 97138336.0,
      "step": 46800
    },
    {
      "entropy": 5.7102274894714355,
      "epoch": 4.245736574746009,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003314101574049732,
      "loss": 4.8462,
      "mean_token_accuracy": 0.23022533804178238,
      "num_tokens": 97149196.0,
      "step": 46805
    },
    {
      "entropy": 5.758026361465454,
      "epoch": 4.246190130624093,
      "grad_norm": 1.453125,
      "learning_rate": 0.00033137883380880847,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22046665251255035,
      "num_tokens": 97159622.0,
      "step": 46810
    },
    {
      "entropy": 5.782643556594849,
      "epoch": 4.2466436865021775,
      "grad_norm": 1.234375,
      "learning_rate": 0.000331347509046715,
      "loss": 4.9631,
      "mean_token_accuracy": 0.21455042958259582,
      "num_tokens": 97170149.0,
      "step": 46815
    },
    {
      "entropy": 5.774175596237183,
      "epoch": 4.247097242380261,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003313161831193408,
      "loss": 4.9014,
      "mean_token_accuracy": 0.21896244287490846,
      "num_tokens": 97180409.0,
      "step": 46820
    },
    {
      "entropy": 5.747155380249024,
      "epoch": 4.247550798258345,
      "grad_norm": 1.359375,
      "learning_rate": 0.00033128485602733354,
      "loss": 4.9177,
      "mean_token_accuracy": 0.2189839780330658,
      "num_tokens": 97190899.0,
      "step": 46825
    },
    {
      "entropy": 5.761643552780152,
      "epoch": 4.248004354136429,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003312535277713411,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22565484046936035,
      "num_tokens": 97200759.0,
      "step": 46830
    },
    {
      "entropy": 5.72773175239563,
      "epoch": 4.248457910014514,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033122219835201135,
      "loss": 4.825,
      "mean_token_accuracy": 0.23168253302574157,
      "num_tokens": 97211476.0,
      "step": 46835
    },
    {
      "entropy": 5.741878032684326,
      "epoch": 4.248911465892598,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003311908677699923,
      "loss": 4.9469,
      "mean_token_accuracy": 0.2210792824625969,
      "num_tokens": 97222374.0,
      "step": 46840
    },
    {
      "entropy": 5.727603721618652,
      "epoch": 4.249365021770682,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003311595360259317,
      "loss": 4.9007,
      "mean_token_accuracy": 0.22446949034929276,
      "num_tokens": 97233079.0,
      "step": 46845
    },
    {
      "entropy": 5.636947345733643,
      "epoch": 4.2498185776487665,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00033112820312047756,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23394824266433717,
      "num_tokens": 97243796.0,
      "step": 46850
    },
    {
      "entropy": 5.669175863265991,
      "epoch": 4.250272133526851,
      "grad_norm": 1.34375,
      "learning_rate": 0.00033109686905427794,
      "loss": 4.772,
      "mean_token_accuracy": 0.24018997102975845,
      "num_tokens": 97253341.0,
      "step": 46855
    },
    {
      "entropy": 5.796775245666504,
      "epoch": 4.250725689404935,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033106553382798075,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21450992822647094,
      "num_tokens": 97264244.0,
      "step": 46860
    },
    {
      "entropy": 5.682430601119995,
      "epoch": 4.251179245283019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033103419744223413,
      "loss": 4.8409,
      "mean_token_accuracy": 0.22392309457063675,
      "num_tokens": 97274239.0,
      "step": 46865
    },
    {
      "entropy": 5.6568714618682865,
      "epoch": 4.251632801161103,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003310028598976859,
      "loss": 4.7761,
      "mean_token_accuracy": 0.23683577477931977,
      "num_tokens": 97284709.0,
      "step": 46870
    },
    {
      "entropy": 5.664611768722534,
      "epoch": 4.252086357039187,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033097152119498425,
      "loss": 4.803,
      "mean_token_accuracy": 0.22921531796455383,
      "num_tokens": 97295262.0,
      "step": 46875
    },
    {
      "entropy": 5.717364931106568,
      "epoch": 4.252539912917271,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003309401813347774,
      "loss": 4.9187,
      "mean_token_accuracy": 0.21575379222631455,
      "num_tokens": 97306143.0,
      "step": 46880
    },
    {
      "entropy": 5.760208225250244,
      "epoch": 4.2529934687953554,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003309088403177132,
      "loss": 4.8199,
      "mean_token_accuracy": 0.2337322935461998,
      "num_tokens": 97316248.0,
      "step": 46885
    },
    {
      "entropy": 5.797776079177856,
      "epoch": 4.25344702467344,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033087749814444,
      "loss": 4.9743,
      "mean_token_accuracy": 0.21983991116285323,
      "num_tokens": 97327704.0,
      "step": 46890
    },
    {
      "entropy": 5.763378620147705,
      "epoch": 4.253900580551524,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033084615481560585,
      "loss": 4.9708,
      "mean_token_accuracy": 0.2128466174006462,
      "num_tokens": 97337817.0,
      "step": 46895
    },
    {
      "entropy": 5.777481317520142,
      "epoch": 4.254354136429608,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033081481033185903,
      "loss": 4.9569,
      "mean_token_accuracy": 0.21995104253292083,
      "num_tokens": 97348315.0,
      "step": 46900
    },
    {
      "entropy": 5.717471361160278,
      "epoch": 4.2548076923076925,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003307834646938477,
      "loss": 4.885,
      "mean_token_accuracy": 0.22637817561626433,
      "num_tokens": 97358842.0,
      "step": 46905
    },
    {
      "entropy": 5.805797672271728,
      "epoch": 4.255261248185777,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003307521179022201,
      "loss": 4.9884,
      "mean_token_accuracy": 0.2185544788837433,
      "num_tokens": 97369247.0,
      "step": 46910
    },
    {
      "entropy": 5.7030072689056395,
      "epoch": 4.255714804063861,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003307207699576246,
      "loss": 4.8551,
      "mean_token_accuracy": 0.22705576121807097,
      "num_tokens": 97379668.0,
      "step": 46915
    },
    {
      "entropy": 5.689864540100098,
      "epoch": 4.256168359941945,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033068942086070924,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2269763633608818,
      "num_tokens": 97389131.0,
      "step": 46920
    },
    {
      "entropy": 5.747761344909668,
      "epoch": 4.256621915820029,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033065807061212254,
      "loss": 4.9932,
      "mean_token_accuracy": 0.2127627581357956,
      "num_tokens": 97400583.0,
      "step": 46925
    },
    {
      "entropy": 5.746261262893677,
      "epoch": 4.257075471698113,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003306267192125128,
      "loss": 4.864,
      "mean_token_accuracy": 0.2233629122376442,
      "num_tokens": 97410828.0,
      "step": 46930
    },
    {
      "entropy": 5.750951862335205,
      "epoch": 4.257529027576197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003305953666625283,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21792038083076476,
      "num_tokens": 97421190.0,
      "step": 46935
    },
    {
      "entropy": 5.737555599212646,
      "epoch": 4.2579825834542815,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033056401296281733,
      "loss": 4.9478,
      "mean_token_accuracy": 0.21584450155496598,
      "num_tokens": 97432257.0,
      "step": 46940
    },
    {
      "entropy": 5.7182825088500975,
      "epoch": 4.258436139332366,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003305326581140286,
      "loss": 4.9284,
      "mean_token_accuracy": 0.21970686167478562,
      "num_tokens": 97442272.0,
      "step": 46945
    },
    {
      "entropy": 5.722885465621948,
      "epoch": 4.25888969521045,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00033050130211681017,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22780009359121323,
      "num_tokens": 97451462.0,
      "step": 46950
    },
    {
      "entropy": 5.683341646194458,
      "epoch": 4.259343251088534,
      "grad_norm": 1.3125,
      "learning_rate": 0.00033046994497181077,
      "loss": 4.7794,
      "mean_token_accuracy": 0.22897925674915315,
      "num_tokens": 97461579.0,
      "step": 46955
    },
    {
      "entropy": 5.790071201324463,
      "epoch": 4.2597968069666186,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00033043858667967875,
      "loss": 4.9623,
      "mean_token_accuracy": 0.21994038224220275,
      "num_tokens": 97472892.0,
      "step": 46960
    },
    {
      "entropy": 5.694077253341675,
      "epoch": 4.260250362844703,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033040722724106257,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2230777159333229,
      "num_tokens": 97484278.0,
      "step": 46965
    },
    {
      "entropy": 5.727041864395142,
      "epoch": 4.260703918722786,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00033037586665661094,
      "loss": 4.907,
      "mean_token_accuracy": 0.21800587177276612,
      "num_tokens": 97494127.0,
      "step": 46970
    },
    {
      "entropy": 5.677419805526734,
      "epoch": 4.2611574746008705,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003303445049269722,
      "loss": 4.8108,
      "mean_token_accuracy": 0.22843805849552154,
      "num_tokens": 97504213.0,
      "step": 46975
    },
    {
      "entropy": 5.762578916549683,
      "epoch": 4.261611030478955,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00033031314205279503,
      "loss": 4.9467,
      "mean_token_accuracy": 0.22187138646841048,
      "num_tokens": 97515242.0,
      "step": 46980
    },
    {
      "entropy": 5.790261602401733,
      "epoch": 4.262064586357039,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033028177803472793,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2273780271410942,
      "num_tokens": 97525956.0,
      "step": 46985
    },
    {
      "entropy": 5.688992595672607,
      "epoch": 4.262518142235123,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003302504128734196,
      "loss": 4.8067,
      "mean_token_accuracy": 0.23281866014003755,
      "num_tokens": 97535973.0,
      "step": 46990
    },
    {
      "entropy": 5.666906929016113,
      "epoch": 4.2629716981132075,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033021904656951867,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2260513037443161,
      "num_tokens": 97547580.0,
      "step": 46995
    },
    {
      "entropy": 5.721551418304443,
      "epoch": 4.263425253991292,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033018767912367376,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22469193041324614,
      "num_tokens": 97558050.0,
      "step": 47000
    },
    {
      "entropy": 5.763391017913818,
      "epoch": 4.263878809869376,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00033015631053653357,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2354947179555893,
      "num_tokens": 97567997.0,
      "step": 47005
    },
    {
      "entropy": 5.690925788879395,
      "epoch": 4.26433236574746,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003301249408087469,
      "loss": 4.844,
      "mean_token_accuracy": 0.23155179768800735,
      "num_tokens": 97577220.0,
      "step": 47010
    },
    {
      "entropy": 5.698481845855713,
      "epoch": 4.264785921625545,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033009356994096237,
      "loss": 4.876,
      "mean_token_accuracy": 0.22941813319921495,
      "num_tokens": 97588421.0,
      "step": 47015
    },
    {
      "entropy": 5.649937963485717,
      "epoch": 4.265239477503629,
      "grad_norm": 1.25,
      "learning_rate": 0.00033006219793382867,
      "loss": 4.7875,
      "mean_token_accuracy": 0.23168975710868836,
      "num_tokens": 97598624.0,
      "step": 47020
    },
    {
      "entropy": 5.729764413833618,
      "epoch": 4.265693033381712,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003300308247879948,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22202031016349794,
      "num_tokens": 97609128.0,
      "step": 47025
    },
    {
      "entropy": 5.744069147109985,
      "epoch": 4.2661465892597965,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003299994505041094,
      "loss": 4.8985,
      "mean_token_accuracy": 0.23035580962896346,
      "num_tokens": 97619549.0,
      "step": 47030
    },
    {
      "entropy": 5.681211471557617,
      "epoch": 4.266600145137881,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032996807508282134,
      "loss": 4.8792,
      "mean_token_accuracy": 0.22345985174179078,
      "num_tokens": 97630184.0,
      "step": 47035
    },
    {
      "entropy": 5.68397011756897,
      "epoch": 4.267053701015965,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032993669852477954,
      "loss": 4.8659,
      "mean_token_accuracy": 0.2237844169139862,
      "num_tokens": 97640569.0,
      "step": 47040
    },
    {
      "entropy": 5.685236406326294,
      "epoch": 4.267507256894049,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003299053208306328,
      "loss": 4.8351,
      "mean_token_accuracy": 0.2269110530614853,
      "num_tokens": 97651006.0,
      "step": 47045
    },
    {
      "entropy": 5.714099168777466,
      "epoch": 4.267960812772134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003298739420010299,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22519755959510804,
      "num_tokens": 97661567.0,
      "step": 47050
    },
    {
      "entropy": 5.7084578514099125,
      "epoch": 4.268414368650218,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003298425620366201,
      "loss": 4.8527,
      "mean_token_accuracy": 0.2324652522802353,
      "num_tokens": 97673058.0,
      "step": 47055
    },
    {
      "entropy": 5.833694553375244,
      "epoch": 4.268867924528302,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003298111809380521,
      "loss": 5.0148,
      "mean_token_accuracy": 0.21293772011995316,
      "num_tokens": 97684509.0,
      "step": 47060
    },
    {
      "entropy": 5.786008548736572,
      "epoch": 4.269321480406386,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032977979870597477,
      "loss": 4.891,
      "mean_token_accuracy": 0.22159728705883025,
      "num_tokens": 97693954.0,
      "step": 47065
    },
    {
      "entropy": 5.729776382446289,
      "epoch": 4.269775036284471,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003297484153410374,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2310206934809685,
      "num_tokens": 97704345.0,
      "step": 47070
    },
    {
      "entropy": 5.6537864208221436,
      "epoch": 4.270228592162554,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032971703084388885,
      "loss": 4.7496,
      "mean_token_accuracy": 0.24231005609035491,
      "num_tokens": 97714857.0,
      "step": 47075
    },
    {
      "entropy": 5.717009258270264,
      "epoch": 4.270682148040638,
      "grad_norm": 1.265625,
      "learning_rate": 0.000329685645215178,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22007610499858857,
      "num_tokens": 97725921.0,
      "step": 47080
    },
    {
      "entropy": 5.697414398193359,
      "epoch": 4.2711357039187225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032965425845555413,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23545854091644286,
      "num_tokens": 97735361.0,
      "step": 47085
    },
    {
      "entropy": 5.694683313369751,
      "epoch": 4.271589259796807,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032962287056566633,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22214128226041793,
      "num_tokens": 97745915.0,
      "step": 47090
    },
    {
      "entropy": 5.7234491348266605,
      "epoch": 4.272042815674891,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003295914815461637,
      "loss": 4.8672,
      "mean_token_accuracy": 0.232294300198555,
      "num_tokens": 97755663.0,
      "step": 47095
    },
    {
      "entropy": 5.7602471828460695,
      "epoch": 4.272496371552975,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032956009139769514,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21778757721185685,
      "num_tokens": 97766300.0,
      "step": 47100
    },
    {
      "entropy": 5.7362018585205075,
      "epoch": 4.27294992743106,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032952870012091013,
      "loss": 4.8851,
      "mean_token_accuracy": 0.22367059737443923,
      "num_tokens": 97776584.0,
      "step": 47105
    },
    {
      "entropy": 5.686241865158081,
      "epoch": 4.273403483309144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032949730771645767,
      "loss": 4.819,
      "mean_token_accuracy": 0.22706236839294433,
      "num_tokens": 97786230.0,
      "step": 47110
    },
    {
      "entropy": 5.727143383026123,
      "epoch": 4.273857039187228,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000329465914184987,
      "loss": 4.9347,
      "mean_token_accuracy": 0.22130536735057832,
      "num_tokens": 97796671.0,
      "step": 47115
    },
    {
      "entropy": 5.663255500793457,
      "epoch": 4.274310595065312,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003294345195271473,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22972838580608368,
      "num_tokens": 97807569.0,
      "step": 47120
    },
    {
      "entropy": 5.793165779113769,
      "epoch": 4.274764150943396,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032940312374358794,
      "loss": 5.0029,
      "mean_token_accuracy": 0.21767988950014114,
      "num_tokens": 97818680.0,
      "step": 47125
    },
    {
      "entropy": 5.7469761848449705,
      "epoch": 4.27521770682148,
      "grad_norm": 1.4375,
      "learning_rate": 0.000329371726834958,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22748010605573654,
      "num_tokens": 97828955.0,
      "step": 47130
    },
    {
      "entropy": 5.70758023262024,
      "epoch": 4.275671262699564,
      "grad_norm": 1.265625,
      "learning_rate": 0.000329340328801907,
      "loss": 4.8677,
      "mean_token_accuracy": 0.23439973890781401,
      "num_tokens": 97839000.0,
      "step": 47135
    },
    {
      "entropy": 5.7121889114379885,
      "epoch": 4.276124818577649,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003293089296450841,
      "loss": 4.9334,
      "mean_token_accuracy": 0.21423581540584563,
      "num_tokens": 97850140.0,
      "step": 47140
    },
    {
      "entropy": 5.682531690597534,
      "epoch": 4.276578374455733,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003292775293651388,
      "loss": 4.7897,
      "mean_token_accuracy": 0.22980619817972184,
      "num_tokens": 97860887.0,
      "step": 47145
    },
    {
      "entropy": 5.734488296508789,
      "epoch": 4.277031930333817,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003292461279627203,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22494918256998062,
      "num_tokens": 97871190.0,
      "step": 47150
    },
    {
      "entropy": 5.705101156234742,
      "epoch": 4.277485486211901,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032921472543847804,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23014242351055145,
      "num_tokens": 97881479.0,
      "step": 47155
    },
    {
      "entropy": 5.6697064399719235,
      "epoch": 4.277939042089986,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003291833217930614,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22658446431159973,
      "num_tokens": 97892340.0,
      "step": 47160
    },
    {
      "entropy": 5.732178544998169,
      "epoch": 4.27839259796807,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032915191702711995,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22466536462306977,
      "num_tokens": 97902882.0,
      "step": 47165
    },
    {
      "entropy": 5.682470178604126,
      "epoch": 4.278846153846154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032912051114130297,
      "loss": 4.7816,
      "mean_token_accuracy": 0.2334643632173538,
      "num_tokens": 97913187.0,
      "step": 47170
    },
    {
      "entropy": 5.669349670410156,
      "epoch": 4.2792997097242385,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00032908910413626,
      "loss": 4.8131,
      "mean_token_accuracy": 0.23599857091903687,
      "num_tokens": 97922434.0,
      "step": 47175
    },
    {
      "entropy": 5.603707885742187,
      "epoch": 4.279753265602322,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003290576960126407,
      "loss": 4.7419,
      "mean_token_accuracy": 0.23732489496469497,
      "num_tokens": 97933020.0,
      "step": 47180
    },
    {
      "entropy": 5.65603494644165,
      "epoch": 4.280206821480406,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032902628677109434,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22634703069925308,
      "num_tokens": 97944103.0,
      "step": 47185
    },
    {
      "entropy": 5.734215116500854,
      "epoch": 4.28066037735849,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003289948764122707,
      "loss": 4.8615,
      "mean_token_accuracy": 0.23005225509405136,
      "num_tokens": 97953692.0,
      "step": 47190
    },
    {
      "entropy": 5.741902589797974,
      "epoch": 4.281113933236575,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003289634649368193,
      "loss": 4.88,
      "mean_token_accuracy": 0.2304357573390007,
      "num_tokens": 97963592.0,
      "step": 47195
    },
    {
      "entropy": 5.750173568725586,
      "epoch": 4.281567489114659,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003289320523453895,
      "loss": 5.0446,
      "mean_token_accuracy": 0.2159873977303505,
      "num_tokens": 97974092.0,
      "step": 47200
    },
    {
      "entropy": 5.657929468154907,
      "epoch": 4.282021044992743,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032890063863863126,
      "loss": 4.7749,
      "mean_token_accuracy": 0.23314350545406343,
      "num_tokens": 97984175.0,
      "step": 47205
    },
    {
      "entropy": 5.701040267944336,
      "epoch": 4.282474600870827,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003288692238171941,
      "loss": 4.898,
      "mean_token_accuracy": 0.24038833230733872,
      "num_tokens": 97994993.0,
      "step": 47210
    },
    {
      "entropy": 5.723231554031372,
      "epoch": 4.282928156748912,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032883780788172756,
      "loss": 4.8935,
      "mean_token_accuracy": 0.2186749517917633,
      "num_tokens": 98005892.0,
      "step": 47215
    },
    {
      "entropy": 5.6950726985931395,
      "epoch": 4.283381712626996,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003288063908328814,
      "loss": 4.8123,
      "mean_token_accuracy": 0.2300744742155075,
      "num_tokens": 98016726.0,
      "step": 47220
    },
    {
      "entropy": 5.767842435836792,
      "epoch": 4.28383526850508,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003287749726713055,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2261524200439453,
      "num_tokens": 98028500.0,
      "step": 47225
    },
    {
      "entropy": 5.7863058090209964,
      "epoch": 4.284288824383164,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003287435533976494,
      "loss": 4.9846,
      "mean_token_accuracy": 0.21539334058761597,
      "num_tokens": 98039304.0,
      "step": 47230
    },
    {
      "entropy": 5.725332355499267,
      "epoch": 4.284742380261248,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003287121330125629,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22087255418300628,
      "num_tokens": 98049448.0,
      "step": 47235
    },
    {
      "entropy": 5.71489372253418,
      "epoch": 4.285195936139332,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032868071151669585,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2340805187821388,
      "num_tokens": 98059532.0,
      "step": 47240
    },
    {
      "entropy": 5.668030643463135,
      "epoch": 4.285649492017416,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000328649288910698,
      "loss": 4.7926,
      "mean_token_accuracy": 0.23101378381252288,
      "num_tokens": 98069385.0,
      "step": 47245
    },
    {
      "entropy": 5.7362464427947994,
      "epoch": 4.286103047895501,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003286178651952192,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2297645777463913,
      "num_tokens": 98079354.0,
      "step": 47250
    },
    {
      "entropy": 5.753766870498657,
      "epoch": 4.286556603773585,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032858644037090924,
      "loss": 4.977,
      "mean_token_accuracy": 0.21726818531751632,
      "num_tokens": 98090864.0,
      "step": 47255
    },
    {
      "entropy": 5.70212836265564,
      "epoch": 4.287010159651669,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003285550144384181,
      "loss": 4.835,
      "mean_token_accuracy": 0.2268703892827034,
      "num_tokens": 98101515.0,
      "step": 47260
    },
    {
      "entropy": 5.741011142730713,
      "epoch": 4.2874637155297535,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032852358739839555,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22431121319532393,
      "num_tokens": 98111786.0,
      "step": 47265
    },
    {
      "entropy": 5.6865620613098145,
      "epoch": 4.287917271407838,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032849215925149155,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23117925822734833,
      "num_tokens": 98121732.0,
      "step": 47270
    },
    {
      "entropy": 5.7583647727966305,
      "epoch": 4.288370827285922,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003284607299983561,
      "loss": 4.9906,
      "mean_token_accuracy": 0.2214345693588257,
      "num_tokens": 98131417.0,
      "step": 47275
    },
    {
      "entropy": 5.75811800956726,
      "epoch": 4.288824383164005,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032842929963963915,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2263333022594452,
      "num_tokens": 98141664.0,
      "step": 47280
    },
    {
      "entropy": 5.712303829193115,
      "epoch": 4.28927793904209,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032839786817599065,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2232130214571953,
      "num_tokens": 98150959.0,
      "step": 47285
    },
    {
      "entropy": 5.723396825790405,
      "epoch": 4.289731494920174,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003283664356080606,
      "loss": 4.9615,
      "mean_token_accuracy": 0.22543786019086837,
      "num_tokens": 98161957.0,
      "step": 47290
    },
    {
      "entropy": 5.6535321235656735,
      "epoch": 4.290185050798258,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032833500193649916,
      "loss": 4.8093,
      "mean_token_accuracy": 0.22701679915189743,
      "num_tokens": 98172053.0,
      "step": 47295
    },
    {
      "entropy": 5.723155784606933,
      "epoch": 4.2906386066763424,
      "grad_norm": 1.375,
      "learning_rate": 0.00032830356716195624,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2278629556298256,
      "num_tokens": 98182074.0,
      "step": 47300
    },
    {
      "entropy": 5.789839887619019,
      "epoch": 4.291092162554427,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003282721312850819,
      "loss": 4.884,
      "mean_token_accuracy": 0.22491135448217392,
      "num_tokens": 98192591.0,
      "step": 47305
    },
    {
      "entropy": 5.694799137115479,
      "epoch": 4.291545718432511,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003282406943065264,
      "loss": 4.7918,
      "mean_token_accuracy": 0.2349269136786461,
      "num_tokens": 98203426.0,
      "step": 47310
    },
    {
      "entropy": 5.697488451004029,
      "epoch": 4.291999274310595,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032820925622693977,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22565727233886718,
      "num_tokens": 98214342.0,
      "step": 47315
    },
    {
      "entropy": 5.742821788787841,
      "epoch": 4.2924528301886795,
      "grad_norm": 1.25,
      "learning_rate": 0.00032817781704697204,
      "loss": 4.8438,
      "mean_token_accuracy": 0.22346201688051223,
      "num_tokens": 98224155.0,
      "step": 47320
    },
    {
      "entropy": 5.7035619735717775,
      "epoch": 4.292906386066764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003281463767672736,
      "loss": 4.829,
      "mean_token_accuracy": 0.23137144446372987,
      "num_tokens": 98233613.0,
      "step": 47325
    },
    {
      "entropy": 5.70261516571045,
      "epoch": 4.293359941944848,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003281149353884946,
      "loss": 4.8609,
      "mean_token_accuracy": 0.23343997150659562,
      "num_tokens": 98243117.0,
      "step": 47330
    },
    {
      "entropy": 5.7136945724487305,
      "epoch": 4.293813497822931,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003280834929112851,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2285476416349411,
      "num_tokens": 98253619.0,
      "step": 47335
    },
    {
      "entropy": 5.696676635742188,
      "epoch": 4.294267053701016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003280520493362955,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22670871019363403,
      "num_tokens": 98264221.0,
      "step": 47340
    },
    {
      "entropy": 5.691981172561645,
      "epoch": 4.2947206095791,
      "grad_norm": 1.25,
      "learning_rate": 0.0003280206046641759,
      "loss": 4.8188,
      "mean_token_accuracy": 0.2265792340040207,
      "num_tokens": 98275530.0,
      "step": 47345
    },
    {
      "entropy": 5.7876814842224125,
      "epoch": 4.295174165457184,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032798915889557687,
      "loss": 4.9507,
      "mean_token_accuracy": 0.22129734754562377,
      "num_tokens": 98285937.0,
      "step": 47350
    },
    {
      "entropy": 5.743341445922852,
      "epoch": 4.2956277213352685,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003279577120311483,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22573418021202088,
      "num_tokens": 98295690.0,
      "step": 47355
    },
    {
      "entropy": 5.655327796936035,
      "epoch": 4.296081277213353,
      "grad_norm": 1.40625,
      "learning_rate": 0.000327926264071541,
      "loss": 4.8476,
      "mean_token_accuracy": 0.2268459051847458,
      "num_tokens": 98305073.0,
      "step": 47360
    },
    {
      "entropy": 5.712704992294311,
      "epoch": 4.296534833091437,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003278948150174049,
      "loss": 4.946,
      "mean_token_accuracy": 0.21838900595903396,
      "num_tokens": 98316008.0,
      "step": 47365
    },
    {
      "entropy": 5.699159240722656,
      "epoch": 4.296988388969521,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003278633648693906,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22644217759370805,
      "num_tokens": 98327032.0,
      "step": 47370
    },
    {
      "entropy": 5.770933151245117,
      "epoch": 4.2974419448476056,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003278319136281484,
      "loss": 4.8778,
      "mean_token_accuracy": 0.23105017691850663,
      "num_tokens": 98336085.0,
      "step": 47375
    },
    {
      "entropy": 5.749867153167725,
      "epoch": 4.297895500725689,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032780046129432885,
      "loss": 4.888,
      "mean_token_accuracy": 0.23064714223146437,
      "num_tokens": 98346180.0,
      "step": 47380
    },
    {
      "entropy": 5.6983458518981935,
      "epoch": 4.298349056603773,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032776900786858214,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2288931667804718,
      "num_tokens": 98356001.0,
      "step": 47385
    },
    {
      "entropy": 5.750356578826905,
      "epoch": 4.2988026124818575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000327737553351559,
      "loss": 4.9619,
      "mean_token_accuracy": 0.22224070280790328,
      "num_tokens": 98366744.0,
      "step": 47390
    },
    {
      "entropy": 5.664729356765747,
      "epoch": 4.299256168359942,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032770609774390985,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23239726126194,
      "num_tokens": 98377350.0,
      "step": 47395
    },
    {
      "entropy": 5.669324731826782,
      "epoch": 4.299709724238026,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003276746410462852,
      "loss": 4.8618,
      "mean_token_accuracy": 0.23059386014938354,
      "num_tokens": 98387926.0,
      "step": 47400
    },
    {
      "entropy": 5.776981019973755,
      "epoch": 4.30016328011611,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032764318325933546,
      "loss": 4.9345,
      "mean_token_accuracy": 0.22492991983890534,
      "num_tokens": 98397117.0,
      "step": 47405
    },
    {
      "entropy": 5.756745386123657,
      "epoch": 4.3006168359941945,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003276117243837113,
      "loss": 4.9016,
      "mean_token_accuracy": 0.22146118432283401,
      "num_tokens": 98407983.0,
      "step": 47410
    },
    {
      "entropy": 5.70599479675293,
      "epoch": 4.301070391872279,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003275802644200633,
      "loss": 4.901,
      "mean_token_accuracy": 0.2178119793534279,
      "num_tokens": 98419940.0,
      "step": 47415
    },
    {
      "entropy": 5.6656585216522215,
      "epoch": 4.301523947750363,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000327548803369042,
      "loss": 4.8245,
      "mean_token_accuracy": 0.2306269034743309,
      "num_tokens": 98430487.0,
      "step": 47420
    },
    {
      "entropy": 5.702292633056641,
      "epoch": 4.301977503628447,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032751734123129814,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2365352228283882,
      "num_tokens": 98440231.0,
      "step": 47425
    },
    {
      "entropy": 5.741506910324096,
      "epoch": 4.302431059506532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003274858780074822,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22336503863334656,
      "num_tokens": 98449741.0,
      "step": 47430
    },
    {
      "entropy": 5.69879264831543,
      "epoch": 4.302884615384615,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032745441369824485,
      "loss": 4.801,
      "mean_token_accuracy": 0.23827111274003981,
      "num_tokens": 98461068.0,
      "step": 47435
    },
    {
      "entropy": 5.684864187240601,
      "epoch": 4.303338171262699,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032742294830423696,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22948870360851287,
      "num_tokens": 98472585.0,
      "step": 47440
    },
    {
      "entropy": 5.7274088859558105,
      "epoch": 4.3037917271407835,
      "grad_norm": 1.25,
      "learning_rate": 0.0003273914818261092,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22651124894618987,
      "num_tokens": 98482874.0,
      "step": 47445
    },
    {
      "entropy": 5.715044927597046,
      "epoch": 4.304245283018868,
      "grad_norm": 1.25,
      "learning_rate": 0.00032736001426451216,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22375308573246003,
      "num_tokens": 98492995.0,
      "step": 47450
    },
    {
      "entropy": 5.798288488388062,
      "epoch": 4.304698838896952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003273285456200966,
      "loss": 4.9513,
      "mean_token_accuracy": 0.223180890083313,
      "num_tokens": 98503965.0,
      "step": 47455
    },
    {
      "entropy": 5.730708503723145,
      "epoch": 4.305152394775036,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003272970758935135,
      "loss": 4.8809,
      "mean_token_accuracy": 0.22410071343183519,
      "num_tokens": 98515584.0,
      "step": 47460
    },
    {
      "entropy": 5.744846153259277,
      "epoch": 4.305605950653121,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032726560508541356,
      "loss": 4.9204,
      "mean_token_accuracy": 0.22682396620512008,
      "num_tokens": 98527134.0,
      "step": 47465
    },
    {
      "entropy": 5.751050233840942,
      "epoch": 4.306059506531205,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032723413319644756,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22252132147550582,
      "num_tokens": 98537189.0,
      "step": 47470
    },
    {
      "entropy": 5.72584867477417,
      "epoch": 4.306513062409289,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003272026602272663,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22800536304712296,
      "num_tokens": 98547803.0,
      "step": 47475
    },
    {
      "entropy": 5.7046489238739015,
      "epoch": 4.306966618287373,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032717118617852086,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22359438240528107,
      "num_tokens": 98558332.0,
      "step": 47480
    },
    {
      "entropy": 5.74920768737793,
      "epoch": 4.307420174165458,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003271397110508619,
      "loss": 4.985,
      "mean_token_accuracy": 0.21742497384548187,
      "num_tokens": 98569046.0,
      "step": 47485
    },
    {
      "entropy": 5.694894647598266,
      "epoch": 4.307873730043541,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003271082348449405,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22749540954828262,
      "num_tokens": 98579358.0,
      "step": 47490
    },
    {
      "entropy": 5.703712511062622,
      "epoch": 4.308327285921625,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003270767575614074,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22223605513572692,
      "num_tokens": 98589224.0,
      "step": 47495
    },
    {
      "entropy": 5.741110134124756,
      "epoch": 4.3087808417997095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003270452792009138,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22975554019212724,
      "num_tokens": 98598954.0,
      "step": 47500
    },
    {
      "entropy": 5.698847770690918,
      "epoch": 4.309234397677794,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032701379976411043,
      "loss": 4.8087,
      "mean_token_accuracy": 0.2242337003350258,
      "num_tokens": 98608813.0,
      "step": 47505
    },
    {
      "entropy": 5.711156225204467,
      "epoch": 4.309687953555878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003269823192516486,
      "loss": 4.8561,
      "mean_token_accuracy": 0.22608546465635299,
      "num_tokens": 98618746.0,
      "step": 47510
    },
    {
      "entropy": 5.730131387710571,
      "epoch": 4.310141509433962,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032695083766417913,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2333596333861351,
      "num_tokens": 98628804.0,
      "step": 47515
    },
    {
      "entropy": 5.708068037033081,
      "epoch": 4.310595065312047,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000326919355002353,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23179402202367783,
      "num_tokens": 98640056.0,
      "step": 47520
    },
    {
      "entropy": 5.681007337570191,
      "epoch": 4.311048621190131,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00032688787126682143,
      "loss": 4.829,
      "mean_token_accuracy": 0.23254396617412568,
      "num_tokens": 98651042.0,
      "step": 47525
    },
    {
      "entropy": 5.7097352027893065,
      "epoch": 4.311502177068215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003268563864582354,
      "loss": 4.8478,
      "mean_token_accuracy": 0.2293147101998329,
      "num_tokens": 98661405.0,
      "step": 47530
    },
    {
      "entropy": 5.7599334716796875,
      "epoch": 4.3119557329462985,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032682490057724616,
      "loss": 4.9438,
      "mean_token_accuracy": 0.22137796133756638,
      "num_tokens": 98672323.0,
      "step": 47535
    },
    {
      "entropy": 5.7729079246521,
      "epoch": 4.312409288824383,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003267934136245047,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2238161489367485,
      "num_tokens": 98681431.0,
      "step": 47540
    },
    {
      "entropy": 5.6334991455078125,
      "epoch": 4.312862844702467,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003267619256006623,
      "loss": 4.8025,
      "mean_token_accuracy": 0.2418549880385399,
      "num_tokens": 98692345.0,
      "step": 47545
    },
    {
      "entropy": 5.696433019638062,
      "epoch": 4.313316400580551,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003267304365063701,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22480279952287674,
      "num_tokens": 98703425.0,
      "step": 47550
    },
    {
      "entropy": 5.727588796615601,
      "epoch": 4.313769956458636,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003266989463422793,
      "loss": 4.8884,
      "mean_token_accuracy": 0.2285721406340599,
      "num_tokens": 98714193.0,
      "step": 47555
    },
    {
      "entropy": 5.782414674758911,
      "epoch": 4.31422351233672,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032666745510904105,
      "loss": 4.8611,
      "mean_token_accuracy": 0.23327391743659973,
      "num_tokens": 98723550.0,
      "step": 47560
    },
    {
      "entropy": 5.71943302154541,
      "epoch": 4.314677068214804,
      "grad_norm": 1.375,
      "learning_rate": 0.00032663596280730673,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2234220489859581,
      "num_tokens": 98734075.0,
      "step": 47565
    },
    {
      "entropy": 5.691188764572144,
      "epoch": 4.315130624092888,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032660446943772755,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22979795932769775,
      "num_tokens": 98745447.0,
      "step": 47570
    },
    {
      "entropy": 5.738022422790527,
      "epoch": 4.315584179970973,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003265729750009547,
      "loss": 4.9491,
      "mean_token_accuracy": 0.22035609632730485,
      "num_tokens": 98756169.0,
      "step": 47575
    },
    {
      "entropy": 5.741964864730835,
      "epoch": 4.316037735849057,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003265414794976397,
      "loss": 4.926,
      "mean_token_accuracy": 0.21834061443805694,
      "num_tokens": 98767112.0,
      "step": 47580
    },
    {
      "entropy": 5.767176723480224,
      "epoch": 4.316491291727141,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003265099829284337,
      "loss": 4.9325,
      "mean_token_accuracy": 0.2247194916009903,
      "num_tokens": 98777533.0,
      "step": 47585
    },
    {
      "entropy": 5.7947431087493895,
      "epoch": 4.316944847605225,
      "grad_norm": 1.25,
      "learning_rate": 0.0003264784852939881,
      "loss": 4.9357,
      "mean_token_accuracy": 0.2207162544131279,
      "num_tokens": 98788101.0,
      "step": 47590
    },
    {
      "entropy": 5.820633172988892,
      "epoch": 4.317398403483309,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032644698659495437,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22303320169448854,
      "num_tokens": 98799438.0,
      "step": 47595
    },
    {
      "entropy": 5.681515884399414,
      "epoch": 4.317851959361393,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032641548683198384,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22426000237464905,
      "num_tokens": 98810325.0,
      "step": 47600
    },
    {
      "entropy": 5.648531246185303,
      "epoch": 4.318305515239477,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003263839860057279,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22797728776931764,
      "num_tokens": 98821897.0,
      "step": 47605
    },
    {
      "entropy": 5.729699659347534,
      "epoch": 4.318759071117562,
      "grad_norm": 1.390625,
      "learning_rate": 0.000326352484116838,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22475551068782806,
      "num_tokens": 98833338.0,
      "step": 47610
    },
    {
      "entropy": 5.758721494674683,
      "epoch": 4.319212626995646,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032632098116596576,
      "loss": 4.8888,
      "mean_token_accuracy": 0.22402835339307786,
      "num_tokens": 98844308.0,
      "step": 47615
    },
    {
      "entropy": 5.727911376953125,
      "epoch": 4.31966618287373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032628947715376247,
      "loss": 4.8811,
      "mean_token_accuracy": 0.22754756808280946,
      "num_tokens": 98854491.0,
      "step": 47620
    },
    {
      "entropy": 5.751237392425537,
      "epoch": 4.320119738751814,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003262579720808797,
      "loss": 4.9302,
      "mean_token_accuracy": 0.21626027971506118,
      "num_tokens": 98864702.0,
      "step": 47625
    },
    {
      "entropy": 5.708958482742309,
      "epoch": 4.320573294629899,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032622646594796905,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22711183577775956,
      "num_tokens": 98874682.0,
      "step": 47630
    },
    {
      "entropy": 5.7701669216156,
      "epoch": 4.321026850507983,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003261949587556821,
      "loss": 4.9797,
      "mean_token_accuracy": 0.2161133259534836,
      "num_tokens": 98885130.0,
      "step": 47635
    },
    {
      "entropy": 5.704041051864624,
      "epoch": 4.321480406386067,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032616345050467016,
      "loss": 4.8834,
      "mean_token_accuracy": 0.2269508048892021,
      "num_tokens": 98896398.0,
      "step": 47640
    },
    {
      "entropy": 5.671356821060181,
      "epoch": 4.321933962264151,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003261319411955852,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23080590963363648,
      "num_tokens": 98906571.0,
      "step": 47645
    },
    {
      "entropy": 5.692090702056885,
      "epoch": 4.322387518142235,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003261004308290786,
      "loss": 4.8185,
      "mean_token_accuracy": 0.2323712855577469,
      "num_tokens": 98916666.0,
      "step": 47650
    },
    {
      "entropy": 5.752804088592529,
      "epoch": 4.322841074020319,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003260689194058021,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22465049773454665,
      "num_tokens": 98926599.0,
      "step": 47655
    },
    {
      "entropy": 5.676303243637085,
      "epoch": 4.323294629898403,
      "grad_norm": 1.421875,
      "learning_rate": 0.00032603740692640733,
      "loss": 4.8154,
      "mean_token_accuracy": 0.2302648216485977,
      "num_tokens": 98936512.0,
      "step": 47660
    },
    {
      "entropy": 5.759439516067505,
      "epoch": 4.323748185776488,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003260058933915459,
      "loss": 4.9876,
      "mean_token_accuracy": 0.2114409923553467,
      "num_tokens": 98946604.0,
      "step": 47665
    },
    {
      "entropy": 5.7701033592224125,
      "epoch": 4.324201741654572,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003259743788018697,
      "loss": 4.9024,
      "mean_token_accuracy": 0.21640635281801224,
      "num_tokens": 98956257.0,
      "step": 47670
    },
    {
      "entropy": 5.766636991500855,
      "epoch": 4.324655297532656,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032594286315803026,
      "loss": 4.9335,
      "mean_token_accuracy": 0.22376431077718734,
      "num_tokens": 98966346.0,
      "step": 47675
    },
    {
      "entropy": 5.718514823913575,
      "epoch": 4.3251088534107405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003259113464606795,
      "loss": 4.8333,
      "mean_token_accuracy": 0.22950637191534043,
      "num_tokens": 98977410.0,
      "step": 47680
    },
    {
      "entropy": 5.779750013351441,
      "epoch": 4.325562409288825,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003258798287104691,
      "loss": 4.9315,
      "mean_token_accuracy": 0.2258617624640465,
      "num_tokens": 98987488.0,
      "step": 47685
    },
    {
      "entropy": 5.704332685470581,
      "epoch": 4.326015965166908,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032584830990805094,
      "loss": 4.81,
      "mean_token_accuracy": 0.23527394831180573,
      "num_tokens": 98996792.0,
      "step": 47690
    },
    {
      "entropy": 5.715037536621094,
      "epoch": 4.326469521044992,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032581679005407667,
      "loss": 4.801,
      "mean_token_accuracy": 0.23227770626544952,
      "num_tokens": 99007425.0,
      "step": 47695
    },
    {
      "entropy": 5.690401697158814,
      "epoch": 4.326923076923077,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003257852691491983,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22581492215394974,
      "num_tokens": 99018275.0,
      "step": 47700
    },
    {
      "entropy": 5.742451143264771,
      "epoch": 4.327376632801161,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032575374719406755,
      "loss": 4.9206,
      "mean_token_accuracy": 0.22387272268533706,
      "num_tokens": 99028853.0,
      "step": 47705
    },
    {
      "entropy": 5.757655763626099,
      "epoch": 4.327830188679245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032572222418933644,
      "loss": 4.9019,
      "mean_token_accuracy": 0.21916379779577255,
      "num_tokens": 99039423.0,
      "step": 47710
    },
    {
      "entropy": 5.739642667770386,
      "epoch": 4.3282837445573294,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003256907001356568,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21374572217464446,
      "num_tokens": 99050041.0,
      "step": 47715
    },
    {
      "entropy": 5.720286893844604,
      "epoch": 4.328737300435414,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003256591750336806,
      "loss": 4.9145,
      "mean_token_accuracy": 0.22291525304317475,
      "num_tokens": 99062050.0,
      "step": 47720
    },
    {
      "entropy": 5.640514373779297,
      "epoch": 4.329190856313498,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003256276488840596,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2281416654586792,
      "num_tokens": 99072337.0,
      "step": 47725
    },
    {
      "entropy": 5.6560389518737795,
      "epoch": 4.329644412191582,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032559612168744613,
      "loss": 4.8002,
      "mean_token_accuracy": 0.22825632393360137,
      "num_tokens": 99082468.0,
      "step": 47730
    },
    {
      "entropy": 5.749736070632935,
      "epoch": 4.3300979680696665,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003255645934444918,
      "loss": 4.8933,
      "mean_token_accuracy": 0.21808244585990905,
      "num_tokens": 99091757.0,
      "step": 47735
    },
    {
      "entropy": 5.7687595844268795,
      "epoch": 4.330551523947751,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003255330641558489,
      "loss": 4.9487,
      "mean_token_accuracy": 0.2170223370194435,
      "num_tokens": 99103192.0,
      "step": 47740
    },
    {
      "entropy": 5.6675224781036375,
      "epoch": 4.331005079825834,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003255015338221693,
      "loss": 4.7973,
      "mean_token_accuracy": 0.22336716204881668,
      "num_tokens": 99112660.0,
      "step": 47745
    },
    {
      "entropy": 5.776683807373047,
      "epoch": 4.331458635703918,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032547000244410515,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22384812235832213,
      "num_tokens": 99123867.0,
      "step": 47750
    },
    {
      "entropy": 5.722028684616089,
      "epoch": 4.331912191582003,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032543847002230845,
      "loss": 4.8864,
      "mean_token_accuracy": 0.2220388263463974,
      "num_tokens": 99134445.0,
      "step": 47755
    },
    {
      "entropy": 5.688880681991577,
      "epoch": 4.332365747460087,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003254069365574313,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22736697793006896,
      "num_tokens": 99143483.0,
      "step": 47760
    },
    {
      "entropy": 5.687357425689697,
      "epoch": 4.332819303338171,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0003253754020501259,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23522282242774964,
      "num_tokens": 99154514.0,
      "step": 47765
    },
    {
      "entropy": 5.7095154285430905,
      "epoch": 4.3332728592162555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032534386650104444,
      "loss": 4.9218,
      "mean_token_accuracy": 0.21722199618816376,
      "num_tokens": 99165969.0,
      "step": 47770
    },
    {
      "entropy": 5.634279203414917,
      "epoch": 4.33372641509434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003253123299108388,
      "loss": 4.7788,
      "mean_token_accuracy": 0.23705280274152757,
      "num_tokens": 99176595.0,
      "step": 47775
    },
    {
      "entropy": 5.76306185722351,
      "epoch": 4.334179970972424,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003252807922801615,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22384551912546158,
      "num_tokens": 99186916.0,
      "step": 47780
    },
    {
      "entropy": 5.770965671539306,
      "epoch": 4.334633526850508,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003252492536096646,
      "loss": 5.0121,
      "mean_token_accuracy": 0.2129335254430771,
      "num_tokens": 99197892.0,
      "step": 47785
    },
    {
      "entropy": 5.784752464294433,
      "epoch": 4.3350870827285926,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003252177139000003,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22883645594120025,
      "num_tokens": 99208373.0,
      "step": 47790
    },
    {
      "entropy": 5.736116123199463,
      "epoch": 4.335540638606676,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003251861731518209,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22651212215423583,
      "num_tokens": 99218788.0,
      "step": 47795
    },
    {
      "entropy": 5.768899583816529,
      "epoch": 4.33599419448476,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00032515463136577864,
      "loss": 4.9503,
      "mean_token_accuracy": 0.2246137425303459,
      "num_tokens": 99228653.0,
      "step": 47800
    },
    {
      "entropy": 5.685198545455933,
      "epoch": 4.3364477503628445,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032512308854252577,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2311001554131508,
      "num_tokens": 99239925.0,
      "step": 47805
    },
    {
      "entropy": 5.675279426574707,
      "epoch": 4.336901306240929,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003250915446827147,
      "loss": 4.7614,
      "mean_token_accuracy": 0.23587924987077713,
      "num_tokens": 99250091.0,
      "step": 47810
    },
    {
      "entropy": 5.68445930480957,
      "epoch": 4.337354862119013,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003250599997869977,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2100244343280792,
      "num_tokens": 99260799.0,
      "step": 47815
    },
    {
      "entropy": 5.67267336845398,
      "epoch": 4.337808417997097,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003250284538560272,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23603229373693466,
      "num_tokens": 99270338.0,
      "step": 47820
    },
    {
      "entropy": 5.692536544799805,
      "epoch": 4.3382619738751815,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003249969068904554,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22658213675022126,
      "num_tokens": 99280419.0,
      "step": 47825
    },
    {
      "entropy": 5.780430841445923,
      "epoch": 4.338715529753266,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032496535889093496,
      "loss": 4.9541,
      "mean_token_accuracy": 0.21733846217393876,
      "num_tokens": 99291590.0,
      "step": 47830
    },
    {
      "entropy": 5.7695725440979,
      "epoch": 4.33916908563135,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032493380985811814,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22431772202253342,
      "num_tokens": 99301987.0,
      "step": 47835
    },
    {
      "entropy": 5.7824286937713625,
      "epoch": 4.339622641509434,
      "grad_norm": 1.453125,
      "learning_rate": 0.00032490225979265734,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22356190979480745,
      "num_tokens": 99312231.0,
      "step": 47840
    },
    {
      "entropy": 5.743990850448609,
      "epoch": 4.340076197387518,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032487070869520517,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22332664132118224,
      "num_tokens": 99322784.0,
      "step": 47845
    },
    {
      "entropy": 5.764790105819702,
      "epoch": 4.340529753265602,
      "grad_norm": 1.3125,
      "learning_rate": 0.000324839156566414,
      "loss": 4.9318,
      "mean_token_accuracy": 0.224173004925251,
      "num_tokens": 99333989.0,
      "step": 47850
    },
    {
      "entropy": 5.67326865196228,
      "epoch": 4.340983309143686,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003248076034069364,
      "loss": 4.865,
      "mean_token_accuracy": 0.22917208224534988,
      "num_tokens": 99345034.0,
      "step": 47855
    },
    {
      "entropy": 5.696638679504394,
      "epoch": 4.3414368650217705,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003247760492174248,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2253072217106819,
      "num_tokens": 99356062.0,
      "step": 47860
    },
    {
      "entropy": 5.717177534103394,
      "epoch": 4.341890420899855,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003247444939985319,
      "loss": 4.8393,
      "mean_token_accuracy": 0.23249341994524003,
      "num_tokens": 99365315.0,
      "step": 47865
    },
    {
      "entropy": 5.81448073387146,
      "epoch": 4.342343976777939,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032471293775091013,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22708537578582763,
      "num_tokens": 99374897.0,
      "step": 47870
    },
    {
      "entropy": 5.712912511825562,
      "epoch": 4.342797532656023,
      "grad_norm": 1.46875,
      "learning_rate": 0.00032468138047521216,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23007194846868514,
      "num_tokens": 99383624.0,
      "step": 47875
    },
    {
      "entropy": 5.666673564910889,
      "epoch": 4.343251088534108,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003246498221720906,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22094326466321945,
      "num_tokens": 99393791.0,
      "step": 47880
    },
    {
      "entropy": 5.7130790710449215,
      "epoch": 4.343704644412192,
      "grad_norm": 1.328125,
      "learning_rate": 0.000324618262842198,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22551727145910264,
      "num_tokens": 99403687.0,
      "step": 47885
    },
    {
      "entropy": 5.745731544494629,
      "epoch": 4.344158200290276,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032458670248618714,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2284923955798149,
      "num_tokens": 99413411.0,
      "step": 47890
    },
    {
      "entropy": 5.77409782409668,
      "epoch": 4.34461175616836,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003245551411047106,
      "loss": 4.913,
      "mean_token_accuracy": 0.22405630499124526,
      "num_tokens": 99425563.0,
      "step": 47895
    },
    {
      "entropy": 5.70238242149353,
      "epoch": 4.345065312046444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032452357869842117,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23720509856939315,
      "num_tokens": 99436297.0,
      "step": 47900
    },
    {
      "entropy": 5.746962594985962,
      "epoch": 4.345518867924528,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003244920152679715,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22353681325912475,
      "num_tokens": 99445745.0,
      "step": 47905
    },
    {
      "entropy": 5.688229703903199,
      "epoch": 4.345972423802612,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032446045081401426,
      "loss": 4.9036,
      "mean_token_accuracy": 0.22453925460577012,
      "num_tokens": 99456915.0,
      "step": 47910
    },
    {
      "entropy": 5.7792518615722654,
      "epoch": 4.3464259796806965,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003244288853372024,
      "loss": 4.9033,
      "mean_token_accuracy": 0.2239854469895363,
      "num_tokens": 99466778.0,
      "step": 47915
    },
    {
      "entropy": 5.7454277038574215,
      "epoch": 4.346879535558781,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003243973188381885,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2268712192773819,
      "num_tokens": 99475984.0,
      "step": 47920
    },
    {
      "entropy": 5.651070547103882,
      "epoch": 4.347333091436865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032436575131762553,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22353974282741546,
      "num_tokens": 99486238.0,
      "step": 47925
    },
    {
      "entropy": 5.826613235473633,
      "epoch": 4.347786647314949,
      "grad_norm": 1.5625,
      "learning_rate": 0.00032433418277616623,
      "loss": 4.9517,
      "mean_token_accuracy": 0.22032860219478606,
      "num_tokens": 99497806.0,
      "step": 47930
    },
    {
      "entropy": 5.733721828460693,
      "epoch": 4.348240203193034,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003243026132144635,
      "loss": 4.8787,
      "mean_token_accuracy": 0.21953198909759522,
      "num_tokens": 99507214.0,
      "step": 47935
    },
    {
      "entropy": 5.764279222488403,
      "epoch": 4.348693759071118,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003242710426331701,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22084222435951234,
      "num_tokens": 99517025.0,
      "step": 47940
    },
    {
      "entropy": 5.752401447296142,
      "epoch": 4.349147314949202,
      "grad_norm": 1.375,
      "learning_rate": 0.00032423947103293893,
      "loss": 4.9156,
      "mean_token_accuracy": 0.22640835046768187,
      "num_tokens": 99526972.0,
      "step": 47945
    },
    {
      "entropy": 5.763329029083252,
      "epoch": 4.3496008708272855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032420789841442306,
      "loss": 4.9143,
      "mean_token_accuracy": 0.2207396373152733,
      "num_tokens": 99538583.0,
      "step": 47950
    },
    {
      "entropy": 5.688599681854248,
      "epoch": 4.35005442670537,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032417632477827526,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2311137467622757,
      "num_tokens": 99549178.0,
      "step": 47955
    },
    {
      "entropy": 5.692799234390259,
      "epoch": 4.350507982583454,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032414475012514847,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23171629309654235,
      "num_tokens": 99558491.0,
      "step": 47960
    },
    {
      "entropy": 5.701320648193359,
      "epoch": 4.350961538461538,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003241131744556958,
      "loss": 4.9062,
      "mean_token_accuracy": 0.22377326488494872,
      "num_tokens": 99568711.0,
      "step": 47965
    },
    {
      "entropy": 5.688068389892578,
      "epoch": 4.351415094339623,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003240815977705702,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22050018906593322,
      "num_tokens": 99578329.0,
      "step": 47970
    },
    {
      "entropy": 5.676171970367432,
      "epoch": 4.351868650217707,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032405002007042453,
      "loss": 4.7417,
      "mean_token_accuracy": 0.23265634030103682,
      "num_tokens": 99588646.0,
      "step": 47975
    },
    {
      "entropy": 5.792137002944946,
      "epoch": 4.352322206095791,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032401844135591213,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2183767557144165,
      "num_tokens": 99598780.0,
      "step": 47980
    },
    {
      "entropy": 5.770745849609375,
      "epoch": 4.352775761973875,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003239868616276857,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22764685451984407,
      "num_tokens": 99609528.0,
      "step": 47985
    },
    {
      "entropy": 5.664984655380249,
      "epoch": 4.35322931785196,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003239552808863986,
      "loss": 4.7989,
      "mean_token_accuracy": 0.232413212954998,
      "num_tokens": 99619928.0,
      "step": 47990
    },
    {
      "entropy": 5.636775541305542,
      "epoch": 4.353682873730044,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032392369913270376,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23174843192100525,
      "num_tokens": 99630135.0,
      "step": 47995
    },
    {
      "entropy": 5.744085264205933,
      "epoch": 4.354136429608127,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003238921163672544,
      "loss": 4.9871,
      "mean_token_accuracy": 0.21075623631477355,
      "num_tokens": 99641557.0,
      "step": 48000
    },
    {
      "epoch": 4.354136429608127,
      "eval_entropy": 5.4334194552305055,
      "eval_loss": 4.966175556182861,
      "eval_mean_token_accuracy": 0.2291430791022416,
      "eval_num_tokens": 99641557.0,
      "eval_runtime": 20.5982,
      "eval_samples_per_second": 1716.657,
      "eval_steps_per_second": 214.582,
      "step": 48000
    },
    {
      "entropy": 5.754856300354004,
      "epoch": 4.354589985486212,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003238605325907035,
      "loss": 4.8867,
      "mean_token_accuracy": 0.2252671867609024,
      "num_tokens": 99651244.0,
      "step": 48005
    },
    {
      "entropy": 5.747967147827149,
      "epoch": 4.355043541364296,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032382894780370446,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22163437008857728,
      "num_tokens": 99662317.0,
      "step": 48010
    },
    {
      "entropy": 5.7282407760620115,
      "epoch": 4.35549709724238,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032379736200691036,
      "loss": 4.8256,
      "mean_token_accuracy": 0.22684383243322373,
      "num_tokens": 99671624.0,
      "step": 48015
    },
    {
      "entropy": 5.7553548336029055,
      "epoch": 4.355950653120464,
      "grad_norm": 1.34375,
      "learning_rate": 0.00032376577520097427,
      "loss": 4.9889,
      "mean_token_accuracy": 0.21615587621927262,
      "num_tokens": 99682076.0,
      "step": 48020
    },
    {
      "entropy": 5.714172077178955,
      "epoch": 4.356404208998549,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032373418738654953,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22813450992107392,
      "num_tokens": 99692076.0,
      "step": 48025
    },
    {
      "entropy": 5.73134446144104,
      "epoch": 4.356857764876633,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003237025985642895,
      "loss": 4.8503,
      "mean_token_accuracy": 0.22799849957227708,
      "num_tokens": 99701455.0,
      "step": 48030
    },
    {
      "entropy": 5.757724094390869,
      "epoch": 4.357311320754717,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032367100873484725,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2214389145374298,
      "num_tokens": 99712034.0,
      "step": 48035
    },
    {
      "entropy": 5.722414207458496,
      "epoch": 4.357764876632801,
      "grad_norm": 1.21875,
      "learning_rate": 0.000323639417898876,
      "loss": 4.9125,
      "mean_token_accuracy": 0.22011184692382812,
      "num_tokens": 99722965.0,
      "step": 48040
    },
    {
      "entropy": 5.651888799667359,
      "epoch": 4.358218432510886,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003236078260570293,
      "loss": 4.7875,
      "mean_token_accuracy": 0.22616953104734422,
      "num_tokens": 99732847.0,
      "step": 48045
    },
    {
      "entropy": 5.696615934371948,
      "epoch": 4.35867198838897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003235762332099604,
      "loss": 4.8159,
      "mean_token_accuracy": 0.2372762680053711,
      "num_tokens": 99743639.0,
      "step": 48050
    },
    {
      "entropy": 5.743232154846192,
      "epoch": 4.359125544267053,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003235446393583227,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23649220913648605,
      "num_tokens": 99753374.0,
      "step": 48055
    },
    {
      "entropy": 5.709448719024659,
      "epoch": 4.359579100145138,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003235130445027693,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2287297561764717,
      "num_tokens": 99763361.0,
      "step": 48060
    },
    {
      "entropy": 5.766731882095337,
      "epoch": 4.360032656023222,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032348144864395383,
      "loss": 4.9466,
      "mean_token_accuracy": 0.2200195699930191,
      "num_tokens": 99773813.0,
      "step": 48065
    },
    {
      "entropy": 5.749612522125244,
      "epoch": 4.360486211901306,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003234498517825297,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22215845435857773,
      "num_tokens": 99784812.0,
      "step": 48070
    },
    {
      "entropy": 5.726060581207276,
      "epoch": 4.36093976777939,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032341825391915026,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2261236146092415,
      "num_tokens": 99795197.0,
      "step": 48075
    },
    {
      "entropy": 5.633527612686157,
      "epoch": 4.361393323657475,
      "grad_norm": 1.296875,
      "learning_rate": 0.000323386655054469,
      "loss": 4.9022,
      "mean_token_accuracy": 0.2268037885427475,
      "num_tokens": 99805883.0,
      "step": 48080
    },
    {
      "entropy": 5.728684377670288,
      "epoch": 4.361846879535559,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003233550551891394,
      "loss": 4.8689,
      "mean_token_accuracy": 0.2257583975791931,
      "num_tokens": 99816940.0,
      "step": 48085
    },
    {
      "entropy": 5.724533653259277,
      "epoch": 4.362300435413643,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032332345432381493,
      "loss": 4.8514,
      "mean_token_accuracy": 0.22337026596069337,
      "num_tokens": 99827926.0,
      "step": 48090
    },
    {
      "entropy": 5.665515613555908,
      "epoch": 4.3627539912917275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003232918524591491,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23082994818687438,
      "num_tokens": 99837900.0,
      "step": 48095
    },
    {
      "entropy": 5.7546039581298825,
      "epoch": 4.363207547169811,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003232602495957955,
      "loss": 4.9227,
      "mean_token_accuracy": 0.21770776212215423,
      "num_tokens": 99848655.0,
      "step": 48100
    },
    {
      "entropy": 5.689410400390625,
      "epoch": 4.363661103047895,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032322864573440757,
      "loss": 4.8317,
      "mean_token_accuracy": 0.2268327444791794,
      "num_tokens": 99859715.0,
      "step": 48105
    },
    {
      "entropy": 5.721135520935059,
      "epoch": 4.364114658925979,
      "grad_norm": 1.25,
      "learning_rate": 0.0003231970408756389,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2370438739657402,
      "num_tokens": 99870175.0,
      "step": 48110
    },
    {
      "entropy": 5.78285174369812,
      "epoch": 4.364568214804064,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003231654350201432,
      "loss": 4.9066,
      "mean_token_accuracy": 0.21762568652629852,
      "num_tokens": 99879794.0,
      "step": 48115
    },
    {
      "entropy": 5.741653728485107,
      "epoch": 4.365021770682148,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003231338281685741,
      "loss": 4.8642,
      "mean_token_accuracy": 0.23267504721879959,
      "num_tokens": 99889418.0,
      "step": 48120
    },
    {
      "entropy": 5.757284545898438,
      "epoch": 4.365475326560232,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003231022203215851,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22092608213424683,
      "num_tokens": 99898741.0,
      "step": 48125
    },
    {
      "entropy": 5.692172956466675,
      "epoch": 4.3659288824383164,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032307061147982994,
      "loss": 4.8355,
      "mean_token_accuracy": 0.22363526970148087,
      "num_tokens": 99909167.0,
      "step": 48130
    },
    {
      "entropy": 5.690251874923706,
      "epoch": 4.366382438316401,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003230390016439623,
      "loss": 4.8726,
      "mean_token_accuracy": 0.22755857259035112,
      "num_tokens": 99918786.0,
      "step": 48135
    },
    {
      "entropy": 5.716893577575684,
      "epoch": 4.366835994194485,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003230073908146359,
      "loss": 4.9113,
      "mean_token_accuracy": 0.2218858003616333,
      "num_tokens": 99928758.0,
      "step": 48140
    },
    {
      "entropy": 5.7125037670135494,
      "epoch": 4.367289550072569,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003229757789925044,
      "loss": 4.8574,
      "mean_token_accuracy": 0.224515800178051,
      "num_tokens": 99939721.0,
      "step": 48145
    },
    {
      "entropy": 5.695094442367553,
      "epoch": 4.3677431059506535,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003229441661782215,
      "loss": 4.803,
      "mean_token_accuracy": 0.2288145750761032,
      "num_tokens": 99949308.0,
      "step": 48150
    },
    {
      "entropy": 5.721691703796386,
      "epoch": 4.368196661828737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032291255237244104,
      "loss": 4.8706,
      "mean_token_accuracy": 0.22912891805171967,
      "num_tokens": 99960284.0,
      "step": 48155
    },
    {
      "entropy": 5.669543313980102,
      "epoch": 4.368650217706821,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032288093757581684,
      "loss": 4.7814,
      "mean_token_accuracy": 0.2303463637828827,
      "num_tokens": 99970701.0,
      "step": 48160
    },
    {
      "entropy": 5.737131595611572,
      "epoch": 4.369103773584905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003228493217890026,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2272432416677475,
      "num_tokens": 99981362.0,
      "step": 48165
    },
    {
      "entropy": 5.754561042785644,
      "epoch": 4.36955732946299,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003228177050126523,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22978359460830688,
      "num_tokens": 99991383.0,
      "step": 48170
    },
    {
      "entropy": 5.725299882888794,
      "epoch": 4.370010885341074,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032278608724741956,
      "loss": 4.8714,
      "mean_token_accuracy": 0.2239371970295906,
      "num_tokens": 100001176.0,
      "step": 48175
    },
    {
      "entropy": 5.674046516418457,
      "epoch": 4.370464441219158,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003227544684939584,
      "loss": 4.8143,
      "mean_token_accuracy": 0.2336648315191269,
      "num_tokens": 100011467.0,
      "step": 48180
    },
    {
      "entropy": 5.709420585632325,
      "epoch": 4.3709179970972425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003227228487529226,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22808006554841995,
      "num_tokens": 100020976.0,
      "step": 48185
    },
    {
      "entropy": 5.657899856567383,
      "epoch": 4.371371552975327,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003226912280249662,
      "loss": 4.8508,
      "mean_token_accuracy": 0.23185340613126754,
      "num_tokens": 100032250.0,
      "step": 48190
    },
    {
      "entropy": 5.745649337768555,
      "epoch": 4.371825108853411,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000322659606310743,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2268432155251503,
      "num_tokens": 100043383.0,
      "step": 48195
    },
    {
      "entropy": 5.7041624069213865,
      "epoch": 4.372278664731495,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032262798361090705,
      "loss": 4.7942,
      "mean_token_accuracy": 0.2333431139588356,
      "num_tokens": 100053396.0,
      "step": 48200
    },
    {
      "entropy": 5.677447700500489,
      "epoch": 4.37273222060958,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003225963599261123,
      "loss": 4.8753,
      "mean_token_accuracy": 0.2277991235256195,
      "num_tokens": 100064159.0,
      "step": 48205
    },
    {
      "entropy": 5.664836454391479,
      "epoch": 4.373185776487663,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032256473525701257,
      "loss": 4.8218,
      "mean_token_accuracy": 0.2378368854522705,
      "num_tokens": 100074722.0,
      "step": 48210
    },
    {
      "entropy": 5.676998853683472,
      "epoch": 4.373639332365747,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032253310960426215,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2330252543091774,
      "num_tokens": 100085067.0,
      "step": 48215
    },
    {
      "entropy": 5.722400665283203,
      "epoch": 4.3740928882438315,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032250148296851484,
      "loss": 4.8856,
      "mean_token_accuracy": 0.22940223664045334,
      "num_tokens": 100095806.0,
      "step": 48220
    },
    {
      "entropy": 5.800817203521729,
      "epoch": 4.374546444121916,
      "grad_norm": 1.25,
      "learning_rate": 0.0003224698553504247,
      "loss": 4.9511,
      "mean_token_accuracy": 0.2180701270699501,
      "num_tokens": 100106069.0,
      "step": 48225
    },
    {
      "entropy": 5.799606561660767,
      "epoch": 4.375,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032243822675064585,
      "loss": 4.9345,
      "mean_token_accuracy": 0.21652537286281587,
      "num_tokens": 100117136.0,
      "step": 48230
    },
    {
      "entropy": 5.764972305297851,
      "epoch": 4.375453555878084,
      "grad_norm": 1.484375,
      "learning_rate": 0.0003224065971698324,
      "loss": 4.8916,
      "mean_token_accuracy": 0.21425021737813948,
      "num_tokens": 100126757.0,
      "step": 48235
    },
    {
      "entropy": 5.8532147884368895,
      "epoch": 4.3759071117561685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003223749666086385,
      "loss": 4.9961,
      "mean_token_accuracy": 0.2211802512407303,
      "num_tokens": 100138431.0,
      "step": 48240
    },
    {
      "entropy": 5.758647489547729,
      "epoch": 4.376360667634253,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003223433350677182,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22224764227867128,
      "num_tokens": 100148919.0,
      "step": 48245
    },
    {
      "entropy": 5.783418369293213,
      "epoch": 4.376814223512337,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003223117025477256,
      "loss": 5.0069,
      "mean_token_accuracy": 0.2232340931892395,
      "num_tokens": 100159175.0,
      "step": 48250
    },
    {
      "entropy": 5.709730052947998,
      "epoch": 4.37726777939042,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003222800690493151,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2249506324529648,
      "num_tokens": 100169451.0,
      "step": 48255
    },
    {
      "entropy": 5.736291933059692,
      "epoch": 4.377721335268505,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003222484345731405,
      "loss": 4.8735,
      "mean_token_accuracy": 0.2290205702185631,
      "num_tokens": 100180077.0,
      "step": 48260
    },
    {
      "entropy": 5.736315727233887,
      "epoch": 4.378174891146589,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032221679911985634,
      "loss": 4.8659,
      "mean_token_accuracy": 0.2299245297908783,
      "num_tokens": 100190175.0,
      "step": 48265
    },
    {
      "entropy": 5.69259204864502,
      "epoch": 4.378628447024673,
      "grad_norm": 1.25,
      "learning_rate": 0.00032218516269011676,
      "loss": 4.801,
      "mean_token_accuracy": 0.23384876400232316,
      "num_tokens": 100200285.0,
      "step": 48270
    },
    {
      "entropy": 5.599824047088623,
      "epoch": 4.3790820029027575,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032215352528457597,
      "loss": 4.706,
      "mean_token_accuracy": 0.23516691327095032,
      "num_tokens": 100210685.0,
      "step": 48275
    },
    {
      "entropy": 5.7016339778900145,
      "epoch": 4.379535558780842,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032212188690388825,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23042042404413224,
      "num_tokens": 100220486.0,
      "step": 48280
    },
    {
      "entropy": 5.671523761749268,
      "epoch": 4.379989114658926,
      "grad_norm": 1.40625,
      "learning_rate": 0.00032209024754870796,
      "loss": 4.7888,
      "mean_token_accuracy": 0.2213226318359375,
      "num_tokens": 100229007.0,
      "step": 48285
    },
    {
      "entropy": 5.678337049484253,
      "epoch": 4.38044267053701,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003220586072196893,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23170037865638732,
      "num_tokens": 100240363.0,
      "step": 48290
    },
    {
      "entropy": 5.6542713165283205,
      "epoch": 4.380896226415095,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032202696591748667,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22967141568660737,
      "num_tokens": 100251193.0,
      "step": 48295
    },
    {
      "entropy": 5.711236238479614,
      "epoch": 4.381349782293179,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032199532364275436,
      "loss": 4.8528,
      "mean_token_accuracy": 0.2314735844731331,
      "num_tokens": 100261248.0,
      "step": 48300
    },
    {
      "entropy": 5.719952201843261,
      "epoch": 4.381803338171263,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003219636803961468,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22158954292535782,
      "num_tokens": 100271174.0,
      "step": 48305
    },
    {
      "entropy": 5.7072759628295895,
      "epoch": 4.3822568940493465,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032193203617831835,
      "loss": 4.9133,
      "mean_token_accuracy": 0.223773692548275,
      "num_tokens": 100282428.0,
      "step": 48310
    },
    {
      "entropy": 5.736277341842651,
      "epoch": 4.382710449927431,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032190039098992345,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22376007437705994,
      "num_tokens": 100292559.0,
      "step": 48315
    },
    {
      "entropy": 5.71068811416626,
      "epoch": 4.383164005805515,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003218687448316166,
      "loss": 4.792,
      "mean_token_accuracy": 0.23089179545640945,
      "num_tokens": 100303956.0,
      "step": 48320
    },
    {
      "entropy": 5.656160020828247,
      "epoch": 4.383617561683599,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032183709770405204,
      "loss": 4.8631,
      "mean_token_accuracy": 0.22648233771324158,
      "num_tokens": 100315281.0,
      "step": 48325
    },
    {
      "entropy": 5.750617742538452,
      "epoch": 4.3840711175616836,
      "grad_norm": 1.25,
      "learning_rate": 0.0003218054496078844,
      "loss": 4.9056,
      "mean_token_accuracy": 0.21881963312625885,
      "num_tokens": 100326761.0,
      "step": 48330
    },
    {
      "entropy": 5.761556482315063,
      "epoch": 4.384524673439768,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032177380054376813,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22471027970314025,
      "num_tokens": 100337219.0,
      "step": 48335
    },
    {
      "entropy": 5.7537596225738525,
      "epoch": 4.384978229317852,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032174215051235776,
      "loss": 4.8677,
      "mean_token_accuracy": 0.23234803378582,
      "num_tokens": 100347135.0,
      "step": 48340
    },
    {
      "entropy": 5.6106483936309814,
      "epoch": 4.385431785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00032171049951430775,
      "loss": 4.7351,
      "mean_token_accuracy": 0.23387365341186522,
      "num_tokens": 100357476.0,
      "step": 48345
    },
    {
      "entropy": 5.694971418380737,
      "epoch": 4.385885341074021,
      "grad_norm": 1.3125,
      "learning_rate": 0.00032167884755027285,
      "loss": 4.8241,
      "mean_token_accuracy": 0.22466493397951126,
      "num_tokens": 100367860.0,
      "step": 48350
    },
    {
      "entropy": 5.702750015258789,
      "epoch": 4.386338896952105,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003216471946209074,
      "loss": 4.8316,
      "mean_token_accuracy": 0.2327662631869316,
      "num_tokens": 100378613.0,
      "step": 48355
    },
    {
      "entropy": 5.723435926437378,
      "epoch": 4.386792452830189,
      "grad_norm": 1.296875,
      "learning_rate": 0.000321615540726866,
      "loss": 4.8903,
      "mean_token_accuracy": 0.23172685205936433,
      "num_tokens": 100389499.0,
      "step": 48360
    },
    {
      "entropy": 5.777131366729736,
      "epoch": 4.3872460087082725,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032158388586880346,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22684430778026582,
      "num_tokens": 100400538.0,
      "step": 48365
    },
    {
      "entropy": 5.7159693241119385,
      "epoch": 4.387699564586357,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032155223004737423,
      "loss": 4.8985,
      "mean_token_accuracy": 0.23512910604476928,
      "num_tokens": 100411528.0,
      "step": 48370
    },
    {
      "entropy": 5.756238126754761,
      "epoch": 4.388153120464441,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00032152057326323305,
      "loss": 4.874,
      "mean_token_accuracy": 0.2201366037130356,
      "num_tokens": 100421349.0,
      "step": 48375
    },
    {
      "entropy": 5.690658378601074,
      "epoch": 4.388606676342525,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00032148891551703445,
      "loss": 4.8007,
      "mean_token_accuracy": 0.22748851031064987,
      "num_tokens": 100431132.0,
      "step": 48380
    },
    {
      "entropy": 5.675035715103149,
      "epoch": 4.38906023222061,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003214572568094334,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22728347182273864,
      "num_tokens": 100441436.0,
      "step": 48385
    },
    {
      "entropy": 5.64617805480957,
      "epoch": 4.389513788098694,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003214255971410843,
      "loss": 4.7354,
      "mean_token_accuracy": 0.23412152528762817,
      "num_tokens": 100451711.0,
      "step": 48390
    },
    {
      "entropy": 5.825428438186646,
      "epoch": 4.389967343976778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000321393936512642,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21960613131523132,
      "num_tokens": 100461177.0,
      "step": 48395
    },
    {
      "entropy": 5.71170392036438,
      "epoch": 4.390420899854862,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032136227492476145,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2264800935983658,
      "num_tokens": 100471391.0,
      "step": 48400
    },
    {
      "entropy": 5.753666734695434,
      "epoch": 4.390874455732947,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003213306123780971,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22868386507034302,
      "num_tokens": 100480747.0,
      "step": 48405
    },
    {
      "entropy": 5.734088516235351,
      "epoch": 4.39132801161103,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032129894887330396,
      "loss": 4.8942,
      "mean_token_accuracy": 0.23103619664907454,
      "num_tokens": 100491451.0,
      "step": 48410
    },
    {
      "entropy": 5.777245664596558,
      "epoch": 4.391781567489114,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003212672844110367,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22804169803857804,
      "num_tokens": 100501801.0,
      "step": 48415
    },
    {
      "entropy": 5.675481557846069,
      "epoch": 4.392235123367199,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00032123561899195026,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23377628177404403,
      "num_tokens": 100511857.0,
      "step": 48420
    },
    {
      "entropy": 5.758440399169922,
      "epoch": 4.392688679245283,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003212039526166994,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22407146990299226,
      "num_tokens": 100523654.0,
      "step": 48425
    },
    {
      "entropy": 5.826967096328735,
      "epoch": 4.393142235123367,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032117228528593907,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22362145632505417,
      "num_tokens": 100533030.0,
      "step": 48430
    },
    {
      "entropy": 5.764245986938477,
      "epoch": 4.393595791001451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000321140617000324,
      "loss": 4.8892,
      "mean_token_accuracy": 0.2179689958691597,
      "num_tokens": 100543414.0,
      "step": 48435
    },
    {
      "entropy": 5.7333083152771,
      "epoch": 4.394049346879536,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032110894776050936,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2297167330980301,
      "num_tokens": 100553303.0,
      "step": 48440
    },
    {
      "entropy": 5.655999374389649,
      "epoch": 4.39450290275762,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032107727756714984,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22627147585153579,
      "num_tokens": 100563030.0,
      "step": 48445
    },
    {
      "entropy": 5.757625246047974,
      "epoch": 4.394956458635704,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003210456064209005,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21693530678749084,
      "num_tokens": 100573216.0,
      "step": 48450
    },
    {
      "entropy": 5.669421052932739,
      "epoch": 4.395410014513788,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032101393432241635,
      "loss": 4.7943,
      "mean_token_accuracy": 0.22678820341825484,
      "num_tokens": 100582646.0,
      "step": 48455
    },
    {
      "entropy": 5.721813821792603,
      "epoch": 4.395863570391873,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032098226127235215,
      "loss": 4.9077,
      "mean_token_accuracy": 0.21810466051101685,
      "num_tokens": 100593689.0,
      "step": 48460
    },
    {
      "entropy": 5.707226848602295,
      "epoch": 4.396317126269956,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003209505872713632,
      "loss": 4.9363,
      "mean_token_accuracy": 0.2195977419614792,
      "num_tokens": 100604721.0,
      "step": 48465
    },
    {
      "entropy": 5.734987020492554,
      "epoch": 4.39677068214804,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032091891232010434,
      "loss": 4.9292,
      "mean_token_accuracy": 0.22342269569635392,
      "num_tokens": 100614684.0,
      "step": 48470
    },
    {
      "entropy": 5.718614387512207,
      "epoch": 4.397224238026125,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00032088723641923066,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22566639333963395,
      "num_tokens": 100624434.0,
      "step": 48475
    },
    {
      "entropy": 5.81403923034668,
      "epoch": 4.397677793904209,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032085555956939726,
      "loss": 5.003,
      "mean_token_accuracy": 0.21456727385520935,
      "num_tokens": 100635656.0,
      "step": 48480
    },
    {
      "entropy": 5.625837230682373,
      "epoch": 4.398131349782293,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003208238817712593,
      "loss": 4.745,
      "mean_token_accuracy": 0.2379722014069557,
      "num_tokens": 100645749.0,
      "step": 48485
    },
    {
      "entropy": 5.666671991348267,
      "epoch": 4.398584905660377,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003207922030254717,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22660772502422333,
      "num_tokens": 100655814.0,
      "step": 48490
    },
    {
      "entropy": 5.6751667022705075,
      "epoch": 4.399038461538462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032076052333268964,
      "loss": 4.847,
      "mean_token_accuracy": 0.22633007764816285,
      "num_tokens": 100665579.0,
      "step": 48495
    },
    {
      "entropy": 5.799668598175049,
      "epoch": 4.399492017416546,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003207288426935683,
      "loss": 5.0181,
      "mean_token_accuracy": 0.21416420936584474,
      "num_tokens": 100676421.0,
      "step": 48500
    },
    {
      "entropy": 5.734559631347656,
      "epoch": 4.39994557329463,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003206971611087629,
      "loss": 4.8454,
      "mean_token_accuracy": 0.23191233575344086,
      "num_tokens": 100686689.0,
      "step": 48505
    },
    {
      "entropy": 5.82573504447937,
      "epoch": 4.4003991291727145,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003206654785789285,
      "loss": 4.9497,
      "mean_token_accuracy": 0.2198496326804161,
      "num_tokens": 100697324.0,
      "step": 48510
    },
    {
      "entropy": 5.663131952285767,
      "epoch": 4.400852685050798,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00032063379510472036,
      "loss": 4.7943,
      "mean_token_accuracy": 0.2340845510363579,
      "num_tokens": 100706003.0,
      "step": 48515
    },
    {
      "entropy": 5.749132776260376,
      "epoch": 4.401306240928882,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003206021106867938,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2311435952782631,
      "num_tokens": 100716755.0,
      "step": 48520
    },
    {
      "entropy": 5.688873147964477,
      "epoch": 4.401759796806966,
      "grad_norm": 1.25,
      "learning_rate": 0.0003205704253258039,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2198081538081169,
      "num_tokens": 100727846.0,
      "step": 48525
    },
    {
      "entropy": 5.710984802246093,
      "epoch": 4.402213352685051,
      "grad_norm": 1.328125,
      "learning_rate": 0.000320538739022406,
      "loss": 4.8451,
      "mean_token_accuracy": 0.2221707284450531,
      "num_tokens": 100738293.0,
      "step": 48530
    },
    {
      "entropy": 5.697389936447143,
      "epoch": 4.402666908563135,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003205070517772554,
      "loss": 4.8047,
      "mean_token_accuracy": 0.2288490578532219,
      "num_tokens": 100748031.0,
      "step": 48535
    },
    {
      "entropy": 5.679288148880005,
      "epoch": 4.403120464441219,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003204753635910073,
      "loss": 4.8302,
      "mean_token_accuracy": 0.2339565262198448,
      "num_tokens": 100758825.0,
      "step": 48540
    },
    {
      "entropy": 5.725550317764283,
      "epoch": 4.4035740203193035,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032044367446431714,
      "loss": 4.8327,
      "mean_token_accuracy": 0.23177336454391478,
      "num_tokens": 100768226.0,
      "step": 48545
    },
    {
      "entropy": 5.691484689712524,
      "epoch": 4.404027576197388,
      "grad_norm": 1.375,
      "learning_rate": 0.00032041198439784025,
      "loss": 4.8106,
      "mean_token_accuracy": 0.22982343435287475,
      "num_tokens": 100778125.0,
      "step": 48550
    },
    {
      "entropy": 5.724103832244873,
      "epoch": 4.404481132075472,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003203802933922319,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22714536786079406,
      "num_tokens": 100788670.0,
      "step": 48555
    },
    {
      "entropy": 5.737030935287476,
      "epoch": 4.404934687953556,
      "grad_norm": 1.25,
      "learning_rate": 0.00032034860144814756,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22203867733478547,
      "num_tokens": 100800077.0,
      "step": 48560
    },
    {
      "entropy": 5.706910610198975,
      "epoch": 4.40538824383164,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003203169085662425,
      "loss": 4.856,
      "mean_token_accuracy": 0.22357385009527206,
      "num_tokens": 100809432.0,
      "step": 48565
    },
    {
      "entropy": 5.689004707336426,
      "epoch": 4.405841799709724,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003202852147471723,
      "loss": 4.8397,
      "mean_token_accuracy": 0.23546531945466995,
      "num_tokens": 100819675.0,
      "step": 48570
    },
    {
      "entropy": 5.735198974609375,
      "epoch": 4.406295355587808,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00032025351999159237,
      "loss": 4.8819,
      "mean_token_accuracy": 0.2250776395201683,
      "num_tokens": 100829779.0,
      "step": 48575
    },
    {
      "entropy": 5.706893682479858,
      "epoch": 4.406748911465892,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032022182430015795,
      "loss": 4.9104,
      "mean_token_accuracy": 0.22198135852813722,
      "num_tokens": 100839840.0,
      "step": 48580
    },
    {
      "entropy": 5.682071304321289,
      "epoch": 4.407202467343977,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003201901276735248,
      "loss": 4.9099,
      "mean_token_accuracy": 0.22511858940124513,
      "num_tokens": 100849495.0,
      "step": 48585
    },
    {
      "entropy": 5.681952905654907,
      "epoch": 4.407656023222061,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032015843011234826,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2260376736521721,
      "num_tokens": 100860214.0,
      "step": 48590
    },
    {
      "entropy": 5.824224376678467,
      "epoch": 4.408109579100145,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032012673161728396,
      "loss": 5.0003,
      "mean_token_accuracy": 0.20943001955747603,
      "num_tokens": 100870773.0,
      "step": 48595
    },
    {
      "entropy": 5.779804706573486,
      "epoch": 4.4085631349782295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032009503218898716,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23541587144136428,
      "num_tokens": 100880654.0,
      "step": 48600
    },
    {
      "entropy": 5.661440658569336,
      "epoch": 4.409016690856314,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003200633318281138,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22257371842861176,
      "num_tokens": 100891217.0,
      "step": 48605
    },
    {
      "entropy": 5.688356113433838,
      "epoch": 4.409470246734398,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00032003163053531913,
      "loss": 4.7769,
      "mean_token_accuracy": 0.2285538762807846,
      "num_tokens": 100901424.0,
      "step": 48610
    },
    {
      "entropy": 5.754046249389648,
      "epoch": 4.409923802612482,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031999992831125893,
      "loss": 4.9494,
      "mean_token_accuracy": 0.22497936636209487,
      "num_tokens": 100912069.0,
      "step": 48615
    },
    {
      "entropy": 5.716155052185059,
      "epoch": 4.410377358490566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003199682251565887,
      "loss": 4.8573,
      "mean_token_accuracy": 0.21721750497817993,
      "num_tokens": 100922750.0,
      "step": 48620
    },
    {
      "entropy": 5.774622678756714,
      "epoch": 4.41083091436865,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031993652107196416,
      "loss": 4.9092,
      "mean_token_accuracy": 0.22655668407678603,
      "num_tokens": 100933052.0,
      "step": 48625
    },
    {
      "entropy": 5.741427326202393,
      "epoch": 4.411284470246734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031990481605804084,
      "loss": 4.8785,
      "mean_token_accuracy": 0.23104341179132462,
      "num_tokens": 100944519.0,
      "step": 48630
    },
    {
      "entropy": 5.694859313964844,
      "epoch": 4.4117380261248185,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031987311011547455,
      "loss": 4.9111,
      "mean_token_accuracy": 0.21859900653362274,
      "num_tokens": 100954901.0,
      "step": 48635
    },
    {
      "entropy": 5.731744480133057,
      "epoch": 4.412191582002903,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003198414032449209,
      "loss": 4.8558,
      "mean_token_accuracy": 0.2289525181055069,
      "num_tokens": 100964931.0,
      "step": 48640
    },
    {
      "entropy": 5.7788983345031735,
      "epoch": 4.412645137880987,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003198096954470356,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22776162922382354,
      "num_tokens": 100974877.0,
      "step": 48645
    },
    {
      "entropy": 5.769122695922851,
      "epoch": 4.413098693759071,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031977798672247444,
      "loss": 4.9429,
      "mean_token_accuracy": 0.2264244332909584,
      "num_tokens": 100986122.0,
      "step": 48650
    },
    {
      "entropy": 5.756955337524414,
      "epoch": 4.4135522496371555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031974627707189303,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2257692486047745,
      "num_tokens": 100996489.0,
      "step": 48655
    },
    {
      "entropy": 5.719209146499634,
      "epoch": 4.41400580551524,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031971456649594723,
      "loss": 4.9146,
      "mean_token_accuracy": 0.22705331295728684,
      "num_tokens": 101006453.0,
      "step": 48660
    },
    {
      "entropy": 5.77767972946167,
      "epoch": 4.414459361393324,
      "grad_norm": 1.5,
      "learning_rate": 0.0003196828549952927,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22257055193185807,
      "num_tokens": 101015609.0,
      "step": 48665
    },
    {
      "entropy": 5.741404294967651,
      "epoch": 4.4149129172714074,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003196511425705855,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22399526536464692,
      "num_tokens": 101027406.0,
      "step": 48670
    },
    {
      "entropy": 5.678369760513306,
      "epoch": 4.415366473149492,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003196194292224812,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22254468947649003,
      "num_tokens": 101038255.0,
      "step": 48675
    },
    {
      "entropy": 5.701490116119385,
      "epoch": 4.415820029027576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031958771495163574,
      "loss": 4.8607,
      "mean_token_accuracy": 0.2265022873878479,
      "num_tokens": 101048857.0,
      "step": 48680
    },
    {
      "entropy": 5.769374704360962,
      "epoch": 4.41627358490566,
      "grad_norm": 1.375,
      "learning_rate": 0.000319555999758705,
      "loss": 4.936,
      "mean_token_accuracy": 0.22176032662391662,
      "num_tokens": 101058640.0,
      "step": 48685
    },
    {
      "entropy": 5.779002380371094,
      "epoch": 4.4167271407837445,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031952428364434484,
      "loss": 5.0129,
      "mean_token_accuracy": 0.22030138969421387,
      "num_tokens": 101069537.0,
      "step": 48690
    },
    {
      "entropy": 5.774582481384277,
      "epoch": 4.417180696661829,
      "grad_norm": 1.25,
      "learning_rate": 0.00031949256660921104,
      "loss": 4.9129,
      "mean_token_accuracy": 0.21567132771015168,
      "num_tokens": 101080202.0,
      "step": 48695
    },
    {
      "entropy": 5.7222387313842775,
      "epoch": 4.417634252539913,
      "grad_norm": 1.375,
      "learning_rate": 0.00031946084865395975,
      "loss": 4.8652,
      "mean_token_accuracy": 0.229398275911808,
      "num_tokens": 101089500.0,
      "step": 48700
    },
    {
      "entropy": 5.81341142654419,
      "epoch": 4.418087808417997,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00031942912977924674,
      "loss": 4.988,
      "mean_token_accuracy": 0.21714771091938018,
      "num_tokens": 101099858.0,
      "step": 48705
    },
    {
      "entropy": 5.735785341262817,
      "epoch": 4.418541364296082,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031939740998572806,
      "loss": 4.9423,
      "mean_token_accuracy": 0.22007596343755723,
      "num_tokens": 101110834.0,
      "step": 48710
    },
    {
      "entropy": 5.763248682022095,
      "epoch": 4.418994920174166,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003193656892740595,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2264731228351593,
      "num_tokens": 101121003.0,
      "step": 48715
    },
    {
      "entropy": 5.770125675201416,
      "epoch": 4.419448476052249,
      "grad_norm": 1.25,
      "learning_rate": 0.0003193339676448973,
      "loss": 4.9251,
      "mean_token_accuracy": 0.22552232891321183,
      "num_tokens": 101131925.0,
      "step": 48720
    },
    {
      "entropy": 5.714857387542724,
      "epoch": 4.4199020319303335,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003193022450988973,
      "loss": 4.8077,
      "mean_token_accuracy": 0.2292596533894539,
      "num_tokens": 101143130.0,
      "step": 48725
    },
    {
      "entropy": 5.683552026748657,
      "epoch": 4.420355587808418,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003192705216367156,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22891725897789,
      "num_tokens": 101155524.0,
      "step": 48730
    },
    {
      "entropy": 5.691310930252075,
      "epoch": 4.420809143686502,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031923879725900817,
      "loss": 4.866,
      "mean_token_accuracy": 0.23438010811805726,
      "num_tokens": 101164815.0,
      "step": 48735
    },
    {
      "entropy": 5.761245918273926,
      "epoch": 4.421262699564586,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031920707196643125,
      "loss": 4.9729,
      "mean_token_accuracy": 0.21845511496067047,
      "num_tokens": 101174910.0,
      "step": 48740
    },
    {
      "entropy": 5.674001598358155,
      "epoch": 4.4217162554426706,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003191753457596407,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23273660689592363,
      "num_tokens": 101185698.0,
      "step": 48745
    },
    {
      "entropy": 5.710898113250733,
      "epoch": 4.422169811320755,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031914361863929276,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22427260577678682,
      "num_tokens": 101197001.0,
      "step": 48750
    },
    {
      "entropy": 5.702236795425415,
      "epoch": 4.422623367198839,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003191118906060436,
      "loss": 4.8378,
      "mean_token_accuracy": 0.22464142590761185,
      "num_tokens": 101206705.0,
      "step": 48755
    },
    {
      "entropy": 5.742019510269165,
      "epoch": 4.423076923076923,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031908016166054933,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22605873495340348,
      "num_tokens": 101216484.0,
      "step": 48760
    },
    {
      "entropy": 5.713755035400391,
      "epoch": 4.423530478955008,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003190484318034659,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22195085436105727,
      "num_tokens": 101227169.0,
      "step": 48765
    },
    {
      "entropy": 5.74940447807312,
      "epoch": 4.423984034833092,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003190167010354499,
      "loss": 4.8418,
      "mean_token_accuracy": 0.22887637019157409,
      "num_tokens": 101236956.0,
      "step": 48770
    },
    {
      "entropy": 5.721866846084595,
      "epoch": 4.424437590711175,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003189849693571572,
      "loss": 4.89,
      "mean_token_accuracy": 0.22266714572906493,
      "num_tokens": 101248681.0,
      "step": 48775
    },
    {
      "entropy": 5.748760938644409,
      "epoch": 4.4248911465892595,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003189532367692441,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22438401281833648,
      "num_tokens": 101258375.0,
      "step": 48780
    },
    {
      "entropy": 5.696539497375488,
      "epoch": 4.425344702467344,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031892150327236693,
      "loss": 4.8561,
      "mean_token_accuracy": 0.23162932246923446,
      "num_tokens": 101268513.0,
      "step": 48785
    },
    {
      "entropy": 5.74338321685791,
      "epoch": 4.425798258345428,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003188897688671818,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21645080745220185,
      "num_tokens": 101278998.0,
      "step": 48790
    },
    {
      "entropy": 5.799574279785157,
      "epoch": 4.426251814223512,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031885803355434516,
      "loss": 4.924,
      "mean_token_accuracy": 0.22367642372846602,
      "num_tokens": 101289090.0,
      "step": 48795
    },
    {
      "entropy": 5.730767726898193,
      "epoch": 4.426705370101597,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003188262973345131,
      "loss": 4.8829,
      "mean_token_accuracy": 0.22427576929330825,
      "num_tokens": 101299697.0,
      "step": 48800
    },
    {
      "entropy": 5.759106540679932,
      "epoch": 4.427158925979681,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031879456020834213,
      "loss": 4.9627,
      "mean_token_accuracy": 0.21225423663854598,
      "num_tokens": 101310470.0,
      "step": 48805
    },
    {
      "entropy": 5.774667978286743,
      "epoch": 4.427612481857765,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031876282217648845,
      "loss": 4.866,
      "mean_token_accuracy": 0.22527284771203995,
      "num_tokens": 101321864.0,
      "step": 48810
    },
    {
      "entropy": 5.7451409816741945,
      "epoch": 4.428066037735849,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031873108323960845,
      "loss": 4.865,
      "mean_token_accuracy": 0.22762885689735413,
      "num_tokens": 101331302.0,
      "step": 48815
    },
    {
      "entropy": 5.725645351409912,
      "epoch": 4.428519593613933,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031869934339835856,
      "loss": 4.9349,
      "mean_token_accuracy": 0.22271935641765594,
      "num_tokens": 101341387.0,
      "step": 48820
    },
    {
      "entropy": 5.732679653167724,
      "epoch": 4.428973149492017,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031866760265339514,
      "loss": 4.8407,
      "mean_token_accuracy": 0.22871918678283693,
      "num_tokens": 101351235.0,
      "step": 48825
    },
    {
      "entropy": 5.740983295440674,
      "epoch": 4.429426705370101,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031863586100537453,
      "loss": 4.9071,
      "mean_token_accuracy": 0.21940479427576065,
      "num_tokens": 101361289.0,
      "step": 48830
    },
    {
      "entropy": 5.7149275779724125,
      "epoch": 4.429880261248186,
      "grad_norm": 1.25,
      "learning_rate": 0.0003186041184549531,
      "loss": 4.883,
      "mean_token_accuracy": 0.22640427500009536,
      "num_tokens": 101371666.0,
      "step": 48835
    },
    {
      "entropy": 5.721884775161743,
      "epoch": 4.43033381712627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003185723750027875,
      "loss": 4.8637,
      "mean_token_accuracy": 0.22990185171365737,
      "num_tokens": 101381195.0,
      "step": 48840
    },
    {
      "entropy": 5.799325561523437,
      "epoch": 4.430787373004354,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031854063064953403,
      "loss": 4.9506,
      "mean_token_accuracy": 0.22153268456459047,
      "num_tokens": 101392478.0,
      "step": 48845
    },
    {
      "entropy": 5.756163263320923,
      "epoch": 4.431240928882438,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003185088853958492,
      "loss": 4.9606,
      "mean_token_accuracy": 0.220029553771019,
      "num_tokens": 101405005.0,
      "step": 48850
    },
    {
      "entropy": 5.700409173965454,
      "epoch": 4.431694484760523,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031847713924238965,
      "loss": 4.8871,
      "mean_token_accuracy": 0.22249876856803893,
      "num_tokens": 101416588.0,
      "step": 48855
    },
    {
      "entropy": 5.703930473327636,
      "epoch": 4.432148040638607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031844539218981165,
      "loss": 4.8881,
      "mean_token_accuracy": 0.21804315447807313,
      "num_tokens": 101426784.0,
      "step": 48860
    },
    {
      "entropy": 5.698753738403321,
      "epoch": 4.432601596516691,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003184136442387719,
      "loss": 4.8258,
      "mean_token_accuracy": 0.22600532025098802,
      "num_tokens": 101437919.0,
      "step": 48865
    },
    {
      "entropy": 5.7561897277832035,
      "epoch": 4.433055152394775,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031838189538992693,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22611398994922638,
      "num_tokens": 101448654.0,
      "step": 48870
    },
    {
      "entropy": 5.764741325378418,
      "epoch": 4.433508708272859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003183501456439334,
      "loss": 4.823,
      "mean_token_accuracy": 0.22498007416725158,
      "num_tokens": 101458648.0,
      "step": 48875
    },
    {
      "entropy": 5.691751098632812,
      "epoch": 4.433962264150943,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031831839500144767,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22342903912067413,
      "num_tokens": 101469339.0,
      "step": 48880
    },
    {
      "entropy": 5.747754335403442,
      "epoch": 4.434415820029027,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031828664346312657,
      "loss": 4.8753,
      "mean_token_accuracy": 0.22656873911619185,
      "num_tokens": 101479452.0,
      "step": 48885
    },
    {
      "entropy": 5.721105909347534,
      "epoch": 4.434869375907112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003182548910296267,
      "loss": 4.85,
      "mean_token_accuracy": 0.22663989961147307,
      "num_tokens": 101489147.0,
      "step": 48890
    },
    {
      "entropy": 5.676346302032471,
      "epoch": 4.435322931785196,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003182231377016046,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23143116682767867,
      "num_tokens": 101498958.0,
      "step": 48895
    },
    {
      "entropy": 5.7676667213439945,
      "epoch": 4.43577648766328,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000318191383479717,
      "loss": 4.9651,
      "mean_token_accuracy": 0.2150108277797699,
      "num_tokens": 101509197.0,
      "step": 48900
    },
    {
      "entropy": 5.740822982788086,
      "epoch": 4.436230043541364,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003181596283646205,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22519649416208268,
      "num_tokens": 101520018.0,
      "step": 48905
    },
    {
      "entropy": 5.7448571681976315,
      "epoch": 4.436683599419449,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031812787235697206,
      "loss": 4.9051,
      "mean_token_accuracy": 0.2252930447459221,
      "num_tokens": 101530755.0,
      "step": 48910
    },
    {
      "entropy": 5.714315605163574,
      "epoch": 4.437137155297533,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031809611545742804,
      "loss": 4.9384,
      "mean_token_accuracy": 0.21533193439245224,
      "num_tokens": 101541028.0,
      "step": 48915
    },
    {
      "entropy": 5.736772871017456,
      "epoch": 4.437590711175617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003180643576666455,
      "loss": 4.8477,
      "mean_token_accuracy": 0.226355642080307,
      "num_tokens": 101550826.0,
      "step": 48920
    },
    {
      "entropy": 5.7620837688446045,
      "epoch": 4.4380442670537015,
      "grad_norm": 1.375,
      "learning_rate": 0.0003180325989852811,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2267605632543564,
      "num_tokens": 101560657.0,
      "step": 48925
    },
    {
      "entropy": 5.693183851242066,
      "epoch": 4.438497822931785,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003180008394139914,
      "loss": 4.8639,
      "mean_token_accuracy": 0.22989336550235748,
      "num_tokens": 101570504.0,
      "step": 48930
    },
    {
      "entropy": 5.744651794433594,
      "epoch": 4.438951378809869,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031796907895343356,
      "loss": 5.0022,
      "mean_token_accuracy": 0.21743665635585785,
      "num_tokens": 101581729.0,
      "step": 48935
    },
    {
      "entropy": 5.794562101364136,
      "epoch": 4.439404934687953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031793731760426414,
      "loss": 4.959,
      "mean_token_accuracy": 0.21915254294872283,
      "num_tokens": 101592649.0,
      "step": 48940
    },
    {
      "entropy": 5.860422849655151,
      "epoch": 4.439858490566038,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031790555536714013,
      "loss": 5.008,
      "mean_token_accuracy": 0.22090056240558625,
      "num_tokens": 101603824.0,
      "step": 48945
    },
    {
      "entropy": 5.760408687591553,
      "epoch": 4.440312046444122,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003178737922427182,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22545047253370284,
      "num_tokens": 101615659.0,
      "step": 48950
    },
    {
      "entropy": 5.82798581123352,
      "epoch": 4.440765602322206,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003178420282316554,
      "loss": 4.9644,
      "mean_token_accuracy": 0.21739823818206788,
      "num_tokens": 101625937.0,
      "step": 48955
    },
    {
      "entropy": 5.739239501953125,
      "epoch": 4.4412191582002905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003178102633346085,
      "loss": 4.858,
      "mean_token_accuracy": 0.2284615993499756,
      "num_tokens": 101637352.0,
      "step": 48960
    },
    {
      "entropy": 5.737011051177978,
      "epoch": 4.441672714078375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003177784975522344,
      "loss": 4.8235,
      "mean_token_accuracy": 0.2323366031050682,
      "num_tokens": 101647939.0,
      "step": 48965
    },
    {
      "entropy": 5.798782539367676,
      "epoch": 4.442126269956459,
      "grad_norm": 1.5,
      "learning_rate": 0.00031774673088519014,
      "loss": 4.9266,
      "mean_token_accuracy": 0.22166964411735535,
      "num_tokens": 101658200.0,
      "step": 48970
    },
    {
      "entropy": 5.746160221099854,
      "epoch": 4.442579825834542,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003177149633341326,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2204037219285965,
      "num_tokens": 101668624.0,
      "step": 48975
    },
    {
      "entropy": 5.801127529144287,
      "epoch": 4.443033381712627,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003176831948997187,
      "loss": 5.0195,
      "mean_token_accuracy": 0.21150715351104737,
      "num_tokens": 101678939.0,
      "step": 48980
    },
    {
      "entropy": 5.725743818283081,
      "epoch": 4.443486937590711,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003176514255826054,
      "loss": 4.9223,
      "mean_token_accuracy": 0.22739204317331313,
      "num_tokens": 101689365.0,
      "step": 48985
    },
    {
      "entropy": 5.7228700637817385,
      "epoch": 4.443940493468795,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003176196553834498,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22164728492498398,
      "num_tokens": 101699429.0,
      "step": 48990
    },
    {
      "entropy": 5.729933214187622,
      "epoch": 4.444394049346879,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031758788430290893,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2244092971086502,
      "num_tokens": 101709452.0,
      "step": 48995
    },
    {
      "entropy": 5.688572883605957,
      "epoch": 4.444847605224964,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031755611234163966,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22563354820013046,
      "num_tokens": 101719783.0,
      "step": 49000
    },
    {
      "entropy": 5.703445768356323,
      "epoch": 4.445301161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003175243395002992,
      "loss": 4.818,
      "mean_token_accuracy": 0.2353411450982094,
      "num_tokens": 101729100.0,
      "step": 49005
    },
    {
      "entropy": 5.716486024856567,
      "epoch": 4.445754716981132,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031749256577954453,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2259763613343239,
      "num_tokens": 101738711.0,
      "step": 49010
    },
    {
      "entropy": 5.691438293457031,
      "epoch": 4.4462082728592165,
      "grad_norm": 1.375,
      "learning_rate": 0.0003174607911800328,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2269422233104706,
      "num_tokens": 101749624.0,
      "step": 49015
    },
    {
      "entropy": 5.703809213638306,
      "epoch": 4.446661828737301,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031742901570242107,
      "loss": 4.8325,
      "mean_token_accuracy": 0.22933371514081954,
      "num_tokens": 101760951.0,
      "step": 49020
    },
    {
      "entropy": 5.724318885803223,
      "epoch": 4.447115384615385,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00031739723934736655,
      "loss": 4.8004,
      "mean_token_accuracy": 0.2307110235095024,
      "num_tokens": 101770773.0,
      "step": 49025
    },
    {
      "entropy": 5.752127408981323,
      "epoch": 4.447568940493468,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003173654621155262,
      "loss": 4.9014,
      "mean_token_accuracy": 0.2192749947309494,
      "num_tokens": 101781566.0,
      "step": 49030
    },
    {
      "entropy": 5.693458795547485,
      "epoch": 4.448022496371553,
      "grad_norm": 1.375,
      "learning_rate": 0.00031733368400755735,
      "loss": 4.8964,
      "mean_token_accuracy": 0.23380762189626694,
      "num_tokens": 101792268.0,
      "step": 49035
    },
    {
      "entropy": 5.755898666381836,
      "epoch": 4.448476052249637,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031730190502411715,
      "loss": 4.934,
      "mean_token_accuracy": 0.21993439495563508,
      "num_tokens": 101802500.0,
      "step": 49040
    },
    {
      "entropy": 5.773810338973999,
      "epoch": 4.448929608127721,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003172701251658628,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2251031905412674,
      "num_tokens": 101813851.0,
      "step": 49045
    },
    {
      "entropy": 5.807064008712769,
      "epoch": 4.4493831640058055,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031723834443345134,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21961747109889984,
      "num_tokens": 101824774.0,
      "step": 49050
    },
    {
      "entropy": 5.689844179153442,
      "epoch": 4.44983671988389,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00031720656282754027,
      "loss": 4.7681,
      "mean_token_accuracy": 0.24463718384504318,
      "num_tokens": 101835975.0,
      "step": 49055
    },
    {
      "entropy": 5.673431921005249,
      "epoch": 4.450290275761974,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031717478034878676,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23135290294885635,
      "num_tokens": 101846764.0,
      "step": 49060
    },
    {
      "entropy": 5.696204423904419,
      "epoch": 4.450743831640058,
      "grad_norm": 1.453125,
      "learning_rate": 0.00031714299699784796,
      "loss": 4.8869,
      "mean_token_accuracy": 0.22151027768850326,
      "num_tokens": 101857968.0,
      "step": 49065
    },
    {
      "entropy": 5.709985065460205,
      "epoch": 4.4511973875181425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003171112127753813,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22729405760765076,
      "num_tokens": 101868484.0,
      "step": 49070
    },
    {
      "entropy": 5.767145490646362,
      "epoch": 4.451650943396227,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003170794276820441,
      "loss": 4.8821,
      "mean_token_accuracy": 0.22729930877685547,
      "num_tokens": 101877969.0,
      "step": 49075
    },
    {
      "entropy": 5.6975836753845215,
      "epoch": 4.452104499274311,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031704764171849344,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22894588857889175,
      "num_tokens": 101888635.0,
      "step": 49080
    },
    {
      "entropy": 5.726966762542725,
      "epoch": 4.4525580551523944,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031701585488538695,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22839662730693816,
      "num_tokens": 101898340.0,
      "step": 49085
    },
    {
      "entropy": 5.783838510513306,
      "epoch": 4.453011611030479,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003169840671833819,
      "loss": 4.9477,
      "mean_token_accuracy": 0.2159852385520935,
      "num_tokens": 101908993.0,
      "step": 49090
    },
    {
      "entropy": 5.6716022968292235,
      "epoch": 4.453465166908563,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003169522786131355,
      "loss": 4.8047,
      "mean_token_accuracy": 0.22992924451828003,
      "num_tokens": 101919675.0,
      "step": 49095
    },
    {
      "entropy": 5.762811136245728,
      "epoch": 4.453918722786647,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031692048917530535,
      "loss": 4.9518,
      "mean_token_accuracy": 0.22302173376083373,
      "num_tokens": 101929949.0,
      "step": 49100
    },
    {
      "entropy": 5.705183792114258,
      "epoch": 4.4543722786647315,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031688869887054887,
      "loss": 4.7594,
      "mean_token_accuracy": 0.24064428508281707,
      "num_tokens": 101940025.0,
      "step": 49105
    },
    {
      "entropy": 5.7144303798675535,
      "epoch": 4.454825834542816,
      "grad_norm": 1.25,
      "learning_rate": 0.00031685690769952336,
      "loss": 4.8125,
      "mean_token_accuracy": 0.23339764177799224,
      "num_tokens": 101950344.0,
      "step": 49110
    },
    {
      "entropy": 5.775491666793823,
      "epoch": 4.4552793904209,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003168251156628863,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22839764803647994,
      "num_tokens": 101961405.0,
      "step": 49115
    },
    {
      "entropy": 5.678362607955933,
      "epoch": 4.455732946298984,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00031679332276129523,
      "loss": 4.7631,
      "mean_token_accuracy": 0.2325708195567131,
      "num_tokens": 101972164.0,
      "step": 49120
    },
    {
      "entropy": 5.688220548629761,
      "epoch": 4.456186502177069,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031676152899540763,
      "loss": 4.8557,
      "mean_token_accuracy": 0.22484685182571412,
      "num_tokens": 101981573.0,
      "step": 49125
    },
    {
      "entropy": 5.7138659954071045,
      "epoch": 4.456640058055152,
      "grad_norm": 1.484375,
      "learning_rate": 0.000316729734365881,
      "loss": 4.8437,
      "mean_token_accuracy": 0.22443462610244752,
      "num_tokens": 101991051.0,
      "step": 49130
    },
    {
      "entropy": 5.736844158172607,
      "epoch": 4.457093613933236,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003166979388733727,
      "loss": 4.8527,
      "mean_token_accuracy": 0.23277666568756103,
      "num_tokens": 102000003.0,
      "step": 49135
    },
    {
      "entropy": 5.753327703475952,
      "epoch": 4.4575471698113205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003166661425185405,
      "loss": 4.9246,
      "mean_token_accuracy": 0.21641442477703093,
      "num_tokens": 102010489.0,
      "step": 49140
    },
    {
      "entropy": 5.731748294830322,
      "epoch": 4.458000725689405,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031663434530204183,
      "loss": 4.933,
      "mean_token_accuracy": 0.21470169126987457,
      "num_tokens": 102020486.0,
      "step": 49145
    },
    {
      "entropy": 5.7492810726165775,
      "epoch": 4.458454281567489,
      "grad_norm": 1.25,
      "learning_rate": 0.0003166025472245342,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2260649934411049,
      "num_tokens": 102030098.0,
      "step": 49150
    },
    {
      "entropy": 5.762627458572387,
      "epoch": 4.458907837445573,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003165707482866754,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22536997348070145,
      "num_tokens": 102040850.0,
      "step": 49155
    },
    {
      "entropy": 5.696212673187256,
      "epoch": 4.4593613933236576,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031653894848912296,
      "loss": 4.8144,
      "mean_token_accuracy": 0.2278676763176918,
      "num_tokens": 102051056.0,
      "step": 49160
    },
    {
      "entropy": 5.776078081130981,
      "epoch": 4.459814949201742,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003165071478325344,
      "loss": 4.8825,
      "mean_token_accuracy": 0.21998630464076996,
      "num_tokens": 102061727.0,
      "step": 49165
    },
    {
      "entropy": 5.731180191040039,
      "epoch": 4.460268505079826,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003164753463175675,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22445057481527328,
      "num_tokens": 102071501.0,
      "step": 49170
    },
    {
      "entropy": 5.732015609741211,
      "epoch": 4.46072206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031644354394487977,
      "loss": 4.82,
      "mean_token_accuracy": 0.2372695341706276,
      "num_tokens": 102082511.0,
      "step": 49175
    },
    {
      "entropy": 5.772749900817871,
      "epoch": 4.461175616835995,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031641174071512905,
      "loss": 4.876,
      "mean_token_accuracy": 0.23044679164886475,
      "num_tokens": 102093010.0,
      "step": 49180
    },
    {
      "entropy": 5.693564939498901,
      "epoch": 4.461629172714078,
      "grad_norm": 1.421875,
      "learning_rate": 0.000316379936628973,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2262322798371315,
      "num_tokens": 102102691.0,
      "step": 49185
    },
    {
      "entropy": 5.728352689743042,
      "epoch": 4.462082728592162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031634813168706933,
      "loss": 4.8414,
      "mean_token_accuracy": 0.2285582035779953,
      "num_tokens": 102112726.0,
      "step": 49190
    },
    {
      "entropy": 5.716710710525513,
      "epoch": 4.4625362844702465,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00031631632589007575,
      "loss": 4.8859,
      "mean_token_accuracy": 0.23214833587408065,
      "num_tokens": 102122879.0,
      "step": 49195
    },
    {
      "entropy": 5.748193788528442,
      "epoch": 4.462989840348331,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031628451923864997,
      "loss": 4.8954,
      "mean_token_accuracy": 0.22496644258499146,
      "num_tokens": 102133485.0,
      "step": 49200
    },
    {
      "entropy": 5.7516378402709964,
      "epoch": 4.463443396226415,
      "grad_norm": 1.375,
      "learning_rate": 0.00031625271173344984,
      "loss": 4.9688,
      "mean_token_accuracy": 0.2203364849090576,
      "num_tokens": 102143439.0,
      "step": 49205
    },
    {
      "entropy": 5.676785564422607,
      "epoch": 4.463896952104499,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031622090337513314,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23121090233325958,
      "num_tokens": 102153808.0,
      "step": 49210
    },
    {
      "entropy": 5.815995931625366,
      "epoch": 4.464350507982584,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031618909416435766,
      "loss": 4.9643,
      "mean_token_accuracy": 0.2159768134355545,
      "num_tokens": 102163375.0,
      "step": 49215
    },
    {
      "entropy": 5.7431704044342045,
      "epoch": 4.464804063860668,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003161572841017811,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22446552515029908,
      "num_tokens": 102174402.0,
      "step": 49220
    },
    {
      "entropy": 5.645090198516845,
      "epoch": 4.465257619738752,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031612547318806154,
      "loss": 4.7654,
      "mean_token_accuracy": 0.2315055936574936,
      "num_tokens": 102184446.0,
      "step": 49225
    },
    {
      "entropy": 5.721493530273437,
      "epoch": 4.465711175616836,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031609366142385675,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22369036078453064,
      "num_tokens": 102194398.0,
      "step": 49230
    },
    {
      "entropy": 5.682028532028198,
      "epoch": 4.46616473149492,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003160618488098245,
      "loss": 4.823,
      "mean_token_accuracy": 0.23682934492826463,
      "num_tokens": 102205953.0,
      "step": 49235
    },
    {
      "entropy": 5.703773307800293,
      "epoch": 4.466618287373004,
      "grad_norm": 1.4375,
      "learning_rate": 0.00031603003534662273,
      "loss": 4.8151,
      "mean_token_accuracy": 0.23893739581108092,
      "num_tokens": 102215815.0,
      "step": 49240
    },
    {
      "entropy": 5.735068368911743,
      "epoch": 4.467071843251088,
      "grad_norm": 1.25,
      "learning_rate": 0.00031599822103490946,
      "loss": 4.8136,
      "mean_token_accuracy": 0.2313885882496834,
      "num_tokens": 102225703.0,
      "step": 49245
    },
    {
      "entropy": 5.788575983047485,
      "epoch": 4.467525399129173,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003159664058753425,
      "loss": 5.0049,
      "mean_token_accuracy": 0.20858490020036696,
      "num_tokens": 102236627.0,
      "step": 49250
    },
    {
      "entropy": 5.704589939117431,
      "epoch": 4.467978955007257,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003159345898685797,
      "loss": 4.892,
      "mean_token_accuracy": 0.22961045056581497,
      "num_tokens": 102246539.0,
      "step": 49255
    },
    {
      "entropy": 5.747065639495849,
      "epoch": 4.468432510885341,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003159027730152793,
      "loss": 4.8962,
      "mean_token_accuracy": 0.22550927400588988,
      "num_tokens": 102256734.0,
      "step": 49260
    },
    {
      "entropy": 5.786070680618286,
      "epoch": 4.468886066763425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003158709553160991,
      "loss": 4.9956,
      "mean_token_accuracy": 0.21524903923273087,
      "num_tokens": 102267469.0,
      "step": 49265
    },
    {
      "entropy": 5.7830558776855465,
      "epoch": 4.46933962264151,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031583913677169705,
      "loss": 4.9262,
      "mean_token_accuracy": 0.22066075950860978,
      "num_tokens": 102278435.0,
      "step": 49270
    },
    {
      "entropy": 5.804447650909424,
      "epoch": 4.469793178519594,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031580731738273135,
      "loss": 4.9179,
      "mean_token_accuracy": 0.2189957618713379,
      "num_tokens": 102289171.0,
      "step": 49275
    },
    {
      "entropy": 5.732480716705322,
      "epoch": 4.470246734397678,
      "grad_norm": 1.25,
      "learning_rate": 0.0003157754971498599,
      "loss": 4.8324,
      "mean_token_accuracy": 0.2376865953207016,
      "num_tokens": 102299254.0,
      "step": 49280
    },
    {
      "entropy": 5.786828851699829,
      "epoch": 4.4707002902757615,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003157436760737408,
      "loss": 4.9181,
      "mean_token_accuracy": 0.219608810544014,
      "num_tokens": 102308471.0,
      "step": 49285
    },
    {
      "entropy": 5.760569143295288,
      "epoch": 4.471153846153846,
      "grad_norm": 1.375,
      "learning_rate": 0.000315711854155032,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22248756885528564,
      "num_tokens": 102319005.0,
      "step": 49290
    },
    {
      "entropy": 5.73257212638855,
      "epoch": 4.47160740203193,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031568003139439185,
      "loss": 4.8664,
      "mean_token_accuracy": 0.22900616377592087,
      "num_tokens": 102329568.0,
      "step": 49295
    },
    {
      "entropy": 5.818240070343018,
      "epoch": 4.472060957910014,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003156482077924781,
      "loss": 4.996,
      "mean_token_accuracy": 0.21725784838199616,
      "num_tokens": 102340810.0,
      "step": 49300
    },
    {
      "entropy": 5.689255714416504,
      "epoch": 4.472514513788099,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003156163833499491,
      "loss": 4.877,
      "mean_token_accuracy": 0.22169329822063447,
      "num_tokens": 102351305.0,
      "step": 49305
    },
    {
      "entropy": 5.72937126159668,
      "epoch": 4.472968069666183,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031558455806746306,
      "loss": 4.901,
      "mean_token_accuracy": 0.2195478931069374,
      "num_tokens": 102362057.0,
      "step": 49310
    },
    {
      "entropy": 5.802380132675171,
      "epoch": 4.473421625544267,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003155527319456779,
      "loss": 5.002,
      "mean_token_accuracy": 0.21766447275877,
      "num_tokens": 102372612.0,
      "step": 49315
    },
    {
      "entropy": 5.702709436416626,
      "epoch": 4.473875181422351,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031552090498525195,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22574374377727507,
      "num_tokens": 102383337.0,
      "step": 49320
    },
    {
      "entropy": 5.752618217468262,
      "epoch": 4.474328737300436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003154890771868433,
      "loss": 4.8939,
      "mean_token_accuracy": 0.22221102565526962,
      "num_tokens": 102394196.0,
      "step": 49325
    },
    {
      "entropy": 5.7886061668396,
      "epoch": 4.47478229317852,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031545724855111025,
      "loss": 4.9099,
      "mean_token_accuracy": 0.2263663738965988,
      "num_tokens": 102404677.0,
      "step": 49330
    },
    {
      "entropy": 5.754249620437622,
      "epoch": 4.475235849056604,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031542541907871095,
      "loss": 4.8498,
      "mean_token_accuracy": 0.23645256459712982,
      "num_tokens": 102413445.0,
      "step": 49335
    },
    {
      "entropy": 5.689929819107055,
      "epoch": 4.475689404934688,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003153935887703038,
      "loss": 4.8844,
      "mean_token_accuracy": 0.2274179071187973,
      "num_tokens": 102424527.0,
      "step": 49340
    },
    {
      "entropy": 5.741129970550537,
      "epoch": 4.476142960812772,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003153617576265469,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22165639102458953,
      "num_tokens": 102434940.0,
      "step": 49345
    },
    {
      "entropy": 5.797518396377564,
      "epoch": 4.476596516690856,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00031532992564809847,
      "loss": 4.9525,
      "mean_token_accuracy": 0.228712859749794,
      "num_tokens": 102446678.0,
      "step": 49350
    },
    {
      "entropy": 5.710380125045776,
      "epoch": 4.47705007256894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000315298092835617,
      "loss": 4.86,
      "mean_token_accuracy": 0.22885988056659698,
      "num_tokens": 102457255.0,
      "step": 49355
    },
    {
      "entropy": 5.776284122467041,
      "epoch": 4.477503628447025,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031526625918976066,
      "loss": 5.0188,
      "mean_token_accuracy": 0.22064112722873688,
      "num_tokens": 102468385.0,
      "step": 49360
    },
    {
      "entropy": 5.7420470237731935,
      "epoch": 4.477957184325109,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031523442471118783,
      "loss": 4.8408,
      "mean_token_accuracy": 0.22347273230552672,
      "num_tokens": 102478204.0,
      "step": 49365
    },
    {
      "entropy": 5.7758753299713135,
      "epoch": 4.478410740203193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031520258940055684,
      "loss": 4.9129,
      "mean_token_accuracy": 0.22564130872488022,
      "num_tokens": 102488492.0,
      "step": 49370
    },
    {
      "entropy": 5.714055585861206,
      "epoch": 4.4788642960812775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031517075325852606,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22744178026914597,
      "num_tokens": 102500309.0,
      "step": 49375
    },
    {
      "entropy": 5.677381563186645,
      "epoch": 4.479317851959362,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003151389162857539,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22619787156581878,
      "num_tokens": 102511196.0,
      "step": 49380
    },
    {
      "entropy": 5.722521209716797,
      "epoch": 4.479771407837446,
      "grad_norm": 1.25,
      "learning_rate": 0.00031510707848289864,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22958897948265075,
      "num_tokens": 102522100.0,
      "step": 49385
    },
    {
      "entropy": 5.710387897491455,
      "epoch": 4.480224963715529,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031507523985061886,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23330094814300537,
      "num_tokens": 102531677.0,
      "step": 49390
    },
    {
      "entropy": 5.685977220535278,
      "epoch": 4.480678519593614,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031504340038957294,
      "loss": 4.7689,
      "mean_token_accuracy": 0.24128725230693818,
      "num_tokens": 102541854.0,
      "step": 49395
    },
    {
      "entropy": 5.718473720550537,
      "epoch": 4.481132075471698,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003150115601004192,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22959458231925964,
      "num_tokens": 102551187.0,
      "step": 49400
    },
    {
      "entropy": 5.753218412399292,
      "epoch": 4.481585631349782,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003149797189838162,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22136911898851394,
      "num_tokens": 102562568.0,
      "step": 49405
    },
    {
      "entropy": 5.751283502578735,
      "epoch": 4.482039187227866,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031494787704042243,
      "loss": 4.9237,
      "mean_token_accuracy": 0.21781037449836732,
      "num_tokens": 102573170.0,
      "step": 49410
    },
    {
      "entropy": 5.65772500038147,
      "epoch": 4.482492743105951,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031491603427089644,
      "loss": 4.7331,
      "mean_token_accuracy": 0.23799602836370468,
      "num_tokens": 102583950.0,
      "step": 49415
    },
    {
      "entropy": 5.702438497543335,
      "epoch": 4.482946298984035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031488419067589665,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22685442864894867,
      "num_tokens": 102594370.0,
      "step": 49420
    },
    {
      "entropy": 5.763774585723877,
      "epoch": 4.483399854862119,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003148523462560816,
      "loss": 4.849,
      "mean_token_accuracy": 0.23233221769332885,
      "num_tokens": 102604837.0,
      "step": 49425
    },
    {
      "entropy": 5.757041358947754,
      "epoch": 4.4838534107402035,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003148205010121099,
      "loss": 4.9396,
      "mean_token_accuracy": 0.22388678640127183,
      "num_tokens": 102616170.0,
      "step": 49430
    },
    {
      "entropy": 5.6953002452850345,
      "epoch": 4.484306966618288,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003147886549446401,
      "loss": 4.881,
      "mean_token_accuracy": 0.22613926380872726,
      "num_tokens": 102627283.0,
      "step": 49435
    },
    {
      "entropy": 5.777172231674195,
      "epoch": 4.484760522496371,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003147568080543308,
      "loss": 4.919,
      "mean_token_accuracy": 0.2193957597017288,
      "num_tokens": 102637087.0,
      "step": 49440
    },
    {
      "entropy": 5.7826484680175785,
      "epoch": 4.485214078374455,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003147249603418406,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2259744182229042,
      "num_tokens": 102646818.0,
      "step": 49445
    },
    {
      "entropy": 5.708488512039184,
      "epoch": 4.48566763425254,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003146931118078281,
      "loss": 4.8171,
      "mean_token_accuracy": 0.2375792533159256,
      "num_tokens": 102656940.0,
      "step": 49450
    },
    {
      "entropy": 5.69768123626709,
      "epoch": 4.486121190130624,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003146612624529518,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22342426478862762,
      "num_tokens": 102666841.0,
      "step": 49455
    },
    {
      "entropy": 5.728089237213135,
      "epoch": 4.486574746008708,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031462941227787056,
      "loss": 4.8861,
      "mean_token_accuracy": 0.22213256359100342,
      "num_tokens": 102676821.0,
      "step": 49460
    },
    {
      "entropy": 5.833434820175171,
      "epoch": 4.4870283018867925,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003145975612832431,
      "loss": 4.9733,
      "mean_token_accuracy": 0.22181185334920883,
      "num_tokens": 102687340.0,
      "step": 49465
    },
    {
      "entropy": 5.71254677772522,
      "epoch": 4.487481857764877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003145657094697278,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22571255266666412,
      "num_tokens": 102698320.0,
      "step": 49470
    },
    {
      "entropy": 5.773758792877198,
      "epoch": 4.487935413642961,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003145338568379836,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22482330054044725,
      "num_tokens": 102708685.0,
      "step": 49475
    },
    {
      "entropy": 5.722334480285644,
      "epoch": 4.488388969521045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003145020033886692,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22183025479316712,
      "num_tokens": 102719680.0,
      "step": 49480
    },
    {
      "entropy": 5.670054483413696,
      "epoch": 4.4888425253991295,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031447014912244325,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23035450279712677,
      "num_tokens": 102731156.0,
      "step": 49485
    },
    {
      "entropy": 5.706424236297607,
      "epoch": 4.489296081277214,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003144382940399646,
      "loss": 4.8202,
      "mean_token_accuracy": 0.2296672135591507,
      "num_tokens": 102740790.0,
      "step": 49490
    },
    {
      "entropy": 5.701715421676636,
      "epoch": 4.489749637155297,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000314406438141892,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22680283188819886,
      "num_tokens": 102750672.0,
      "step": 49495
    },
    {
      "entropy": 5.666266012191772,
      "epoch": 4.4902031930333814,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003143745814288842,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2217583879828453,
      "num_tokens": 102760364.0,
      "step": 49500
    },
    {
      "entropy": 5.828448867797851,
      "epoch": 4.490656748911466,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003143427239016001,
      "loss": 4.9614,
      "mean_token_accuracy": 0.222805655002594,
      "num_tokens": 102770415.0,
      "step": 49505
    },
    {
      "entropy": 5.7486738681793215,
      "epoch": 4.49111030478955,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003143108655606983,
      "loss": 4.8599,
      "mean_token_accuracy": 0.23174078911542892,
      "num_tokens": 102781203.0,
      "step": 49510
    },
    {
      "entropy": 5.704688501358032,
      "epoch": 4.491563860667634,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00031427900640683783,
      "loss": 4.8239,
      "mean_token_accuracy": 0.2291004851460457,
      "num_tokens": 102791071.0,
      "step": 49515
    },
    {
      "entropy": 5.7474946022033695,
      "epoch": 4.4920174165457185,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00031424714644067754,
      "loss": 4.9576,
      "mean_token_accuracy": 0.21674390584230424,
      "num_tokens": 102801755.0,
      "step": 49520
    },
    {
      "entropy": 5.75717716217041,
      "epoch": 4.492470972423803,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031421528566287623,
      "loss": 4.9181,
      "mean_token_accuracy": 0.223215688765049,
      "num_tokens": 102812947.0,
      "step": 49525
    },
    {
      "entropy": 5.708867168426513,
      "epoch": 4.492924528301887,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003141834240740928,
      "loss": 4.8766,
      "mean_token_accuracy": 0.22716851383447648,
      "num_tokens": 102822917.0,
      "step": 49530
    },
    {
      "entropy": 5.752626943588257,
      "epoch": 4.493378084179971,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003141515616749862,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22236350327730178,
      "num_tokens": 102833102.0,
      "step": 49535
    },
    {
      "entropy": 5.780746555328369,
      "epoch": 4.493831640058055,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003141196984662153,
      "loss": 4.9045,
      "mean_token_accuracy": 0.23009061217308044,
      "num_tokens": 102843888.0,
      "step": 49540
    },
    {
      "entropy": 5.751823806762696,
      "epoch": 4.494285195936139,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003140878344484391,
      "loss": 4.9364,
      "mean_token_accuracy": 0.2282428875565529,
      "num_tokens": 102854417.0,
      "step": 49545
    },
    {
      "entropy": 5.686506366729736,
      "epoch": 4.494738751814223,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003140559696223165,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23007576912641525,
      "num_tokens": 102865206.0,
      "step": 49550
    },
    {
      "entropy": 5.778037166595459,
      "epoch": 4.4951923076923075,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003140241039885065,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22605589628219605,
      "num_tokens": 102874967.0,
      "step": 49555
    },
    {
      "entropy": 5.693698740005493,
      "epoch": 4.495645863570392,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003139922375476681,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22454414367675782,
      "num_tokens": 102884808.0,
      "step": 49560
    },
    {
      "entropy": 5.658512020111084,
      "epoch": 4.496099419448476,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003139603703004603,
      "loss": 4.7869,
      "mean_token_accuracy": 0.23047556281089782,
      "num_tokens": 102895206.0,
      "step": 49565
    },
    {
      "entropy": 5.782391834259033,
      "epoch": 4.49655297532656,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003139285022475421,
      "loss": 4.919,
      "mean_token_accuracy": 0.22480683773756027,
      "num_tokens": 102906658.0,
      "step": 49570
    },
    {
      "entropy": 5.719053649902344,
      "epoch": 4.4970065312046446,
      "grad_norm": 1.34375,
      "learning_rate": 0.00031389663338957254,
      "loss": 4.8748,
      "mean_token_accuracy": 0.221991603076458,
      "num_tokens": 102917071.0,
      "step": 49575
    },
    {
      "entropy": 5.707975673675537,
      "epoch": 4.497460087082729,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031386476372721067,
      "loss": 4.8283,
      "mean_token_accuracy": 0.22875060886144638,
      "num_tokens": 102927280.0,
      "step": 49580
    },
    {
      "entropy": 5.730584669113159,
      "epoch": 4.497913642960813,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003138328932611156,
      "loss": 4.951,
      "mean_token_accuracy": 0.2172462597489357,
      "num_tokens": 102937670.0,
      "step": 49585
    },
    {
      "entropy": 5.711715841293335,
      "epoch": 4.498367198838897,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031380102199194646,
      "loss": 4.8943,
      "mean_token_accuracy": 0.22236790060997008,
      "num_tokens": 102947874.0,
      "step": 49590
    },
    {
      "entropy": 5.779786682128906,
      "epoch": 4.498820754716981,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003137691499203622,
      "loss": 4.9264,
      "mean_token_accuracy": 0.2165897086262703,
      "num_tokens": 102959311.0,
      "step": 49595
    },
    {
      "entropy": 5.734201860427857,
      "epoch": 4.499274310595065,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031373727704702217,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22832054793834686,
      "num_tokens": 102969866.0,
      "step": 49600
    },
    {
      "entropy": 5.727122783660889,
      "epoch": 4.499727866473149,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031370540337258523,
      "loss": 4.8184,
      "mean_token_accuracy": 0.23201760798692703,
      "num_tokens": 102979672.0,
      "step": 49605
    },
    {
      "entropy": 5.688215112686157,
      "epoch": 4.5001814223512335,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003136735288977108,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22349701225757598,
      "num_tokens": 102991050.0,
      "step": 49610
    },
    {
      "entropy": 5.72427773475647,
      "epoch": 4.500634978229318,
      "grad_norm": 1.359375,
      "learning_rate": 0.00031364165362305783,
      "loss": 4.9141,
      "mean_token_accuracy": 0.22163047194480895,
      "num_tokens": 103001700.0,
      "step": 49615
    },
    {
      "entropy": 5.753887081146241,
      "epoch": 4.501088534107402,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003136097775492857,
      "loss": 4.9142,
      "mean_token_accuracy": 0.22287559807300567,
      "num_tokens": 103013013.0,
      "step": 49620
    },
    {
      "entropy": 5.7316002368927,
      "epoch": 4.501542089985486,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003135779006770535,
      "loss": 4.8908,
      "mean_token_accuracy": 0.2285076290369034,
      "num_tokens": 103024110.0,
      "step": 49625
    },
    {
      "entropy": 5.78897066116333,
      "epoch": 4.501995645863571,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003135460230070205,
      "loss": 4.92,
      "mean_token_accuracy": 0.21923914551734924,
      "num_tokens": 103034991.0,
      "step": 49630
    },
    {
      "entropy": 5.782354879379272,
      "epoch": 4.502449201741655,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031351414453984595,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22388073652982712,
      "num_tokens": 103045634.0,
      "step": 49635
    },
    {
      "entropy": 5.711866092681885,
      "epoch": 4.502902757619739,
      "grad_norm": 1.3125,
      "learning_rate": 0.000313482265276189,
      "loss": 4.8872,
      "mean_token_accuracy": 0.2268739104270935,
      "num_tokens": 103054999.0,
      "step": 49640
    },
    {
      "entropy": 5.721697092056274,
      "epoch": 4.503356313497823,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031345038521670907,
      "loss": 4.8965,
      "mean_token_accuracy": 0.21726747304201127,
      "num_tokens": 103065163.0,
      "step": 49645
    },
    {
      "entropy": 5.778688240051269,
      "epoch": 4.503809869375907,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003134185043620654,
      "loss": 4.8673,
      "mean_token_accuracy": 0.235277396440506,
      "num_tokens": 103075979.0,
      "step": 49650
    },
    {
      "entropy": 5.67758617401123,
      "epoch": 4.504263425253991,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003133866227129171,
      "loss": 4.7699,
      "mean_token_accuracy": 0.22777113914489747,
      "num_tokens": 103086124.0,
      "step": 49655
    },
    {
      "entropy": 5.66511754989624,
      "epoch": 4.504716981132075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031335474026992377,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22301192581653595,
      "num_tokens": 103095330.0,
      "step": 49660
    },
    {
      "entropy": 5.719798755645752,
      "epoch": 4.50517053701016,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003133228570337447,
      "loss": 4.9148,
      "mean_token_accuracy": 0.22455461621284484,
      "num_tokens": 103106863.0,
      "step": 49665
    },
    {
      "entropy": 5.795332956314087,
      "epoch": 4.505624092888244,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003132909730050391,
      "loss": 4.9212,
      "mean_token_accuracy": 0.22256566733121871,
      "num_tokens": 103117256.0,
      "step": 49670
    },
    {
      "entropy": 5.766191577911377,
      "epoch": 4.506077648766328,
      "grad_norm": 1.25,
      "learning_rate": 0.00031325908818446643,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22910598516464234,
      "num_tokens": 103127601.0,
      "step": 49675
    },
    {
      "entropy": 5.742695999145508,
      "epoch": 4.506531204644412,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00031322720257268607,
      "loss": 4.8205,
      "mean_token_accuracy": 0.22380074560642244,
      "num_tokens": 103136868.0,
      "step": 49680
    },
    {
      "entropy": 5.724640893936157,
      "epoch": 4.506984760522497,
      "grad_norm": 1.40625,
      "learning_rate": 0.00031319531617035747,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22870322763919831,
      "num_tokens": 103146908.0,
      "step": 49685
    },
    {
      "entropy": 5.661006021499634,
      "epoch": 4.507438316400581,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031316342897813994,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22425456196069718,
      "num_tokens": 103157244.0,
      "step": 49690
    },
    {
      "entropy": 5.66667799949646,
      "epoch": 4.507891872278664,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031313154099669294,
      "loss": 4.8312,
      "mean_token_accuracy": 0.2216276228427887,
      "num_tokens": 103168502.0,
      "step": 49695
    },
    {
      "entropy": 5.7232307434082035,
      "epoch": 4.5083454281567485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003130996522266761,
      "loss": 4.9009,
      "mean_token_accuracy": 0.21727980822324752,
      "num_tokens": 103179015.0,
      "step": 49700
    },
    {
      "entropy": 5.798306512832641,
      "epoch": 4.508798984034833,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003130677626687485,
      "loss": 4.883,
      "mean_token_accuracy": 0.23127828687429428,
      "num_tokens": 103189200.0,
      "step": 49705
    },
    {
      "entropy": 5.81449875831604,
      "epoch": 4.509252539912917,
      "grad_norm": 1.25,
      "learning_rate": 0.0003130358723235702,
      "loss": 5.0251,
      "mean_token_accuracy": 0.21739578992128372,
      "num_tokens": 103200683.0,
      "step": 49710
    },
    {
      "entropy": 5.758195590972901,
      "epoch": 4.509706095791001,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031300398119180013,
      "loss": 4.9054,
      "mean_token_accuracy": 0.23061269968748094,
      "num_tokens": 103210992.0,
      "step": 49715
    },
    {
      "entropy": 5.683659505844116,
      "epoch": 4.510159651669086,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003129720892740981,
      "loss": 4.8122,
      "mean_token_accuracy": 0.22380020916461946,
      "num_tokens": 103220783.0,
      "step": 49720
    },
    {
      "entropy": 5.673583221435547,
      "epoch": 4.51061320754717,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003129401965711236,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22866446971893312,
      "num_tokens": 103231593.0,
      "step": 49725
    },
    {
      "entropy": 5.75624041557312,
      "epoch": 4.511066763425254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003129083030835361,
      "loss": 4.926,
      "mean_token_accuracy": 0.22086962610483168,
      "num_tokens": 103242416.0,
      "step": 49730
    },
    {
      "entropy": 5.697023248672485,
      "epoch": 4.511520319303338,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00031287640881199534,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2353135511279106,
      "num_tokens": 103252535.0,
      "step": 49735
    },
    {
      "entropy": 5.784514760971069,
      "epoch": 4.511973875181423,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003128445137571607,
      "loss": 4.9673,
      "mean_token_accuracy": 0.21717362254858016,
      "num_tokens": 103263039.0,
      "step": 49740
    },
    {
      "entropy": 5.737369537353516,
      "epoch": 4.512427431059507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031281261791969196,
      "loss": 4.8485,
      "mean_token_accuracy": 0.22262033969163894,
      "num_tokens": 103273304.0,
      "step": 49745
    },
    {
      "entropy": 5.779254913330078,
      "epoch": 4.51288098693759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031278072130024856,
      "loss": 4.9671,
      "mean_token_accuracy": 0.2282001033425331,
      "num_tokens": 103283688.0,
      "step": 49750
    },
    {
      "entropy": 5.796020221710205,
      "epoch": 4.513334542815675,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003127488238994902,
      "loss": 4.921,
      "mean_token_accuracy": 0.22493835538625717,
      "num_tokens": 103293572.0,
      "step": 49755
    },
    {
      "entropy": 5.75352635383606,
      "epoch": 4.513788098693759,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003127169257180766,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22426873445510864,
      "num_tokens": 103303474.0,
      "step": 49760
    },
    {
      "entropy": 5.819457197189331,
      "epoch": 4.514241654571843,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003126850267566674,
      "loss": 5.0145,
      "mean_token_accuracy": 0.21593862473964692,
      "num_tokens": 103314343.0,
      "step": 49765
    },
    {
      "entropy": 5.7433233737945555,
      "epoch": 4.514695210449927,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003126531270159221,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23230137228965758,
      "num_tokens": 103323942.0,
      "step": 49770
    },
    {
      "entropy": 5.764509773254394,
      "epoch": 4.515148766328012,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003126212264965007,
      "loss": 4.8953,
      "mean_token_accuracy": 0.2212221473455429,
      "num_tokens": 103333785.0,
      "step": 49775
    },
    {
      "entropy": 5.645817613601684,
      "epoch": 4.515602322206096,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031258932519906266,
      "loss": 4.8069,
      "mean_token_accuracy": 0.2326994091272354,
      "num_tokens": 103344184.0,
      "step": 49780
    },
    {
      "entropy": 5.7546608448028564,
      "epoch": 4.51605587808418,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031255742312426786,
      "loss": 4.9492,
      "mean_token_accuracy": 0.223319810628891,
      "num_tokens": 103355085.0,
      "step": 49785
    },
    {
      "entropy": 5.6999842643737795,
      "epoch": 4.5165094339622645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031252552027277595,
      "loss": 4.8308,
      "mean_token_accuracy": 0.22542963325977325,
      "num_tokens": 103365942.0,
      "step": 49790
    },
    {
      "entropy": 5.685839653015137,
      "epoch": 4.516962989840349,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003124936166452467,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23788347095251083,
      "num_tokens": 103375844.0,
      "step": 49795
    },
    {
      "entropy": 5.752854108810425,
      "epoch": 4.517416545718433,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003124617122423399,
      "loss": 4.911,
      "mean_token_accuracy": 0.2295355975627899,
      "num_tokens": 103385871.0,
      "step": 49800
    },
    {
      "entropy": 5.746824407577515,
      "epoch": 4.517870101596516,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003124298070647154,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22367191463708877,
      "num_tokens": 103396888.0,
      "step": 49805
    },
    {
      "entropy": 5.7360187530517575,
      "epoch": 4.518323657474601,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003123979011130329,
      "loss": 4.888,
      "mean_token_accuracy": 0.22583985030651094,
      "num_tokens": 103406913.0,
      "step": 49810
    },
    {
      "entropy": 5.740308380126953,
      "epoch": 4.518777213352685,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003123659943879523,
      "loss": 4.8308,
      "mean_token_accuracy": 0.23073086738586426,
      "num_tokens": 103416485.0,
      "step": 49815
    },
    {
      "entropy": 5.669770908355713,
      "epoch": 4.519230769230769,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003123340868901334,
      "loss": 4.8007,
      "mean_token_accuracy": 0.2263145089149475,
      "num_tokens": 103427476.0,
      "step": 49820
    },
    {
      "entropy": 5.668594169616699,
      "epoch": 4.519684325108853,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031230217862023604,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22919078022241593,
      "num_tokens": 103437500.0,
      "step": 49825
    },
    {
      "entropy": 5.711445713043213,
      "epoch": 4.520137880986938,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031227026957892013,
      "loss": 4.8127,
      "mean_token_accuracy": 0.23074695020914077,
      "num_tokens": 103447646.0,
      "step": 49830
    },
    {
      "entropy": 5.683657789230347,
      "epoch": 4.520591436865022,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031223835976684555,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23071045577526092,
      "num_tokens": 103457698.0,
      "step": 49835
    },
    {
      "entropy": 5.779767227172852,
      "epoch": 4.521044992743106,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031220644918467224,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22683031409978865,
      "num_tokens": 103467566.0,
      "step": 49840
    },
    {
      "entropy": 5.754968690872192,
      "epoch": 4.5214985486211905,
      "grad_norm": 1.5,
      "learning_rate": 0.00031217453783306007,
      "loss": 4.891,
      "mean_token_accuracy": 0.22214778661727905,
      "num_tokens": 103477724.0,
      "step": 49845
    },
    {
      "entropy": 5.739178657531738,
      "epoch": 4.521952104499274,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003121426257126689,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22487276196479797,
      "num_tokens": 103487864.0,
      "step": 49850
    },
    {
      "entropy": 5.693035745620728,
      "epoch": 4.522405660377358,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031211071282415885,
      "loss": 4.84,
      "mean_token_accuracy": 0.23057712614536285,
      "num_tokens": 103497229.0,
      "step": 49855
    },
    {
      "entropy": 5.69840030670166,
      "epoch": 4.522859216255442,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003120787991681898,
      "loss": 4.8123,
      "mean_token_accuracy": 0.23646138459444047,
      "num_tokens": 103506984.0,
      "step": 49860
    },
    {
      "entropy": 5.675708198547364,
      "epoch": 4.523312772133527,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003120468847454217,
      "loss": 4.8255,
      "mean_token_accuracy": 0.22769038528203964,
      "num_tokens": 103517287.0,
      "step": 49865
    },
    {
      "entropy": 5.689247989654541,
      "epoch": 4.523766328011611,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031201496955651466,
      "loss": 4.8521,
      "mean_token_accuracy": 0.22845977544784546,
      "num_tokens": 103528575.0,
      "step": 49870
    },
    {
      "entropy": 5.781920385360718,
      "epoch": 4.524219883889695,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031198305360212854,
      "loss": 4.9803,
      "mean_token_accuracy": 0.22158444076776504,
      "num_tokens": 103538406.0,
      "step": 49875
    },
    {
      "entropy": 5.758143424987793,
      "epoch": 4.5246734397677795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003119511368829235,
      "loss": 4.9285,
      "mean_token_accuracy": 0.22243400663137436,
      "num_tokens": 103548738.0,
      "step": 49880
    },
    {
      "entropy": 5.660760593414307,
      "epoch": 4.525126995645864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031191921939955947,
      "loss": 4.8121,
      "mean_token_accuracy": 0.23056754171848298,
      "num_tokens": 103558755.0,
      "step": 49885
    },
    {
      "entropy": 5.771086263656616,
      "epoch": 4.525580551523948,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003118873011526966,
      "loss": 4.889,
      "mean_token_accuracy": 0.22589824199676514,
      "num_tokens": 103569115.0,
      "step": 49890
    },
    {
      "entropy": 5.809561395645142,
      "epoch": 4.526034107402032,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003118553821429949,
      "loss": 4.9235,
      "mean_token_accuracy": 0.22746070176362992,
      "num_tokens": 103580453.0,
      "step": 49895
    },
    {
      "entropy": 5.720018196105957,
      "epoch": 4.5264876632801165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003118234623711146,
      "loss": 4.925,
      "mean_token_accuracy": 0.22402493804693221,
      "num_tokens": 103592995.0,
      "step": 49900
    },
    {
      "entropy": 5.676813411712646,
      "epoch": 4.5269412191582,
      "grad_norm": 1.421875,
      "learning_rate": 0.00031179154183771576,
      "loss": 4.7958,
      "mean_token_accuracy": 0.23417182862758637,
      "num_tokens": 103603496.0,
      "step": 49905
    },
    {
      "entropy": 5.797978115081787,
      "epoch": 4.527394775036284,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003117596205434583,
      "loss": 4.9027,
      "mean_token_accuracy": 0.2270574763417244,
      "num_tokens": 103615028.0,
      "step": 49910
    },
    {
      "entropy": 5.741382694244384,
      "epoch": 4.5278483309143684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031172769848900256,
      "loss": 4.9021,
      "mean_token_accuracy": 0.22519079595804214,
      "num_tokens": 103625430.0,
      "step": 49915
    },
    {
      "entropy": 5.778798770904541,
      "epoch": 4.528301886792453,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003116957756750087,
      "loss": 4.9086,
      "mean_token_accuracy": 0.22614155560731888,
      "num_tokens": 103636409.0,
      "step": 49920
    },
    {
      "entropy": 5.72736496925354,
      "epoch": 4.528755442670537,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00031166385210213684,
      "loss": 4.9345,
      "mean_token_accuracy": 0.23438194692134856,
      "num_tokens": 103646869.0,
      "step": 49925
    },
    {
      "entropy": 5.668318271636963,
      "epoch": 4.529208998548621,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031163192777104715,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23003604561090468,
      "num_tokens": 103657098.0,
      "step": 49930
    },
    {
      "entropy": 5.765071058273316,
      "epoch": 4.5296625544267055,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003116000026823999,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22787152379751205,
      "num_tokens": 103667331.0,
      "step": 49935
    },
    {
      "entropy": 5.692436456680298,
      "epoch": 4.53011611030479,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003115680768368552,
      "loss": 4.8109,
      "mean_token_accuracy": 0.23010306060314178,
      "num_tokens": 103676755.0,
      "step": 49940
    },
    {
      "entropy": 5.687757921218872,
      "epoch": 4.530569666182874,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003115361502350734,
      "loss": 4.8054,
      "mean_token_accuracy": 0.22737136483192444,
      "num_tokens": 103686334.0,
      "step": 49945
    },
    {
      "entropy": 5.711759233474732,
      "epoch": 4.531023222060957,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003115042228777148,
      "loss": 4.9286,
      "mean_token_accuracy": 0.22573889046907425,
      "num_tokens": 103697127.0,
      "step": 49950
    },
    {
      "entropy": 5.665952682495117,
      "epoch": 4.531476777939043,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031147229476543935,
      "loss": 4.854,
      "mean_token_accuracy": 0.2164224624633789,
      "num_tokens": 103707860.0,
      "step": 49955
    },
    {
      "entropy": 5.788259601593017,
      "epoch": 4.531930333817126,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031144036589890765,
      "loss": 4.9457,
      "mean_token_accuracy": 0.2188169166445732,
      "num_tokens": 103718816.0,
      "step": 49960
    },
    {
      "entropy": 5.696492147445679,
      "epoch": 4.53238388969521,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031140843627878003,
      "loss": 4.8113,
      "mean_token_accuracy": 0.22625518292188646,
      "num_tokens": 103729203.0,
      "step": 49965
    },
    {
      "entropy": 5.816153192520142,
      "epoch": 4.5328374455732945,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031137650590571655,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22485844641923905,
      "num_tokens": 103740830.0,
      "step": 49970
    },
    {
      "entropy": 5.731826686859131,
      "epoch": 4.533291001451379,
      "grad_norm": 1.25,
      "learning_rate": 0.0003113445747803776,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22353940457105637,
      "num_tokens": 103752881.0,
      "step": 49975
    },
    {
      "entropy": 5.731249141693115,
      "epoch": 4.533744557329463,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003113126429034237,
      "loss": 4.9063,
      "mean_token_accuracy": 0.22638477087020875,
      "num_tokens": 103762875.0,
      "step": 49980
    },
    {
      "entropy": 5.746651840209961,
      "epoch": 4.534198113207547,
      "grad_norm": 1.375,
      "learning_rate": 0.00031128071027551504,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22543585896492005,
      "num_tokens": 103772147.0,
      "step": 49985
    },
    {
      "entropy": 5.67318434715271,
      "epoch": 4.5346516690856316,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000311248776897312,
      "loss": 4.8241,
      "mean_token_accuracy": 0.2269103929400444,
      "num_tokens": 103782904.0,
      "step": 49990
    },
    {
      "entropy": 5.652976703643799,
      "epoch": 4.535105224963716,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031121684276947504,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23416225463151932,
      "num_tokens": 103793202.0,
      "step": 49995
    },
    {
      "entropy": 5.84495358467102,
      "epoch": 4.5355587808418,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0003111849078926646,
      "loss": 5.0816,
      "mean_token_accuracy": 0.21418097317218782,
      "num_tokens": 103805796.0,
      "step": 50000
    },
    {
      "entropy": 5.794236469268799,
      "epoch": 4.5360123367198835,
      "grad_norm": 1.25,
      "learning_rate": 0.000311152972267541,
      "loss": 4.914,
      "mean_token_accuracy": 0.21915741562843322,
      "num_tokens": 103816622.0,
      "step": 50005
    },
    {
      "entropy": 5.7376467227935795,
      "epoch": 4.536465892597968,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003111210358947647,
      "loss": 4.8179,
      "mean_token_accuracy": 0.22978282272815703,
      "num_tokens": 103826174.0,
      "step": 50010
    },
    {
      "entropy": 5.781724500656128,
      "epoch": 4.536919448476052,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003110890987749962,
      "loss": 4.9357,
      "mean_token_accuracy": 0.22643920332193374,
      "num_tokens": 103836217.0,
      "step": 50015
    },
    {
      "entropy": 5.718146181106567,
      "epoch": 4.537373004354136,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003110571609088959,
      "loss": 4.7887,
      "mean_token_accuracy": 0.23197061419487,
      "num_tokens": 103846661.0,
      "step": 50020
    },
    {
      "entropy": 5.741133117675782,
      "epoch": 4.5378265602322205,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003110252222971244,
      "loss": 4.9455,
      "mean_token_accuracy": 0.2180018365383148,
      "num_tokens": 103857323.0,
      "step": 50025
    },
    {
      "entropy": 5.787078619003296,
      "epoch": 4.538280116110305,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003109932829403421,
      "loss": 4.9601,
      "mean_token_accuracy": 0.22172957062721252,
      "num_tokens": 103867409.0,
      "step": 50030
    },
    {
      "entropy": 5.808265352249146,
      "epoch": 4.538733671988389,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003109613428392095,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22470245212316514,
      "num_tokens": 103877603.0,
      "step": 50035
    },
    {
      "entropy": 5.7337103366851805,
      "epoch": 4.539187227866473,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031092940199438715,
      "loss": 4.9362,
      "mean_token_accuracy": 0.22386628836393357,
      "num_tokens": 103888236.0,
      "step": 50040
    },
    {
      "entropy": 5.726613140106201,
      "epoch": 4.539640783744558,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003108974604065357,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22800807803869247,
      "num_tokens": 103898474.0,
      "step": 50045
    },
    {
      "entropy": 5.7041627883911135,
      "epoch": 4.540094339622642,
      "grad_norm": 1.125,
      "learning_rate": 0.00031086551807631553,
      "loss": 4.8589,
      "mean_token_accuracy": 0.2315766155719757,
      "num_tokens": 103909854.0,
      "step": 50050
    },
    {
      "entropy": 5.715964698791504,
      "epoch": 4.540547895500726,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031083357500438736,
      "loss": 4.8794,
      "mean_token_accuracy": 0.2238117679953575,
      "num_tokens": 103919227.0,
      "step": 50055
    },
    {
      "entropy": 5.679849004745483,
      "epoch": 4.5410014513788095,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031080163119141167,
      "loss": 4.8407,
      "mean_token_accuracy": 0.23604931980371474,
      "num_tokens": 103929022.0,
      "step": 50060
    },
    {
      "entropy": 5.753259325027466,
      "epoch": 4.541455007256894,
      "grad_norm": 1.390625,
      "learning_rate": 0.00031076968663804917,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22267629206180573,
      "num_tokens": 103939441.0,
      "step": 50065
    },
    {
      "entropy": 5.695793581008911,
      "epoch": 4.541908563134978,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031073774134496047,
      "loss": 4.8458,
      "mean_token_accuracy": 0.22569490373134612,
      "num_tokens": 103950312.0,
      "step": 50070
    },
    {
      "entropy": 5.749083518981934,
      "epoch": 4.542362119013062,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003107057953128061,
      "loss": 4.9017,
      "mean_token_accuracy": 0.21957225054502488,
      "num_tokens": 103960415.0,
      "step": 50075
    },
    {
      "entropy": 5.694440078735352,
      "epoch": 4.542815674891147,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003106738485422468,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2323874369263649,
      "num_tokens": 103971183.0,
      "step": 50080
    },
    {
      "entropy": 5.802647686004638,
      "epoch": 4.543269230769231,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003106419010339433,
      "loss": 4.9532,
      "mean_token_accuracy": 0.21906538009643556,
      "num_tokens": 103982120.0,
      "step": 50085
    },
    {
      "entropy": 5.7298903465271,
      "epoch": 4.543722786647315,
      "grad_norm": 1.328125,
      "learning_rate": 0.00031060995278855616,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23485895693302156,
      "num_tokens": 103992116.0,
      "step": 50090
    },
    {
      "entropy": 5.68427267074585,
      "epoch": 4.544176342525399,
      "grad_norm": 1.375,
      "learning_rate": 0.00031057800380674604,
      "loss": 4.9045,
      "mean_token_accuracy": 0.2260415330529213,
      "num_tokens": 104002646.0,
      "step": 50095
    },
    {
      "entropy": 5.774344348907471,
      "epoch": 4.544629898403484,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003105460540891738,
      "loss": 4.9462,
      "mean_token_accuracy": 0.21838510036468506,
      "num_tokens": 104013158.0,
      "step": 50100
    },
    {
      "entropy": 5.703759431838989,
      "epoch": 4.545083454281567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003105141036365001,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22291889935731887,
      "num_tokens": 104023485.0,
      "step": 50105
    },
    {
      "entropy": 5.731867933273316,
      "epoch": 4.545537010159651,
      "grad_norm": 1.375,
      "learning_rate": 0.0003104821524493857,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2286438375711441,
      "num_tokens": 104034588.0,
      "step": 50110
    },
    {
      "entropy": 5.742609882354737,
      "epoch": 4.5459905660377355,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003104502005284913,
      "loss": 4.8903,
      "mean_token_accuracy": 0.22705933451652527,
      "num_tokens": 104045081.0,
      "step": 50115
    },
    {
      "entropy": 5.764215612411499,
      "epoch": 4.54644412191582,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003104182478744778,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22411947846412658,
      "num_tokens": 104056166.0,
      "step": 50120
    },
    {
      "entropy": 5.740949201583862,
      "epoch": 4.546897677793904,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003103862944880058,
      "loss": 4.8533,
      "mean_token_accuracy": 0.23020352572202682,
      "num_tokens": 104066699.0,
      "step": 50125
    },
    {
      "entropy": 5.719545221328735,
      "epoch": 4.547351233671988,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003103543403697363,
      "loss": 4.8843,
      "mean_token_accuracy": 0.2307618021965027,
      "num_tokens": 104077061.0,
      "step": 50130
    },
    {
      "entropy": 5.716215753555298,
      "epoch": 4.547804789550073,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003103223855203301,
      "loss": 4.9184,
      "mean_token_accuracy": 0.22220320403575897,
      "num_tokens": 104087877.0,
      "step": 50135
    },
    {
      "entropy": 5.784110498428345,
      "epoch": 4.548258345428157,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00031029042994044784,
      "loss": 4.9659,
      "mean_token_accuracy": 0.2186102256178856,
      "num_tokens": 104099483.0,
      "step": 50140
    },
    {
      "entropy": 5.718633842468262,
      "epoch": 4.548711901306241,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00031025847363075056,
      "loss": 4.8658,
      "mean_token_accuracy": 0.22049227803945542,
      "num_tokens": 104109530.0,
      "step": 50145
    },
    {
      "entropy": 5.72745532989502,
      "epoch": 4.549165457184325,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031022651659189907,
      "loss": 4.8891,
      "mean_token_accuracy": 0.22603793442249298,
      "num_tokens": 104119610.0,
      "step": 50150
    },
    {
      "entropy": 5.777479887008667,
      "epoch": 4.54961901306241,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003101945588245541,
      "loss": 4.9358,
      "mean_token_accuracy": 0.21965746730566024,
      "num_tokens": 104130337.0,
      "step": 50155
    },
    {
      "entropy": 5.802640390396118,
      "epoch": 4.550072568940493,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003101626003293768,
      "loss": 4.9295,
      "mean_token_accuracy": 0.2191159501671791,
      "num_tokens": 104139462.0,
      "step": 50160
    },
    {
      "entropy": 5.714025020599365,
      "epoch": 4.550526124818577,
      "grad_norm": 1.5,
      "learning_rate": 0.0003101306411070279,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23609573245048524,
      "num_tokens": 104148640.0,
      "step": 50165
    },
    {
      "entropy": 5.725752496719361,
      "epoch": 4.550979680696662,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031009868115816847,
      "loss": 4.9157,
      "mean_token_accuracy": 0.23090744018554688,
      "num_tokens": 104159400.0,
      "step": 50170
    },
    {
      "entropy": 5.7263177871704105,
      "epoch": 4.551433236574746,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031006672048345936,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23536117523908615,
      "num_tokens": 104170149.0,
      "step": 50175
    },
    {
      "entropy": 5.716387176513672,
      "epoch": 4.55188679245283,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003100347590835615,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23469439297914504,
      "num_tokens": 104180787.0,
      "step": 50180
    },
    {
      "entropy": 5.759949541091919,
      "epoch": 4.552340348330914,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003100027969591359,
      "loss": 4.9017,
      "mean_token_accuracy": 0.2258258879184723,
      "num_tokens": 104190961.0,
      "step": 50185
    },
    {
      "entropy": 5.693611907958984,
      "epoch": 4.552793904208999,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030997083411084356,
      "loss": 4.8348,
      "mean_token_accuracy": 0.23324265331029892,
      "num_tokens": 104201886.0,
      "step": 50190
    },
    {
      "entropy": 5.766423559188842,
      "epoch": 4.553247460087083,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003099388705393454,
      "loss": 4.8929,
      "mean_token_accuracy": 0.2297290548682213,
      "num_tokens": 104212834.0,
      "step": 50195
    },
    {
      "entropy": 5.7537524700164795,
      "epoch": 4.553701015965167,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003099069062453025,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2225659266114235,
      "num_tokens": 104222384.0,
      "step": 50200
    },
    {
      "entropy": 5.717786693572998,
      "epoch": 4.5541545718432515,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030987494122937604,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22774314284324645,
      "num_tokens": 104232971.0,
      "step": 50205
    },
    {
      "entropy": 5.764561653137207,
      "epoch": 4.554608127721336,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003098429754922267,
      "loss": 4.8005,
      "mean_token_accuracy": 0.22952667772769927,
      "num_tokens": 104245239.0,
      "step": 50210
    },
    {
      "entropy": 5.749413347244262,
      "epoch": 4.555061683599419,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030981100903451587,
      "loss": 4.9116,
      "mean_token_accuracy": 0.2224345713853836,
      "num_tokens": 104256323.0,
      "step": 50215
    },
    {
      "entropy": 5.770790767669678,
      "epoch": 4.555515239477503,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003097790418569045,
      "loss": 4.9013,
      "mean_token_accuracy": 0.22240694016218185,
      "num_tokens": 104267106.0,
      "step": 50220
    },
    {
      "entropy": 5.728108549118042,
      "epoch": 4.555968795355588,
      "grad_norm": 1.375,
      "learning_rate": 0.0003097470739600536,
      "loss": 4.8852,
      "mean_token_accuracy": 0.23102462142705918,
      "num_tokens": 104277336.0,
      "step": 50225
    },
    {
      "entropy": 5.733521318435669,
      "epoch": 4.556422351233672,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003097151053446244,
      "loss": 4.8669,
      "mean_token_accuracy": 0.2299989327788353,
      "num_tokens": 104288809.0,
      "step": 50230
    },
    {
      "entropy": 5.738927841186523,
      "epoch": 4.556875907111756,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003096831360112781,
      "loss": 4.8957,
      "mean_token_accuracy": 0.21860980987548828,
      "num_tokens": 104299434.0,
      "step": 50235
    },
    {
      "entropy": 5.674752616882325,
      "epoch": 4.55732946298984,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003096511659606756,
      "loss": 4.829,
      "mean_token_accuracy": 0.22721750885248185,
      "num_tokens": 104309426.0,
      "step": 50240
    },
    {
      "entropy": 5.650431060791016,
      "epoch": 4.557783018867925,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003096191951934781,
      "loss": 4.8184,
      "mean_token_accuracy": 0.22755895406007767,
      "num_tokens": 104319530.0,
      "step": 50245
    },
    {
      "entropy": 5.700270652770996,
      "epoch": 4.558236574746009,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003095872237103468,
      "loss": 4.8049,
      "mean_token_accuracy": 0.2310262754559517,
      "num_tokens": 104329198.0,
      "step": 50250
    },
    {
      "entropy": 5.811351346969604,
      "epoch": 4.558690130624093,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030955525151194304,
      "loss": 4.9695,
      "mean_token_accuracy": 0.22239771187305452,
      "num_tokens": 104340481.0,
      "step": 50255
    },
    {
      "entropy": 5.812048625946045,
      "epoch": 4.559143686502177,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003095232785989277,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22788890451192856,
      "num_tokens": 104350382.0,
      "step": 50260
    },
    {
      "entropy": 5.721561527252197,
      "epoch": 4.559597242380261,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003094913049719623,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23036703914403917,
      "num_tokens": 104360967.0,
      "step": 50265
    },
    {
      "entropy": 5.700926017761231,
      "epoch": 4.560050798258345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030945933063170793,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22879558950662612,
      "num_tokens": 104371283.0,
      "step": 50270
    },
    {
      "entropy": 5.725731039047242,
      "epoch": 4.560504354136429,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030942735557882575,
      "loss": 4.8632,
      "mean_token_accuracy": 0.2262589529156685,
      "num_tokens": 104381545.0,
      "step": 50275
    },
    {
      "entropy": 5.827846908569336,
      "epoch": 4.560957910014514,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030939537981397716,
      "loss": 4.9738,
      "mean_token_accuracy": 0.2192406177520752,
      "num_tokens": 104392533.0,
      "step": 50280
    },
    {
      "entropy": 5.788604402542115,
      "epoch": 4.561411465892598,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030936340333782326,
      "loss": 4.9315,
      "mean_token_accuracy": 0.22171240597963332,
      "num_tokens": 104402230.0,
      "step": 50285
    },
    {
      "entropy": 5.7593740940094,
      "epoch": 4.561865021770682,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003093314261510254,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2340518817305565,
      "num_tokens": 104412029.0,
      "step": 50290
    },
    {
      "entropy": 5.6812254905700685,
      "epoch": 4.5623185776487665,
      "grad_norm": 1.375,
      "learning_rate": 0.000309299448254245,
      "loss": 4.84,
      "mean_token_accuracy": 0.22231726497411727,
      "num_tokens": 104422107.0,
      "step": 50295
    },
    {
      "entropy": 5.749126005172729,
      "epoch": 4.562772133526851,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00030926746964814324,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22375029176473618,
      "num_tokens": 104432073.0,
      "step": 50300
    },
    {
      "entropy": 5.745212507247925,
      "epoch": 4.563225689404935,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030923549033338147,
      "loss": 4.8626,
      "mean_token_accuracy": 0.22688350677490235,
      "num_tokens": 104441998.0,
      "step": 50305
    },
    {
      "entropy": 5.646826839447021,
      "epoch": 4.563679245283019,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000309203510310621,
      "loss": 4.7547,
      "mean_token_accuracy": 0.23745787590742112,
      "num_tokens": 104451474.0,
      "step": 50310
    },
    {
      "entropy": 5.70024471282959,
      "epoch": 4.564132801161103,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003091715295805233,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23519914150238036,
      "num_tokens": 104461351.0,
      "step": 50315
    },
    {
      "entropy": 5.771061658859253,
      "epoch": 4.564586357039187,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003091395481437496,
      "loss": 4.8707,
      "mean_token_accuracy": 0.23113802522420884,
      "num_tokens": 104471013.0,
      "step": 50320
    },
    {
      "entropy": 5.789152526855469,
      "epoch": 4.565039912917271,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003091075660009614,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22432982325553893,
      "num_tokens": 104481863.0,
      "step": 50325
    },
    {
      "entropy": 5.7878491401672365,
      "epoch": 4.5654934687953554,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030907558315282,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22079543620347977,
      "num_tokens": 104492718.0,
      "step": 50330
    },
    {
      "entropy": 5.751968050003052,
      "epoch": 4.56594702467344,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030904359959998685,
      "loss": 4.865,
      "mean_token_accuracy": 0.22506439089775085,
      "num_tokens": 104502837.0,
      "step": 50335
    },
    {
      "entropy": 5.7714637279510494,
      "epoch": 4.566400580551524,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003090116153431234,
      "loss": 4.9287,
      "mean_token_accuracy": 0.22780078202486037,
      "num_tokens": 104514202.0,
      "step": 50340
    },
    {
      "entropy": 5.776850318908691,
      "epoch": 4.566854136429608,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000308979630382891,
      "loss": 4.8833,
      "mean_token_accuracy": 0.22231459468603135,
      "num_tokens": 104525037.0,
      "step": 50345
    },
    {
      "entropy": 5.676826524734497,
      "epoch": 4.5673076923076925,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030894764471995137,
      "loss": 4.8187,
      "mean_token_accuracy": 0.2285290077328682,
      "num_tokens": 104534883.0,
      "step": 50350
    },
    {
      "entropy": 5.698248052597046,
      "epoch": 4.567761248185777,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003089156583549657,
      "loss": 4.8706,
      "mean_token_accuracy": 0.23079768270254136,
      "num_tokens": 104544845.0,
      "step": 50355
    },
    {
      "entropy": 5.7458850860595705,
      "epoch": 4.568214804063861,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003088836712885955,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22135435342788695,
      "num_tokens": 104555654.0,
      "step": 50360
    },
    {
      "entropy": 5.7017249584198,
      "epoch": 4.568668359941945,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030885168352150243,
      "loss": 4.914,
      "mean_token_accuracy": 0.2130612313747406,
      "num_tokens": 104565015.0,
      "step": 50365
    },
    {
      "entropy": 5.649281978607178,
      "epoch": 4.569121915820029,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030881969505434795,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2314806029200554,
      "num_tokens": 104575934.0,
      "step": 50370
    },
    {
      "entropy": 5.675646638870239,
      "epoch": 4.569575471698113,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030878770588779335,
      "loss": 4.7849,
      "mean_token_accuracy": 0.22835865765810012,
      "num_tokens": 104585915.0,
      "step": 50375
    },
    {
      "entropy": 5.802739048004151,
      "epoch": 4.570029027576197,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030875571602250064,
      "loss": 4.9177,
      "mean_token_accuracy": 0.22024765163660048,
      "num_tokens": 104595605.0,
      "step": 50380
    },
    {
      "entropy": 5.768214273452759,
      "epoch": 4.5704825834542815,
      "grad_norm": 1.1875,
      "learning_rate": 0.000308723725459131,
      "loss": 4.9827,
      "mean_token_accuracy": 0.21410881876945495,
      "num_tokens": 104608251.0,
      "step": 50385
    },
    {
      "entropy": 5.786833381652832,
      "epoch": 4.570936139332366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003086917341983461,
      "loss": 4.8537,
      "mean_token_accuracy": 0.23453160673379897,
      "num_tokens": 104618586.0,
      "step": 50390
    },
    {
      "entropy": 5.728103828430176,
      "epoch": 4.57138969521045,
      "grad_norm": 1.8125,
      "learning_rate": 0.0003086597422408076,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2323728993535042,
      "num_tokens": 104629718.0,
      "step": 50395
    },
    {
      "entropy": 5.699656009674072,
      "epoch": 4.571843251088534,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000308627749587177,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22991064339876174,
      "num_tokens": 104639923.0,
      "step": 50400
    },
    {
      "entropy": 5.740402793884277,
      "epoch": 4.5722968069666186,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000308595756238116,
      "loss": 4.9007,
      "mean_token_accuracy": 0.22124334871768953,
      "num_tokens": 104650890.0,
      "step": 50405
    },
    {
      "entropy": 5.739603900909424,
      "epoch": 4.572750362844703,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003085637621942862,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2284027263522148,
      "num_tokens": 104661231.0,
      "step": 50410
    },
    {
      "entropy": 5.6997518062591555,
      "epoch": 4.573203918722786,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030853176745634925,
      "loss": 4.8073,
      "mean_token_accuracy": 0.22897140234708785,
      "num_tokens": 104671641.0,
      "step": 50415
    },
    {
      "entropy": 5.724748516082764,
      "epoch": 4.5736574746008705,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003084997720249668,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22331537902355195,
      "num_tokens": 104682310.0,
      "step": 50420
    },
    {
      "entropy": 5.812520360946655,
      "epoch": 4.574111030478955,
      "grad_norm": 1.40625,
      "learning_rate": 0.00030846777590080057,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22325566858053209,
      "num_tokens": 104692414.0,
      "step": 50425
    },
    {
      "entropy": 5.722196865081787,
      "epoch": 4.574564586357039,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030843577908451215,
      "loss": 4.9379,
      "mean_token_accuracy": 0.21883358657360077,
      "num_tokens": 104703617.0,
      "step": 50430
    },
    {
      "entropy": 5.668055725097656,
      "epoch": 4.575018142235123,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003084037815767633,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22917782366275788,
      "num_tokens": 104713494.0,
      "step": 50435
    },
    {
      "entropy": 5.7093949794769285,
      "epoch": 4.5754716981132075,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003083717833782158,
      "loss": 4.8374,
      "mean_token_accuracy": 0.22634795904159546,
      "num_tokens": 104723349.0,
      "step": 50440
    },
    {
      "entropy": 5.710498476028443,
      "epoch": 4.575925253991292,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003083397844895312,
      "loss": 4.8066,
      "mean_token_accuracy": 0.24164293706417084,
      "num_tokens": 104732903.0,
      "step": 50445
    },
    {
      "entropy": 5.639994478225708,
      "epoch": 4.576378809869376,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003083077849113715,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23235463351011276,
      "num_tokens": 104742533.0,
      "step": 50450
    },
    {
      "entropy": 5.627281665802002,
      "epoch": 4.57683236574746,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030827578464439823,
      "loss": 4.7577,
      "mean_token_accuracy": 0.22817343473434448,
      "num_tokens": 104752161.0,
      "step": 50455
    },
    {
      "entropy": 5.677188587188721,
      "epoch": 4.577285921625545,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030824378368927324,
      "loss": 4.8539,
      "mean_token_accuracy": 0.23169560879468917,
      "num_tokens": 104762630.0,
      "step": 50460
    },
    {
      "entropy": 5.697794198989868,
      "epoch": 4.577739477503629,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030821178204665836,
      "loss": 4.8189,
      "mean_token_accuracy": 0.23142397552728652,
      "num_tokens": 104773529.0,
      "step": 50465
    },
    {
      "entropy": 5.697564220428466,
      "epoch": 4.578193033381712,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003081797797172154,
      "loss": 4.8034,
      "mean_token_accuracy": 0.22479477226734162,
      "num_tokens": 104783386.0,
      "step": 50470
    },
    {
      "entropy": 5.717767190933228,
      "epoch": 4.5786465892597965,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003081477767016061,
      "loss": 4.8997,
      "mean_token_accuracy": 0.223509381711483,
      "num_tokens": 104793254.0,
      "step": 50475
    },
    {
      "entropy": 5.712533760070801,
      "epoch": 4.579100145137881,
      "grad_norm": 1.25,
      "learning_rate": 0.00030811577300049226,
      "loss": 4.9031,
      "mean_token_accuracy": 0.22852091938257219,
      "num_tokens": 104804345.0,
      "step": 50480
    },
    {
      "entropy": 5.720235443115234,
      "epoch": 4.579553701015965,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003080837686145359,
      "loss": 4.8544,
      "mean_token_accuracy": 0.22794366180896758,
      "num_tokens": 104814891.0,
      "step": 50485
    },
    {
      "entropy": 5.69306378364563,
      "epoch": 4.580007256894049,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003080517635443987,
      "loss": 4.8211,
      "mean_token_accuracy": 0.22631726711988448,
      "num_tokens": 104825015.0,
      "step": 50490
    },
    {
      "entropy": 5.7541601181030275,
      "epoch": 4.580460812772134,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003080197577907427,
      "loss": 4.894,
      "mean_token_accuracy": 0.21835893839597703,
      "num_tokens": 104835735.0,
      "step": 50495
    },
    {
      "entropy": 5.719085550308227,
      "epoch": 4.580914368650218,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030798775135422966,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22196639329195023,
      "num_tokens": 104847484.0,
      "step": 50500
    },
    {
      "entropy": 5.692076015472412,
      "epoch": 4.581367924528302,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003079557442355215,
      "loss": 4.8987,
      "mean_token_accuracy": 0.2269342839717865,
      "num_tokens": 104857783.0,
      "step": 50505
    },
    {
      "entropy": 5.6760204315185545,
      "epoch": 4.581821480406386,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003079237364352801,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2395823448896408,
      "num_tokens": 104867498.0,
      "step": 50510
    },
    {
      "entropy": 5.731898927688599,
      "epoch": 4.58227503628447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003078917279541675,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22833222895860672,
      "num_tokens": 104876505.0,
      "step": 50515
    },
    {
      "entropy": 5.726686477661133,
      "epoch": 4.582728592162555,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003078597187928456,
      "loss": 4.8939,
      "mean_token_accuracy": 0.21878527104854584,
      "num_tokens": 104886932.0,
      "step": 50520
    },
    {
      "entropy": 5.756633520126343,
      "epoch": 4.583182148040638,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0003078277089519764,
      "loss": 4.8603,
      "mean_token_accuracy": 0.23295507729053497,
      "num_tokens": 104896733.0,
      "step": 50525
    },
    {
      "entropy": 5.736713314056397,
      "epoch": 4.5836357039187225,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030779569843222165,
      "loss": 4.8973,
      "mean_token_accuracy": 0.2182561442255974,
      "num_tokens": 104906846.0,
      "step": 50530
    },
    {
      "entropy": 5.60648775100708,
      "epoch": 4.584089259796807,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030776368723424365,
      "loss": 4.6865,
      "mean_token_accuracy": 0.2444351375102997,
      "num_tokens": 104916136.0,
      "step": 50535
    },
    {
      "entropy": 5.686765956878662,
      "epoch": 4.584542815674891,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003077316753587042,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23324964344501495,
      "num_tokens": 104926145.0,
      "step": 50540
    },
    {
      "entropy": 5.739676189422608,
      "epoch": 4.584996371552975,
      "grad_norm": 1.484375,
      "learning_rate": 0.00030769966280626526,
      "loss": 4.9425,
      "mean_token_accuracy": 0.2180771127343178,
      "num_tokens": 104935576.0,
      "step": 50545
    },
    {
      "entropy": 5.6791750431060795,
      "epoch": 4.58544992743106,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030766764957758906,
      "loss": 4.7684,
      "mean_token_accuracy": 0.2326889991760254,
      "num_tokens": 104944846.0,
      "step": 50550
    },
    {
      "entropy": 5.712737894058227,
      "epoch": 4.585903483309144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030763563567333754,
      "loss": 4.8211,
      "mean_token_accuracy": 0.22921747714281082,
      "num_tokens": 104954604.0,
      "step": 50555
    },
    {
      "entropy": 5.696177625656128,
      "epoch": 4.586357039187228,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030760362109417277,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2266802206635475,
      "num_tokens": 104964858.0,
      "step": 50560
    },
    {
      "entropy": 5.757247066497802,
      "epoch": 4.586810595065312,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003075716058407568,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22390396296977996,
      "num_tokens": 104974997.0,
      "step": 50565
    },
    {
      "entropy": 5.720857429504394,
      "epoch": 4.587264150943396,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030753958991375176,
      "loss": 4.829,
      "mean_token_accuracy": 0.23295016586780548,
      "num_tokens": 104984585.0,
      "step": 50570
    },
    {
      "entropy": 5.69962477684021,
      "epoch": 4.58771770682148,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030750757331381964,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23231568783521653,
      "num_tokens": 104994843.0,
      "step": 50575
    },
    {
      "entropy": 5.712131452560425,
      "epoch": 4.588171262699564,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003074755560416226,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2266097754240036,
      "num_tokens": 105006082.0,
      "step": 50580
    },
    {
      "entropy": 5.710009145736694,
      "epoch": 4.588624818577649,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030744353809782283,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22944409698247908,
      "num_tokens": 105016704.0,
      "step": 50585
    },
    {
      "entropy": 5.773340940475464,
      "epoch": 4.589078374455733,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030741151948308237,
      "loss": 4.9499,
      "mean_token_accuracy": 0.2229883998632431,
      "num_tokens": 105027810.0,
      "step": 50590
    },
    {
      "entropy": 5.681578350067139,
      "epoch": 4.589531930333817,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003073795001980635,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2307019606232643,
      "num_tokens": 105037488.0,
      "step": 50595
    },
    {
      "entropy": 5.718272876739502,
      "epoch": 4.589985486211901,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003073474802434282,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22971002012491226,
      "num_tokens": 105047461.0,
      "step": 50600
    },
    {
      "entropy": 5.737926626205445,
      "epoch": 4.590439042089986,
      "grad_norm": 1.296875,
      "learning_rate": 0.000307315459619839,
      "loss": 4.8974,
      "mean_token_accuracy": 0.22312204539775848,
      "num_tokens": 105058020.0,
      "step": 50605
    },
    {
      "entropy": 5.806051635742188,
      "epoch": 4.59089259796807,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003072834383279576,
      "loss": 4.9177,
      "mean_token_accuracy": 0.22150880545377732,
      "num_tokens": 105069638.0,
      "step": 50610
    },
    {
      "entropy": 5.7252692699432375,
      "epoch": 4.591346153846154,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003072514163684466,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22764186114072799,
      "num_tokens": 105078686.0,
      "step": 50615
    },
    {
      "entropy": 5.726677703857422,
      "epoch": 4.5917997097242385,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003072193937419681,
      "loss": 4.8812,
      "mean_token_accuracy": 0.2259785622358322,
      "num_tokens": 105089739.0,
      "step": 50620
    },
    {
      "entropy": 5.724665212631225,
      "epoch": 4.592253265602322,
      "grad_norm": 1.25,
      "learning_rate": 0.00030718737044918435,
      "loss": 4.8689,
      "mean_token_accuracy": 0.22451457232236863,
      "num_tokens": 105100673.0,
      "step": 50625
    },
    {
      "entropy": 5.719394159317017,
      "epoch": 4.592706821480406,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030715534649075745,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22857394814491272,
      "num_tokens": 105110750.0,
      "step": 50630
    },
    {
      "entropy": 5.655692672729492,
      "epoch": 4.59316037735849,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030712332186734985,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2374560132622719,
      "num_tokens": 105120809.0,
      "step": 50635
    },
    {
      "entropy": 5.728529787063598,
      "epoch": 4.593613933236575,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003070912965796238,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22816620916128158,
      "num_tokens": 105131106.0,
      "step": 50640
    },
    {
      "entropy": 5.74159688949585,
      "epoch": 4.594067489114659,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030705927062824154,
      "loss": 4.8983,
      "mean_token_accuracy": 0.2243645027279854,
      "num_tokens": 105140970.0,
      "step": 50645
    },
    {
      "entropy": 5.784335708618164,
      "epoch": 4.594521044992743,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003070272440138654,
      "loss": 4.8912,
      "mean_token_accuracy": 0.219778810441494,
      "num_tokens": 105152429.0,
      "step": 50650
    },
    {
      "entropy": 5.79698371887207,
      "epoch": 4.594974600870827,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003069952167371577,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2215821400284767,
      "num_tokens": 105163534.0,
      "step": 50655
    },
    {
      "entropy": 5.677051830291748,
      "epoch": 4.595428156748912,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030696318879878064,
      "loss": 4.7935,
      "mean_token_accuracy": 0.22679104954004287,
      "num_tokens": 105172798.0,
      "step": 50660
    },
    {
      "entropy": 5.782365846633911,
      "epoch": 4.595881712626996,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030693116019939686,
      "loss": 4.9019,
      "mean_token_accuracy": 0.22739467024803162,
      "num_tokens": 105182576.0,
      "step": 50665
    },
    {
      "entropy": 5.75454535484314,
      "epoch": 4.596335268505079,
      "grad_norm": 1.578125,
      "learning_rate": 0.00030689913093966847,
      "loss": 4.89,
      "mean_token_accuracy": 0.22836508601903915,
      "num_tokens": 105192070.0,
      "step": 50670
    },
    {
      "entropy": 5.7551398277282715,
      "epoch": 4.5967888243831645,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000306867101020258,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2271262988448143,
      "num_tokens": 105203151.0,
      "step": 50675
    },
    {
      "entropy": 5.7578428268432615,
      "epoch": 4.597242380261248,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003068350704418277,
      "loss": 4.9341,
      "mean_token_accuracy": 0.21755018681287766,
      "num_tokens": 105213217.0,
      "step": 50680
    },
    {
      "entropy": 5.6717977046966555,
      "epoch": 4.597695936139332,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030680303920504006,
      "loss": 4.7807,
      "mean_token_accuracy": 0.2327335759997368,
      "num_tokens": 105223450.0,
      "step": 50685
    },
    {
      "entropy": 5.779776763916016,
      "epoch": 4.598149492017416,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003067710073105574,
      "loss": 4.9502,
      "mean_token_accuracy": 0.22339384108781815,
      "num_tokens": 105234705.0,
      "step": 50690
    },
    {
      "entropy": 5.825994920730591,
      "epoch": 4.598603047895501,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030673897475904223,
      "loss": 5.0726,
      "mean_token_accuracy": 0.20613370686769486,
      "num_tokens": 105245042.0,
      "step": 50695
    },
    {
      "entropy": 5.746119785308838,
      "epoch": 4.599056603773585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030670694155115704,
      "loss": 4.8752,
      "mean_token_accuracy": 0.21929134726524352,
      "num_tokens": 105255830.0,
      "step": 50700
    },
    {
      "entropy": 5.709834480285645,
      "epoch": 4.599510159651669,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030667490768756416,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22649133652448655,
      "num_tokens": 105265844.0,
      "step": 50705
    },
    {
      "entropy": 5.691987371444702,
      "epoch": 4.5999637155297535,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003066428731689262,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2294115349650383,
      "num_tokens": 105276846.0,
      "step": 50710
    },
    {
      "entropy": 5.777865791320801,
      "epoch": 4.600417271407838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030661083799590544,
      "loss": 4.9257,
      "mean_token_accuracy": 0.22410153448581696,
      "num_tokens": 105286740.0,
      "step": 50715
    },
    {
      "entropy": 5.686180639266968,
      "epoch": 4.600870827285922,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030657880216916454,
      "loss": 4.8527,
      "mean_token_accuracy": 0.22997837662696838,
      "num_tokens": 105296710.0,
      "step": 50720
    },
    {
      "entropy": 5.7363523006439205,
      "epoch": 4.601324383164005,
      "grad_norm": 1.3125,
      "learning_rate": 0.000306546765689366,
      "loss": 4.9013,
      "mean_token_accuracy": 0.23310864418745042,
      "num_tokens": 105307076.0,
      "step": 50725
    },
    {
      "entropy": 5.73448486328125,
      "epoch": 4.60177793904209,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003065147285571722,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2290446564555168,
      "num_tokens": 105317486.0,
      "step": 50730
    },
    {
      "entropy": 5.689556455612182,
      "epoch": 4.602231494920174,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003064826907732459,
      "loss": 4.8669,
      "mean_token_accuracy": 0.22711561024188995,
      "num_tokens": 105327771.0,
      "step": 50735
    },
    {
      "entropy": 5.712956428527832,
      "epoch": 4.602685050798258,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030645065233824955,
      "loss": 4.9111,
      "mean_token_accuracy": 0.2213507354259491,
      "num_tokens": 105339191.0,
      "step": 50740
    },
    {
      "entropy": 5.747328662872315,
      "epoch": 4.6031386066763424,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030641861325284557,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22443721145391465,
      "num_tokens": 105349949.0,
      "step": 50745
    },
    {
      "entropy": 5.697824239730835,
      "epoch": 4.603592162554427,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030638657351769673,
      "loss": 4.868,
      "mean_token_accuracy": 0.2170468747615814,
      "num_tokens": 105360357.0,
      "step": 50750
    },
    {
      "entropy": 5.804789113998413,
      "epoch": 4.604045718432511,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030635453313346554,
      "loss": 4.9058,
      "mean_token_accuracy": 0.22727405428886413,
      "num_tokens": 105370765.0,
      "step": 50755
    },
    {
      "entropy": 5.76274881362915,
      "epoch": 4.604499274310595,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003063224921008146,
      "loss": 4.8946,
      "mean_token_accuracy": 0.22163957357406616,
      "num_tokens": 105380797.0,
      "step": 50760
    },
    {
      "entropy": 5.731989336013794,
      "epoch": 4.6049528301886795,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003062904504204066,
      "loss": 4.869,
      "mean_token_accuracy": 0.22493635714054108,
      "num_tokens": 105390653.0,
      "step": 50765
    },
    {
      "entropy": 5.812828540802002,
      "epoch": 4.605406386066764,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030625840809290404,
      "loss": 5.0219,
      "mean_token_accuracy": 0.2197385013103485,
      "num_tokens": 105401308.0,
      "step": 50770
    },
    {
      "entropy": 5.708014249801636,
      "epoch": 4.605859941944848,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030622636511896977,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2193480297923088,
      "num_tokens": 105412412.0,
      "step": 50775
    },
    {
      "entropy": 5.810382413864136,
      "epoch": 4.606313497822931,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003061943214992662,
      "loss": 4.9531,
      "mean_token_accuracy": 0.21983499675989152,
      "num_tokens": 105424789.0,
      "step": 50780
    },
    {
      "entropy": 5.733348798751831,
      "epoch": 4.606767053701016,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003061622772344561,
      "loss": 4.9684,
      "mean_token_accuracy": 0.22892388701438904,
      "num_tokens": 105434380.0,
      "step": 50785
    },
    {
      "entropy": 5.745625638961792,
      "epoch": 4.6072206095791,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003061302323252022,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2303740069270134,
      "num_tokens": 105443889.0,
      "step": 50790
    },
    {
      "entropy": 5.7573741436004635,
      "epoch": 4.607674165457184,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003060981867721671,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22812665104866028,
      "num_tokens": 105454336.0,
      "step": 50795
    },
    {
      "entropy": 5.7990312576293945,
      "epoch": 4.6081277213352685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030606614057601374,
      "loss": 4.9774,
      "mean_token_accuracy": 0.22085154354572295,
      "num_tokens": 105464601.0,
      "step": 50800
    },
    {
      "entropy": 5.801136016845703,
      "epoch": 4.608581277213353,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030603409373740456,
      "loss": 4.9035,
      "mean_token_accuracy": 0.21659974455833436,
      "num_tokens": 105475265.0,
      "step": 50805
    },
    {
      "entropy": 5.6631824493408205,
      "epoch": 4.609034833091437,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030600204625700234,
      "loss": 4.7763,
      "mean_token_accuracy": 0.23541960418224334,
      "num_tokens": 105485491.0,
      "step": 50810
    },
    {
      "entropy": 5.68995361328125,
      "epoch": 4.609488388969521,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030596999813547007,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22991372346878053,
      "num_tokens": 105495859.0,
      "step": 50815
    },
    {
      "entropy": 5.6559645652771,
      "epoch": 4.6099419448476056,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003059379493734702,
      "loss": 4.7909,
      "mean_token_accuracy": 0.23309634923934935,
      "num_tokens": 105506385.0,
      "step": 50820
    },
    {
      "entropy": 5.714199352264404,
      "epoch": 4.610395500725689,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030590589997166575,
      "loss": 4.7946,
      "mean_token_accuracy": 0.2307930439710617,
      "num_tokens": 105515829.0,
      "step": 50825
    },
    {
      "entropy": 5.69038200378418,
      "epoch": 4.610849056603773,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003058738499307193,
      "loss": 4.7958,
      "mean_token_accuracy": 0.2402627632021904,
      "num_tokens": 105525950.0,
      "step": 50830
    },
    {
      "entropy": 5.681950187683105,
      "epoch": 4.6113026124818575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030584179925129384,
      "loss": 4.8848,
      "mean_token_accuracy": 0.22489263713359833,
      "num_tokens": 105536536.0,
      "step": 50835
    },
    {
      "entropy": 5.6439512252807615,
      "epoch": 4.611756168359942,
      "grad_norm": 1.375,
      "learning_rate": 0.00030580974793405213,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23475583344697953,
      "num_tokens": 105546069.0,
      "step": 50840
    },
    {
      "entropy": 5.681201648712158,
      "epoch": 4.612209724238026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030577769597965696,
      "loss": 4.858,
      "mean_token_accuracy": 0.22450230568647384,
      "num_tokens": 105556522.0,
      "step": 50845
    },
    {
      "entropy": 5.700773859024048,
      "epoch": 4.61266328011611,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003057456433887712,
      "loss": 4.7982,
      "mean_token_accuracy": 0.22775616943836213,
      "num_tokens": 105566431.0,
      "step": 50850
    },
    {
      "entropy": 5.690542507171631,
      "epoch": 4.6131168359941945,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030571359016205767,
      "loss": 4.8216,
      "mean_token_accuracy": 0.2372843623161316,
      "num_tokens": 105576270.0,
      "step": 50855
    },
    {
      "entropy": 5.700283718109131,
      "epoch": 4.613570391872279,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003056815363001793,
      "loss": 4.9028,
      "mean_token_accuracy": 0.22280809730291368,
      "num_tokens": 105587144.0,
      "step": 50860
    },
    {
      "entropy": 5.66920223236084,
      "epoch": 4.614023947750363,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030564948180379897,
      "loss": 4.8521,
      "mean_token_accuracy": 0.2290820524096489,
      "num_tokens": 105596636.0,
      "step": 50865
    },
    {
      "entropy": 5.721886396408081,
      "epoch": 4.614477503628447,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003056174266735795,
      "loss": 4.8729,
      "mean_token_accuracy": 0.23001186847686766,
      "num_tokens": 105605825.0,
      "step": 50870
    },
    {
      "entropy": 5.805042934417725,
      "epoch": 4.614931059506532,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003055853709101839,
      "loss": 4.9541,
      "mean_token_accuracy": 0.22222080528736116,
      "num_tokens": 105615990.0,
      "step": 50875
    },
    {
      "entropy": 5.748851156234741,
      "epoch": 4.615384615384615,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030555331451427494,
      "loss": 4.8962,
      "mean_token_accuracy": 0.220198330283165,
      "num_tokens": 105626673.0,
      "step": 50880
    },
    {
      "entropy": 5.767621994018555,
      "epoch": 4.615838171262699,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030552125748651573,
      "loss": 4.9595,
      "mean_token_accuracy": 0.21889046281576158,
      "num_tokens": 105636752.0,
      "step": 50885
    },
    {
      "entropy": 5.803873062133789,
      "epoch": 4.6162917271407835,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030548919982756914,
      "loss": 4.9431,
      "mean_token_accuracy": 0.22010902017354966,
      "num_tokens": 105647476.0,
      "step": 50890
    },
    {
      "entropy": 5.731475448608398,
      "epoch": 4.616745283018868,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030545714153809805,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22778620421886445,
      "num_tokens": 105658135.0,
      "step": 50895
    },
    {
      "entropy": 5.684292650222778,
      "epoch": 4.617198838896952,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003054250826187655,
      "loss": 4.7929,
      "mean_token_accuracy": 0.23635137677192689,
      "num_tokens": 105667802.0,
      "step": 50900
    },
    {
      "entropy": 5.759898090362549,
      "epoch": 4.617652394775036,
      "grad_norm": 1.5625,
      "learning_rate": 0.0003053930230702346,
      "loss": 4.9255,
      "mean_token_accuracy": 0.22656637132167817,
      "num_tokens": 105679161.0,
      "step": 50905
    },
    {
      "entropy": 5.742093658447265,
      "epoch": 4.618105950653121,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0003053609628931682,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2112136334180832,
      "num_tokens": 105688093.0,
      "step": 50910
    },
    {
      "entropy": 5.777035474777222,
      "epoch": 4.618559506531205,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003053289020882293,
      "loss": 4.941,
      "mean_token_accuracy": 0.21587659567594528,
      "num_tokens": 105697982.0,
      "step": 50915
    },
    {
      "entropy": 5.764431667327881,
      "epoch": 4.619013062409289,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030529684065608103,
      "loss": 4.7984,
      "mean_token_accuracy": 0.23378394544124603,
      "num_tokens": 105708729.0,
      "step": 50920
    },
    {
      "entropy": 5.685211992263794,
      "epoch": 4.619466618287373,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030526477859738635,
      "loss": 4.8108,
      "mean_token_accuracy": 0.22944063693284988,
      "num_tokens": 105718919.0,
      "step": 50925
    },
    {
      "entropy": 5.688052701950073,
      "epoch": 4.619920174165458,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003052327159128083,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23148417621850967,
      "num_tokens": 105727787.0,
      "step": 50930
    },
    {
      "entropy": 5.692604112625122,
      "epoch": 4.620373730043541,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030520065260301003,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22446772307157517,
      "num_tokens": 105738282.0,
      "step": 50935
    },
    {
      "entropy": 5.714572477340698,
      "epoch": 4.620827285921625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003051685886686546,
      "loss": 4.845,
      "mean_token_accuracy": 0.22455475330352784,
      "num_tokens": 105748786.0,
      "step": 50940
    },
    {
      "entropy": 5.71193561553955,
      "epoch": 4.6212808417997095,
      "grad_norm": 1.3125,
      "learning_rate": 0.000305136524110405,
      "loss": 4.8628,
      "mean_token_accuracy": 0.22552423775196076,
      "num_tokens": 105759156.0,
      "step": 50945
    },
    {
      "entropy": 5.699240350723267,
      "epoch": 4.621734397677794,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003051044589289245,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21297800689935684,
      "num_tokens": 105769434.0,
      "step": 50950
    },
    {
      "entropy": 5.793159008026123,
      "epoch": 4.622187953555878,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030507239312487595,
      "loss": 4.9915,
      "mean_token_accuracy": 0.22028448879718782,
      "num_tokens": 105779408.0,
      "step": 50955
    },
    {
      "entropy": 5.790826892852783,
      "epoch": 4.622641509433962,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030504032669892276,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22406589090824128,
      "num_tokens": 105789517.0,
      "step": 50960
    },
    {
      "entropy": 5.750369548797607,
      "epoch": 4.623095065312047,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030500825965172797,
      "loss": 4.8535,
      "mean_token_accuracy": 0.2297673612833023,
      "num_tokens": 105799823.0,
      "step": 50965
    },
    {
      "entropy": 5.728070211410523,
      "epoch": 4.623548621190131,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030497619198395466,
      "loss": 4.902,
      "mean_token_accuracy": 0.22499565929174423,
      "num_tokens": 105810038.0,
      "step": 50970
    },
    {
      "entropy": 5.741662979125977,
      "epoch": 4.624002177068215,
      "grad_norm": 1.375,
      "learning_rate": 0.00030494412369626615,
      "loss": 4.9121,
      "mean_token_accuracy": 0.22257521897554397,
      "num_tokens": 105819940.0,
      "step": 50975
    },
    {
      "entropy": 5.70567569732666,
      "epoch": 4.6244557329462985,
      "grad_norm": 1.421875,
      "learning_rate": 0.0003049120547893255,
      "loss": 4.8436,
      "mean_token_accuracy": 0.22992804199457167,
      "num_tokens": 105829829.0,
      "step": 50980
    },
    {
      "entropy": 5.698531150817871,
      "epoch": 4.624909288824383,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003048799852637958,
      "loss": 4.8288,
      "mean_token_accuracy": 0.23445278406143188,
      "num_tokens": 105839463.0,
      "step": 50985
    },
    {
      "entropy": 5.808040618896484,
      "epoch": 4.625362844702467,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030484791512034063,
      "loss": 5.0536,
      "mean_token_accuracy": 0.21022400707006456,
      "num_tokens": 105849916.0,
      "step": 50990
    },
    {
      "entropy": 5.72371244430542,
      "epoch": 4.625816400580551,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003048158443596228,
      "loss": 4.8488,
      "mean_token_accuracy": 0.22297966629266738,
      "num_tokens": 105860699.0,
      "step": 50995
    },
    {
      "entropy": 5.772852230072021,
      "epoch": 4.626269956458636,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003047837729823058,
      "loss": 4.8419,
      "mean_token_accuracy": 0.22186854779720305,
      "num_tokens": 105870332.0,
      "step": 51000
    },
    {
      "epoch": 4.626269956458636,
      "eval_entropy": 5.421642386967241,
      "eval_loss": 4.962176322937012,
      "eval_mean_token_accuracy": 0.23030492847456652,
      "eval_num_tokens": 105870332.0,
      "eval_runtime": 20.5273,
      "eval_samples_per_second": 1722.586,
      "eval_steps_per_second": 215.323,
      "step": 51000
    },
    {
      "entropy": 5.691490554809571,
      "epoch": 4.62672351233672,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003047517009890527,
      "loss": 4.8627,
      "mean_token_accuracy": 0.23618475943803788,
      "num_tokens": 105879364.0,
      "step": 51005
    },
    {
      "entropy": 5.76324610710144,
      "epoch": 4.627177068214804,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030471962838052693,
      "loss": 4.9694,
      "mean_token_accuracy": 0.2156776487827301,
      "num_tokens": 105889554.0,
      "step": 51010
    },
    {
      "entropy": 5.735954713821411,
      "epoch": 4.627630624092888,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003046875551573916,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23680986613035201,
      "num_tokens": 105900426.0,
      "step": 51015
    },
    {
      "entropy": 5.607890510559082,
      "epoch": 4.628084179970973,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003046554813203101,
      "loss": 4.6788,
      "mean_token_accuracy": 0.23841808140277862,
      "num_tokens": 105909541.0,
      "step": 51020
    },
    {
      "entropy": 5.712298154830933,
      "epoch": 4.628537735849057,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003046234068699457,
      "loss": 4.9212,
      "mean_token_accuracy": 0.22040753066539764,
      "num_tokens": 105919951.0,
      "step": 51025
    },
    {
      "entropy": 5.680587720870972,
      "epoch": 4.628991291727141,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003045913318069616,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23020151108503342,
      "num_tokens": 105929048.0,
      "step": 51030
    },
    {
      "entropy": 5.743010807037353,
      "epoch": 4.629444847605225,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003045592561320213,
      "loss": 4.8124,
      "mean_token_accuracy": 0.22900669127702714,
      "num_tokens": 105938105.0,
      "step": 51035
    },
    {
      "entropy": 5.738589859008789,
      "epoch": 4.629898403483309,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000304527179845788,
      "loss": 4.8908,
      "mean_token_accuracy": 0.21967400461435319,
      "num_tokens": 105948167.0,
      "step": 51040
    },
    {
      "entropy": 5.7477319717407225,
      "epoch": 4.630351959361393,
      "grad_norm": 1.375,
      "learning_rate": 0.0003044951029489252,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2264263167977333,
      "num_tokens": 105959387.0,
      "step": 51045
    },
    {
      "entropy": 5.636099433898925,
      "epoch": 4.630805515239477,
      "grad_norm": 1.390625,
      "learning_rate": 0.000304463025442096,
      "loss": 4.7857,
      "mean_token_accuracy": 0.2318551629781723,
      "num_tokens": 105969436.0,
      "step": 51050
    },
    {
      "entropy": 5.754473304748535,
      "epoch": 4.631259071117562,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003044309473259641,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2283968910574913,
      "num_tokens": 105979102.0,
      "step": 51055
    },
    {
      "entropy": 5.678262662887573,
      "epoch": 4.631712626995646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030439886860119265,
      "loss": 4.7773,
      "mean_token_accuracy": 0.23815197497606277,
      "num_tokens": 105989796.0,
      "step": 51060
    },
    {
      "entropy": 5.703248977661133,
      "epoch": 4.63216618287373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000304366789268445,
      "loss": 4.877,
      "mean_token_accuracy": 0.22709328383207322,
      "num_tokens": 106000194.0,
      "step": 51065
    },
    {
      "entropy": 5.728683567047119,
      "epoch": 4.632619738751814,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003043347093283848,
      "loss": 4.8698,
      "mean_token_accuracy": 0.22817789167165756,
      "num_tokens": 106009108.0,
      "step": 51070
    },
    {
      "entropy": 5.679253816604614,
      "epoch": 4.633073294629899,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030430262878167533,
      "loss": 4.821,
      "mean_token_accuracy": 0.228461392223835,
      "num_tokens": 106019342.0,
      "step": 51075
    },
    {
      "entropy": 5.796146059036255,
      "epoch": 4.633526850507983,
      "grad_norm": 1.375,
      "learning_rate": 0.0003042705476289799,
      "loss": 4.9444,
      "mean_token_accuracy": 0.21710583567619324,
      "num_tokens": 106029079.0,
      "step": 51080
    },
    {
      "entropy": 5.677875947952271,
      "epoch": 4.633980406386067,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003042384658709622,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23600248843431473,
      "num_tokens": 106039406.0,
      "step": 51085
    },
    {
      "entropy": 5.7401172637939455,
      "epoch": 4.634433962264151,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003042063835082856,
      "loss": 4.889,
      "mean_token_accuracy": 0.2226448342204094,
      "num_tokens": 106049603.0,
      "step": 51090
    },
    {
      "entropy": 5.725952768325806,
      "epoch": 4.634887518142235,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003041743005416135,
      "loss": 4.8391,
      "mean_token_accuracy": 0.23287599980831147,
      "num_tokens": 106058692.0,
      "step": 51095
    },
    {
      "entropy": 5.6505567073822025,
      "epoch": 4.635341074020319,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003041422169716094,
      "loss": 4.8374,
      "mean_token_accuracy": 0.23563594073057176,
      "num_tokens": 106069307.0,
      "step": 51100
    },
    {
      "entropy": 5.783578824996948,
      "epoch": 4.635794629898403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003041101327989368,
      "loss": 4.8967,
      "mean_token_accuracy": 0.230181585252285,
      "num_tokens": 106080831.0,
      "step": 51105
    },
    {
      "entropy": 5.744425296783447,
      "epoch": 4.636248185776488,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003040780480242593,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22509552538394928,
      "num_tokens": 106090007.0,
      "step": 51110
    },
    {
      "entropy": 5.748862218856812,
      "epoch": 4.636701741654572,
      "grad_norm": 1.421875,
      "learning_rate": 0.00030404596264824033,
      "loss": 4.8879,
      "mean_token_accuracy": 0.23150358945131302,
      "num_tokens": 106099790.0,
      "step": 51115
    },
    {
      "entropy": 5.728938579559326,
      "epoch": 4.637155297532656,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030401387667154347,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2310151845216751,
      "num_tokens": 106110364.0,
      "step": 51120
    },
    {
      "entropy": 5.630988121032715,
      "epoch": 4.6376088534107405,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003039817900948322,
      "loss": 4.7842,
      "mean_token_accuracy": 0.2287372425198555,
      "num_tokens": 106121353.0,
      "step": 51125
    },
    {
      "entropy": 5.688761425018311,
      "epoch": 4.638062409288825,
      "grad_norm": 1.5,
      "learning_rate": 0.0003039497029187702,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22362796515226363,
      "num_tokens": 106130750.0,
      "step": 51130
    },
    {
      "entropy": 5.737570285797119,
      "epoch": 4.638515965166908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030391761514402096,
      "loss": 4.8499,
      "mean_token_accuracy": 0.2324984163045883,
      "num_tokens": 106142014.0,
      "step": 51135
    },
    {
      "entropy": 5.736945343017578,
      "epoch": 4.638969521044992,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030388552677124804,
      "loss": 4.8486,
      "mean_token_accuracy": 0.23522557318210602,
      "num_tokens": 106152419.0,
      "step": 51140
    },
    {
      "entropy": 5.6908103942871096,
      "epoch": 4.639423076923077,
      "grad_norm": 1.375,
      "learning_rate": 0.000303853437801115,
      "loss": 4.7931,
      "mean_token_accuracy": 0.23080360293388366,
      "num_tokens": 106163365.0,
      "step": 51145
    },
    {
      "entropy": 5.71789698600769,
      "epoch": 4.639876632801161,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0003038213482342856,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23759062737226486,
      "num_tokens": 106173324.0,
      "step": 51150
    },
    {
      "entropy": 5.809322166442871,
      "epoch": 4.640330188679245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030378925807142334,
      "loss": 4.965,
      "mean_token_accuracy": 0.22410084307193756,
      "num_tokens": 106183427.0,
      "step": 51155
    },
    {
      "entropy": 5.765050888061523,
      "epoch": 4.6407837445573294,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030375716731319193,
      "loss": 4.9245,
      "mean_token_accuracy": 0.21763210892677307,
      "num_tokens": 106192894.0,
      "step": 51160
    },
    {
      "entropy": 5.687732076644897,
      "epoch": 4.641237300435414,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030372507596025493,
      "loss": 4.844,
      "mean_token_accuracy": 0.22444216907024384,
      "num_tokens": 106203185.0,
      "step": 51165
    },
    {
      "entropy": 5.826116561889648,
      "epoch": 4.641690856313498,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000303692984013276,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21821703612804413,
      "num_tokens": 106213817.0,
      "step": 51170
    },
    {
      "entropy": 5.68279619216919,
      "epoch": 4.642144412191582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003036608914729189,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23179415613412857,
      "num_tokens": 106224367.0,
      "step": 51175
    },
    {
      "entropy": 5.608373308181763,
      "epoch": 4.6425979680696665,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003036287983398473,
      "loss": 4.7832,
      "mean_token_accuracy": 0.23380963653326034,
      "num_tokens": 106234146.0,
      "step": 51180
    },
    {
      "entropy": 5.791401481628418,
      "epoch": 4.643051523947751,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030359670461472475,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2178636834025383,
      "num_tokens": 106244910.0,
      "step": 51185
    },
    {
      "entropy": 5.830815172195434,
      "epoch": 4.643505079825834,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030356461029821516,
      "loss": 4.9701,
      "mean_token_accuracy": 0.21234198957681655,
      "num_tokens": 106254267.0,
      "step": 51190
    },
    {
      "entropy": 5.790170288085937,
      "epoch": 4.643958635703918,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003035325153909821,
      "loss": 4.9124,
      "mean_token_accuracy": 0.227203893661499,
      "num_tokens": 106264755.0,
      "step": 51195
    },
    {
      "entropy": 5.682299709320068,
      "epoch": 4.644412191582003,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003035004198936894,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22440162003040315,
      "num_tokens": 106273964.0,
      "step": 51200
    },
    {
      "entropy": 5.747400760650635,
      "epoch": 4.644865747460087,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003034683238070008,
      "loss": 4.9244,
      "mean_token_accuracy": 0.22444498538970947,
      "num_tokens": 106283933.0,
      "step": 51205
    },
    {
      "entropy": 5.771704196929932,
      "epoch": 4.645319303338171,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030343622713157985,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23507773578166963,
      "num_tokens": 106294075.0,
      "step": 51210
    },
    {
      "entropy": 5.7363502979278564,
      "epoch": 4.6457728592162555,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003034041298680906,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23270546197891234,
      "num_tokens": 106304970.0,
      "step": 51215
    },
    {
      "entropy": 5.694227838516236,
      "epoch": 4.64622641509434,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003033720320171966,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2235713094472885,
      "num_tokens": 106315610.0,
      "step": 51220
    },
    {
      "entropy": 5.752142286300659,
      "epoch": 4.646679970972424,
      "grad_norm": 1.453125,
      "learning_rate": 0.0003033399335795619,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2222737893462181,
      "num_tokens": 106326302.0,
      "step": 51225
    },
    {
      "entropy": 5.734094715118408,
      "epoch": 4.647133526850508,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030330783455585,
      "loss": 4.8473,
      "mean_token_accuracy": 0.22540082186460494,
      "num_tokens": 106336350.0,
      "step": 51230
    },
    {
      "entropy": 5.712200069427491,
      "epoch": 4.647587082728592,
      "grad_norm": 1.25,
      "learning_rate": 0.0003032757349467249,
      "loss": 4.9235,
      "mean_token_accuracy": 0.21756965070962905,
      "num_tokens": 106348233.0,
      "step": 51235
    },
    {
      "entropy": 5.690881252288818,
      "epoch": 4.648040638606677,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030324363475285046,
      "loss": 4.8002,
      "mean_token_accuracy": 0.22656178772449492,
      "num_tokens": 106357992.0,
      "step": 51240
    },
    {
      "entropy": 5.7035199165344235,
      "epoch": 4.64849419448476,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003032115339748904,
      "loss": 4.7714,
      "mean_token_accuracy": 0.23435565531253816,
      "num_tokens": 106367683.0,
      "step": 51245
    },
    {
      "entropy": 5.726685285568237,
      "epoch": 4.6489477503628445,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003031794326135086,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2258929118514061,
      "num_tokens": 106378275.0,
      "step": 51250
    },
    {
      "entropy": 5.745105600357055,
      "epoch": 4.649401306240929,
      "grad_norm": 1.375,
      "learning_rate": 0.00030314733066936903,
      "loss": 4.8697,
      "mean_token_accuracy": 0.22825475931167602,
      "num_tokens": 106387811.0,
      "step": 51255
    },
    {
      "entropy": 5.7219610691070555,
      "epoch": 4.649854862119013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003031152281431355,
      "loss": 4.8977,
      "mean_token_accuracy": 0.2250053733587265,
      "num_tokens": 106398327.0,
      "step": 51260
    },
    {
      "entropy": 5.677974796295166,
      "epoch": 4.650308417997097,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030308312503547183,
      "loss": 4.8386,
      "mean_token_accuracy": 0.22450734227895736,
      "num_tokens": 106408602.0,
      "step": 51265
    },
    {
      "entropy": 5.694329118728637,
      "epoch": 4.6507619738751815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000303051021347042,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23311095982789992,
      "num_tokens": 106418717.0,
      "step": 51270
    },
    {
      "entropy": 5.665132761001587,
      "epoch": 4.651215529753266,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00030301891707850987,
      "loss": 4.7494,
      "mean_token_accuracy": 0.23315250724554062,
      "num_tokens": 106428311.0,
      "step": 51275
    },
    {
      "entropy": 5.7392912864685055,
      "epoch": 4.65166908563135,
      "grad_norm": 1.375,
      "learning_rate": 0.0003029868122305395,
      "loss": 4.872,
      "mean_token_accuracy": 0.2200079396367073,
      "num_tokens": 106438772.0,
      "step": 51280
    },
    {
      "entropy": 5.744803476333618,
      "epoch": 4.652122641509434,
      "grad_norm": 1.375,
      "learning_rate": 0.00030295470680379456,
      "loss": 4.858,
      "mean_token_accuracy": 0.2317780703306198,
      "num_tokens": 106448895.0,
      "step": 51285
    },
    {
      "entropy": 5.717853164672851,
      "epoch": 4.652576197387518,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003029226007989393,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22507114112377166,
      "num_tokens": 106459711.0,
      "step": 51290
    },
    {
      "entropy": 5.737358951568604,
      "epoch": 4.653029753265602,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00030289049421663744,
      "loss": 4.8444,
      "mean_token_accuracy": 0.22744954824447633,
      "num_tokens": 106469691.0,
      "step": 51295
    },
    {
      "entropy": 5.747418451309204,
      "epoch": 4.653483309143686,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003028583870575531,
      "loss": 4.8785,
      "mean_token_accuracy": 0.21811689734458922,
      "num_tokens": 106479894.0,
      "step": 51300
    },
    {
      "entropy": 5.726578903198242,
      "epoch": 4.6539368650217705,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003028262793223502,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23601107746362687,
      "num_tokens": 106491444.0,
      "step": 51305
    },
    {
      "entropy": 5.732763671875,
      "epoch": 4.654390420899855,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030279417101169276,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22891258895397187,
      "num_tokens": 106501633.0,
      "step": 51310
    },
    {
      "entropy": 5.676821851730347,
      "epoch": 4.654843976777939,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0003027620621262448,
      "loss": 4.823,
      "mean_token_accuracy": 0.2372549891471863,
      "num_tokens": 106511272.0,
      "step": 51315
    },
    {
      "entropy": 5.731031274795532,
      "epoch": 4.655297532656023,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003027299526666703,
      "loss": 4.9558,
      "mean_token_accuracy": 0.2241523265838623,
      "num_tokens": 106521977.0,
      "step": 51320
    },
    {
      "entropy": 5.751341724395752,
      "epoch": 4.655751088534108,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030269784263363327,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22731841504573821,
      "num_tokens": 106532802.0,
      "step": 51325
    },
    {
      "entropy": 5.728298044204712,
      "epoch": 4.656204644412192,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030266573202779784,
      "loss": 4.8687,
      "mean_token_accuracy": 0.22426507472991944,
      "num_tokens": 106542249.0,
      "step": 51330
    },
    {
      "entropy": 5.733735799789429,
      "epoch": 4.656658200290276,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030263362084982805,
      "loss": 4.8733,
      "mean_token_accuracy": 0.2296021357178688,
      "num_tokens": 106552501.0,
      "step": 51335
    },
    {
      "entropy": 5.672136926651001,
      "epoch": 4.65711175616836,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030260150910038793,
      "loss": 4.8101,
      "mean_token_accuracy": 0.23922621011734008,
      "num_tokens": 106561901.0,
      "step": 51340
    },
    {
      "entropy": 5.6833045959472654,
      "epoch": 4.657565312046444,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030256939678014155,
      "loss": 4.8563,
      "mean_token_accuracy": 0.2306545615196228,
      "num_tokens": 106572104.0,
      "step": 51345
    },
    {
      "entropy": 5.718745756149292,
      "epoch": 4.658018867924528,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003025372838897529,
      "loss": 4.8789,
      "mean_token_accuracy": 0.22054257541894912,
      "num_tokens": 106583436.0,
      "step": 51350
    },
    {
      "entropy": 5.771311521530151,
      "epoch": 4.658472423802612,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003025051704298864,
      "loss": 4.9567,
      "mean_token_accuracy": 0.22069106847047806,
      "num_tokens": 106593879.0,
      "step": 51355
    },
    {
      "entropy": 5.768088531494141,
      "epoch": 4.6589259796806965,
      "grad_norm": 1.25,
      "learning_rate": 0.00030247305640120585,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22021767348051072,
      "num_tokens": 106604774.0,
      "step": 51360
    },
    {
      "entropy": 5.789374589920044,
      "epoch": 4.659379535558781,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003024409418043756,
      "loss": 4.9577,
      "mean_token_accuracy": 0.21943669021129608,
      "num_tokens": 106616125.0,
      "step": 51365
    },
    {
      "entropy": 5.770188665390014,
      "epoch": 4.659833091436865,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003024088266400595,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22560051083564758,
      "num_tokens": 106625171.0,
      "step": 51370
    },
    {
      "entropy": 5.672677946090698,
      "epoch": 4.660286647314949,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030237671090892204,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2263270989060402,
      "num_tokens": 106636455.0,
      "step": 51375
    },
    {
      "entropy": 5.6992431640625,
      "epoch": 4.660740203193034,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003023445946116272,
      "loss": 4.8574,
      "mean_token_accuracy": 0.22096627950668335,
      "num_tokens": 106645956.0,
      "step": 51380
    },
    {
      "entropy": 5.7739386558532715,
      "epoch": 4.661193759071118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00030231247774883913,
      "loss": 4.9279,
      "mean_token_accuracy": 0.22541023641824723,
      "num_tokens": 106655562.0,
      "step": 51385
    },
    {
      "entropy": 5.777066946029663,
      "epoch": 4.661647314949201,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003022803603212221,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22195705473423005,
      "num_tokens": 106666620.0,
      "step": 51390
    },
    {
      "entropy": 5.7384789943695065,
      "epoch": 4.662100870827286,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003022482423294403,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2265644371509552,
      "num_tokens": 106676881.0,
      "step": 51395
    },
    {
      "entropy": 5.724926996231079,
      "epoch": 4.66255442670537,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003022161237741578,
      "loss": 4.8183,
      "mean_token_accuracy": 0.22830760776996611,
      "num_tokens": 106687765.0,
      "step": 51400
    },
    {
      "entropy": 5.698446369171142,
      "epoch": 4.663007982583454,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030218400465603895,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22499510943889617,
      "num_tokens": 106698796.0,
      "step": 51405
    },
    {
      "entropy": 5.736961221694946,
      "epoch": 4.663461538461538,
      "grad_norm": 1.453125,
      "learning_rate": 0.000302151884975748,
      "loss": 4.8363,
      "mean_token_accuracy": 0.2283301755785942,
      "num_tokens": 106709439.0,
      "step": 51410
    },
    {
      "entropy": 5.757413005828857,
      "epoch": 4.663915094339623,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003021197647339492,
      "loss": 4.9668,
      "mean_token_accuracy": 0.21872442960739136,
      "num_tokens": 106720325.0,
      "step": 51415
    },
    {
      "entropy": 5.772698593139649,
      "epoch": 4.664368650217707,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003020876439313066,
      "loss": 4.8886,
      "mean_token_accuracy": 0.2293073296546936,
      "num_tokens": 106730526.0,
      "step": 51420
    },
    {
      "entropy": 5.760725212097168,
      "epoch": 4.664822206095791,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030205552256848476,
      "loss": 4.8597,
      "mean_token_accuracy": 0.22671469002962114,
      "num_tokens": 106741142.0,
      "step": 51425
    },
    {
      "entropy": 5.8102867603302,
      "epoch": 4.665275761973875,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030202340064614783,
      "loss": 4.9596,
      "mean_token_accuracy": 0.21923359781503676,
      "num_tokens": 106751843.0,
      "step": 51430
    },
    {
      "entropy": 5.650503921508789,
      "epoch": 4.66572931785196,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003019912781649599,
      "loss": 4.7291,
      "mean_token_accuracy": 0.2369423493742943,
      "num_tokens": 106761726.0,
      "step": 51435
    },
    {
      "entropy": 5.704082298278808,
      "epoch": 4.666182873730044,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030195915512558564,
      "loss": 4.8434,
      "mean_token_accuracy": 0.23520651906728746,
      "num_tokens": 106771829.0,
      "step": 51440
    },
    {
      "entropy": 5.662385034561157,
      "epoch": 4.666636429608127,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003019270315286892,
      "loss": 4.7743,
      "mean_token_accuracy": 0.23354535102844237,
      "num_tokens": 106783457.0,
      "step": 51445
    },
    {
      "entropy": 5.722894668579102,
      "epoch": 4.667089985486212,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00030189490737493477,
      "loss": 4.8523,
      "mean_token_accuracy": 0.23609187304973603,
      "num_tokens": 106793290.0,
      "step": 51450
    },
    {
      "entropy": 5.7074357032775875,
      "epoch": 4.667543541364296,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030186278266498677,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22614172250032424,
      "num_tokens": 106804412.0,
      "step": 51455
    },
    {
      "entropy": 5.787492752075195,
      "epoch": 4.66799709724238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003018306573995097,
      "loss": 4.9649,
      "mean_token_accuracy": 0.22380151897668837,
      "num_tokens": 106815353.0,
      "step": 51460
    },
    {
      "entropy": 5.746248197555542,
      "epoch": 4.668450653120464,
      "grad_norm": 1.390625,
      "learning_rate": 0.00030179853157916777,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22762187272310258,
      "num_tokens": 106826350.0,
      "step": 51465
    },
    {
      "entropy": 5.757683563232422,
      "epoch": 4.668904208998549,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003017664052046253,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2256131038069725,
      "num_tokens": 106837434.0,
      "step": 51470
    },
    {
      "entropy": 5.703148651123047,
      "epoch": 4.669357764876633,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030173427827654683,
      "loss": 4.8925,
      "mean_token_accuracy": 0.22307565808296204,
      "num_tokens": 106847829.0,
      "step": 51475
    },
    {
      "entropy": 5.7936779975891115,
      "epoch": 4.669811320754717,
      "grad_norm": 1.375,
      "learning_rate": 0.0003017021507955967,
      "loss": 4.9899,
      "mean_token_accuracy": 0.21836089640855788,
      "num_tokens": 106858519.0,
      "step": 51480
    },
    {
      "entropy": 5.781712007522583,
      "epoch": 4.670264876632801,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030167002276243915,
      "loss": 4.8452,
      "mean_token_accuracy": 0.23784094899892808,
      "num_tokens": 106868614.0,
      "step": 51485
    },
    {
      "entropy": 5.785002183914185,
      "epoch": 4.670718432510886,
      "grad_norm": 1.3125,
      "learning_rate": 0.00030163789417773885,
      "loss": 4.9952,
      "mean_token_accuracy": 0.21650078296661376,
      "num_tokens": 106879592.0,
      "step": 51490
    },
    {
      "entropy": 5.73831696510315,
      "epoch": 4.67117198838897,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003016057650421601,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22140124440193176,
      "num_tokens": 106890083.0,
      "step": 51495
    },
    {
      "entropy": 5.80596661567688,
      "epoch": 4.671625544267053,
      "grad_norm": 1.46875,
      "learning_rate": 0.00030157363535636726,
      "loss": 4.9504,
      "mean_token_accuracy": 0.2256872057914734,
      "num_tokens": 106900672.0,
      "step": 51500
    },
    {
      "entropy": 5.761689567565918,
      "epoch": 4.672079100145138,
      "grad_norm": 1.40625,
      "learning_rate": 0.000301541505121025,
      "loss": 4.8315,
      "mean_token_accuracy": 0.22799703627824783,
      "num_tokens": 106910396.0,
      "step": 51505
    },
    {
      "entropy": 5.7893431186676025,
      "epoch": 4.672532656023222,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003015093743367976,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22177671045064926,
      "num_tokens": 106919742.0,
      "step": 51510
    },
    {
      "entropy": 5.680334663391113,
      "epoch": 4.672986211901306,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003014772430043495,
      "loss": 4.7985,
      "mean_token_accuracy": 0.23326301723718643,
      "num_tokens": 106930837.0,
      "step": 51515
    },
    {
      "entropy": 5.687160873413086,
      "epoch": 4.67343976777939,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030144511112434536,
      "loss": 4.826,
      "mean_token_accuracy": 0.23338584005832672,
      "num_tokens": 106942023.0,
      "step": 51520
    },
    {
      "entropy": 5.767622423171997,
      "epoch": 4.673893323657475,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030141297869744956,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22334835380315782,
      "num_tokens": 106951374.0,
      "step": 51525
    },
    {
      "entropy": 5.837855672836303,
      "epoch": 4.674346879535559,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030138084572432663,
      "loss": 4.9668,
      "mean_token_accuracy": 0.2175795242190361,
      "num_tokens": 106961714.0,
      "step": 51530
    },
    {
      "entropy": 5.752449703216553,
      "epoch": 4.674800435413643,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003013487122056411,
      "loss": 4.8789,
      "mean_token_accuracy": 0.23010564148426055,
      "num_tokens": 106971783.0,
      "step": 51535
    },
    {
      "entropy": 5.7074363231658936,
      "epoch": 4.6752539912917275,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003013165781420575,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2339213639497757,
      "num_tokens": 106982047.0,
      "step": 51540
    },
    {
      "entropy": 5.7486108303070065,
      "epoch": 4.675707547169811,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030128444353424034,
      "loss": 4.8822,
      "mean_token_accuracy": 0.22810571044683456,
      "num_tokens": 106993063.0,
      "step": 51545
    },
    {
      "entropy": 5.772168445587158,
      "epoch": 4.676161103047895,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030125230838285415,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22180558294057845,
      "num_tokens": 107002865.0,
      "step": 51550
    },
    {
      "entropy": 5.7699689865112305,
      "epoch": 4.676614658925979,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030122017268856353,
      "loss": 4.8396,
      "mean_token_accuracy": 0.23305536061525345,
      "num_tokens": 107013007.0,
      "step": 51555
    },
    {
      "entropy": 5.761817073822021,
      "epoch": 4.677068214804064,
      "grad_norm": 1.34375,
      "learning_rate": 0.00030118803645203315,
      "loss": 4.9301,
      "mean_token_accuracy": 0.21818890124559404,
      "num_tokens": 107023109.0,
      "step": 51560
    },
    {
      "entropy": 5.697683811187744,
      "epoch": 4.677521770682148,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003011558996739274,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22514154613018036,
      "num_tokens": 107033824.0,
      "step": 51565
    },
    {
      "entropy": 5.747349643707276,
      "epoch": 4.677975326560232,
      "grad_norm": 1.34375,
      "learning_rate": 0.000301123762354911,
      "loss": 4.9014,
      "mean_token_accuracy": 0.22396245151758193,
      "num_tokens": 107043529.0,
      "step": 51570
    },
    {
      "entropy": 5.768860292434693,
      "epoch": 4.6784288824383164,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003010916244956486,
      "loss": 4.9154,
      "mean_token_accuracy": 0.23263935297727584,
      "num_tokens": 107054378.0,
      "step": 51575
    },
    {
      "entropy": 5.702558088302612,
      "epoch": 4.678882438316401,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003010594860968048,
      "loss": 4.7946,
      "mean_token_accuracy": 0.22707646191120148,
      "num_tokens": 107064300.0,
      "step": 51580
    },
    {
      "entropy": 5.785502862930298,
      "epoch": 4.679335994194485,
      "grad_norm": 1.34375,
      "learning_rate": 0.000301027347159044,
      "loss": 4.9399,
      "mean_token_accuracy": 0.22154979705810546,
      "num_tokens": 107074560.0,
      "step": 51585
    },
    {
      "entropy": 5.7197455883026125,
      "epoch": 4.679789550072569,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003009952076830312,
      "loss": 4.8554,
      "mean_token_accuracy": 0.21966522186994553,
      "num_tokens": 107084574.0,
      "step": 51590
    },
    {
      "entropy": 5.710264492034912,
      "epoch": 4.6802431059506535,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003009630676694308,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2239594802260399,
      "num_tokens": 107095207.0,
      "step": 51595
    },
    {
      "entropy": 5.7067742347717285,
      "epoch": 4.680696661828737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00030093092711890756,
      "loss": 4.8594,
      "mean_token_accuracy": 0.22674519568681717,
      "num_tokens": 107106091.0,
      "step": 51600
    },
    {
      "entropy": 5.762162542343139,
      "epoch": 4.681150217706821,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003008987860321261,
      "loss": 4.9233,
      "mean_token_accuracy": 0.220753014087677,
      "num_tokens": 107117753.0,
      "step": 51605
    },
    {
      "entropy": 5.685170745849609,
      "epoch": 4.681603773584905,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003008666444097512,
      "loss": 4.7872,
      "mean_token_accuracy": 0.23885023593902588,
      "num_tokens": 107128131.0,
      "step": 51610
    },
    {
      "entropy": 5.77716326713562,
      "epoch": 4.68205732946299,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003008345022524475,
      "loss": 5.0156,
      "mean_token_accuracy": 0.2097482427954674,
      "num_tokens": 107139447.0,
      "step": 51615
    },
    {
      "entropy": 5.735281705856323,
      "epoch": 4.682510885341074,
      "grad_norm": 1.296875,
      "learning_rate": 0.00030080235956087963,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22154488563537597,
      "num_tokens": 107149721.0,
      "step": 51620
    },
    {
      "entropy": 5.753556251525879,
      "epoch": 4.682964441219158,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003007702163357125,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22360310405492784,
      "num_tokens": 107159814.0,
      "step": 51625
    },
    {
      "entropy": 5.696829986572266,
      "epoch": 4.6834179970972425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00030073807257761074,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2265480563044548,
      "num_tokens": 107169012.0,
      "step": 51630
    },
    {
      "entropy": 5.775127410888672,
      "epoch": 4.683871552975327,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00030070592828723903,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22296603322029113,
      "num_tokens": 107179571.0,
      "step": 51635
    },
    {
      "entropy": 5.733376216888428,
      "epoch": 4.684325108853411,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003006737834652621,
      "loss": 4.8714,
      "mean_token_accuracy": 0.22372306734323502,
      "num_tokens": 107190535.0,
      "step": 51640
    },
    {
      "entropy": 5.793765211105347,
      "epoch": 4.684778664731495,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003006416381123449,
      "loss": 4.8803,
      "mean_token_accuracy": 0.2318541243672371,
      "num_tokens": 107200361.0,
      "step": 51645
    },
    {
      "entropy": 5.773046398162842,
      "epoch": 4.68523222060958,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030060949222915205,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22367767691612245,
      "num_tokens": 107211080.0,
      "step": 51650
    },
    {
      "entropy": 5.7543402194976805,
      "epoch": 4.685685776487663,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003005773458163484,
      "loss": 4.915,
      "mean_token_accuracy": 0.22826526761054994,
      "num_tokens": 107220556.0,
      "step": 51655
    },
    {
      "entropy": 5.732157039642334,
      "epoch": 4.686139332365747,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030054519887459864,
      "loss": 4.8547,
      "mean_token_accuracy": 0.22265018820762633,
      "num_tokens": 107230693.0,
      "step": 51660
    },
    {
      "entropy": 5.754461812973022,
      "epoch": 4.6865928882438315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030051305140456775,
      "loss": 4.9283,
      "mean_token_accuracy": 0.220540788769722,
      "num_tokens": 107240926.0,
      "step": 51665
    },
    {
      "entropy": 5.726976203918457,
      "epoch": 4.687046444121916,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030048090340692034,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22420696914196014,
      "num_tokens": 107251425.0,
      "step": 51670
    },
    {
      "entropy": 5.747324514389038,
      "epoch": 4.6875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00030044875488232145,
      "loss": 4.8823,
      "mean_token_accuracy": 0.23226401805877686,
      "num_tokens": 107262046.0,
      "step": 51675
    },
    {
      "entropy": 5.647211694717408,
      "epoch": 4.687953555878084,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003004166058314358,
      "loss": 4.8268,
      "mean_token_accuracy": 0.23222002536058425,
      "num_tokens": 107273141.0,
      "step": 51680
    },
    {
      "entropy": 5.749797582626343,
      "epoch": 4.6884071117561685,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003003844562549282,
      "loss": 4.8951,
      "mean_token_accuracy": 0.23144533336162568,
      "num_tokens": 107282497.0,
      "step": 51685
    },
    {
      "entropy": 5.682872867584228,
      "epoch": 4.688860667634253,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00030035230615346373,
      "loss": 4.8291,
      "mean_token_accuracy": 0.23311278522014617,
      "num_tokens": 107292710.0,
      "step": 51690
    },
    {
      "entropy": 5.666674423217773,
      "epoch": 4.689314223512337,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00030032015552770695,
      "loss": 4.7449,
      "mean_token_accuracy": 0.23501755446195602,
      "num_tokens": 107303157.0,
      "step": 51695
    },
    {
      "entropy": 5.717285680770874,
      "epoch": 4.68976777939042,
      "grad_norm": 1.375,
      "learning_rate": 0.00030028800437832296,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2264223203063011,
      "num_tokens": 107313503.0,
      "step": 51700
    },
    {
      "entropy": 5.699522399902344,
      "epoch": 4.690221335268505,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003002558527059766,
      "loss": 4.853,
      "mean_token_accuracy": 0.2275082662701607,
      "num_tokens": 107322276.0,
      "step": 51705
    },
    {
      "entropy": 5.713011264801025,
      "epoch": 4.690674891146589,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003002237005113328,
      "loss": 4.9189,
      "mean_token_accuracy": 0.22197742760181427,
      "num_tokens": 107332876.0,
      "step": 51710
    },
    {
      "entropy": 5.724290275573731,
      "epoch": 4.691128447024673,
      "grad_norm": 1.4375,
      "learning_rate": 0.00030019154779505644,
      "loss": 4.8608,
      "mean_token_accuracy": 0.22795587331056594,
      "num_tokens": 107343262.0,
      "step": 51715
    },
    {
      "entropy": 5.669172525405884,
      "epoch": 4.6915820029027575,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003001593945578124,
      "loss": 4.841,
      "mean_token_accuracy": 0.21951988935470582,
      "num_tokens": 107354281.0,
      "step": 51720
    },
    {
      "entropy": 5.654275035858154,
      "epoch": 4.692035558780842,
      "grad_norm": 1.25,
      "learning_rate": 0.0003001272408002657,
      "loss": 4.8237,
      "mean_token_accuracy": 0.23599613308906556,
      "num_tokens": 107364011.0,
      "step": 51725
    },
    {
      "entropy": 5.691574668884277,
      "epoch": 4.692489114658926,
      "grad_norm": 1.375,
      "learning_rate": 0.00030009508652308124,
      "loss": 4.8464,
      "mean_token_accuracy": 0.22606113851070403,
      "num_tokens": 107372943.0,
      "step": 51730
    },
    {
      "entropy": 5.726454019546509,
      "epoch": 4.69294267053701,
      "grad_norm": 1.359375,
      "learning_rate": 0.00030006293172692405,
      "loss": 4.9014,
      "mean_token_accuracy": 0.23088294863700867,
      "num_tokens": 107382838.0,
      "step": 51735
    },
    {
      "entropy": 5.784696006774903,
      "epoch": 4.693396226415095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030003077641245897,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2261417552828789,
      "num_tokens": 107393250.0,
      "step": 51740
    },
    {
      "entropy": 5.766187953948974,
      "epoch": 4.693849782293179,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029999862058035116,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23392524421215058,
      "num_tokens": 107403153.0,
      "step": 51745
    },
    {
      "entropy": 5.749983072280884,
      "epoch": 4.694303338171263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029996646423126544,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22529706954956055,
      "num_tokens": 107413092.0,
      "step": 51750
    },
    {
      "entropy": 5.730692911148071,
      "epoch": 4.6947568940493465,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002999343073658669,
      "loss": 4.8965,
      "mean_token_accuracy": 0.22397483736276627,
      "num_tokens": 107423561.0,
      "step": 51755
    },
    {
      "entropy": 5.775936603546143,
      "epoch": 4.695210449927431,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029990214998482046,
      "loss": 4.8999,
      "mean_token_accuracy": 0.22763929814100264,
      "num_tokens": 107433954.0,
      "step": 51760
    },
    {
      "entropy": 5.757132863998413,
      "epoch": 4.695664005805515,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002998699920887913,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2286432206630707,
      "num_tokens": 107443150.0,
      "step": 51765
    },
    {
      "entropy": 5.691089391708374,
      "epoch": 4.696117561683599,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002998378336784443,
      "loss": 4.8127,
      "mean_token_accuracy": 0.22819581180810927,
      "num_tokens": 107454250.0,
      "step": 51770
    },
    {
      "entropy": 5.665602397918701,
      "epoch": 4.6965711175616836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002998056747544446,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23844688087701799,
      "num_tokens": 107465058.0,
      "step": 51775
    },
    {
      "entropy": 5.654711818695068,
      "epoch": 4.697024673439768,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002997735153174572,
      "loss": 4.8044,
      "mean_token_accuracy": 0.23512857556343078,
      "num_tokens": 107475449.0,
      "step": 51780
    },
    {
      "entropy": 5.847668695449829,
      "epoch": 4.697478229317852,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029974135536814725,
      "loss": 5.0656,
      "mean_token_accuracy": 0.21303813606500627,
      "num_tokens": 107486645.0,
      "step": 51785
    },
    {
      "entropy": 5.7273622989654545,
      "epoch": 4.697931785195936,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002997091949071797,
      "loss": 4.8769,
      "mean_token_accuracy": 0.22586718201637268,
      "num_tokens": 107497242.0,
      "step": 51790
    },
    {
      "entropy": 5.782224416732788,
      "epoch": 4.698385341074021,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002996770339352197,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2227385625243187,
      "num_tokens": 107507596.0,
      "step": 51795
    },
    {
      "entropy": 5.79605016708374,
      "epoch": 4.698838896952105,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002996448724529324,
      "loss": 4.9578,
      "mean_token_accuracy": 0.2206279531121254,
      "num_tokens": 107517795.0,
      "step": 51800
    },
    {
      "entropy": 5.749842643737793,
      "epoch": 4.699292452830189,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029961271046098284,
      "loss": 4.8513,
      "mean_token_accuracy": 0.23080481588840485,
      "num_tokens": 107527805.0,
      "step": 51805
    },
    {
      "entropy": 5.721256113052368,
      "epoch": 4.6997460087082725,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029958054796003615,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22594361305236815,
      "num_tokens": 107537396.0,
      "step": 51810
    },
    {
      "entropy": 5.700278854370117,
      "epoch": 4.700199564586357,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002995483849507575,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22637415677309036,
      "num_tokens": 107546876.0,
      "step": 51815
    },
    {
      "entropy": 5.743438053131103,
      "epoch": 4.700653120464441,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000299516221433812,
      "loss": 4.8438,
      "mean_token_accuracy": 0.2330237179994583,
      "num_tokens": 107556865.0,
      "step": 51820
    },
    {
      "entropy": 5.683301305770874,
      "epoch": 4.701106676342525,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029948405740986466,
      "loss": 4.8596,
      "mean_token_accuracy": 0.22216011732816696,
      "num_tokens": 107566305.0,
      "step": 51825
    },
    {
      "entropy": 5.7773661613464355,
      "epoch": 4.70156023222061,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002994518928795809,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22165110409259797,
      "num_tokens": 107575640.0,
      "step": 51830
    },
    {
      "entropy": 5.7924439907073975,
      "epoch": 4.702013788098694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029941972784362574,
      "loss": 4.9474,
      "mean_token_accuracy": 0.23148273676633835,
      "num_tokens": 107585955.0,
      "step": 51835
    },
    {
      "entropy": 5.785369443893432,
      "epoch": 4.702467343976778,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029938756230266436,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22488485425710678,
      "num_tokens": 107596911.0,
      "step": 51840
    },
    {
      "entropy": 5.668543815612793,
      "epoch": 4.702920899854862,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000299355396257362,
      "loss": 4.8184,
      "mean_token_accuracy": 0.23126240372657775,
      "num_tokens": 107606117.0,
      "step": 51845
    },
    {
      "entropy": 5.664534568786621,
      "epoch": 4.703374455732947,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002993232297083838,
      "loss": 4.7322,
      "mean_token_accuracy": 0.24791700690984725,
      "num_tokens": 107616363.0,
      "step": 51850
    },
    {
      "entropy": 5.690454149246216,
      "epoch": 4.70382801161103,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002992910626563951,
      "loss": 4.8027,
      "mean_token_accuracy": 0.22667231857776643,
      "num_tokens": 107627402.0,
      "step": 51855
    },
    {
      "entropy": 5.713854789733887,
      "epoch": 4.704281567489114,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029925889510206094,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22881426066160201,
      "num_tokens": 107637446.0,
      "step": 51860
    },
    {
      "entropy": 5.674842691421508,
      "epoch": 4.704735123367199,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002992267270460467,
      "loss": 4.8307,
      "mean_token_accuracy": 0.2307622566819191,
      "num_tokens": 107647386.0,
      "step": 51865
    },
    {
      "entropy": 5.7519636154174805,
      "epoch": 4.705188679245283,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002991945584890175,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2296588599681854,
      "num_tokens": 107657881.0,
      "step": 51870
    },
    {
      "entropy": 5.78574800491333,
      "epoch": 4.705642235123367,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029916238943163874,
      "loss": 4.8832,
      "mean_token_accuracy": 0.2325642704963684,
      "num_tokens": 107668478.0,
      "step": 51875
    },
    {
      "entropy": 5.86171145439148,
      "epoch": 4.706095791001451,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002991302198745756,
      "loss": 4.9996,
      "mean_token_accuracy": 0.21979895532131194,
      "num_tokens": 107679149.0,
      "step": 51880
    },
    {
      "entropy": 5.784580993652344,
      "epoch": 4.706549346879536,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002990980498184934,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22261294573545456,
      "num_tokens": 107689471.0,
      "step": 51885
    },
    {
      "entropy": 5.74379448890686,
      "epoch": 4.70700290275762,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029906587926405734,
      "loss": 4.8495,
      "mean_token_accuracy": 0.23520029932260514,
      "num_tokens": 107700604.0,
      "step": 51890
    },
    {
      "entropy": 5.643253755569458,
      "epoch": 4.707456458635704,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002990337082119327,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23258235156536103,
      "num_tokens": 107710268.0,
      "step": 51895
    },
    {
      "entropy": 5.762271595001221,
      "epoch": 4.707910014513788,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029900153666278503,
      "loss": 4.8644,
      "mean_token_accuracy": 0.2238776430487633,
      "num_tokens": 107720328.0,
      "step": 51900
    },
    {
      "entropy": 5.737943172454834,
      "epoch": 4.708363570391873,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029896936461727924,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2259916692972183,
      "num_tokens": 107730295.0,
      "step": 51905
    },
    {
      "entropy": 5.6963259220123295,
      "epoch": 4.708817126269956,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029893719207608105,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2251191794872284,
      "num_tokens": 107741334.0,
      "step": 51910
    },
    {
      "entropy": 5.734709119796753,
      "epoch": 4.70927068214804,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029890501903985555,
      "loss": 4.8514,
      "mean_token_accuracy": 0.2236187055706978,
      "num_tokens": 107752060.0,
      "step": 51915
    },
    {
      "entropy": 5.7470423698425295,
      "epoch": 4.709724238026125,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029887284550926817,
      "loss": 4.8348,
      "mean_token_accuracy": 0.22658580988645555,
      "num_tokens": 107762188.0,
      "step": 51920
    },
    {
      "entropy": 5.769175243377686,
      "epoch": 4.710177793904209,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002988406714849843,
      "loss": 4.8946,
      "mean_token_accuracy": 0.2219429209828377,
      "num_tokens": 107771670.0,
      "step": 51925
    },
    {
      "entropy": 5.705093479156494,
      "epoch": 4.710631349782293,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002988084969676692,
      "loss": 4.8349,
      "mean_token_accuracy": 0.23165858834981917,
      "num_tokens": 107781734.0,
      "step": 51930
    },
    {
      "entropy": 5.751940107345581,
      "epoch": 4.711084905660377,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029877632195798843,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22066578567028045,
      "num_tokens": 107792159.0,
      "step": 51935
    },
    {
      "entropy": 5.735274076461792,
      "epoch": 4.711538461538462,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029874414645660713,
      "loss": 4.8776,
      "mean_token_accuracy": 0.22660958021879196,
      "num_tokens": 107803153.0,
      "step": 51940
    },
    {
      "entropy": 5.68940839767456,
      "epoch": 4.711992017416546,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002987119704641909,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2292718455195427,
      "num_tokens": 107813214.0,
      "step": 51945
    },
    {
      "entropy": 5.718183183670044,
      "epoch": 4.71244557329463,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002986797939814051,
      "loss": 4.8759,
      "mean_token_accuracy": 0.22532938569784164,
      "num_tokens": 107823976.0,
      "step": 51950
    },
    {
      "entropy": 5.741864347457886,
      "epoch": 4.712899129172714,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029864761700891507,
      "loss": 4.888,
      "mean_token_accuracy": 0.23026739060878754,
      "num_tokens": 107834127.0,
      "step": 51955
    },
    {
      "entropy": 5.789374256134034,
      "epoch": 4.713352685050799,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029861543954738635,
      "loss": 4.9907,
      "mean_token_accuracy": 0.21568645536899567,
      "num_tokens": 107843753.0,
      "step": 51960
    },
    {
      "entropy": 5.780705976486206,
      "epoch": 4.713806240928882,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029858326159748434,
      "loss": 4.8757,
      "mean_token_accuracy": 0.23155114948749542,
      "num_tokens": 107854690.0,
      "step": 51965
    },
    {
      "entropy": 5.735806322097778,
      "epoch": 4.714259796806966,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029855108315987437,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2294026106595993,
      "num_tokens": 107865235.0,
      "step": 51970
    },
    {
      "entropy": 5.741643714904785,
      "epoch": 4.714713352685051,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000298518904235222,
      "loss": 4.8397,
      "mean_token_accuracy": 0.22992040514945983,
      "num_tokens": 107876052.0,
      "step": 51975
    },
    {
      "entropy": 5.701871395111084,
      "epoch": 4.715166908563135,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002984867248241928,
      "loss": 4.8855,
      "mean_token_accuracy": 0.2299584299325943,
      "num_tokens": 107886187.0,
      "step": 51980
    },
    {
      "entropy": 5.743469953536987,
      "epoch": 4.715620464441219,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002984545449274521,
      "loss": 4.9054,
      "mean_token_accuracy": 0.22963334918022155,
      "num_tokens": 107896995.0,
      "step": 51985
    },
    {
      "entropy": 5.771836042404175,
      "epoch": 4.7160740203193035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002984223645456655,
      "loss": 4.9094,
      "mean_token_accuracy": 0.22716859430074693,
      "num_tokens": 107907868.0,
      "step": 51990
    },
    {
      "entropy": 5.7437724590301515,
      "epoch": 4.716527576197388,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029839018367949825,
      "loss": 4.8997,
      "mean_token_accuracy": 0.22353609800338745,
      "num_tokens": 107917564.0,
      "step": 51995
    },
    {
      "entropy": 5.736751508712769,
      "epoch": 4.716981132075472,
      "grad_norm": 1.375,
      "learning_rate": 0.0002983580023296162,
      "loss": 4.8333,
      "mean_token_accuracy": 0.23266978412866593,
      "num_tokens": 107926853.0,
      "step": 52000
    },
    {
      "entropy": 5.677763891220093,
      "epoch": 4.717434687953556,
      "grad_norm": 1.25,
      "learning_rate": 0.00029832582049668465,
      "loss": 4.8798,
      "mean_token_accuracy": 0.220980966091156,
      "num_tokens": 107937440.0,
      "step": 52005
    },
    {
      "entropy": 5.737581443786621,
      "epoch": 4.71788824383164,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029829363818136917,
      "loss": 4.8979,
      "mean_token_accuracy": 0.22756726741790773,
      "num_tokens": 107947750.0,
      "step": 52010
    },
    {
      "entropy": 5.774404239654541,
      "epoch": 4.718341799709724,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002982614553843354,
      "loss": 4.8991,
      "mean_token_accuracy": 0.2292641744017601,
      "num_tokens": 107959646.0,
      "step": 52015
    },
    {
      "entropy": 5.786757850646973,
      "epoch": 4.718795355587808,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029822927210624873,
      "loss": 4.9131,
      "mean_token_accuracy": 0.2245540738105774,
      "num_tokens": 107969678.0,
      "step": 52020
    },
    {
      "entropy": 5.786465454101562,
      "epoch": 4.719248911465892,
      "grad_norm": 1.375,
      "learning_rate": 0.00029819708834777483,
      "loss": 4.9041,
      "mean_token_accuracy": 0.23132051527500153,
      "num_tokens": 107979589.0,
      "step": 52025
    },
    {
      "entropy": 5.651255083084107,
      "epoch": 4.719702467343977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002981649041095792,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2315720275044441,
      "num_tokens": 107990682.0,
      "step": 52030
    },
    {
      "entropy": 5.752179718017578,
      "epoch": 4.720156023222061,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002981327193923275,
      "loss": 4.8312,
      "mean_token_accuracy": 0.22878191620111465,
      "num_tokens": 108000999.0,
      "step": 52035
    },
    {
      "entropy": 5.7009344577789305,
      "epoch": 4.720609579100145,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002981005341966852,
      "loss": 4.787,
      "mean_token_accuracy": 0.2282201424241066,
      "num_tokens": 108010709.0,
      "step": 52040
    },
    {
      "entropy": 5.6669587135314945,
      "epoch": 4.7210631349782295,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000298068348523318,
      "loss": 4.8175,
      "mean_token_accuracy": 0.2310306280851364,
      "num_tokens": 108020522.0,
      "step": 52045
    },
    {
      "entropy": 5.669869375228882,
      "epoch": 4.721516690856314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002980361623728915,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23346278220415115,
      "num_tokens": 108031387.0,
      "step": 52050
    },
    {
      "entropy": 5.735886716842652,
      "epoch": 4.721970246734398,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002980039757460713,
      "loss": 4.8734,
      "mean_token_accuracy": 0.2248528629541397,
      "num_tokens": 108041653.0,
      "step": 52055
    },
    {
      "entropy": 5.815509891510009,
      "epoch": 4.722423802612482,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029797178864352306,
      "loss": 5.0166,
      "mean_token_accuracy": 0.22266253530979158,
      "num_tokens": 108051969.0,
      "step": 52060
    },
    {
      "entropy": 5.743517208099365,
      "epoch": 4.722877358490566,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029793960106591226,
      "loss": 4.866,
      "mean_token_accuracy": 0.2275713488459587,
      "num_tokens": 108061364.0,
      "step": 52065
    },
    {
      "entropy": 5.737391948699951,
      "epoch": 4.72333091436865,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002979074130139047,
      "loss": 4.9294,
      "mean_token_accuracy": 0.2266415536403656,
      "num_tokens": 108071140.0,
      "step": 52070
    },
    {
      "entropy": 5.724600791931152,
      "epoch": 4.723784470246734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029787522448816603,
      "loss": 4.9579,
      "mean_token_accuracy": 0.22318881899118423,
      "num_tokens": 108082599.0,
      "step": 52075
    },
    {
      "entropy": 5.770811557769775,
      "epoch": 4.7242380261248185,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029784303548936183,
      "loss": 4.8842,
      "mean_token_accuracy": 0.2288818374276161,
      "num_tokens": 108092017.0,
      "step": 52080
    },
    {
      "entropy": 5.789920520782471,
      "epoch": 4.724691582002903,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002978108460181579,
      "loss": 4.9512,
      "mean_token_accuracy": 0.2202690526843071,
      "num_tokens": 108103853.0,
      "step": 52085
    },
    {
      "entropy": 5.725744438171387,
      "epoch": 4.725145137880987,
      "grad_norm": 1.25,
      "learning_rate": 0.0002977786560752198,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22786682546138765,
      "num_tokens": 108114727.0,
      "step": 52090
    },
    {
      "entropy": 5.744351577758789,
      "epoch": 4.725598693759071,
      "grad_norm": 1.25,
      "learning_rate": 0.0002977464656612133,
      "loss": 4.8443,
      "mean_token_accuracy": 0.23432247191667557,
      "num_tokens": 108125653.0,
      "step": 52095
    },
    {
      "entropy": 5.759162378311157,
      "epoch": 4.7260522496371555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029771427477680415,
      "loss": 4.8705,
      "mean_token_accuracy": 0.22985611855983734,
      "num_tokens": 108135281.0,
      "step": 52100
    },
    {
      "entropy": 5.76783242225647,
      "epoch": 4.72650580551524,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029768208342265796,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22482046186923982,
      "num_tokens": 108145005.0,
      "step": 52105
    },
    {
      "entropy": 5.777819490432739,
      "epoch": 4.726959361393323,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002976498915994406,
      "loss": 4.9591,
      "mean_token_accuracy": 0.22065050899982452,
      "num_tokens": 108156477.0,
      "step": 52110
    },
    {
      "entropy": 5.7595220565795895,
      "epoch": 4.727412917271408,
      "grad_norm": 1.546875,
      "learning_rate": 0.00029761769930781757,
      "loss": 4.8948,
      "mean_token_accuracy": 0.22768924087285997,
      "num_tokens": 108166947.0,
      "step": 52115
    },
    {
      "entropy": 5.6929737567901615,
      "epoch": 4.727866473149492,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002975855065484548,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2424178034067154,
      "num_tokens": 108175889.0,
      "step": 52120
    },
    {
      "entropy": 5.81324872970581,
      "epoch": 4.728320029027576,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000297553313322018,
      "loss": 4.9655,
      "mean_token_accuracy": 0.2161299467086792,
      "num_tokens": 108186276.0,
      "step": 52125
    },
    {
      "entropy": 5.713753747940063,
      "epoch": 4.72877358490566,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029752111962917297,
      "loss": 4.8865,
      "mean_token_accuracy": 0.22509230375289918,
      "num_tokens": 108196960.0,
      "step": 52130
    },
    {
      "entropy": 5.778022050857544,
      "epoch": 4.7292271407837445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002974889254705854,
      "loss": 4.8752,
      "mean_token_accuracy": 0.2242655262351036,
      "num_tokens": 108207442.0,
      "step": 52135
    },
    {
      "entropy": 5.78687801361084,
      "epoch": 4.729680696661829,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029745673084692116,
      "loss": 4.8418,
      "mean_token_accuracy": 0.24009375274181366,
      "num_tokens": 108217919.0,
      "step": 52140
    },
    {
      "entropy": 5.706824398040771,
      "epoch": 4.730134252539913,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000297424535758846,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23449888974428176,
      "num_tokens": 108228106.0,
      "step": 52145
    },
    {
      "entropy": 5.71440544128418,
      "epoch": 4.730587808417997,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029739234020702564,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22749220132827758,
      "num_tokens": 108238398.0,
      "step": 52150
    },
    {
      "entropy": 5.730269193649292,
      "epoch": 4.731041364296082,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029736014419212605,
      "loss": 4.9172,
      "mean_token_accuracy": 0.22391206324100493,
      "num_tokens": 108248825.0,
      "step": 52155
    },
    {
      "entropy": 5.671224784851074,
      "epoch": 4.731494920174166,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029732794771481305,
      "loss": 4.816,
      "mean_token_accuracy": 0.23091934323310853,
      "num_tokens": 108258989.0,
      "step": 52160
    },
    {
      "entropy": 5.677763748168945,
      "epoch": 4.731948476052249,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002972957507757523,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22958742082118988,
      "num_tokens": 108268251.0,
      "step": 52165
    },
    {
      "entropy": 5.729031324386597,
      "epoch": 4.7324020319303335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029726355337560985,
      "loss": 4.8868,
      "mean_token_accuracy": 0.223382368683815,
      "num_tokens": 108278290.0,
      "step": 52170
    },
    {
      "entropy": 5.760158538818359,
      "epoch": 4.732855587808418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029723135551505133,
      "loss": 4.887,
      "mean_token_accuracy": 0.22393913120031356,
      "num_tokens": 108288727.0,
      "step": 52175
    },
    {
      "entropy": 5.724992847442627,
      "epoch": 4.733309143686502,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029719915719474274,
      "loss": 4.8662,
      "mean_token_accuracy": 0.2265891671180725,
      "num_tokens": 108299781.0,
      "step": 52180
    },
    {
      "entropy": 5.771351861953735,
      "epoch": 4.733762699564586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002971669584153499,
      "loss": 4.8938,
      "mean_token_accuracy": 0.22107615172863007,
      "num_tokens": 108310703.0,
      "step": 52185
    },
    {
      "entropy": 5.694336175918579,
      "epoch": 4.7342162554426706,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002971347591775388,
      "loss": 4.8322,
      "mean_token_accuracy": 0.2315443843603134,
      "num_tokens": 108321096.0,
      "step": 52190
    },
    {
      "entropy": 5.757145261764526,
      "epoch": 4.734669811320755,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002971025594819751,
      "loss": 4.9048,
      "mean_token_accuracy": 0.2233284130692482,
      "num_tokens": 108331899.0,
      "step": 52195
    },
    {
      "entropy": 5.667433881759644,
      "epoch": 4.735123367198839,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029707035932932495,
      "loss": 4.7446,
      "mean_token_accuracy": 0.23520841002464293,
      "num_tokens": 108342248.0,
      "step": 52200
    },
    {
      "entropy": 5.738564682006836,
      "epoch": 4.735576923076923,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029703815872025413,
      "loss": 4.934,
      "mean_token_accuracy": 0.2259595960378647,
      "num_tokens": 108355264.0,
      "step": 52205
    },
    {
      "entropy": 5.693443679809571,
      "epoch": 4.736030478955008,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002970059576554285,
      "loss": 4.8553,
      "mean_token_accuracy": 0.2256321743130684,
      "num_tokens": 108365929.0,
      "step": 52210
    },
    {
      "entropy": 5.746431303024292,
      "epoch": 4.736484034833092,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002969737561355142,
      "loss": 4.8693,
      "mean_token_accuracy": 0.2266010597348213,
      "num_tokens": 108376874.0,
      "step": 52215
    },
    {
      "entropy": 5.733250284194947,
      "epoch": 4.736937590711175,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002969415541611769,
      "loss": 4.8264,
      "mean_token_accuracy": 0.22342591136693954,
      "num_tokens": 108387277.0,
      "step": 52220
    },
    {
      "entropy": 5.77848801612854,
      "epoch": 4.7373911465892595,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002969093517330827,
      "loss": 4.983,
      "mean_token_accuracy": 0.2213229551911354,
      "num_tokens": 108399088.0,
      "step": 52225
    },
    {
      "entropy": 5.773819541931152,
      "epoch": 4.737844702467344,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002968771488518975,
      "loss": 4.9662,
      "mean_token_accuracy": 0.22060526311397552,
      "num_tokens": 108409875.0,
      "step": 52230
    },
    {
      "entropy": 5.765353870391846,
      "epoch": 4.738298258345428,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002968449455182874,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2288182109594345,
      "num_tokens": 108421133.0,
      "step": 52235
    },
    {
      "entropy": 5.890233373641967,
      "epoch": 4.738751814223512,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029681274173291817,
      "loss": 5.071,
      "mean_token_accuracy": 0.21142694652080535,
      "num_tokens": 108433408.0,
      "step": 52240
    },
    {
      "entropy": 5.737804651260376,
      "epoch": 4.739205370101597,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029678053749645583,
      "loss": 4.801,
      "mean_token_accuracy": 0.23129211068153382,
      "num_tokens": 108443783.0,
      "step": 52245
    },
    {
      "entropy": 5.613408136367798,
      "epoch": 4.739658925979681,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002967483328095665,
      "loss": 4.7505,
      "mean_token_accuracy": 0.23271640837192537,
      "num_tokens": 108454770.0,
      "step": 52250
    },
    {
      "entropy": 5.694039916992187,
      "epoch": 4.740112481857765,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002967161276729161,
      "loss": 4.8489,
      "mean_token_accuracy": 0.23269549906253814,
      "num_tokens": 108464812.0,
      "step": 52255
    },
    {
      "entropy": 5.775166845321655,
      "epoch": 4.740566037735849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002966839220871707,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22233758568763734,
      "num_tokens": 108474642.0,
      "step": 52260
    },
    {
      "entropy": 5.6916748046875,
      "epoch": 4.741019593613933,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002966517160529962,
      "loss": 4.8404,
      "mean_token_accuracy": 0.2297011747956276,
      "num_tokens": 108484143.0,
      "step": 52265
    },
    {
      "entropy": 5.795123434066772,
      "epoch": 4.741473149492017,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002966195095710587,
      "loss": 4.9976,
      "mean_token_accuracy": 0.22040179818868638,
      "num_tokens": 108494850.0,
      "step": 52270
    },
    {
      "entropy": 5.767268562316895,
      "epoch": 4.741926705370101,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002965873026420242,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21693125516176223,
      "num_tokens": 108505441.0,
      "step": 52275
    },
    {
      "entropy": 5.8028419494628904,
      "epoch": 4.742380261248186,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029655509526655885,
      "loss": 4.9599,
      "mean_token_accuracy": 0.21425091475248337,
      "num_tokens": 108517350.0,
      "step": 52280
    },
    {
      "entropy": 5.703174591064453,
      "epoch": 4.74283381712627,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029652288744532856,
      "loss": 4.8443,
      "mean_token_accuracy": 0.22947938740253448,
      "num_tokens": 108528472.0,
      "step": 52285
    },
    {
      "entropy": 5.747810554504395,
      "epoch": 4.743287373004354,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002964906791789995,
      "loss": 4.9618,
      "mean_token_accuracy": 0.21974271535873413,
      "num_tokens": 108539834.0,
      "step": 52290
    },
    {
      "entropy": 5.797972297668457,
      "epoch": 4.743740928882438,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002964584704682377,
      "loss": 4.9681,
      "mean_token_accuracy": 0.22473911345005035,
      "num_tokens": 108551478.0,
      "step": 52295
    },
    {
      "entropy": 5.660841417312622,
      "epoch": 4.744194484760523,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029642626131370936,
      "loss": 4.7397,
      "mean_token_accuracy": 0.24317233264446259,
      "num_tokens": 108560957.0,
      "step": 52300
    },
    {
      "entropy": 5.75013484954834,
      "epoch": 4.744648040638607,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002963940517160804,
      "loss": 4.9138,
      "mean_token_accuracy": 0.2233869895339012,
      "num_tokens": 108571146.0,
      "step": 52305
    },
    {
      "entropy": 5.742275857925415,
      "epoch": 4.745101596516691,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029636184167601695,
      "loss": 4.846,
      "mean_token_accuracy": 0.22846123874187468,
      "num_tokens": 108581085.0,
      "step": 52310
    },
    {
      "entropy": 5.710595417022705,
      "epoch": 4.745555152394775,
      "grad_norm": 1.53125,
      "learning_rate": 0.00029632963119418526,
      "loss": 4.8369,
      "mean_token_accuracy": 0.2280794844031334,
      "num_tokens": 108591103.0,
      "step": 52315
    },
    {
      "entropy": 5.7569968700408936,
      "epoch": 4.746008708272859,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029629742027125134,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22824612706899644,
      "num_tokens": 108601433.0,
      "step": 52320
    },
    {
      "entropy": 5.728277778625488,
      "epoch": 4.746462264150943,
      "grad_norm": 1.453125,
      "learning_rate": 0.00029626520890788127,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22648436427116395,
      "num_tokens": 108611728.0,
      "step": 52325
    },
    {
      "entropy": 5.754777097702027,
      "epoch": 4.746915820029027,
      "grad_norm": 1.375,
      "learning_rate": 0.00029623299710474134,
      "loss": 4.8834,
      "mean_token_accuracy": 0.22566175311803818,
      "num_tokens": 108621521.0,
      "step": 52330
    },
    {
      "entropy": 5.728661346435547,
      "epoch": 4.747369375907112,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029620078486249763,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22050722390413285,
      "num_tokens": 108632921.0,
      "step": 52335
    },
    {
      "entropy": 5.742632675170898,
      "epoch": 4.747822931785196,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002961685721818163,
      "loss": 4.8849,
      "mean_token_accuracy": 0.23169320672750474,
      "num_tokens": 108643244.0,
      "step": 52340
    },
    {
      "entropy": 5.711321783065796,
      "epoch": 4.74827648766328,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002961363590633634,
      "loss": 4.8617,
      "mean_token_accuracy": 0.23022657334804536,
      "num_tokens": 108653036.0,
      "step": 52345
    },
    {
      "entropy": 5.79217758178711,
      "epoch": 4.748730043541364,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002961041455078053,
      "loss": 4.9156,
      "mean_token_accuracy": 0.2285692423582077,
      "num_tokens": 108663317.0,
      "step": 52350
    },
    {
      "entropy": 5.717902898788452,
      "epoch": 4.749183599419449,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002960719315158081,
      "loss": 4.821,
      "mean_token_accuracy": 0.22509679943323135,
      "num_tokens": 108673657.0,
      "step": 52355
    },
    {
      "entropy": 5.718699836730957,
      "epoch": 4.749637155297533,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002960397170880378,
      "loss": 4.846,
      "mean_token_accuracy": 0.22199005484580994,
      "num_tokens": 108683831.0,
      "step": 52360
    },
    {
      "entropy": 5.708472824096679,
      "epoch": 4.750090711175617,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029600750222516097,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2232380375266075,
      "num_tokens": 108694062.0,
      "step": 52365
    },
    {
      "entropy": 5.7141705513000485,
      "epoch": 4.7505442670537015,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029597528692784365,
      "loss": 4.8574,
      "mean_token_accuracy": 0.2315657466650009,
      "num_tokens": 108704712.0,
      "step": 52370
    },
    {
      "entropy": 5.723170328140259,
      "epoch": 4.750997822931785,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029594307119675195,
      "loss": 4.8353,
      "mean_token_accuracy": 0.22942208498716354,
      "num_tokens": 108714741.0,
      "step": 52375
    },
    {
      "entropy": 5.78348708152771,
      "epoch": 4.751451378809869,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002959108550325523,
      "loss": 4.9543,
      "mean_token_accuracy": 0.22819010615348817,
      "num_tokens": 108724860.0,
      "step": 52380
    },
    {
      "entropy": 5.706610822677613,
      "epoch": 4.751904934687953,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029587863843591076,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2414358675479889,
      "num_tokens": 108735527.0,
      "step": 52385
    },
    {
      "entropy": 5.7310089588165285,
      "epoch": 4.752358490566038,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029584642140749375,
      "loss": 4.9335,
      "mean_token_accuracy": 0.21850105226039887,
      "num_tokens": 108745748.0,
      "step": 52390
    },
    {
      "entropy": 5.740648365020752,
      "epoch": 4.752812046444122,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029581420394796733,
      "loss": 4.9023,
      "mean_token_accuracy": 0.22876232564449311,
      "num_tokens": 108756635.0,
      "step": 52395
    },
    {
      "entropy": 5.7120973587036135,
      "epoch": 4.753265602322206,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029578198605799796,
      "loss": 4.8139,
      "mean_token_accuracy": 0.23021832555532457,
      "num_tokens": 108765912.0,
      "step": 52400
    },
    {
      "entropy": 5.756437206268311,
      "epoch": 4.7537191582002905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029574976773825176,
      "loss": 4.8506,
      "mean_token_accuracy": 0.22535753697156907,
      "num_tokens": 108775831.0,
      "step": 52405
    },
    {
      "entropy": 5.735830020904541,
      "epoch": 4.754172714078375,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000295717548989395,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22719539999961852,
      "num_tokens": 108785937.0,
      "step": 52410
    },
    {
      "entropy": 5.718538236618042,
      "epoch": 4.754626269956459,
      "grad_norm": 1.375,
      "learning_rate": 0.0002956853298120942,
      "loss": 4.8507,
      "mean_token_accuracy": 0.22875372916460038,
      "num_tokens": 108796289.0,
      "step": 52415
    },
    {
      "entropy": 5.761146211624146,
      "epoch": 4.755079825834542,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029565311020701544,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22345309853553771,
      "num_tokens": 108806040.0,
      "step": 52420
    },
    {
      "entropy": 5.766274929046631,
      "epoch": 4.755533381712627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029562089017482515,
      "loss": 4.94,
      "mean_token_accuracy": 0.22644669860601424,
      "num_tokens": 108815440.0,
      "step": 52425
    },
    {
      "entropy": 5.789774084091187,
      "epoch": 4.755986937590711,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002955886697161895,
      "loss": 4.9442,
      "mean_token_accuracy": 0.2199125587940216,
      "num_tokens": 108826059.0,
      "step": 52430
    },
    {
      "entropy": 5.7417079448699955,
      "epoch": 4.756440493468795,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029555644883177496,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2301152914762497,
      "num_tokens": 108836373.0,
      "step": 52435
    },
    {
      "entropy": 5.662623882293701,
      "epoch": 4.756894049346879,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029552422752224787,
      "loss": 4.8042,
      "mean_token_accuracy": 0.2371852919459343,
      "num_tokens": 108847364.0,
      "step": 52440
    },
    {
      "entropy": 5.702552747726441,
      "epoch": 4.757347605224964,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029549200578827444,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22611523270606995,
      "num_tokens": 108856893.0,
      "step": 52445
    },
    {
      "entropy": 5.772364950180053,
      "epoch": 4.757801161103048,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002954597836305212,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23520751744508744,
      "num_tokens": 108867218.0,
      "step": 52450
    },
    {
      "entropy": 5.729339122772217,
      "epoch": 4.758254716981132,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029542756104965434,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2261333644390106,
      "num_tokens": 108876938.0,
      "step": 52455
    },
    {
      "entropy": 5.742070436477661,
      "epoch": 4.7587082728592165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029539533804634027,
      "loss": 4.8783,
      "mean_token_accuracy": 0.22794810086488723,
      "num_tokens": 108887831.0,
      "step": 52460
    },
    {
      "entropy": 5.707121181488037,
      "epoch": 4.759161828737301,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029536311462124553,
      "loss": 4.8393,
      "mean_token_accuracy": 0.23405114263296128,
      "num_tokens": 108897634.0,
      "step": 52465
    },
    {
      "entropy": 5.692772722244262,
      "epoch": 4.759615384615385,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002953308907750363,
      "loss": 4.8203,
      "mean_token_accuracy": 0.2269541010260582,
      "num_tokens": 108907976.0,
      "step": 52470
    },
    {
      "entropy": 5.705157279968262,
      "epoch": 4.760068940493468,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002952986665083792,
      "loss": 4.8791,
      "mean_token_accuracy": 0.22447418570518493,
      "num_tokens": 108918755.0,
      "step": 52475
    },
    {
      "entropy": 5.739822149276733,
      "epoch": 4.760522496371553,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002952664418219404,
      "loss": 4.8754,
      "mean_token_accuracy": 0.2341247245669365,
      "num_tokens": 108928349.0,
      "step": 52480
    },
    {
      "entropy": 5.730664777755737,
      "epoch": 4.760976052249637,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029523421671638644,
      "loss": 4.9292,
      "mean_token_accuracy": 0.21494185477495192,
      "num_tokens": 108938429.0,
      "step": 52485
    },
    {
      "entropy": 5.733451461791992,
      "epoch": 4.761429608127721,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002952019911923837,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22621866911649705,
      "num_tokens": 108948378.0,
      "step": 52490
    },
    {
      "entropy": 5.667526769638061,
      "epoch": 4.7618831640058055,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029516976525059864,
      "loss": 4.801,
      "mean_token_accuracy": 0.2310488760471344,
      "num_tokens": 108958667.0,
      "step": 52495
    },
    {
      "entropy": 5.808361339569092,
      "epoch": 4.76233671988389,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002951375388916977,
      "loss": 4.9443,
      "mean_token_accuracy": 0.21914156675338745,
      "num_tokens": 108968955.0,
      "step": 52500
    },
    {
      "entropy": 5.764720296859741,
      "epoch": 4.762790275761974,
      "grad_norm": 1.375,
      "learning_rate": 0.0002951053121163474,
      "loss": 4.8322,
      "mean_token_accuracy": 0.2263290673494339,
      "num_tokens": 108979565.0,
      "step": 52505
    },
    {
      "entropy": 5.744795417785644,
      "epoch": 4.763243831640058,
      "grad_norm": 1.3125,
      "learning_rate": 0.000295073084925214,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22735028713941574,
      "num_tokens": 108989547.0,
      "step": 52510
    },
    {
      "entropy": 5.707319974899292,
      "epoch": 4.7636973875181425,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029504085731896413,
      "loss": 4.8601,
      "mean_token_accuracy": 0.2308280050754547,
      "num_tokens": 109000286.0,
      "step": 52515
    },
    {
      "entropy": 5.762074422836304,
      "epoch": 4.764150943396227,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002950086292982642,
      "loss": 4.8268,
      "mean_token_accuracy": 0.2346460625529289,
      "num_tokens": 109010676.0,
      "step": 52520
    },
    {
      "entropy": 5.8159809589385985,
      "epoch": 4.764604499274311,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029497640086378077,
      "loss": 4.9423,
      "mean_token_accuracy": 0.21900980025529862,
      "num_tokens": 109020800.0,
      "step": 52525
    },
    {
      "entropy": 5.744914579391479,
      "epoch": 4.7650580551523944,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029494417201618027,
      "loss": 4.9805,
      "mean_token_accuracy": 0.22348010390996934,
      "num_tokens": 109031158.0,
      "step": 52530
    },
    {
      "entropy": 5.708102893829346,
      "epoch": 4.765511611030479,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002949119427561292,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23181158900260926,
      "num_tokens": 109041664.0,
      "step": 52535
    },
    {
      "entropy": 5.724632835388183,
      "epoch": 4.765965166908563,
      "grad_norm": 1.375,
      "learning_rate": 0.00029487971308429415,
      "loss": 4.8819,
      "mean_token_accuracy": 0.23153376579284668,
      "num_tokens": 109052633.0,
      "step": 52540
    },
    {
      "entropy": 5.767268562316895,
      "epoch": 4.766418722786647,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002948474830013414,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2277809426188469,
      "num_tokens": 109062687.0,
      "step": 52545
    },
    {
      "entropy": 5.809348249435425,
      "epoch": 4.7668722786647315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029481525250793777,
      "loss": 4.9149,
      "mean_token_accuracy": 0.23216542303562165,
      "num_tokens": 109073348.0,
      "step": 52550
    },
    {
      "entropy": 5.695405340194702,
      "epoch": 4.767325834542816,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002947830216047497,
      "loss": 4.8877,
      "mean_token_accuracy": 0.21938156932592393,
      "num_tokens": 109083276.0,
      "step": 52555
    },
    {
      "entropy": 5.686476945877075,
      "epoch": 4.7677793904209,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029475079029244356,
      "loss": 4.8799,
      "mean_token_accuracy": 0.22765453159809113,
      "num_tokens": 109094342.0,
      "step": 52560
    },
    {
      "entropy": 5.728580522537231,
      "epoch": 4.768232946298984,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029471855857168616,
      "loss": 4.9095,
      "mean_token_accuracy": 0.2347014605998993,
      "num_tokens": 109105178.0,
      "step": 52565
    },
    {
      "entropy": 5.81139645576477,
      "epoch": 4.768686502177069,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002946863264431439,
      "loss": 4.9756,
      "mean_token_accuracy": 0.2202063262462616,
      "num_tokens": 109116205.0,
      "step": 52570
    },
    {
      "entropy": 5.738919115066528,
      "epoch": 4.769140058055152,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002946540939074834,
      "loss": 4.8565,
      "mean_token_accuracy": 0.22417071163654329,
      "num_tokens": 109126299.0,
      "step": 52575
    },
    {
      "entropy": 5.757051706314087,
      "epoch": 4.769593613933236,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029462186096537116,
      "loss": 4.9614,
      "mean_token_accuracy": 0.22575725615024567,
      "num_tokens": 109137675.0,
      "step": 52580
    },
    {
      "entropy": 5.733745145797729,
      "epoch": 4.7700471698113205,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029458962761747394,
      "loss": 4.8711,
      "mean_token_accuracy": 0.22737185806035995,
      "num_tokens": 109147716.0,
      "step": 52585
    },
    {
      "entropy": 5.744874429702759,
      "epoch": 4.770500725689405,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002945573938644582,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22340999245643617,
      "num_tokens": 109159211.0,
      "step": 52590
    },
    {
      "entropy": 5.734386730194092,
      "epoch": 4.770954281567489,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029452515970699054,
      "loss": 4.9294,
      "mean_token_accuracy": 0.22197561115026473,
      "num_tokens": 109169490.0,
      "step": 52595
    },
    {
      "entropy": 5.7923320770263675,
      "epoch": 4.771407837445573,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029449292514573767,
      "loss": 4.9543,
      "mean_token_accuracy": 0.22022728621959686,
      "num_tokens": 109179687.0,
      "step": 52600
    },
    {
      "entropy": 5.74944372177124,
      "epoch": 4.7718613933236576,
      "grad_norm": 1.5625,
      "learning_rate": 0.00029446069018136613,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22369553446769713,
      "num_tokens": 109189415.0,
      "step": 52605
    },
    {
      "entropy": 5.739500713348389,
      "epoch": 4.772314949201742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002944284548145424,
      "loss": 4.7674,
      "mean_token_accuracy": 0.23019570261240005,
      "num_tokens": 109199281.0,
      "step": 52610
    },
    {
      "entropy": 5.718282461166382,
      "epoch": 4.772768505079826,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002943962190459335,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22490357607603073,
      "num_tokens": 109208347.0,
      "step": 52615
    },
    {
      "entropy": 5.690493154525757,
      "epoch": 4.77322206095791,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029436398287620573,
      "loss": 4.8666,
      "mean_token_accuracy": 0.224729260802269,
      "num_tokens": 109218026.0,
      "step": 52620
    },
    {
      "entropy": 5.676104640960693,
      "epoch": 4.773675616835995,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002943317463060259,
      "loss": 4.8491,
      "mean_token_accuracy": 0.22595081627368926,
      "num_tokens": 109229457.0,
      "step": 52625
    },
    {
      "entropy": 5.722496891021729,
      "epoch": 4.774129172714078,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002942995093360606,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23145319670438766,
      "num_tokens": 109239504.0,
      "step": 52630
    },
    {
      "entropy": 5.64672474861145,
      "epoch": 4.774582728592162,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002942672719669765,
      "loss": 4.7583,
      "mean_token_accuracy": 0.23258443921804428,
      "num_tokens": 109249745.0,
      "step": 52635
    },
    {
      "entropy": 5.692960500717163,
      "epoch": 4.7750362844702465,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002942350341994404,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22905230522155762,
      "num_tokens": 109259767.0,
      "step": 52640
    },
    {
      "entropy": 5.855855226516724,
      "epoch": 4.775489840348331,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029420279603411875,
      "loss": 5.0422,
      "mean_token_accuracy": 0.2175777167081833,
      "num_tokens": 109269611.0,
      "step": 52645
    },
    {
      "entropy": 5.845707893371582,
      "epoch": 4.775943396226415,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002941705574716785,
      "loss": 4.9751,
      "mean_token_accuracy": 0.21989833563566208,
      "num_tokens": 109281396.0,
      "step": 52650
    },
    {
      "entropy": 5.783546066284179,
      "epoch": 4.776396952104499,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002941383185127862,
      "loss": 4.8397,
      "mean_token_accuracy": 0.22753327637910842,
      "num_tokens": 109292346.0,
      "step": 52655
    },
    {
      "entropy": 5.8295070171356205,
      "epoch": 4.776850507982584,
      "grad_norm": 1.515625,
      "learning_rate": 0.00029410607915810854,
      "loss": 4.9384,
      "mean_token_accuracy": 0.22112050652503967,
      "num_tokens": 109302966.0,
      "step": 52660
    },
    {
      "entropy": 5.66844072341919,
      "epoch": 4.777304063860668,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002940738394083124,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2288457915186882,
      "num_tokens": 109312558.0,
      "step": 52665
    },
    {
      "entropy": 5.76115345954895,
      "epoch": 4.777757619738752,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029404159926406436,
      "loss": 4.9404,
      "mean_token_accuracy": 0.22699193358421327,
      "num_tokens": 109322364.0,
      "step": 52670
    },
    {
      "entropy": 5.825018119812012,
      "epoch": 4.7782111756168355,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002940093587260312,
      "loss": 4.949,
      "mean_token_accuracy": 0.22610754072666167,
      "num_tokens": 109332449.0,
      "step": 52675
    },
    {
      "entropy": 5.796683073043823,
      "epoch": 4.778664731494921,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029397711779487954,
      "loss": 4.8505,
      "mean_token_accuracy": 0.22866918295621871,
      "num_tokens": 109343262.0,
      "step": 52680
    },
    {
      "entropy": 5.740981578826904,
      "epoch": 4.779118287373004,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029394487647127645,
      "loss": 4.8355,
      "mean_token_accuracy": 0.2261508047580719,
      "num_tokens": 109352951.0,
      "step": 52685
    },
    {
      "entropy": 5.674746561050415,
      "epoch": 4.779571843251088,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00029391263475588844,
      "loss": 4.9531,
      "mean_token_accuracy": 0.22419815957546235,
      "num_tokens": 109363913.0,
      "step": 52690
    },
    {
      "entropy": 5.689483737945556,
      "epoch": 4.780025399129173,
      "grad_norm": 1.375,
      "learning_rate": 0.00029388039264938224,
      "loss": 4.8623,
      "mean_token_accuracy": 0.22807313948869706,
      "num_tokens": 109374405.0,
      "step": 52695
    },
    {
      "entropy": 5.791430187225342,
      "epoch": 4.780478955007257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002938481501524247,
      "loss": 4.9316,
      "mean_token_accuracy": 0.22100772112607955,
      "num_tokens": 109385696.0,
      "step": 52700
    },
    {
      "entropy": 5.753338623046875,
      "epoch": 4.780932510885341,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029381590726568267,
      "loss": 4.904,
      "mean_token_accuracy": 0.22558448314666749,
      "num_tokens": 109395061.0,
      "step": 52705
    },
    {
      "entropy": 5.757851791381836,
      "epoch": 4.781386066763425,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029378366398982286,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2237846225500107,
      "num_tokens": 109406152.0,
      "step": 52710
    },
    {
      "entropy": 5.81395320892334,
      "epoch": 4.78183962264151,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029375142032551216,
      "loss": 4.9334,
      "mean_token_accuracy": 0.22286365628242494,
      "num_tokens": 109416596.0,
      "step": 52715
    },
    {
      "entropy": 5.84960618019104,
      "epoch": 4.782293178519594,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029371917627341726,
      "loss": 4.9428,
      "mean_token_accuracy": 0.222938472032547,
      "num_tokens": 109426169.0,
      "step": 52720
    },
    {
      "entropy": 5.806995058059693,
      "epoch": 4.782746734397678,
      "grad_norm": 1.390625,
      "learning_rate": 0.000293686931834205,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2283649444580078,
      "num_tokens": 109436359.0,
      "step": 52725
    },
    {
      "entropy": 5.755242252349854,
      "epoch": 4.7832002902757615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002936546870085422,
      "loss": 4.8879,
      "mean_token_accuracy": 0.2239805206656456,
      "num_tokens": 109446050.0,
      "step": 52730
    },
    {
      "entropy": 5.762327432632446,
      "epoch": 4.783653846153846,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029362244179709576,
      "loss": 4.919,
      "mean_token_accuracy": 0.2306077793240547,
      "num_tokens": 109457124.0,
      "step": 52735
    },
    {
      "entropy": 5.677143383026123,
      "epoch": 4.78410740203193,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029359019620053253,
      "loss": 4.7957,
      "mean_token_accuracy": 0.23078289180994033,
      "num_tokens": 109467642.0,
      "step": 52740
    },
    {
      "entropy": 5.749522924423218,
      "epoch": 4.784560957910014,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002935579502195192,
      "loss": 4.9204,
      "mean_token_accuracy": 0.2186886578798294,
      "num_tokens": 109477877.0,
      "step": 52745
    },
    {
      "entropy": 5.755407094955444,
      "epoch": 4.785014513788099,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029352570385472276,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22464905232191085,
      "num_tokens": 109489302.0,
      "step": 52750
    },
    {
      "entropy": 5.740735626220703,
      "epoch": 4.785468069666183,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002934934571068101,
      "loss": 4.8655,
      "mean_token_accuracy": 0.22303910851478576,
      "num_tokens": 109500023.0,
      "step": 52755
    },
    {
      "entropy": 5.7222541809082035,
      "epoch": 4.785921625544267,
      "grad_norm": 1.296875,
      "learning_rate": 0.000293461209976448,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2247626468539238,
      "num_tokens": 109510588.0,
      "step": 52760
    },
    {
      "entropy": 5.7856663227081295,
      "epoch": 4.786375181422351,
      "grad_norm": 1.5,
      "learning_rate": 0.0002934289624643034,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2211928114295006,
      "num_tokens": 109520162.0,
      "step": 52765
    },
    {
      "entropy": 5.756727313995361,
      "epoch": 4.786828737300436,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002933967145710431,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22797905653715134,
      "num_tokens": 109530155.0,
      "step": 52770
    },
    {
      "entropy": 5.7529150485992435,
      "epoch": 4.78728229317852,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002933644662973341,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22876431196928024,
      "num_tokens": 109539897.0,
      "step": 52775
    },
    {
      "entropy": 5.717057847976685,
      "epoch": 4.787735849056604,
      "grad_norm": 1.390625,
      "learning_rate": 0.00029333221764384326,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22598471343517304,
      "num_tokens": 109549995.0,
      "step": 52780
    },
    {
      "entropy": 5.7824548244476315,
      "epoch": 4.788189404934688,
      "grad_norm": 1.25,
      "learning_rate": 0.00029329996861123745,
      "loss": 4.9003,
      "mean_token_accuracy": 0.22696855068206787,
      "num_tokens": 109560446.0,
      "step": 52785
    },
    {
      "entropy": 5.681630277633667,
      "epoch": 4.788642960812772,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029326771920018367,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22900890558958054,
      "num_tokens": 109570864.0,
      "step": 52790
    },
    {
      "entropy": 5.795296907424927,
      "epoch": 4.789096516690856,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00029323546941134875,
      "loss": 4.9195,
      "mean_token_accuracy": 0.22732480764389038,
      "num_tokens": 109581015.0,
      "step": 52795
    },
    {
      "entropy": 5.730304431915283,
      "epoch": 4.78955007256894,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029320321924539974,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2248670294880867,
      "num_tokens": 109591695.0,
      "step": 52800
    },
    {
      "entropy": 5.693893480300903,
      "epoch": 4.790003628447025,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002931709687030035,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2334986075758934,
      "num_tokens": 109602274.0,
      "step": 52805
    },
    {
      "entropy": 5.763409233093261,
      "epoch": 4.790457184325109,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029313871778482693,
      "loss": 4.8958,
      "mean_token_accuracy": 0.23191093802452087,
      "num_tokens": 109612995.0,
      "step": 52810
    },
    {
      "entropy": 5.7470869541168215,
      "epoch": 4.790910740203193,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002931064664915371,
      "loss": 4.8479,
      "mean_token_accuracy": 0.22544656544923783,
      "num_tokens": 109623133.0,
      "step": 52815
    },
    {
      "entropy": 5.6541016578674315,
      "epoch": 4.7913642960812775,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000293074214823801,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23047229945659636,
      "num_tokens": 109634312.0,
      "step": 52820
    },
    {
      "entropy": 5.749652671813965,
      "epoch": 4.791817851959362,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00029304196278228544,
      "loss": 4.9018,
      "mean_token_accuracy": 0.21594303995370864,
      "num_tokens": 109644333.0,
      "step": 52825
    },
    {
      "entropy": 5.698733234405518,
      "epoch": 4.792271407837445,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002930097103676575,
      "loss": 4.8011,
      "mean_token_accuracy": 0.2298659697175026,
      "num_tokens": 109654730.0,
      "step": 52830
    },
    {
      "entropy": 5.685696172714233,
      "epoch": 4.79272496371553,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029297745758058414,
      "loss": 4.8388,
      "mean_token_accuracy": 0.22888888269662858,
      "num_tokens": 109664867.0,
      "step": 52835
    },
    {
      "entropy": 5.784062814712525,
      "epoch": 4.793178519593614,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029294520442173244,
      "loss": 4.9271,
      "mean_token_accuracy": 0.2266416981816292,
      "num_tokens": 109675090.0,
      "step": 52840
    },
    {
      "entropy": 5.699233436584473,
      "epoch": 4.793632075471698,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002929129508917693,
      "loss": 4.8209,
      "mean_token_accuracy": 0.2324760690331459,
      "num_tokens": 109685046.0,
      "step": 52845
    },
    {
      "entropy": 5.801972484588623,
      "epoch": 4.794085631349782,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002928806969913618,
      "loss": 4.9495,
      "mean_token_accuracy": 0.2200348496437073,
      "num_tokens": 109696240.0,
      "step": 52850
    },
    {
      "entropy": 5.7189356803894045,
      "epoch": 4.794539187227866,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002928484427211769,
      "loss": 4.8826,
      "mean_token_accuracy": 0.22079290449619293,
      "num_tokens": 109707676.0,
      "step": 52855
    },
    {
      "entropy": 5.831326723098755,
      "epoch": 4.794992743105951,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002928161880818817,
      "loss": 4.9552,
      "mean_token_accuracy": 0.22163426280021667,
      "num_tokens": 109718168.0,
      "step": 52860
    },
    {
      "entropy": 5.727166271209716,
      "epoch": 4.795446298984035,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002927839330741431,
      "loss": 4.8335,
      "mean_token_accuracy": 0.22625656723976134,
      "num_tokens": 109728624.0,
      "step": 52865
    },
    {
      "entropy": 5.820116424560547,
      "epoch": 4.795899854862119,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00029275167769862834,
      "loss": 4.9754,
      "mean_token_accuracy": 0.21081093549728394,
      "num_tokens": 109738301.0,
      "step": 52870
    },
    {
      "entropy": 5.771745824813843,
      "epoch": 4.7963534107402035,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002927194219560043,
      "loss": 4.9734,
      "mean_token_accuracy": 0.21763087064027786,
      "num_tokens": 109749269.0,
      "step": 52875
    },
    {
      "entropy": 5.712932968139649,
      "epoch": 4.796806966618288,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029268716584693804,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22540998011827468,
      "num_tokens": 109758335.0,
      "step": 52880
    },
    {
      "entropy": 5.7657026767730715,
      "epoch": 4.797260522496371,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002926549093720968,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23037101775407792,
      "num_tokens": 109768848.0,
      "step": 52885
    },
    {
      "entropy": 5.695959949493409,
      "epoch": 4.797714078374455,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029262265253214754,
      "loss": 4.8744,
      "mean_token_accuracy": 0.2238750159740448,
      "num_tokens": 109779331.0,
      "step": 52890
    },
    {
      "entropy": 5.739331579208374,
      "epoch": 4.79816763425254,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029259039532775725,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22899897396564484,
      "num_tokens": 109789331.0,
      "step": 52895
    },
    {
      "entropy": 5.76227650642395,
      "epoch": 4.798621190130624,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002925581377595931,
      "loss": 4.8507,
      "mean_token_accuracy": 0.229796801507473,
      "num_tokens": 109798630.0,
      "step": 52900
    },
    {
      "entropy": 5.730231714248657,
      "epoch": 4.799074746008708,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029252587982832225,
      "loss": 4.8742,
      "mean_token_accuracy": 0.2324410781264305,
      "num_tokens": 109808501.0,
      "step": 52905
    },
    {
      "entropy": 5.747606229782105,
      "epoch": 4.7995283018867925,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002924936215346117,
      "loss": 4.8716,
      "mean_token_accuracy": 0.23293943107128143,
      "num_tokens": 109818847.0,
      "step": 52910
    },
    {
      "entropy": 5.759906816482544,
      "epoch": 4.799981857764877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029246136287912865,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22362373918294906,
      "num_tokens": 109830017.0,
      "step": 52915
    },
    {
      "entropy": 5.722891998291016,
      "epoch": 4.800435413642961,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002924291038625402,
      "loss": 4.8836,
      "mean_token_accuracy": 0.22362721860408782,
      "num_tokens": 109839795.0,
      "step": 52920
    },
    {
      "entropy": 5.7252435207366945,
      "epoch": 4.800888969521045,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029239684448551336,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23332851231098176,
      "num_tokens": 109850202.0,
      "step": 52925
    },
    {
      "entropy": 5.6994056224823,
      "epoch": 4.8013425253991295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002923645847487153,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2278398334980011,
      "num_tokens": 109860625.0,
      "step": 52930
    },
    {
      "entropy": 5.747838401794434,
      "epoch": 4.801796081277214,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002923323246528133,
      "loss": 4.8608,
      "mean_token_accuracy": 0.23380349725484847,
      "num_tokens": 109871241.0,
      "step": 52935
    },
    {
      "entropy": 5.7823169231414795,
      "epoch": 4.802249637155297,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00029230006419847434,
      "loss": 4.9224,
      "mean_token_accuracy": 0.22220810800790786,
      "num_tokens": 109881627.0,
      "step": 52940
    },
    {
      "entropy": 5.763097906112671,
      "epoch": 4.8027031930333814,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002922678033863656,
      "loss": 4.9306,
      "mean_token_accuracy": 0.227706441283226,
      "num_tokens": 109893247.0,
      "step": 52945
    },
    {
      "entropy": 5.727707624435425,
      "epoch": 4.803156748911466,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002922355422171544,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2232788935303688,
      "num_tokens": 109904112.0,
      "step": 52950
    },
    {
      "entropy": 5.704008626937866,
      "epoch": 4.80361030478955,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029220328069150774,
      "loss": 4.7738,
      "mean_token_accuracy": 0.23470181077718735,
      "num_tokens": 109913854.0,
      "step": 52955
    },
    {
      "entropy": 5.811523723602295,
      "epoch": 4.804063860667634,
      "grad_norm": 1.46875,
      "learning_rate": 0.00029217101881009284,
      "loss": 4.935,
      "mean_token_accuracy": 0.2202719733119011,
      "num_tokens": 109923947.0,
      "step": 52960
    },
    {
      "entropy": 5.75461893081665,
      "epoch": 4.8045174165457185,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002921387565735769,
      "loss": 4.864,
      "mean_token_accuracy": 0.2268332675099373,
      "num_tokens": 109934590.0,
      "step": 52965
    },
    {
      "entropy": 5.732774686813355,
      "epoch": 4.804970972423803,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002921064939826272,
      "loss": 4.8954,
      "mean_token_accuracy": 0.2243231028318405,
      "num_tokens": 109944224.0,
      "step": 52970
    },
    {
      "entropy": 5.725294542312622,
      "epoch": 4.805424528301887,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002920742310379108,
      "loss": 4.9269,
      "mean_token_accuracy": 0.22499562352895736,
      "num_tokens": 109954414.0,
      "step": 52975
    },
    {
      "entropy": 5.771968841552734,
      "epoch": 4.805878084179971,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029204196774009476,
      "loss": 4.9131,
      "mean_token_accuracy": 0.22892777770757675,
      "num_tokens": 109964304.0,
      "step": 52980
    },
    {
      "entropy": 5.791261863708496,
      "epoch": 4.806331640058055,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029200970408984673,
      "loss": 4.9892,
      "mean_token_accuracy": 0.21095959097146988,
      "num_tokens": 109975401.0,
      "step": 52985
    },
    {
      "entropy": 5.808906936645508,
      "epoch": 4.806785195936139,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002919774400878336,
      "loss": 4.9327,
      "mean_token_accuracy": 0.2249539837241173,
      "num_tokens": 109985964.0,
      "step": 52990
    },
    {
      "entropy": 5.808730792999268,
      "epoch": 4.807238751814223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002919451757347226,
      "loss": 4.9097,
      "mean_token_accuracy": 0.229017473757267,
      "num_tokens": 109995476.0,
      "step": 52995
    },
    {
      "entropy": 5.765976667404175,
      "epoch": 4.8076923076923075,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002919129110311812,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2249477580189705,
      "num_tokens": 110005238.0,
      "step": 53000
    },
    {
      "entropy": 5.738067293167115,
      "epoch": 4.808145863570392,
      "grad_norm": 1.375,
      "learning_rate": 0.0002918806459778764,
      "loss": 4.9649,
      "mean_token_accuracy": 0.21014612764120102,
      "num_tokens": 110016120.0,
      "step": 53005
    },
    {
      "entropy": 5.736339426040649,
      "epoch": 4.808599419448476,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002918483805754756,
      "loss": 4.8715,
      "mean_token_accuracy": 0.22465234249830246,
      "num_tokens": 110026739.0,
      "step": 53010
    },
    {
      "entropy": 5.833478212356567,
      "epoch": 4.80905297532656,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002918161148246459,
      "loss": 4.9876,
      "mean_token_accuracy": 0.22070140093564988,
      "num_tokens": 110037491.0,
      "step": 53015
    },
    {
      "entropy": 5.782156658172608,
      "epoch": 4.8095065312046446,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029178384872605465,
      "loss": 4.906,
      "mean_token_accuracy": 0.2228652834892273,
      "num_tokens": 110048550.0,
      "step": 53020
    },
    {
      "entropy": 5.72168984413147,
      "epoch": 4.809960087082729,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029175158228036924,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22474730610847474,
      "num_tokens": 110058472.0,
      "step": 53025
    },
    {
      "entropy": 5.710295009613037,
      "epoch": 4.810413642960813,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002917193154882568,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23083682954311371,
      "num_tokens": 110070766.0,
      "step": 53030
    },
    {
      "entropy": 5.785801601409912,
      "epoch": 4.810867198838897,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002916870483503847,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22921282052993774,
      "num_tokens": 110081739.0,
      "step": 53035
    },
    {
      "entropy": 5.7617908954620365,
      "epoch": 4.811320754716981,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029165478086742006,
      "loss": 4.88,
      "mean_token_accuracy": 0.22448290884494781,
      "num_tokens": 110091075.0,
      "step": 53040
    },
    {
      "entropy": 5.738565587997437,
      "epoch": 4.811774310595065,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029162251304003044,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23143628090620041,
      "num_tokens": 110100530.0,
      "step": 53045
    },
    {
      "entropy": 5.68041615486145,
      "epoch": 4.812227866473149,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002915902448688829,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22707736045122145,
      "num_tokens": 110110866.0,
      "step": 53050
    },
    {
      "entropy": 5.696637439727783,
      "epoch": 4.8126814223512335,
      "grad_norm": 1.359375,
      "learning_rate": 0.00029155797635464493,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22970676571130752,
      "num_tokens": 110121344.0,
      "step": 53055
    },
    {
      "entropy": 5.68638801574707,
      "epoch": 4.813134978229318,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002915257074979838,
      "loss": 4.8119,
      "mean_token_accuracy": 0.22787152230739594,
      "num_tokens": 110130529.0,
      "step": 53060
    },
    {
      "entropy": 5.746345043182373,
      "epoch": 4.813588534107402,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002914934382995668,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22884232997894288,
      "num_tokens": 110141536.0,
      "step": 53065
    },
    {
      "entropy": 5.791234254837036,
      "epoch": 4.814042089985486,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002914611687600613,
      "loss": 4.9256,
      "mean_token_accuracy": 0.2279159367084503,
      "num_tokens": 110152364.0,
      "step": 53070
    },
    {
      "entropy": 5.73673825263977,
      "epoch": 4.814495645863571,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002914288988801347,
      "loss": 4.8913,
      "mean_token_accuracy": 0.22803279906511306,
      "num_tokens": 110163800.0,
      "step": 53075
    },
    {
      "entropy": 5.701155090332032,
      "epoch": 4.814949201741655,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002913966286604541,
      "loss": 4.8156,
      "mean_token_accuracy": 0.22749746441841126,
      "num_tokens": 110174590.0,
      "step": 53080
    },
    {
      "entropy": 5.7795202255249025,
      "epoch": 4.815402757619739,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029136435810168726,
      "loss": 4.9363,
      "mean_token_accuracy": 0.21999389976263045,
      "num_tokens": 110185142.0,
      "step": 53085
    },
    {
      "entropy": 5.682797050476074,
      "epoch": 4.815856313497823,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002913320872045012,
      "loss": 4.7503,
      "mean_token_accuracy": 0.24119022041559218,
      "num_tokens": 110194013.0,
      "step": 53090
    },
    {
      "entropy": 5.7149741649627686,
      "epoch": 4.816309869375907,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029129981596956353,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23965335935354232,
      "num_tokens": 110204087.0,
      "step": 53095
    },
    {
      "entropy": 5.777571439743042,
      "epoch": 4.816763425253991,
      "grad_norm": 1.375,
      "learning_rate": 0.00029126754439754137,
      "loss": 4.886,
      "mean_token_accuracy": 0.22528316527605058,
      "num_tokens": 110214066.0,
      "step": 53100
    },
    {
      "entropy": 5.739688348770142,
      "epoch": 4.817216981132075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002912352724891023,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23264508098363876,
      "num_tokens": 110224378.0,
      "step": 53105
    },
    {
      "entropy": 5.679113721847534,
      "epoch": 4.81767053701016,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002912030002449137,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2261831730604172,
      "num_tokens": 110235493.0,
      "step": 53110
    },
    {
      "entropy": 5.711249303817749,
      "epoch": 4.818124092888244,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029117072766564285,
      "loss": 4.9402,
      "mean_token_accuracy": 0.21717481911182404,
      "num_tokens": 110246953.0,
      "step": 53115
    },
    {
      "entropy": 5.7153482913970945,
      "epoch": 4.818577648766328,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002911384547519573,
      "loss": 4.7804,
      "mean_token_accuracy": 0.24153569787740709,
      "num_tokens": 110257387.0,
      "step": 53120
    },
    {
      "entropy": 5.72405514717102,
      "epoch": 4.819031204644412,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029110618150452437,
      "loss": 4.8996,
      "mean_token_accuracy": 0.22707040458917618,
      "num_tokens": 110268152.0,
      "step": 53125
    },
    {
      "entropy": 5.76214017868042,
      "epoch": 4.819484760522497,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029107390792401145,
      "loss": 4.876,
      "mean_token_accuracy": 0.2260974556207657,
      "num_tokens": 110277933.0,
      "step": 53130
    },
    {
      "entropy": 5.674251365661621,
      "epoch": 4.819938316400581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002910416340110861,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23342417925596237,
      "num_tokens": 110288219.0,
      "step": 53135
    },
    {
      "entropy": 5.715783357620239,
      "epoch": 4.820391872278664,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002910093597664156,
      "loss": 4.8921,
      "mean_token_accuracy": 0.23216582089662552,
      "num_tokens": 110299209.0,
      "step": 53140
    },
    {
      "entropy": 5.718314981460571,
      "epoch": 4.8208454281567485,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002909770851906675,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22980009019374847,
      "num_tokens": 110309482.0,
      "step": 53145
    },
    {
      "entropy": 5.749040985107422,
      "epoch": 4.821298984034833,
      "grad_norm": 1.328125,
      "learning_rate": 0.00029094481028450926,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2331783205270767,
      "num_tokens": 110319499.0,
      "step": 53150
    },
    {
      "entropy": 5.728221082687378,
      "epoch": 4.821752539912917,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002909125350486082,
      "loss": 4.7981,
      "mean_token_accuracy": 0.23289269506931304,
      "num_tokens": 110329453.0,
      "step": 53155
    },
    {
      "entropy": 5.6994212627410885,
      "epoch": 4.822206095791001,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002908802594836319,
      "loss": 4.922,
      "mean_token_accuracy": 0.22150215208530427,
      "num_tokens": 110339343.0,
      "step": 53160
    },
    {
      "entropy": 5.770641422271728,
      "epoch": 4.822659651669086,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002908479835902477,
      "loss": 4.9508,
      "mean_token_accuracy": 0.22537464201450347,
      "num_tokens": 110349101.0,
      "step": 53165
    },
    {
      "entropy": 5.695776796340942,
      "epoch": 4.82311320754717,
      "grad_norm": 1.375,
      "learning_rate": 0.0002908157073691233,
      "loss": 4.8386,
      "mean_token_accuracy": 0.2302941009402275,
      "num_tokens": 110359377.0,
      "step": 53170
    },
    {
      "entropy": 5.74040641784668,
      "epoch": 4.823566763425254,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029078343082092594,
      "loss": 4.8457,
      "mean_token_accuracy": 0.22884493619203566,
      "num_tokens": 110370132.0,
      "step": 53175
    },
    {
      "entropy": 5.718785667419434,
      "epoch": 4.824020319303338,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029075115394632325,
      "loss": 4.8608,
      "mean_token_accuracy": 0.21793520599603652,
      "num_tokens": 110381046.0,
      "step": 53180
    },
    {
      "entropy": 5.799324178695679,
      "epoch": 4.824473875181423,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029071887674598265,
      "loss": 4.9363,
      "mean_token_accuracy": 0.22164890319108962,
      "num_tokens": 110392309.0,
      "step": 53185
    },
    {
      "entropy": 5.845611143112182,
      "epoch": 4.824927431059507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002906865992205717,
      "loss": 5.0017,
      "mean_token_accuracy": 0.21458120495080948,
      "num_tokens": 110402464.0,
      "step": 53190
    },
    {
      "entropy": 5.6750571727752686,
      "epoch": 4.82538098693759,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002906543213707579,
      "loss": 4.7715,
      "mean_token_accuracy": 0.23446902483701706,
      "num_tokens": 110412272.0,
      "step": 53195
    },
    {
      "entropy": 5.729290962219238,
      "epoch": 4.825834542815675,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029062204319720865,
      "loss": 4.8832,
      "mean_token_accuracy": 0.22377738058567048,
      "num_tokens": 110423425.0,
      "step": 53200
    },
    {
      "entropy": 5.724271440505982,
      "epoch": 4.826288098693759,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002905897647005917,
      "loss": 4.8907,
      "mean_token_accuracy": 0.2200765073299408,
      "num_tokens": 110434469.0,
      "step": 53205
    },
    {
      "entropy": 5.7465400218963625,
      "epoch": 4.826741654571843,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002905574858815744,
      "loss": 4.8446,
      "mean_token_accuracy": 0.2316815361380577,
      "num_tokens": 110443777.0,
      "step": 53210
    },
    {
      "entropy": 5.735206365585327,
      "epoch": 4.827195210449927,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002905252067408242,
      "loss": 4.905,
      "mean_token_accuracy": 0.21078973412513732,
      "num_tokens": 110454481.0,
      "step": 53215
    },
    {
      "entropy": 5.765194272994995,
      "epoch": 4.827648766328012,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002904929272790089,
      "loss": 4.9628,
      "mean_token_accuracy": 0.219353586435318,
      "num_tokens": 110464789.0,
      "step": 53220
    },
    {
      "entropy": 5.656707668304444,
      "epoch": 4.828102322206096,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002904606474967959,
      "loss": 4.7654,
      "mean_token_accuracy": 0.23455669581890107,
      "num_tokens": 110474276.0,
      "step": 53225
    },
    {
      "entropy": 5.760164260864258,
      "epoch": 4.82855587808418,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029042836739485273,
      "loss": 4.9715,
      "mean_token_accuracy": 0.2227558061480522,
      "num_tokens": 110486193.0,
      "step": 53230
    },
    {
      "entropy": 5.791361904144287,
      "epoch": 4.8290094339622645,
      "grad_norm": 1.234375,
      "learning_rate": 0.000290396086973847,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22579744458198547,
      "num_tokens": 110496619.0,
      "step": 53235
    },
    {
      "entropy": 5.770787286758423,
      "epoch": 4.829462989840349,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029036380623444636,
      "loss": 4.8927,
      "mean_token_accuracy": 0.21923244446516038,
      "num_tokens": 110507103.0,
      "step": 53240
    },
    {
      "entropy": 5.7695599555969235,
      "epoch": 4.829916545718433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002903315251773182,
      "loss": 4.8919,
      "mean_token_accuracy": 0.22444018721580505,
      "num_tokens": 110516855.0,
      "step": 53245
    },
    {
      "entropy": 5.721407890319824,
      "epoch": 4.830370101596516,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00029029924380313024,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23251396417617798,
      "num_tokens": 110526220.0,
      "step": 53250
    },
    {
      "entropy": 5.767763471603393,
      "epoch": 4.830823657474601,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029026696211255,
      "loss": 4.9442,
      "mean_token_accuracy": 0.21945420056581497,
      "num_tokens": 110537029.0,
      "step": 53255
    },
    {
      "entropy": 5.788552379608154,
      "epoch": 4.831277213352685,
      "grad_norm": 1.4375,
      "learning_rate": 0.00029023468010624503,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22443018853664398,
      "num_tokens": 110548334.0,
      "step": 53260
    },
    {
      "entropy": 5.764921569824219,
      "epoch": 4.831730769230769,
      "grad_norm": 1.421875,
      "learning_rate": 0.00029020239778488303,
      "loss": 4.9111,
      "mean_token_accuracy": 0.23109363317489623,
      "num_tokens": 110557999.0,
      "step": 53265
    },
    {
      "entropy": 5.764499187469482,
      "epoch": 4.832184325108853,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002901701151491316,
      "loss": 4.9995,
      "mean_token_accuracy": 0.2255920022726059,
      "num_tokens": 110568743.0,
      "step": 53270
    },
    {
      "entropy": 5.782461404800415,
      "epoch": 4.832637880986938,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00029013783219965835,
      "loss": 4.9032,
      "mean_token_accuracy": 0.2286966398358345,
      "num_tokens": 110578961.0,
      "step": 53275
    },
    {
      "entropy": 5.802322196960449,
      "epoch": 4.833091436865022,
      "grad_norm": 1.375,
      "learning_rate": 0.00029010554893713086,
      "loss": 4.9512,
      "mean_token_accuracy": 0.21878729313611983,
      "num_tokens": 110590496.0,
      "step": 53280
    },
    {
      "entropy": 5.794965219497681,
      "epoch": 4.833544992743106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002900732653622167,
      "loss": 4.9008,
      "mean_token_accuracy": 0.2295579195022583,
      "num_tokens": 110601018.0,
      "step": 53285
    },
    {
      "entropy": 5.726259565353393,
      "epoch": 4.8339985486211905,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002900409814755836,
      "loss": 4.866,
      "mean_token_accuracy": 0.22151411324739456,
      "num_tokens": 110611369.0,
      "step": 53290
    },
    {
      "entropy": 5.7739355087280275,
      "epoch": 4.834452104499274,
      "grad_norm": 1.34375,
      "learning_rate": 0.00029000869727789923,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2327763170003891,
      "num_tokens": 110621223.0,
      "step": 53295
    },
    {
      "entropy": 5.704370403289795,
      "epoch": 4.834905660377358,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028997641276983107,
      "loss": 4.8085,
      "mean_token_accuracy": 0.2310947597026825,
      "num_tokens": 110631033.0,
      "step": 53300
    },
    {
      "entropy": 5.694156217575073,
      "epoch": 4.835359216255442,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028994412795204695,
      "loss": 4.8506,
      "mean_token_accuracy": 0.2273046240210533,
      "num_tokens": 110642368.0,
      "step": 53305
    },
    {
      "entropy": 5.708344841003418,
      "epoch": 4.835812772133527,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002899118428252144,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22234766483306884,
      "num_tokens": 110653317.0,
      "step": 53310
    },
    {
      "entropy": 5.721781492233276,
      "epoch": 4.836266328011611,
      "grad_norm": 1.546875,
      "learning_rate": 0.00028987955739000123,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22608001083135604,
      "num_tokens": 110663143.0,
      "step": 53315
    },
    {
      "entropy": 5.768485403060913,
      "epoch": 4.836719883889695,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000289847271647075,
      "loss": 4.9009,
      "mean_token_accuracy": 0.23089582920074464,
      "num_tokens": 110673383.0,
      "step": 53320
    },
    {
      "entropy": 5.757639932632446,
      "epoch": 4.8371734397677795,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002898149855971034,
      "loss": 4.8296,
      "mean_token_accuracy": 0.22740382552146912,
      "num_tokens": 110683576.0,
      "step": 53325
    },
    {
      "entropy": 5.6917651176452635,
      "epoch": 4.837626995645864,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028978269924075404,
      "loss": 4.8309,
      "mean_token_accuracy": 0.22930834740400313,
      "num_tokens": 110693986.0,
      "step": 53330
    },
    {
      "entropy": 5.697687721252441,
      "epoch": 4.838080551523948,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002897504125786948,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23083695024251938,
      "num_tokens": 110703724.0,
      "step": 53335
    },
    {
      "entropy": 5.690527772903442,
      "epoch": 4.838534107402032,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028971812561159324,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22502065747976302,
      "num_tokens": 110714251.0,
      "step": 53340
    },
    {
      "entropy": 5.741587829589844,
      "epoch": 4.8389876632801165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028968583834011714,
      "loss": 4.9146,
      "mean_token_accuracy": 0.21776103526353835,
      "num_tokens": 110724011.0,
      "step": 53345
    },
    {
      "entropy": 5.7923040866851805,
      "epoch": 4.8394412191582,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028965355076493405,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22317892014980317,
      "num_tokens": 110733627.0,
      "step": 53350
    },
    {
      "entropy": 5.801911163330078,
      "epoch": 4.839894775036284,
      "grad_norm": 1.25,
      "learning_rate": 0.0002896212628867118,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22255603671073915,
      "num_tokens": 110744514.0,
      "step": 53355
    },
    {
      "entropy": 5.832213449478149,
      "epoch": 4.8403483309143684,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028958897470611825,
      "loss": 4.9454,
      "mean_token_accuracy": 0.2263704389333725,
      "num_tokens": 110754257.0,
      "step": 53360
    },
    {
      "entropy": 5.730437135696411,
      "epoch": 4.840801886792453,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002895566862238208,
      "loss": 4.8541,
      "mean_token_accuracy": 0.2295353129506111,
      "num_tokens": 110764383.0,
      "step": 53365
    },
    {
      "entropy": 5.79294490814209,
      "epoch": 4.841255442670537,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002895243974404875,
      "loss": 4.9377,
      "mean_token_accuracy": 0.22077495157718657,
      "num_tokens": 110775006.0,
      "step": 53370
    },
    {
      "entropy": 5.776894474029541,
      "epoch": 4.841708998548621,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002894921083567859,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22676136791706086,
      "num_tokens": 110784764.0,
      "step": 53375
    },
    {
      "entropy": 5.776498794555664,
      "epoch": 4.8421625544267055,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002894598189733838,
      "loss": 4.8789,
      "mean_token_accuracy": 0.23359518200159074,
      "num_tokens": 110794942.0,
      "step": 53380
    },
    {
      "entropy": 5.7135285377502445,
      "epoch": 4.84261611030479,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000289427529290949,
      "loss": 4.8146,
      "mean_token_accuracy": 0.2368596613407135,
      "num_tokens": 110804696.0,
      "step": 53385
    },
    {
      "entropy": 5.670087814331055,
      "epoch": 4.843069666182874,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028939523931014924,
      "loss": 4.7766,
      "mean_token_accuracy": 0.2398675113916397,
      "num_tokens": 110814867.0,
      "step": 53390
    },
    {
      "entropy": 5.752986335754395,
      "epoch": 4.843523222060957,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028936294903165224,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22437357753515244,
      "num_tokens": 110824738.0,
      "step": 53395
    },
    {
      "entropy": 5.772585821151734,
      "epoch": 4.843976777939043,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002893306584561257,
      "loss": 4.9318,
      "mean_token_accuracy": 0.22515154331922532,
      "num_tokens": 110834646.0,
      "step": 53400
    },
    {
      "entropy": 5.78420877456665,
      "epoch": 4.844430333817126,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028929836758423763,
      "loss": 4.9461,
      "mean_token_accuracy": 0.22341522574424744,
      "num_tokens": 110843907.0,
      "step": 53405
    },
    {
      "entropy": 5.761708307266235,
      "epoch": 4.84488388969521,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002892660764166556,
      "loss": 4.91,
      "mean_token_accuracy": 0.22521815448999405,
      "num_tokens": 110854930.0,
      "step": 53410
    },
    {
      "entropy": 5.685295104980469,
      "epoch": 4.8453374455732945,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028923378495404747,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22588998228311538,
      "num_tokens": 110865256.0,
      "step": 53415
    },
    {
      "entropy": 5.765700101852417,
      "epoch": 4.845791001451379,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028920149319708105,
      "loss": 4.94,
      "mean_token_accuracy": 0.2239728555083275,
      "num_tokens": 110875600.0,
      "step": 53420
    },
    {
      "entropy": 5.828576326370239,
      "epoch": 4.846244557329463,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028916920114642415,
      "loss": 5.0032,
      "mean_token_accuracy": 0.22244170904159546,
      "num_tokens": 110886856.0,
      "step": 53425
    },
    {
      "entropy": 5.77738356590271,
      "epoch": 4.846698113207547,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028913690880274454,
      "loss": 4.9099,
      "mean_token_accuracy": 0.23043400943279266,
      "num_tokens": 110896908.0,
      "step": 53430
    },
    {
      "entropy": 5.7430414199829105,
      "epoch": 4.8471516690856316,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028910461616671004,
      "loss": 4.8633,
      "mean_token_accuracy": 0.23142510354518891,
      "num_tokens": 110908987.0,
      "step": 53435
    },
    {
      "entropy": 5.733460235595703,
      "epoch": 4.847605224963716,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028907232323898855,
      "loss": 4.838,
      "mean_token_accuracy": 0.23378176242113113,
      "num_tokens": 110920036.0,
      "step": 53440
    },
    {
      "entropy": 5.79375696182251,
      "epoch": 4.8480587808418,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028904003002024775,
      "loss": 4.9422,
      "mean_token_accuracy": 0.2245466724038124,
      "num_tokens": 110931170.0,
      "step": 53445
    },
    {
      "entropy": 5.797913503646851,
      "epoch": 4.8485123367198835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002890077365111556,
      "loss": 4.9226,
      "mean_token_accuracy": 0.2209007889032364,
      "num_tokens": 110942698.0,
      "step": 53450
    },
    {
      "entropy": 5.763950347900391,
      "epoch": 4.848965892597968,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028897544271237985,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22215773165225983,
      "num_tokens": 110953400.0,
      "step": 53455
    },
    {
      "entropy": 5.768204021453857,
      "epoch": 4.849419448476052,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028894314862458844,
      "loss": 4.8998,
      "mean_token_accuracy": 0.23045941144227983,
      "num_tokens": 110963792.0,
      "step": 53460
    },
    {
      "entropy": 5.800205850601197,
      "epoch": 4.849873004354136,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028891085424844913,
      "loss": 4.8874,
      "mean_token_accuracy": 0.2231293499469757,
      "num_tokens": 110973636.0,
      "step": 53465
    },
    {
      "entropy": 5.747876358032227,
      "epoch": 4.8503265602322205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002888785595846298,
      "loss": 4.8886,
      "mean_token_accuracy": 0.23088424205780028,
      "num_tokens": 110983163.0,
      "step": 53470
    },
    {
      "entropy": 5.7067890644073485,
      "epoch": 4.850780116110305,
      "grad_norm": 1.25,
      "learning_rate": 0.0002888462646337983,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22449941486120223,
      "num_tokens": 110993344.0,
      "step": 53475
    },
    {
      "entropy": 5.766878414154053,
      "epoch": 4.851233671988389,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028881396939662256,
      "loss": 4.9095,
      "mean_token_accuracy": 0.22020067870616913,
      "num_tokens": 111004380.0,
      "step": 53480
    },
    {
      "entropy": 5.761161041259766,
      "epoch": 4.851687227866473,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002887816738737704,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23084161430597305,
      "num_tokens": 111015661.0,
      "step": 53485
    },
    {
      "entropy": 5.730525016784668,
      "epoch": 4.852140783744558,
      "grad_norm": 1.25,
      "learning_rate": 0.0002887493780659097,
      "loss": 4.8543,
      "mean_token_accuracy": 0.2277987316250801,
      "num_tokens": 111026496.0,
      "step": 53490
    },
    {
      "entropy": 5.7003809928894045,
      "epoch": 4.852594339622642,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028871708197370844,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23370823562145232,
      "num_tokens": 111036356.0,
      "step": 53495
    },
    {
      "entropy": 5.759594821929932,
      "epoch": 4.853047895500726,
      "grad_norm": 1.375,
      "learning_rate": 0.00028868478559783433,
      "loss": 4.9478,
      "mean_token_accuracy": 0.22783022969961167,
      "num_tokens": 111045967.0,
      "step": 53500
    },
    {
      "entropy": 5.750672054290772,
      "epoch": 4.8535014513788095,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028865248893895543,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22343815118074417,
      "num_tokens": 111056459.0,
      "step": 53505
    },
    {
      "entropy": 5.72183928489685,
      "epoch": 4.853955007256894,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028862019199773955,
      "loss": 4.8137,
      "mean_token_accuracy": 0.22443224787712096,
      "num_tokens": 111065901.0,
      "step": 53510
    },
    {
      "entropy": 5.710616540908814,
      "epoch": 4.854408563134978,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002885878947748546,
      "loss": 4.8525,
      "mean_token_accuracy": 0.22860313802957535,
      "num_tokens": 111075721.0,
      "step": 53515
    },
    {
      "entropy": 5.695402526855469,
      "epoch": 4.854862119013062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028855559727096854,
      "loss": 4.8496,
      "mean_token_accuracy": 0.23362667709589005,
      "num_tokens": 111086453.0,
      "step": 53520
    },
    {
      "entropy": 5.755975437164307,
      "epoch": 4.855315674891147,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028852329948674926,
      "loss": 4.9478,
      "mean_token_accuracy": 0.2190585806965828,
      "num_tokens": 111099130.0,
      "step": 53525
    },
    {
      "entropy": 5.734133529663086,
      "epoch": 4.855769230769231,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002884910014228648,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22914693057537078,
      "num_tokens": 111108304.0,
      "step": 53530
    },
    {
      "entropy": 5.774022006988526,
      "epoch": 4.856222786647315,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002884587030799829,
      "loss": 4.8401,
      "mean_token_accuracy": 0.22497935295104982,
      "num_tokens": 111118369.0,
      "step": 53535
    },
    {
      "entropy": 5.701703977584839,
      "epoch": 4.856676342525399,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00028842640445877153,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2282060205936432,
      "num_tokens": 111128550.0,
      "step": 53540
    },
    {
      "entropy": 5.7971703052520756,
      "epoch": 4.857129898403484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028839410555989885,
      "loss": 4.9852,
      "mean_token_accuracy": 0.21371220499277116,
      "num_tokens": 111140824.0,
      "step": 53545
    },
    {
      "entropy": 5.750427722930908,
      "epoch": 4.857583454281567,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028836180638403257,
      "loss": 4.8834,
      "mean_token_accuracy": 0.2216004729270935,
      "num_tokens": 111152241.0,
      "step": 53550
    },
    {
      "entropy": 5.77687087059021,
      "epoch": 4.858037010159651,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002883295069318407,
      "loss": 4.9391,
      "mean_token_accuracy": 0.22231267690658568,
      "num_tokens": 111161897.0,
      "step": 53555
    },
    {
      "entropy": 5.7711968421936035,
      "epoch": 4.8584905660377355,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002882972072039914,
      "loss": 4.8817,
      "mean_token_accuracy": 0.2277698650956154,
      "num_tokens": 111172586.0,
      "step": 53560
    },
    {
      "entropy": 5.78973388671875,
      "epoch": 4.85894412191582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028826490720115224,
      "loss": 4.9994,
      "mean_token_accuracy": 0.2181120350956917,
      "num_tokens": 111182734.0,
      "step": 53565
    },
    {
      "entropy": 5.694659328460693,
      "epoch": 4.859397677793904,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028823260692399156,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23609040528535843,
      "num_tokens": 111193551.0,
      "step": 53570
    },
    {
      "entropy": 5.715792608261109,
      "epoch": 4.859851233671988,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002882003063731771,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22991765290498734,
      "num_tokens": 111202604.0,
      "step": 53575
    },
    {
      "entropy": 5.683126163482666,
      "epoch": 4.860304789550073,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028816800554937703,
      "loss": 4.8496,
      "mean_token_accuracy": 0.23206887394189835,
      "num_tokens": 111213675.0,
      "step": 53580
    },
    {
      "entropy": 5.761415004730225,
      "epoch": 4.860758345428157,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028813570445325915,
      "loss": 4.8302,
      "mean_token_accuracy": 0.23576506227254868,
      "num_tokens": 111223354.0,
      "step": 53585
    },
    {
      "entropy": 5.742148065567017,
      "epoch": 4.861211901306241,
      "grad_norm": 1.375,
      "learning_rate": 0.00028810340308549164,
      "loss": 4.8034,
      "mean_token_accuracy": 0.23547853380441666,
      "num_tokens": 111232865.0,
      "step": 53590
    },
    {
      "entropy": 5.676417016983033,
      "epoch": 4.861665457184325,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028807110144674243,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23144350200891495,
      "num_tokens": 111242924.0,
      "step": 53595
    },
    {
      "entropy": 5.752792406082153,
      "epoch": 4.86211901306241,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002880387995376794,
      "loss": 4.8118,
      "mean_token_accuracy": 0.2348112389445305,
      "num_tokens": 111252921.0,
      "step": 53600
    },
    {
      "entropy": 5.762843799591065,
      "epoch": 4.862572568940493,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002880064973589708,
      "loss": 4.9299,
      "mean_token_accuracy": 0.2222815454006195,
      "num_tokens": 111263176.0,
      "step": 53605
    },
    {
      "entropy": 5.749739170074463,
      "epoch": 4.863026124818577,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028797419491128446,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22124443501234053,
      "num_tokens": 111273043.0,
      "step": 53610
    },
    {
      "entropy": 5.752822351455689,
      "epoch": 4.863479680696662,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028794189219528846,
      "loss": 4.8393,
      "mean_token_accuracy": 0.22925104200839996,
      "num_tokens": 111282720.0,
      "step": 53615
    },
    {
      "entropy": 5.666813850402832,
      "epoch": 4.863933236574746,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028790958921165086,
      "loss": 4.8716,
      "mean_token_accuracy": 0.2166697636246681,
      "num_tokens": 111293946.0,
      "step": 53620
    },
    {
      "entropy": 5.736125373840332,
      "epoch": 4.86438679245283,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002878772859610396,
      "loss": 4.949,
      "mean_token_accuracy": 0.21882937252521514,
      "num_tokens": 111305052.0,
      "step": 53625
    },
    {
      "entropy": 5.7842545986175535,
      "epoch": 4.864840348330914,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028784498244412284,
      "loss": 4.8773,
      "mean_token_accuracy": 0.2238449424505234,
      "num_tokens": 111314216.0,
      "step": 53630
    },
    {
      "entropy": 5.725752639770508,
      "epoch": 4.865293904208999,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028781267866156854,
      "loss": 4.8309,
      "mean_token_accuracy": 0.2307440921664238,
      "num_tokens": 111324939.0,
      "step": 53635
    },
    {
      "entropy": 5.792523956298828,
      "epoch": 4.865747460087083,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028778037461404477,
      "loss": 4.8858,
      "mean_token_accuracy": 0.22336280196905137,
      "num_tokens": 111334908.0,
      "step": 53640
    },
    {
      "entropy": 5.706590843200684,
      "epoch": 4.866201015965167,
      "grad_norm": 1.375,
      "learning_rate": 0.00028774807030221966,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22782138884067535,
      "num_tokens": 111345560.0,
      "step": 53645
    },
    {
      "entropy": 5.7174619197845455,
      "epoch": 4.8666545718432515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002877157657267611,
      "loss": 4.8172,
      "mean_token_accuracy": 0.23130905926227568,
      "num_tokens": 111355955.0,
      "step": 53650
    },
    {
      "entropy": 5.781071996688842,
      "epoch": 4.867108127721336,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028768346088833737,
      "loss": 4.9539,
      "mean_token_accuracy": 0.2255904868245125,
      "num_tokens": 111365567.0,
      "step": 53655
    },
    {
      "entropy": 5.681385517120361,
      "epoch": 4.867561683599419,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002876511557876164,
      "loss": 4.7968,
      "mean_token_accuracy": 0.23399462550878525,
      "num_tokens": 111375740.0,
      "step": 53660
    },
    {
      "entropy": 5.730946207046509,
      "epoch": 4.868015239477503,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028761885042526635,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23433686345815657,
      "num_tokens": 111385593.0,
      "step": 53665
    },
    {
      "entropy": 5.7888350009918215,
      "epoch": 4.868468795355588,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028758654480195517,
      "loss": 4.9532,
      "mean_token_accuracy": 0.2168399751186371,
      "num_tokens": 111395731.0,
      "step": 53670
    },
    {
      "entropy": 5.795417833328247,
      "epoch": 4.868922351233672,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002875542389183511,
      "loss": 4.8641,
      "mean_token_accuracy": 0.23456545323133468,
      "num_tokens": 111406140.0,
      "step": 53675
    },
    {
      "entropy": 5.779440546035767,
      "epoch": 4.869375907111756,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028752193277512217,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22753077894449233,
      "num_tokens": 111416853.0,
      "step": 53680
    },
    {
      "entropy": 5.731265830993652,
      "epoch": 4.86982946298984,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002874896263729364,
      "loss": 4.8526,
      "mean_token_accuracy": 0.23307694047689437,
      "num_tokens": 111427043.0,
      "step": 53685
    },
    {
      "entropy": 5.817590045928955,
      "epoch": 4.870283018867925,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002874573197124621,
      "loss": 4.9487,
      "mean_token_accuracy": 0.22352163642644882,
      "num_tokens": 111436791.0,
      "step": 53690
    },
    {
      "entropy": 5.710689496994019,
      "epoch": 4.870736574746009,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002874250127943672,
      "loss": 4.8144,
      "mean_token_accuracy": 0.23217537999153137,
      "num_tokens": 111446603.0,
      "step": 53695
    },
    {
      "entropy": 5.723923110961914,
      "epoch": 4.871190130624093,
      "grad_norm": 1.375,
      "learning_rate": 0.00028739270561931984,
      "loss": 4.7837,
      "mean_token_accuracy": 0.23874378204345703,
      "num_tokens": 111456942.0,
      "step": 53700
    },
    {
      "entropy": 5.730199193954467,
      "epoch": 4.871643686502177,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028736039818798813,
      "loss": 4.8777,
      "mean_token_accuracy": 0.226826074719429,
      "num_tokens": 111467671.0,
      "step": 53705
    },
    {
      "entropy": 5.790156126022339,
      "epoch": 4.872097242380261,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002873280905010403,
      "loss": 4.957,
      "mean_token_accuracy": 0.2184886559844017,
      "num_tokens": 111476495.0,
      "step": 53710
    },
    {
      "entropy": 5.715177583694458,
      "epoch": 4.872550798258345,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028729578255914446,
      "loss": 4.748,
      "mean_token_accuracy": 0.23961493223905564,
      "num_tokens": 111487491.0,
      "step": 53715
    },
    {
      "entropy": 5.708586168289185,
      "epoch": 4.873004354136429,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002872634743629686,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2243741348385811,
      "num_tokens": 111497970.0,
      "step": 53720
    },
    {
      "entropy": 5.73491325378418,
      "epoch": 4.873457910014514,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000287231165913181,
      "loss": 4.8808,
      "mean_token_accuracy": 0.2271100625395775,
      "num_tokens": 111507635.0,
      "step": 53725
    },
    {
      "entropy": 5.726089429855347,
      "epoch": 4.873911465892598,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002871988572104498,
      "loss": 4.8429,
      "mean_token_accuracy": 0.23081888109445572,
      "num_tokens": 111516751.0,
      "step": 53730
    },
    {
      "entropy": 5.712036991119385,
      "epoch": 4.874365021770682,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028716654825544316,
      "loss": 4.8964,
      "mean_token_accuracy": 0.2243110328912735,
      "num_tokens": 111527748.0,
      "step": 53735
    },
    {
      "entropy": 5.748145580291748,
      "epoch": 4.8748185776487665,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002871342390488291,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22323660552501678,
      "num_tokens": 111537718.0,
      "step": 53740
    },
    {
      "entropy": 5.665159082412719,
      "epoch": 4.875272133526851,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028710192959127597,
      "loss": 4.7549,
      "mean_token_accuracy": 0.23617810606956482,
      "num_tokens": 111546631.0,
      "step": 53745
    },
    {
      "entropy": 5.731109571456909,
      "epoch": 4.875725689404935,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002870696198834518,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2244310513138771,
      "num_tokens": 111557276.0,
      "step": 53750
    },
    {
      "entropy": 5.734246397018433,
      "epoch": 4.876179245283019,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028703730992602486,
      "loss": 4.8559,
      "mean_token_accuracy": 0.2262737825512886,
      "num_tokens": 111567146.0,
      "step": 53755
    },
    {
      "entropy": 5.777671575546265,
      "epoch": 4.876632801161103,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028700499971966327,
      "loss": 4.9511,
      "mean_token_accuracy": 0.22320576906204223,
      "num_tokens": 111576887.0,
      "step": 53760
    },
    {
      "entropy": 5.796948957443237,
      "epoch": 4.877086357039187,
      "grad_norm": 1.53125,
      "learning_rate": 0.00028697268926503527,
      "loss": 4.9124,
      "mean_token_accuracy": 0.21903827786445618,
      "num_tokens": 111588212.0,
      "step": 53765
    },
    {
      "entropy": 5.737918901443481,
      "epoch": 4.877539912917271,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028694037856280894,
      "loss": 4.8729,
      "mean_token_accuracy": 0.226279516518116,
      "num_tokens": 111598308.0,
      "step": 53770
    },
    {
      "entropy": 5.762568712234497,
      "epoch": 4.8779934687953554,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002869080676136525,
      "loss": 4.9159,
      "mean_token_accuracy": 0.22893241345882415,
      "num_tokens": 111608252.0,
      "step": 53775
    },
    {
      "entropy": 5.733319711685181,
      "epoch": 4.87844702467344,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002868757564182343,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22345143258571626,
      "num_tokens": 111619555.0,
      "step": 53780
    },
    {
      "entropy": 5.6984107971191404,
      "epoch": 4.878900580551524,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002868434449772223,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2275901973247528,
      "num_tokens": 111629578.0,
      "step": 53785
    },
    {
      "entropy": 5.783961200714112,
      "epoch": 4.879354136429608,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002868111332912849,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2169114425778389,
      "num_tokens": 111639389.0,
      "step": 53790
    },
    {
      "entropy": 5.833168268203735,
      "epoch": 4.8798076923076925,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002867788213610903,
      "loss": 4.9447,
      "mean_token_accuracy": 0.22322158366441727,
      "num_tokens": 111649855.0,
      "step": 53795
    },
    {
      "entropy": 5.820524215698242,
      "epoch": 4.880261248185777,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002867465091873066,
      "loss": 4.9528,
      "mean_token_accuracy": 0.22553272396326066,
      "num_tokens": 111660315.0,
      "step": 53800
    },
    {
      "entropy": 5.742714595794678,
      "epoch": 4.880714804063861,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002867141967706021,
      "loss": 4.9269,
      "mean_token_accuracy": 0.21973207890987395,
      "num_tokens": 111669883.0,
      "step": 53805
    },
    {
      "entropy": 5.677505731582642,
      "epoch": 4.881168359941945,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028668188411164506,
      "loss": 4.9245,
      "mean_token_accuracy": 0.2185431033372879,
      "num_tokens": 111680598.0,
      "step": 53810
    },
    {
      "entropy": 5.749637079238892,
      "epoch": 4.881621915820029,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028664957121110374,
      "loss": 4.8838,
      "mean_token_accuracy": 0.2240924209356308,
      "num_tokens": 111689805.0,
      "step": 53815
    },
    {
      "entropy": 5.781575155258179,
      "epoch": 4.882075471698113,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028661725806964624,
      "loss": 4.82,
      "mean_token_accuracy": 0.23111622929573059,
      "num_tokens": 111700313.0,
      "step": 53820
    },
    {
      "entropy": 5.807082319259644,
      "epoch": 4.882529027576197,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002865849446879409,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22511772960424423,
      "num_tokens": 111711301.0,
      "step": 53825
    },
    {
      "entropy": 5.711468172073364,
      "epoch": 4.8829825834542815,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002865526310666559,
      "loss": 4.8381,
      "mean_token_accuracy": 0.22649310827255248,
      "num_tokens": 111720800.0,
      "step": 53830
    },
    {
      "entropy": 5.70235767364502,
      "epoch": 4.883436139332366,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028652031720645963,
      "loss": 4.8049,
      "mean_token_accuracy": 0.22906875908374785,
      "num_tokens": 111731665.0,
      "step": 53835
    },
    {
      "entropy": 5.721709680557251,
      "epoch": 4.88388969521045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002864880031080202,
      "loss": 4.918,
      "mean_token_accuracy": 0.22464539408683776,
      "num_tokens": 111742149.0,
      "step": 53840
    },
    {
      "entropy": 5.738027620315552,
      "epoch": 4.884343251088534,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028645568877200603,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2311398759484291,
      "num_tokens": 111753089.0,
      "step": 53845
    },
    {
      "entropy": 5.682496213912964,
      "epoch": 4.8847968069666186,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002864233741990852,
      "loss": 4.8338,
      "mean_token_accuracy": 0.22963265031576158,
      "num_tokens": 111762937.0,
      "step": 53850
    },
    {
      "entropy": 5.67499098777771,
      "epoch": 4.885250362844703,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002863910593899261,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23722414225339888,
      "num_tokens": 111772920.0,
      "step": 53855
    },
    {
      "entropy": 5.740085744857788,
      "epoch": 4.885703918722786,
      "grad_norm": 1.359375,
      "learning_rate": 0.000286358744345197,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22990104109048842,
      "num_tokens": 111783117.0,
      "step": 53860
    },
    {
      "entropy": 5.73802433013916,
      "epoch": 4.8861574746008705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002863264290655662,
      "loss": 4.9001,
      "mean_token_accuracy": 0.2213037818670273,
      "num_tokens": 111795033.0,
      "step": 53865
    },
    {
      "entropy": 5.710450839996338,
      "epoch": 4.886611030478955,
      "grad_norm": 1.25,
      "learning_rate": 0.00028629411355170193,
      "loss": 4.8571,
      "mean_token_accuracy": 0.22670791894197465,
      "num_tokens": 111805156.0,
      "step": 53870
    },
    {
      "entropy": 5.7787731170654295,
      "epoch": 4.887064586357039,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028626179780427246,
      "loss": 4.9548,
      "mean_token_accuracy": 0.22337441742420197,
      "num_tokens": 111815363.0,
      "step": 53875
    },
    {
      "entropy": 5.795491743087768,
      "epoch": 4.887518142235123,
      "grad_norm": 1.5625,
      "learning_rate": 0.00028622948182394623,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22941214144229888,
      "num_tokens": 111824983.0,
      "step": 53880
    },
    {
      "entropy": 5.668894958496094,
      "epoch": 4.8879716981132075,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028619716561139143,
      "loss": 4.8768,
      "mean_token_accuracy": 0.2277701511979103,
      "num_tokens": 111836468.0,
      "step": 53885
    },
    {
      "entropy": 5.732281875610352,
      "epoch": 4.888425253991292,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002861648491672764,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22265979051589965,
      "num_tokens": 111847659.0,
      "step": 53890
    },
    {
      "entropy": 5.774370622634888,
      "epoch": 4.888878809869376,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028613253249226944,
      "loss": 4.9509,
      "mean_token_accuracy": 0.21987784951925277,
      "num_tokens": 111858742.0,
      "step": 53895
    },
    {
      "entropy": 5.745674180984497,
      "epoch": 4.88933236574746,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028610021558703885,
      "loss": 4.9115,
      "mean_token_accuracy": 0.2257191628217697,
      "num_tokens": 111869692.0,
      "step": 53900
    },
    {
      "entropy": 5.727071762084961,
      "epoch": 4.889785921625545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028606789845225295,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2237565591931343,
      "num_tokens": 111880532.0,
      "step": 53905
    },
    {
      "entropy": 5.825229024887085,
      "epoch": 4.890239477503629,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028603558108858016,
      "loss": 4.9683,
      "mean_token_accuracy": 0.2175211414694786,
      "num_tokens": 111891178.0,
      "step": 53910
    },
    {
      "entropy": 5.7908172607421875,
      "epoch": 4.890693033381712,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002860032634966887,
      "loss": 4.9294,
      "mean_token_accuracy": 0.2274210587143898,
      "num_tokens": 111902043.0,
      "step": 53915
    },
    {
      "entropy": 5.748470878601074,
      "epoch": 4.8911465892597965,
      "grad_norm": 1.421875,
      "learning_rate": 0.000285970945677247,
      "loss": 4.8493,
      "mean_token_accuracy": 0.2300926148891449,
      "num_tokens": 111912170.0,
      "step": 53920
    },
    {
      "entropy": 5.649572515487671,
      "epoch": 4.891600145137881,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028593862763092327,
      "loss": 4.8561,
      "mean_token_accuracy": 0.23062562346458435,
      "num_tokens": 111922415.0,
      "step": 53925
    },
    {
      "entropy": 5.773506689071655,
      "epoch": 4.892053701015965,
      "grad_norm": 1.25,
      "learning_rate": 0.00028590630935838596,
      "loss": 4.9345,
      "mean_token_accuracy": 0.22205937951803206,
      "num_tokens": 111933338.0,
      "step": 53930
    },
    {
      "entropy": 5.723740434646606,
      "epoch": 4.892507256894049,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002858739908603034,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2271907314658165,
      "num_tokens": 111942225.0,
      "step": 53935
    },
    {
      "entropy": 5.729215478897094,
      "epoch": 4.892960812772134,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002858416721373439,
      "loss": 4.8428,
      "mean_token_accuracy": 0.23198260515928268,
      "num_tokens": 111953356.0,
      "step": 53940
    },
    {
      "entropy": 5.708400440216065,
      "epoch": 4.893414368650218,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028580935319017593,
      "loss": 4.8019,
      "mean_token_accuracy": 0.22415048629045486,
      "num_tokens": 111964275.0,
      "step": 53945
    },
    {
      "entropy": 5.740873908996582,
      "epoch": 4.893867924528302,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028577703401946777,
      "loss": 4.861,
      "mean_token_accuracy": 0.2286331370472908,
      "num_tokens": 111974006.0,
      "step": 53950
    },
    {
      "entropy": 5.774976968765259,
      "epoch": 4.894321480406386,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028574471462588776,
      "loss": 4.9749,
      "mean_token_accuracy": 0.22091459780931472,
      "num_tokens": 111985085.0,
      "step": 53955
    },
    {
      "entropy": 5.779650735855102,
      "epoch": 4.89477503628447,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002857123950101043,
      "loss": 4.9799,
      "mean_token_accuracy": 0.21373831927776338,
      "num_tokens": 111995162.0,
      "step": 53960
    },
    {
      "entropy": 5.7830602645874025,
      "epoch": 4.895228592162555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002856800751727858,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22292605340480803,
      "num_tokens": 112005625.0,
      "step": 53965
    },
    {
      "entropy": 5.752211809158325,
      "epoch": 4.895682148040638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028564775511460066,
      "loss": 4.8568,
      "mean_token_accuracy": 0.23175185471773146,
      "num_tokens": 112015977.0,
      "step": 53970
    },
    {
      "entropy": 5.7450597286224365,
      "epoch": 4.8961357039187225,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002856154348362173,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22360356897115707,
      "num_tokens": 112025655.0,
      "step": 53975
    },
    {
      "entropy": 5.688132667541504,
      "epoch": 4.896589259796807,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028558311433830386,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23755226731300355,
      "num_tokens": 112036986.0,
      "step": 53980
    },
    {
      "entropy": 5.743206119537353,
      "epoch": 4.897042815674891,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000285550793621529,
      "loss": 4.9027,
      "mean_token_accuracy": 0.22871705144643784,
      "num_tokens": 112047541.0,
      "step": 53985
    },
    {
      "entropy": 5.761047220230102,
      "epoch": 4.897496371552975,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002855184726865611,
      "loss": 4.8581,
      "mean_token_accuracy": 0.22381070107221604,
      "num_tokens": 112058125.0,
      "step": 53990
    },
    {
      "entropy": 5.823196029663086,
      "epoch": 4.89794992743106,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028548615153406844,
      "loss": 4.9198,
      "mean_token_accuracy": 0.2228979989886284,
      "num_tokens": 112068712.0,
      "step": 53995
    },
    {
      "entropy": 5.70296196937561,
      "epoch": 4.898403483309144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028545383016471954,
      "loss": 4.86,
      "mean_token_accuracy": 0.2293535441160202,
      "num_tokens": 112078529.0,
      "step": 54000
    },
    {
      "epoch": 4.898403483309144,
      "eval_entropy": 5.453896630619446,
      "eval_loss": 4.9540934562683105,
      "eval_mean_token_accuracy": 0.23154409241217833,
      "eval_num_tokens": 112078529.0,
      "eval_runtime": 20.5226,
      "eval_samples_per_second": 1722.976,
      "eval_steps_per_second": 215.372,
      "step": 54000
    },
    {
      "entropy": 5.801471376419068,
      "epoch": 4.898857039187228,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002854215085791826,
      "loss": 4.9533,
      "mean_token_accuracy": 0.22023424804210662,
      "num_tokens": 112088814.0,
      "step": 54005
    },
    {
      "entropy": 5.761517143249511,
      "epoch": 4.899310595065312,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028538918677812643,
      "loss": 4.8537,
      "mean_token_accuracy": 0.23509639948606492,
      "num_tokens": 112099406.0,
      "step": 54010
    },
    {
      "entropy": 5.738940572738647,
      "epoch": 4.899764150943396,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028535686476221915,
      "loss": 4.866,
      "mean_token_accuracy": 0.2230834349989891,
      "num_tokens": 112110762.0,
      "step": 54015
    },
    {
      "entropy": 5.841030073165894,
      "epoch": 4.90021770682148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002853245425321292,
      "loss": 5.0576,
      "mean_token_accuracy": 0.2125086471438408,
      "num_tokens": 112120636.0,
      "step": 54020
    },
    {
      "entropy": 5.715322113037109,
      "epoch": 4.900671262699564,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028529222008852513,
      "loss": 4.8413,
      "mean_token_accuracy": 0.22996389120817184,
      "num_tokens": 112130756.0,
      "step": 54025
    },
    {
      "entropy": 5.798418378829956,
      "epoch": 4.901124818577649,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002852598974320753,
      "loss": 4.9149,
      "mean_token_accuracy": 0.23377985805273055,
      "num_tokens": 112140881.0,
      "step": 54030
    },
    {
      "entropy": 5.751075887680054,
      "epoch": 4.901578374455733,
      "grad_norm": 1.40625,
      "learning_rate": 0.00028522757456344824,
      "loss": 4.8519,
      "mean_token_accuracy": 0.22824373245239257,
      "num_tokens": 112152337.0,
      "step": 54035
    },
    {
      "entropy": 5.756748342514038,
      "epoch": 4.902031930333817,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002851952514833123,
      "loss": 4.9291,
      "mean_token_accuracy": 0.21992525458335876,
      "num_tokens": 112164199.0,
      "step": 54040
    },
    {
      "entropy": 5.728545379638672,
      "epoch": 4.902485486211901,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028516292819233597,
      "loss": 4.87,
      "mean_token_accuracy": 0.23109887391328812,
      "num_tokens": 112173594.0,
      "step": 54045
    },
    {
      "entropy": 5.741925954818726,
      "epoch": 4.902939042089986,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028513060469118767,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2299194797873497,
      "num_tokens": 112183261.0,
      "step": 54050
    },
    {
      "entropy": 5.753352689743042,
      "epoch": 4.90339259796807,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028509828098053594,
      "loss": 4.844,
      "mean_token_accuracy": 0.2305249586701393,
      "num_tokens": 112193506.0,
      "step": 54055
    },
    {
      "entropy": 5.725646352767944,
      "epoch": 4.903846153846154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002850659570610491,
      "loss": 4.8414,
      "mean_token_accuracy": 0.22922157049179076,
      "num_tokens": 112204665.0,
      "step": 54060
    },
    {
      "entropy": 5.704743528366089,
      "epoch": 4.9042997097242385,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002850336329333958,
      "loss": 4.8892,
      "mean_token_accuracy": 0.23381398469209672,
      "num_tokens": 112214933.0,
      "step": 54065
    },
    {
      "entropy": 5.722425127029419,
      "epoch": 4.904753265602322,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002850013085982443,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2259896755218506,
      "num_tokens": 112224525.0,
      "step": 54070
    },
    {
      "entropy": 5.727838611602783,
      "epoch": 4.905206821480406,
      "grad_norm": 1.375,
      "learning_rate": 0.00028496898405626326,
      "loss": 4.8018,
      "mean_token_accuracy": 0.23350384533405305,
      "num_tokens": 112234363.0,
      "step": 54075
    },
    {
      "entropy": 5.756862497329712,
      "epoch": 4.90566037735849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028493665930812113,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2307159945368767,
      "num_tokens": 112245721.0,
      "step": 54080
    },
    {
      "entropy": 5.752224016189575,
      "epoch": 4.906113933236575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002849043343544863,
      "loss": 4.8737,
      "mean_token_accuracy": 0.23540516793727875,
      "num_tokens": 112255775.0,
      "step": 54085
    },
    {
      "entropy": 5.74810094833374,
      "epoch": 4.906567489114659,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002848720091960273,
      "loss": 4.9272,
      "mean_token_accuracy": 0.22405183166265488,
      "num_tokens": 112266049.0,
      "step": 54090
    },
    {
      "entropy": 5.7281725883483885,
      "epoch": 4.907021044992743,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028483968383341264,
      "loss": 4.8577,
      "mean_token_accuracy": 0.22346789836883546,
      "num_tokens": 112276663.0,
      "step": 54095
    },
    {
      "entropy": 5.733085775375367,
      "epoch": 4.907474600870827,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002848073582673109,
      "loss": 4.8402,
      "mean_token_accuracy": 0.22710599154233932,
      "num_tokens": 112287065.0,
      "step": 54100
    },
    {
      "entropy": 5.748616933822632,
      "epoch": 4.907928156748912,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002847750324983903,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22115947008132936,
      "num_tokens": 112297053.0,
      "step": 54105
    },
    {
      "entropy": 5.742197799682617,
      "epoch": 4.908381712626996,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028474270652731966,
      "loss": 4.88,
      "mean_token_accuracy": 0.2244684174656868,
      "num_tokens": 112306001.0,
      "step": 54110
    },
    {
      "entropy": 5.771730422973633,
      "epoch": 4.908835268505079,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028471038035476735,
      "loss": 4.9057,
      "mean_token_accuracy": 0.23224979788064956,
      "num_tokens": 112315865.0,
      "step": 54115
    },
    {
      "entropy": 5.806035089492798,
      "epoch": 4.9092888243831645,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002846780539814019,
      "loss": 4.9916,
      "mean_token_accuracy": 0.21339087784290314,
      "num_tokens": 112326768.0,
      "step": 54120
    },
    {
      "entropy": 5.7018529891967775,
      "epoch": 4.909742380261248,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028464572740789177,
      "loss": 4.7591,
      "mean_token_accuracy": 0.23550121933221818,
      "num_tokens": 112337272.0,
      "step": 54125
    },
    {
      "entropy": 5.821590852737427,
      "epoch": 4.910195936139332,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002846134006349057,
      "loss": 4.915,
      "mean_token_accuracy": 0.2157830625772476,
      "num_tokens": 112346288.0,
      "step": 54130
    },
    {
      "entropy": 5.724307489395142,
      "epoch": 4.910649492017416,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002845810736631119,
      "loss": 4.7552,
      "mean_token_accuracy": 0.23494693040847778,
      "num_tokens": 112356829.0,
      "step": 54135
    },
    {
      "entropy": 5.6788245677948,
      "epoch": 4.911103047895501,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028454874649317915,
      "loss": 4.8178,
      "mean_token_accuracy": 0.23711263984441758,
      "num_tokens": 112367301.0,
      "step": 54140
    },
    {
      "entropy": 5.779210996627808,
      "epoch": 4.911556603773585,
      "grad_norm": 1.375,
      "learning_rate": 0.00028451641912577587,
      "loss": 4.8685,
      "mean_token_accuracy": 0.2269473284482956,
      "num_tokens": 112377638.0,
      "step": 54145
    },
    {
      "entropy": 5.795755338668823,
      "epoch": 4.912010159651669,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028448409156157055,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22815209180116652,
      "num_tokens": 112387629.0,
      "step": 54150
    },
    {
      "entropy": 5.781544399261475,
      "epoch": 4.9124637155297535,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002844517638012318,
      "loss": 4.9319,
      "mean_token_accuracy": 0.227399480342865,
      "num_tokens": 112397216.0,
      "step": 54155
    },
    {
      "entropy": 5.792121076583863,
      "epoch": 4.912917271407838,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028441943584542827,
      "loss": 4.8804,
      "mean_token_accuracy": 0.23313716053962708,
      "num_tokens": 112407438.0,
      "step": 54160
    },
    {
      "entropy": 5.7351861000061035,
      "epoch": 4.913370827285922,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028438710769482836,
      "loss": 4.8047,
      "mean_token_accuracy": 0.23775531947612763,
      "num_tokens": 112417328.0,
      "step": 54165
    },
    {
      "entropy": 5.736468362808227,
      "epoch": 4.913824383164005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002843547793501007,
      "loss": 4.9198,
      "mean_token_accuracy": 0.21760198771953582,
      "num_tokens": 112427711.0,
      "step": 54170
    },
    {
      "entropy": 5.729829740524292,
      "epoch": 4.91427793904209,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002843224508119137,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22292436063289642,
      "num_tokens": 112438953.0,
      "step": 54175
    },
    {
      "entropy": 5.794517612457275,
      "epoch": 4.914731494920174,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002842901220809362,
      "loss": 4.9241,
      "mean_token_accuracy": 0.2282329946756363,
      "num_tokens": 112449029.0,
      "step": 54180
    },
    {
      "entropy": 5.7146568775177,
      "epoch": 4.915185050798258,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002842577931578365,
      "loss": 4.7871,
      "mean_token_accuracy": 0.2361281096935272,
      "num_tokens": 112458790.0,
      "step": 54185
    },
    {
      "entropy": 5.709454679489136,
      "epoch": 4.9156386066763424,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002842254640432834,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22369466722011566,
      "num_tokens": 112469723.0,
      "step": 54190
    },
    {
      "entropy": 5.754444646835327,
      "epoch": 4.916092162554427,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028419313473794516,
      "loss": 4.8816,
      "mean_token_accuracy": 0.23259878158569336,
      "num_tokens": 112480448.0,
      "step": 54195
    },
    {
      "entropy": 5.720079231262207,
      "epoch": 4.916545718432511,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028416080524249073,
      "loss": 4.8212,
      "mean_token_accuracy": 0.2265756234526634,
      "num_tokens": 112490107.0,
      "step": 54200
    },
    {
      "entropy": 5.737733030319214,
      "epoch": 4.916999274310595,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028412847555758835,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22612987458705902,
      "num_tokens": 112499670.0,
      "step": 54205
    },
    {
      "entropy": 5.7541468143463135,
      "epoch": 4.9174528301886795,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002840961456839069,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23605982065200806,
      "num_tokens": 112509087.0,
      "step": 54210
    },
    {
      "entropy": 5.770773029327392,
      "epoch": 4.917906386066764,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002840638156221148,
      "loss": 4.8183,
      "mean_token_accuracy": 0.22592213600873948,
      "num_tokens": 112518506.0,
      "step": 54215
    },
    {
      "entropy": 5.772821760177612,
      "epoch": 4.918359941944848,
      "grad_norm": 1.359375,
      "learning_rate": 0.00028403148537288063,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22494992911815642,
      "num_tokens": 112528474.0,
      "step": 54220
    },
    {
      "entropy": 5.746809482574463,
      "epoch": 4.918813497822931,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002839991549368731,
      "loss": 4.9151,
      "mean_token_accuracy": 0.2255879446864128,
      "num_tokens": 112538513.0,
      "step": 54225
    },
    {
      "entropy": 5.788671684265137,
      "epoch": 4.919267053701016,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028396682431476074,
      "loss": 4.9681,
      "mean_token_accuracy": 0.22391629219055176,
      "num_tokens": 112548228.0,
      "step": 54230
    },
    {
      "entropy": 5.745562744140625,
      "epoch": 4.9197206095791,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028393449350721216,
      "loss": 4.847,
      "mean_token_accuracy": 0.22771172374486923,
      "num_tokens": 112558218.0,
      "step": 54235
    },
    {
      "entropy": 5.748077297210694,
      "epoch": 4.920174165457184,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000283902162514896,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22349759489297866,
      "num_tokens": 112568103.0,
      "step": 54240
    },
    {
      "entropy": 5.713492918014526,
      "epoch": 4.9206277213352685,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002838698313384809,
      "loss": 4.816,
      "mean_token_accuracy": 0.2304238960146904,
      "num_tokens": 112577729.0,
      "step": 54245
    },
    {
      "entropy": 5.742678165435791,
      "epoch": 4.921081277213353,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002838374999786353,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22795999497175218,
      "num_tokens": 112589103.0,
      "step": 54250
    },
    {
      "entropy": 5.738312292098999,
      "epoch": 4.921534833091437,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028380516843602797,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23537001609802247,
      "num_tokens": 112598935.0,
      "step": 54255
    },
    {
      "entropy": 5.720027494430542,
      "epoch": 4.921988388969521,
      "grad_norm": 1.375,
      "learning_rate": 0.00028377283671132756,
      "loss": 4.8758,
      "mean_token_accuracy": 0.22972720861434937,
      "num_tokens": 112609837.0,
      "step": 54260
    },
    {
      "entropy": 5.723819398880005,
      "epoch": 4.9224419448476056,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002837405048052027,
      "loss": 4.9103,
      "mean_token_accuracy": 0.227642785012722,
      "num_tokens": 112620159.0,
      "step": 54265
    },
    {
      "entropy": 5.704931163787842,
      "epoch": 4.922895500725689,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028370817271832193,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23213533759117128,
      "num_tokens": 112630779.0,
      "step": 54270
    },
    {
      "entropy": 5.681217241287231,
      "epoch": 4.923349056603773,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002836758404513539,
      "loss": 4.8522,
      "mean_token_accuracy": 0.22702087312936783,
      "num_tokens": 112640702.0,
      "step": 54275
    },
    {
      "entropy": 5.760127973556519,
      "epoch": 4.9238026124818575,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002836435080049673,
      "loss": 4.8712,
      "mean_token_accuracy": 0.22558735609054564,
      "num_tokens": 112651294.0,
      "step": 54280
    },
    {
      "entropy": 5.76874885559082,
      "epoch": 4.924256168359942,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002836111753798308,
      "loss": 4.9371,
      "mean_token_accuracy": 0.21805840730667114,
      "num_tokens": 112661570.0,
      "step": 54285
    },
    {
      "entropy": 5.713670253753662,
      "epoch": 4.924709724238026,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028357884257661285,
      "loss": 4.9059,
      "mean_token_accuracy": 0.22709702104330062,
      "num_tokens": 112671996.0,
      "step": 54290
    },
    {
      "entropy": 5.714863300323486,
      "epoch": 4.92516328011611,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002835465095959824,
      "loss": 4.7649,
      "mean_token_accuracy": 0.23726591765880584,
      "num_tokens": 112681661.0,
      "step": 54295
    },
    {
      "entropy": 5.784266090393066,
      "epoch": 4.9256168359941945,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002835141764386079,
      "loss": 4.9302,
      "mean_token_accuracy": 0.22846273481845855,
      "num_tokens": 112692026.0,
      "step": 54300
    },
    {
      "entropy": 5.800254774093628,
      "epoch": 4.926070391872279,
      "grad_norm": 1.140625,
      "learning_rate": 0.000283481843105158,
      "loss": 4.9481,
      "mean_token_accuracy": 0.21604284197092055,
      "num_tokens": 112703011.0,
      "step": 54305
    },
    {
      "entropy": 5.784369993209839,
      "epoch": 4.926523947750363,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002834495095963015,
      "loss": 4.928,
      "mean_token_accuracy": 0.2203434005379677,
      "num_tokens": 112712687.0,
      "step": 54310
    },
    {
      "entropy": 5.80837574005127,
      "epoch": 4.926977503628447,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000283417175912707,
      "loss": 4.9329,
      "mean_token_accuracy": 0.22269609719514846,
      "num_tokens": 112723422.0,
      "step": 54315
    },
    {
      "entropy": 5.728545808792115,
      "epoch": 4.927431059506532,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002833848420550432,
      "loss": 4.86,
      "mean_token_accuracy": 0.22938117235898972,
      "num_tokens": 112733793.0,
      "step": 54320
    },
    {
      "entropy": 5.708316707611084,
      "epoch": 4.927884615384615,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028335250802397863,
      "loss": 4.7995,
      "mean_token_accuracy": 0.230864979326725,
      "num_tokens": 112744175.0,
      "step": 54325
    },
    {
      "entropy": 5.769659233093262,
      "epoch": 4.928338171262699,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028332017382018203,
      "loss": 4.9836,
      "mean_token_accuracy": 0.22063499093055725,
      "num_tokens": 112754484.0,
      "step": 54330
    },
    {
      "entropy": 5.664108562469482,
      "epoch": 4.9287917271407835,
      "grad_norm": 1.5,
      "learning_rate": 0.0002832878394443222,
      "loss": 4.7342,
      "mean_token_accuracy": 0.2407623440027237,
      "num_tokens": 112764245.0,
      "step": 54335
    },
    {
      "entropy": 5.752418756484985,
      "epoch": 4.929245283018868,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028325550489706773,
      "loss": 4.8609,
      "mean_token_accuracy": 0.23124771118164061,
      "num_tokens": 112773629.0,
      "step": 54340
    },
    {
      "entropy": 5.736670255661011,
      "epoch": 4.929698838896952,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002832231701790873,
      "loss": 4.8418,
      "mean_token_accuracy": 0.2273457333445549,
      "num_tokens": 112784091.0,
      "step": 54345
    },
    {
      "entropy": 5.718930196762085,
      "epoch": 4.930152394775036,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028319083529104966,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22538023740053176,
      "num_tokens": 112794116.0,
      "step": 54350
    },
    {
      "entropy": 5.70949501991272,
      "epoch": 4.930605950653121,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028315850023362345,
      "loss": 4.868,
      "mean_token_accuracy": 0.22729205936193467,
      "num_tokens": 112804717.0,
      "step": 54355
    },
    {
      "entropy": 5.691067552566528,
      "epoch": 4.931059506531205,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028312616500747727,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22991494834423065,
      "num_tokens": 112815384.0,
      "step": 54360
    },
    {
      "entropy": 5.733217811584472,
      "epoch": 4.931513062409289,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028309382961328003,
      "loss": 4.9167,
      "mean_token_accuracy": 0.22727010846138002,
      "num_tokens": 112825813.0,
      "step": 54365
    },
    {
      "entropy": 5.704998207092285,
      "epoch": 4.931966618287373,
      "grad_norm": 1.328125,
      "learning_rate": 0.00028306149405170035,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23519957512617112,
      "num_tokens": 112836032.0,
      "step": 54370
    },
    {
      "entropy": 5.682537364959717,
      "epoch": 4.932420174165458,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002830291583234068,
      "loss": 4.866,
      "mean_token_accuracy": 0.22707625925540925,
      "num_tokens": 112846042.0,
      "step": 54375
    },
    {
      "entropy": 5.722149610519409,
      "epoch": 4.932873730043541,
      "grad_norm": 1.25,
      "learning_rate": 0.00028299682242906825,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22373426258563994,
      "num_tokens": 112856409.0,
      "step": 54380
    },
    {
      "entropy": 5.760048484802246,
      "epoch": 4.933327285921625,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028296448636935345,
      "loss": 4.9087,
      "mean_token_accuracy": 0.21881598532199859,
      "num_tokens": 112866788.0,
      "step": 54385
    },
    {
      "entropy": 5.700497817993164,
      "epoch": 4.9337808417997095,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028293215014493095,
      "loss": 4.7864,
      "mean_token_accuracy": 0.23428673595190047,
      "num_tokens": 112876670.0,
      "step": 54390
    },
    {
      "entropy": 5.740596675872803,
      "epoch": 4.934234397677794,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002828998137564697,
      "loss": 4.8874,
      "mean_token_accuracy": 0.22503501027822495,
      "num_tokens": 112887337.0,
      "step": 54395
    },
    {
      "entropy": 5.680146598815918,
      "epoch": 4.934687953555878,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002828674772046382,
      "loss": 4.9,
      "mean_token_accuracy": 0.21560353487730027,
      "num_tokens": 112897693.0,
      "step": 54400
    },
    {
      "entropy": 5.7199924945831295,
      "epoch": 4.935141509433962,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002828351404901052,
      "loss": 4.8603,
      "mean_token_accuracy": 0.22412883937358857,
      "num_tokens": 112907848.0,
      "step": 54405
    },
    {
      "entropy": 5.762948751449585,
      "epoch": 4.935595065312047,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002828028036135396,
      "loss": 4.8976,
      "mean_token_accuracy": 0.2332504704594612,
      "num_tokens": 112917344.0,
      "step": 54410
    },
    {
      "entropy": 5.7344635963439945,
      "epoch": 4.936048621190131,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002827704665756101,
      "loss": 4.8264,
      "mean_token_accuracy": 0.23465348035097122,
      "num_tokens": 112928004.0,
      "step": 54415
    },
    {
      "entropy": 5.722323417663574,
      "epoch": 4.936502177068215,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028273812937698524,
      "loss": 4.8266,
      "mean_token_accuracy": 0.23597555756568908,
      "num_tokens": 112938999.0,
      "step": 54420
    },
    {
      "entropy": 5.752587938308716,
      "epoch": 4.9369557329462985,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028270579201833385,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2287218078970909,
      "num_tokens": 112948246.0,
      "step": 54425
    },
    {
      "entropy": 5.776260137557983,
      "epoch": 4.937409288824383,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002826734545003248,
      "loss": 4.9006,
      "mean_token_accuracy": 0.2271597608923912,
      "num_tokens": 112958430.0,
      "step": 54430
    },
    {
      "entropy": 5.728704166412354,
      "epoch": 4.937862844702467,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002826411168236268,
      "loss": 4.8443,
      "mean_token_accuracy": 0.23063442707061768,
      "num_tokens": 112969327.0,
      "step": 54435
    },
    {
      "entropy": 5.763524723052979,
      "epoch": 4.938316400580551,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028260877898890843,
      "loss": 4.9298,
      "mean_token_accuracy": 0.22594218105077743,
      "num_tokens": 112980413.0,
      "step": 54440
    },
    {
      "entropy": 5.69902777671814,
      "epoch": 4.938769956458636,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00028257644099683864,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2320007637143135,
      "num_tokens": 112990924.0,
      "step": 54445
    },
    {
      "entropy": 5.736195468902588,
      "epoch": 4.93922351233672,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002825441028480862,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21748735755681992,
      "num_tokens": 113001949.0,
      "step": 54450
    },
    {
      "entropy": 5.804226398468018,
      "epoch": 4.939677068214804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002825117645433196,
      "loss": 4.9832,
      "mean_token_accuracy": 0.22305720448493957,
      "num_tokens": 113012149.0,
      "step": 54455
    },
    {
      "entropy": 5.7396880149841305,
      "epoch": 4.940130624092888,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028247942608320797,
      "loss": 4.895,
      "mean_token_accuracy": 0.2291079893708229,
      "num_tokens": 113022783.0,
      "step": 54460
    },
    {
      "entropy": 5.704237985610962,
      "epoch": 4.940584179970973,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00028244708746841984,
      "loss": 4.8936,
      "mean_token_accuracy": 0.23031719774007797,
      "num_tokens": 113032486.0,
      "step": 54465
    },
    {
      "entropy": 5.802404499053955,
      "epoch": 4.941037735849057,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002824147486996241,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2309722274541855,
      "num_tokens": 113042430.0,
      "step": 54470
    },
    {
      "entropy": 5.736430883407593,
      "epoch": 4.941491291727141,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028238240977748936,
      "loss": 4.8802,
      "mean_token_accuracy": 0.21925390213727952,
      "num_tokens": 113052917.0,
      "step": 54475
    },
    {
      "entropy": 5.743439197540283,
      "epoch": 4.941944847605225,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028235007070268454,
      "loss": 4.9052,
      "mean_token_accuracy": 0.2296800509095192,
      "num_tokens": 113063786.0,
      "step": 54480
    },
    {
      "entropy": 5.784452247619629,
      "epoch": 4.942398403483309,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028231773147587843,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23065468072891235,
      "num_tokens": 113073979.0,
      "step": 54485
    },
    {
      "entropy": 5.68681263923645,
      "epoch": 4.942851959361393,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002822853920977397,
      "loss": 4.8439,
      "mean_token_accuracy": 0.22961164861917496,
      "num_tokens": 113085439.0,
      "step": 54490
    },
    {
      "entropy": 5.731515026092529,
      "epoch": 4.943305515239477,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002822530525689373,
      "loss": 4.8526,
      "mean_token_accuracy": 0.22892139554023744,
      "num_tokens": 113096825.0,
      "step": 54495
    },
    {
      "entropy": 5.675727415084839,
      "epoch": 4.943759071117562,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028222071289013984,
      "loss": 4.7879,
      "mean_token_accuracy": 0.2352376401424408,
      "num_tokens": 113107267.0,
      "step": 54500
    },
    {
      "entropy": 5.6934241771698,
      "epoch": 4.944212626995646,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00028218837306201623,
      "loss": 4.8066,
      "mean_token_accuracy": 0.22644503116607667,
      "num_tokens": 113118438.0,
      "step": 54505
    },
    {
      "entropy": 5.732282543182373,
      "epoch": 4.94466618287373,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028215603308523517,
      "loss": 4.846,
      "mean_token_accuracy": 0.23664611130952834,
      "num_tokens": 113128361.0,
      "step": 54510
    },
    {
      "entropy": 5.753063488006592,
      "epoch": 4.945119738751814,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028212369296046566,
      "loss": 4.9217,
      "mean_token_accuracy": 0.2271317034959793,
      "num_tokens": 113138907.0,
      "step": 54515
    },
    {
      "entropy": 5.6624585628509525,
      "epoch": 4.945573294629899,
      "grad_norm": 1.34375,
      "learning_rate": 0.00028209135268837627,
      "loss": 4.9049,
      "mean_token_accuracy": 0.22253224700689317,
      "num_tokens": 113149107.0,
      "step": 54520
    },
    {
      "entropy": 5.726324510574341,
      "epoch": 4.946026850507983,
      "grad_norm": 1.375,
      "learning_rate": 0.00028205901226963586,
      "loss": 4.85,
      "mean_token_accuracy": 0.22862294167280198,
      "num_tokens": 113159300.0,
      "step": 54525
    },
    {
      "entropy": 5.707945966720581,
      "epoch": 4.946480406386067,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028202667170491337,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2323465198278427,
      "num_tokens": 113168911.0,
      "step": 54530
    },
    {
      "entropy": 5.725818824768067,
      "epoch": 4.946933962264151,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002819943309948775,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23115296959877013,
      "num_tokens": 113179514.0,
      "step": 54535
    },
    {
      "entropy": 5.740902662277222,
      "epoch": 4.947387518142235,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000281961990140197,
      "loss": 4.8445,
      "mean_token_accuracy": 0.22858806848526,
      "num_tokens": 113189732.0,
      "step": 54540
    },
    {
      "entropy": 5.7752875804901125,
      "epoch": 4.947841074020319,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028192964914154086,
      "loss": 4.9072,
      "mean_token_accuracy": 0.22615642994642257,
      "num_tokens": 113200012.0,
      "step": 54545
    },
    {
      "entropy": 5.770515584945679,
      "epoch": 4.948294629898403,
      "grad_norm": 1.375,
      "learning_rate": 0.0002818973079995778,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22794753909111024,
      "num_tokens": 113210504.0,
      "step": 54550
    },
    {
      "entropy": 5.726355457305909,
      "epoch": 4.948748185776488,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002818649667149766,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2278631255030632,
      "num_tokens": 113220520.0,
      "step": 54555
    },
    {
      "entropy": 5.6724005222320555,
      "epoch": 4.949201741654572,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00028183262528840616,
      "loss": 4.771,
      "mean_token_accuracy": 0.23719479143619537,
      "num_tokens": 113230355.0,
      "step": 54560
    },
    {
      "entropy": 5.7496178150177,
      "epoch": 4.949655297532656,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028180028372053537,
      "loss": 4.8838,
      "mean_token_accuracy": 0.21794185787439346,
      "num_tokens": 113240184.0,
      "step": 54565
    },
    {
      "entropy": 5.793493413925171,
      "epoch": 4.9501088534107405,
      "grad_norm": 1.515625,
      "learning_rate": 0.00028176794201203293,
      "loss": 4.936,
      "mean_token_accuracy": 0.2265704721212387,
      "num_tokens": 113249748.0,
      "step": 54570
    },
    {
      "entropy": 5.7777729511260985,
      "epoch": 4.950562409288825,
      "grad_norm": 1.375,
      "learning_rate": 0.00028173560016356763,
      "loss": 4.8568,
      "mean_token_accuracy": 0.2284199610352516,
      "num_tokens": 113260018.0,
      "step": 54575
    },
    {
      "entropy": 5.724394083023071,
      "epoch": 4.951015965166908,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002817032581758085,
      "loss": 4.7941,
      "mean_token_accuracy": 0.23294337540864946,
      "num_tokens": 113269682.0,
      "step": 54580
    },
    {
      "entropy": 5.748255729675293,
      "epoch": 4.951469521044992,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028167091604942423,
      "loss": 4.844,
      "mean_token_accuracy": 0.219591124355793,
      "num_tokens": 113279485.0,
      "step": 54585
    },
    {
      "entropy": 5.807943820953369,
      "epoch": 4.951923076923077,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028163857378508386,
      "loss": 4.9436,
      "mean_token_accuracy": 0.22170859426259995,
      "num_tokens": 113290559.0,
      "step": 54590
    },
    {
      "entropy": 5.769555568695068,
      "epoch": 4.952376632801161,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028160623138345585,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2219155490398407,
      "num_tokens": 113300581.0,
      "step": 54595
    },
    {
      "entropy": 5.731149816513062,
      "epoch": 4.952830188679245,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028157388884520953,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22672766894102098,
      "num_tokens": 113311261.0,
      "step": 54600
    },
    {
      "entropy": 5.732681560516357,
      "epoch": 4.9532837445573294,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028154154617101337,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23179541528224945,
      "num_tokens": 113321701.0,
      "step": 54605
    },
    {
      "entropy": 5.690102815628052,
      "epoch": 4.953737300435414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028150920336153637,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22727696895599364,
      "num_tokens": 113332364.0,
      "step": 54610
    },
    {
      "entropy": 5.691157388687134,
      "epoch": 4.954190856313498,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002814768604174475,
      "loss": 4.8246,
      "mean_token_accuracy": 0.2276059240102768,
      "num_tokens": 113342920.0,
      "step": 54615
    },
    {
      "entropy": 5.802882242202759,
      "epoch": 4.954644412191582,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002814445173394154,
      "loss": 4.8774,
      "mean_token_accuracy": 0.22796350717544556,
      "num_tokens": 113353602.0,
      "step": 54620
    },
    {
      "entropy": 5.731995344161987,
      "epoch": 4.9550979680696665,
      "grad_norm": 1.375,
      "learning_rate": 0.00028141217412810905,
      "loss": 4.793,
      "mean_token_accuracy": 0.2314843565225601,
      "num_tokens": 113362413.0,
      "step": 54625
    },
    {
      "entropy": 5.745767545700073,
      "epoch": 4.955551523947751,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002813798307841973,
      "loss": 4.8484,
      "mean_token_accuracy": 0.23250950872898102,
      "num_tokens": 113373016.0,
      "step": 54630
    },
    {
      "entropy": 5.719369411468506,
      "epoch": 4.956005079825834,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002813474873083491,
      "loss": 4.8837,
      "mean_token_accuracy": 0.2250947192311287,
      "num_tokens": 113384798.0,
      "step": 54635
    },
    {
      "entropy": 5.786076402664184,
      "epoch": 4.956458635703918,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002813151437012331,
      "loss": 4.9125,
      "mean_token_accuracy": 0.23347986936569215,
      "num_tokens": 113396266.0,
      "step": 54640
    },
    {
      "entropy": 5.729993391036987,
      "epoch": 4.956912191582003,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002812827999635185,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22602323442697525,
      "num_tokens": 113406475.0,
      "step": 54645
    },
    {
      "entropy": 5.733987617492676,
      "epoch": 4.957365747460087,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028125045609587383,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23314228951931,
      "num_tokens": 113415835.0,
      "step": 54650
    },
    {
      "entropy": 5.7531383514404295,
      "epoch": 4.957819303338171,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002812181120989683,
      "loss": 4.8351,
      "mean_token_accuracy": 0.23115880340337752,
      "num_tokens": 113426359.0,
      "step": 54655
    },
    {
      "entropy": 5.7590407371521,
      "epoch": 4.9582728592162555,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002811857679734705,
      "loss": 4.9834,
      "mean_token_accuracy": 0.22270075529813765,
      "num_tokens": 113437467.0,
      "step": 54660
    },
    {
      "entropy": 5.746186971664429,
      "epoch": 4.95872641509434,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002811534237200494,
      "loss": 4.9382,
      "mean_token_accuracy": 0.22870000451803207,
      "num_tokens": 113447510.0,
      "step": 54665
    },
    {
      "entropy": 5.749554014205932,
      "epoch": 4.959179970972424,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002811210793393741,
      "loss": 4.8598,
      "mean_token_accuracy": 0.22169675379991532,
      "num_tokens": 113457559.0,
      "step": 54670
    },
    {
      "entropy": 5.799225568771362,
      "epoch": 4.959633526850508,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00028108873483211316,
      "loss": 4.9667,
      "mean_token_accuracy": 0.22600997984409332,
      "num_tokens": 113468589.0,
      "step": 54675
    },
    {
      "entropy": 5.7393317222595215,
      "epoch": 4.960087082728592,
      "grad_norm": 1.421875,
      "learning_rate": 0.00028105639019893573,
      "loss": 4.8604,
      "mean_token_accuracy": 0.23185402601957322,
      "num_tokens": 113478585.0,
      "step": 54680
    },
    {
      "entropy": 5.773086166381836,
      "epoch": 4.960540638606677,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028102404544051056,
      "loss": 4.857,
      "mean_token_accuracy": 0.23121411800384523,
      "num_tokens": 113487980.0,
      "step": 54685
    },
    {
      "entropy": 5.732555675506592,
      "epoch": 4.96099419448476,
      "grad_norm": 1.296875,
      "learning_rate": 0.00028099170055750667,
      "loss": 4.8705,
      "mean_token_accuracy": 0.23384444117546083,
      "num_tokens": 113498123.0,
      "step": 54690
    },
    {
      "entropy": 5.737059497833252,
      "epoch": 4.9614477503628445,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028095935555059277,
      "loss": 4.8412,
      "mean_token_accuracy": 0.229848712682724,
      "num_tokens": 113508685.0,
      "step": 54695
    },
    {
      "entropy": 5.7136390686035154,
      "epoch": 4.961901306240929,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00028092701042043794,
      "loss": 4.9046,
      "mean_token_accuracy": 0.23053636103868486,
      "num_tokens": 113519704.0,
      "step": 54700
    },
    {
      "entropy": 5.781048059463501,
      "epoch": 4.962354862119013,
      "grad_norm": 1.3125,
      "learning_rate": 0.000280894665167711,
      "loss": 4.9757,
      "mean_token_accuracy": 0.22466558516025542,
      "num_tokens": 113529770.0,
      "step": 54705
    },
    {
      "entropy": 5.6898702621459964,
      "epoch": 4.962808417997097,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002808623197930809,
      "loss": 4.7432,
      "mean_token_accuracy": 0.24297989904880524,
      "num_tokens": 113538667.0,
      "step": 54710
    },
    {
      "entropy": 5.749663877487182,
      "epoch": 4.9632619738751815,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002808299742972165,
      "loss": 4.8968,
      "mean_token_accuracy": 0.22718949466943741,
      "num_tokens": 113549420.0,
      "step": 54715
    },
    {
      "entropy": 5.746051836013794,
      "epoch": 4.963715529753266,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028079762868078685,
      "loss": 4.8866,
      "mean_token_accuracy": 0.2223042979836464,
      "num_tokens": 113559325.0,
      "step": 54720
    },
    {
      "entropy": 5.749191474914551,
      "epoch": 4.96416908563135,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002807652829444607,
      "loss": 4.8536,
      "mean_token_accuracy": 0.23041940480470657,
      "num_tokens": 113570152.0,
      "step": 54725
    },
    {
      "entropy": 5.695487213134766,
      "epoch": 4.964622641509434,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028073293708890697,
      "loss": 4.7959,
      "mean_token_accuracy": 0.2364317536354065,
      "num_tokens": 113580302.0,
      "step": 54730
    },
    {
      "entropy": 5.750128650665284,
      "epoch": 4.965076197387518,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028070059111479475,
      "loss": 4.9144,
      "mean_token_accuracy": 0.2229754686355591,
      "num_tokens": 113590731.0,
      "step": 54735
    },
    {
      "entropy": 5.767042207717895,
      "epoch": 4.965529753265602,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002806682450227928,
      "loss": 4.8784,
      "mean_token_accuracy": 0.23004196137189864,
      "num_tokens": 113600850.0,
      "step": 54740
    },
    {
      "entropy": 5.744909572601318,
      "epoch": 4.965983309143686,
      "grad_norm": 1.453125,
      "learning_rate": 0.00028063589881357007,
      "loss": 4.9006,
      "mean_token_accuracy": 0.23103539794683456,
      "num_tokens": 113611559.0,
      "step": 54745
    },
    {
      "entropy": 5.7201393127441404,
      "epoch": 4.9664368650217705,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002806035524877955,
      "loss": 4.8024,
      "mean_token_accuracy": 0.22886594384908676,
      "num_tokens": 113621172.0,
      "step": 54750
    },
    {
      "entropy": 5.76043930053711,
      "epoch": 4.966890420899855,
      "grad_norm": 1.375,
      "learning_rate": 0.0002805712060461381,
      "loss": 4.8802,
      "mean_token_accuracy": 0.22635751366615295,
      "num_tokens": 113631573.0,
      "step": 54755
    },
    {
      "entropy": 5.8265608787536625,
      "epoch": 4.967343976777939,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00028053885948926664,
      "loss": 4.9989,
      "mean_token_accuracy": 0.21801652759313583,
      "num_tokens": 113641990.0,
      "step": 54760
    },
    {
      "entropy": 5.778996515274048,
      "epoch": 4.967797532656023,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002805065128178503,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22831969559192658,
      "num_tokens": 113651937.0,
      "step": 54765
    },
    {
      "entropy": 5.7902360439300535,
      "epoch": 4.968251088534108,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028047416603255783,
      "loss": 4.9493,
      "mean_token_accuracy": 0.22351663559675217,
      "num_tokens": 113661150.0,
      "step": 54770
    },
    {
      "entropy": 5.739428615570068,
      "epoch": 4.968704644412192,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002804418191340582,
      "loss": 4.8565,
      "mean_token_accuracy": 0.2283768981695175,
      "num_tokens": 113671636.0,
      "step": 54775
    },
    {
      "entropy": 5.729180145263672,
      "epoch": 4.969158200290276,
      "grad_norm": 1.3125,
      "learning_rate": 0.00028040947212302033,
      "loss": 4.9505,
      "mean_token_accuracy": 0.2197368174791336,
      "num_tokens": 113682051.0,
      "step": 54780
    },
    {
      "entropy": 5.729323244094848,
      "epoch": 4.96961175616836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002803771250001132,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23344462960958481,
      "num_tokens": 113691682.0,
      "step": 54785
    },
    {
      "entropy": 5.843712949752808,
      "epoch": 4.970065312046444,
      "grad_norm": 1.390625,
      "learning_rate": 0.00028034477776600576,
      "loss": 4.974,
      "mean_token_accuracy": 0.22461781650781631,
      "num_tokens": 113702976.0,
      "step": 54790
    },
    {
      "entropy": 5.694817209243775,
      "epoch": 4.970518867924528,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00028031243042136696,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2369515225291252,
      "num_tokens": 113712871.0,
      "step": 54795
    },
    {
      "entropy": 5.715683794021606,
      "epoch": 4.970972423802612,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002802800829668657,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2248109444975853,
      "num_tokens": 113722425.0,
      "step": 54800
    },
    {
      "entropy": 5.763537120819092,
      "epoch": 4.9714259796806965,
      "grad_norm": 1.484375,
      "learning_rate": 0.00028024773540317105,
      "loss": 4.8922,
      "mean_token_accuracy": 0.23119031339883805,
      "num_tokens": 113731851.0,
      "step": 54805
    },
    {
      "entropy": 5.716351318359375,
      "epoch": 4.971879535558781,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00028021538773095184,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22205812036991118,
      "num_tokens": 113742649.0,
      "step": 54810
    },
    {
      "entropy": 5.712519693374634,
      "epoch": 4.972333091436865,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002801830399508771,
      "loss": 4.8707,
      "mean_token_accuracy": 0.22849228829145432,
      "num_tokens": 113752620.0,
      "step": 54815
    },
    {
      "entropy": 5.645544624328613,
      "epoch": 4.972786647314949,
      "grad_norm": 1.4375,
      "learning_rate": 0.00028015069206361584,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23658198863267899,
      "num_tokens": 113762165.0,
      "step": 54820
    },
    {
      "entropy": 5.774818658828735,
      "epoch": 4.973240203193034,
      "grad_norm": 1.375,
      "learning_rate": 0.00028011834406983696,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22925512343645096,
      "num_tokens": 113773342.0,
      "step": 54825
    },
    {
      "entropy": 5.734885311126709,
      "epoch": 4.973693759071118,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002800859959702093,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22588893324136733,
      "num_tokens": 113783159.0,
      "step": 54830
    },
    {
      "entropy": 5.735887908935547,
      "epoch": 4.974147314949201,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000280053647765402,
      "loss": 4.8759,
      "mean_token_accuracy": 0.22946197986602784,
      "num_tokens": 113794089.0,
      "step": 54835
    },
    {
      "entropy": 5.725896739959717,
      "epoch": 4.974600870827286,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00028002129945608404,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22295716106891633,
      "num_tokens": 113804459.0,
      "step": 54840
    },
    {
      "entropy": 5.6756456851959225,
      "epoch": 4.97505442670537,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002799889510429242,
      "loss": 4.8155,
      "mean_token_accuracy": 0.2404129758477211,
      "num_tokens": 113815032.0,
      "step": 54845
    },
    {
      "entropy": 5.714264011383056,
      "epoch": 4.975507982583454,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002799566025265917,
      "loss": 4.8568,
      "mean_token_accuracy": 0.22733055353164672,
      "num_tokens": 113825731.0,
      "step": 54850
    },
    {
      "entropy": 5.774758148193359,
      "epoch": 4.975961538461538,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002799242539077554,
      "loss": 4.8377,
      "mean_token_accuracy": 0.22968682199716567,
      "num_tokens": 113835752.0,
      "step": 54855
    },
    {
      "entropy": 5.774756669998169,
      "epoch": 4.976415094339623,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002798919051870842,
      "loss": 4.8776,
      "mean_token_accuracy": 0.2278080925345421,
      "num_tokens": 113845241.0,
      "step": 54860
    },
    {
      "entropy": 5.734928941726684,
      "epoch": 4.976868650217707,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002798595563652472,
      "loss": 4.9242,
      "mean_token_accuracy": 0.22398622184991837,
      "num_tokens": 113856066.0,
      "step": 54865
    },
    {
      "entropy": 5.685535287857055,
      "epoch": 4.977322206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002798272074429133,
      "loss": 4.8055,
      "mean_token_accuracy": 0.22851866781711577,
      "num_tokens": 113866532.0,
      "step": 54870
    },
    {
      "entropy": 5.686401510238648,
      "epoch": 4.977775761973875,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002797948584207515,
      "loss": 4.8357,
      "mean_token_accuracy": 0.22828309237957,
      "num_tokens": 113876197.0,
      "step": 54875
    },
    {
      "entropy": 5.728666496276856,
      "epoch": 4.97822931785196,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002797625092994309,
      "loss": 4.8666,
      "mean_token_accuracy": 0.22823870480060576,
      "num_tokens": 113886606.0,
      "step": 54880
    },
    {
      "entropy": 5.725904083251953,
      "epoch": 4.978682873730044,
      "grad_norm": 1.375,
      "learning_rate": 0.00027973016007962025,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2268198847770691,
      "num_tokens": 113897010.0,
      "step": 54885
    },
    {
      "entropy": 5.768582534790039,
      "epoch": 4.979136429608127,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002796978107619889,
      "loss": 4.8758,
      "mean_token_accuracy": 0.23651230037212373,
      "num_tokens": 113908762.0,
      "step": 54890
    },
    {
      "entropy": 5.739134979248047,
      "epoch": 4.979589985486212,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002796654613472054,
      "loss": 4.9214,
      "mean_token_accuracy": 0.22390402257442474,
      "num_tokens": 113919164.0,
      "step": 54895
    },
    {
      "entropy": 5.786234998703003,
      "epoch": 4.980043541364296,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002796331118359391,
      "loss": 4.9599,
      "mean_token_accuracy": 0.22539349943399428,
      "num_tokens": 113928975.0,
      "step": 54900
    },
    {
      "entropy": 5.72602334022522,
      "epoch": 4.98049709724238,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002796007622288588,
      "loss": 4.822,
      "mean_token_accuracy": 0.24012700617313384,
      "num_tokens": 113937884.0,
      "step": 54905
    },
    {
      "entropy": 5.765591335296631,
      "epoch": 4.980950653120464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002795684125266336,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23346254229545593,
      "num_tokens": 113948750.0,
      "step": 54910
    },
    {
      "entropy": 5.78844141960144,
      "epoch": 4.981404208998549,
      "grad_norm": 1.375,
      "learning_rate": 0.0002795360627299324,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2291623145341873,
      "num_tokens": 113960166.0,
      "step": 54915
    },
    {
      "entropy": 5.781020212173462,
      "epoch": 4.981857764876633,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027950371283942443,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22311656028032303,
      "num_tokens": 113971279.0,
      "step": 54920
    },
    {
      "entropy": 5.741877794265747,
      "epoch": 4.982311320754717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027947136285577846,
      "loss": 4.9188,
      "mean_token_accuracy": 0.23282488137483598,
      "num_tokens": 113980810.0,
      "step": 54925
    },
    {
      "entropy": 5.717445611953735,
      "epoch": 4.982764876632801,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027943901277966343,
      "loss": 4.8152,
      "mean_token_accuracy": 0.2307525783777237,
      "num_tokens": 113990773.0,
      "step": 54930
    },
    {
      "entropy": 5.742858362197876,
      "epoch": 4.983218432510886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002794066626117486,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2360479399561882,
      "num_tokens": 114001401.0,
      "step": 54935
    },
    {
      "entropy": 5.772112131118774,
      "epoch": 4.98367198838897,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002793743123527029,
      "loss": 4.8765,
      "mean_token_accuracy": 0.22532566338777543,
      "num_tokens": 114012382.0,
      "step": 54940
    },
    {
      "entropy": 5.74694128036499,
      "epoch": 4.984125544267053,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002793419620031952,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22198081016540527,
      "num_tokens": 114022395.0,
      "step": 54945
    },
    {
      "entropy": 5.725828075408936,
      "epoch": 4.984579100145138,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027930961156389473,
      "loss": 4.8459,
      "mean_token_accuracy": 0.2183324560523033,
      "num_tokens": 114032629.0,
      "step": 54950
    },
    {
      "entropy": 5.746423721313477,
      "epoch": 4.985032656023222,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027927726103547034,
      "loss": 4.8784,
      "mean_token_accuracy": 0.22919035255908965,
      "num_tokens": 114042642.0,
      "step": 54955
    },
    {
      "entropy": 5.764694738388061,
      "epoch": 4.985486211901306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002792449104185911,
      "loss": 4.9015,
      "mean_token_accuracy": 0.2203209936618805,
      "num_tokens": 114052972.0,
      "step": 54960
    },
    {
      "entropy": 5.735167694091797,
      "epoch": 4.98593976777939,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027921255971392607,
      "loss": 4.9584,
      "mean_token_accuracy": 0.21946705877780914,
      "num_tokens": 114064422.0,
      "step": 54965
    },
    {
      "entropy": 5.80221381187439,
      "epoch": 4.986393323657475,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002791802089221442,
      "loss": 4.8986,
      "mean_token_accuracy": 0.22269637435674666,
      "num_tokens": 114075611.0,
      "step": 54970
    },
    {
      "entropy": 5.811216163635254,
      "epoch": 4.986846879535559,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027914785804391455,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22919269502162934,
      "num_tokens": 114086686.0,
      "step": 54975
    },
    {
      "entropy": 5.755510759353638,
      "epoch": 4.987300435413643,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027911550707990614,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22961468398571014,
      "num_tokens": 114097968.0,
      "step": 54980
    },
    {
      "entropy": 5.692948722839356,
      "epoch": 4.9877539912917275,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027908315603078796,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22913177460432052,
      "num_tokens": 114108323.0,
      "step": 54985
    },
    {
      "entropy": 5.660962152481079,
      "epoch": 4.988207547169811,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00027905080489722914,
      "loss": 4.8382,
      "mean_token_accuracy": 0.22993566691875458,
      "num_tokens": 114118129.0,
      "step": 54990
    },
    {
      "entropy": 5.735419654846192,
      "epoch": 4.988661103047895,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002790184536798985,
      "loss": 4.9667,
      "mean_token_accuracy": 0.22079207748174667,
      "num_tokens": 114127426.0,
      "step": 54995
    },
    {
      "entropy": 5.698972988128662,
      "epoch": 4.989114658925979,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027898610237946537,
      "loss": 4.8292,
      "mean_token_accuracy": 0.23398627042770387,
      "num_tokens": 114137216.0,
      "step": 55000
    },
    {
      "entropy": 5.7144256114959715,
      "epoch": 4.989568214804064,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002789537509965986,
      "loss": 4.9085,
      "mean_token_accuracy": 0.22366634160280227,
      "num_tokens": 114148617.0,
      "step": 55005
    },
    {
      "entropy": 5.793098258972168,
      "epoch": 4.990021770682148,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002789213995319672,
      "loss": 4.909,
      "mean_token_accuracy": 0.22581785768270493,
      "num_tokens": 114159512.0,
      "step": 55010
    },
    {
      "entropy": 5.746531009674072,
      "epoch": 4.990475326560232,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002788890479862403,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23220344334840776,
      "num_tokens": 114169249.0,
      "step": 55015
    },
    {
      "entropy": 5.7935483932495115,
      "epoch": 4.9909288824383164,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027885669636008685,
      "loss": 4.9104,
      "mean_token_accuracy": 0.2260160118341446,
      "num_tokens": 114179622.0,
      "step": 55020
    },
    {
      "entropy": 5.713989114761352,
      "epoch": 4.991382438316401,
      "grad_norm": 1.5,
      "learning_rate": 0.000278824344654176,
      "loss": 4.7789,
      "mean_token_accuracy": 0.2392644926905632,
      "num_tokens": 114189712.0,
      "step": 55025
    },
    {
      "entropy": 5.788762474060059,
      "epoch": 4.991835994194485,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027879199286917665,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21471820622682572,
      "num_tokens": 114200074.0,
      "step": 55030
    },
    {
      "entropy": 5.743396139144897,
      "epoch": 4.992289550072569,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000278759641005758,
      "loss": 4.9044,
      "mean_token_accuracy": 0.22721798866987228,
      "num_tokens": 114209659.0,
      "step": 55035
    },
    {
      "entropy": 5.73289647102356,
      "epoch": 4.9927431059506535,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000278727289064589,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23142946511507034,
      "num_tokens": 114219765.0,
      "step": 55040
    },
    {
      "entropy": 5.750039052963257,
      "epoch": 4.993196661828737,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002786949370463387,
      "loss": 4.869,
      "mean_token_accuracy": 0.22703836113214493,
      "num_tokens": 114229654.0,
      "step": 55045
    },
    {
      "entropy": 5.745071363449097,
      "epoch": 4.993650217706821,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002786625849516762,
      "loss": 4.8756,
      "mean_token_accuracy": 0.2236989662051201,
      "num_tokens": 114240504.0,
      "step": 55050
    },
    {
      "entropy": 5.765949726104736,
      "epoch": 4.994103773584905,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002786302327812704,
      "loss": 4.9298,
      "mean_token_accuracy": 0.21185849308967591,
      "num_tokens": 114250748.0,
      "step": 55055
    },
    {
      "entropy": 5.782671403884888,
      "epoch": 4.99455732946299,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027859788053579055,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22130464017391205,
      "num_tokens": 114261655.0,
      "step": 55060
    },
    {
      "entropy": 5.722372436523438,
      "epoch": 4.995010885341074,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027856552821590557,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22793878018856048,
      "num_tokens": 114271193.0,
      "step": 55065
    },
    {
      "entropy": 5.720895195007325,
      "epoch": 4.995464441219158,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002785331758222846,
      "loss": 4.8732,
      "mean_token_accuracy": 0.22497608363628388,
      "num_tokens": 114281461.0,
      "step": 55070
    },
    {
      "entropy": 5.742755126953125,
      "epoch": 4.9959179970972425,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002785008233555967,
      "loss": 4.922,
      "mean_token_accuracy": 0.21810490787029266,
      "num_tokens": 114290865.0,
      "step": 55075
    },
    {
      "entropy": 5.796778774261474,
      "epoch": 4.996371552975327,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002784684708165107,
      "loss": 4.9218,
      "mean_token_accuracy": 0.22173603624105453,
      "num_tokens": 114301811.0,
      "step": 55080
    },
    {
      "entropy": 5.812321472167969,
      "epoch": 4.996825108853411,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027843611820569595,
      "loss": 4.901,
      "mean_token_accuracy": 0.22010699808597564,
      "num_tokens": 114313070.0,
      "step": 55085
    },
    {
      "entropy": 5.777211332321167,
      "epoch": 4.997278664731495,
      "grad_norm": 1.546875,
      "learning_rate": 0.00027840376552382145,
      "loss": 4.9355,
      "mean_token_accuracy": 0.21981727629899978,
      "num_tokens": 114323670.0,
      "step": 55090
    },
    {
      "entropy": 5.675423192977905,
      "epoch": 4.99773222060958,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027837141277155616,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22305832356214522,
      "num_tokens": 114334367.0,
      "step": 55095
    },
    {
      "entropy": 5.692883443832398,
      "epoch": 4.998185776487663,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027833905994956913,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23493298143148422,
      "num_tokens": 114344759.0,
      "step": 55100
    },
    {
      "entropy": 5.744275188446045,
      "epoch": 4.998639332365747,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002783067070585296,
      "loss": 4.855,
      "mean_token_accuracy": 0.220083786547184,
      "num_tokens": 114355200.0,
      "step": 55105
    },
    {
      "entropy": 5.732779216766358,
      "epoch": 4.9990928882438315,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027827435409910644,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2371242180466652,
      "num_tokens": 114365333.0,
      "step": 55110
    },
    {
      "entropy": 5.701779508590699,
      "epoch": 4.999546444121916,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002782420010719689,
      "loss": 4.8838,
      "mean_token_accuracy": 0.22834892570972443,
      "num_tokens": 114375965.0,
      "step": 55115
    },
    {
      "entropy": 5.769346237182617,
      "epoch": 5.0,
      "grad_norm": 1.46875,
      "learning_rate": 0.00027820964797778587,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22744541019201278,
      "num_tokens": 114386175.0,
      "step": 55120
    },
    {
      "entropy": 5.738558530807495,
      "epoch": 5.000453555878084,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027817729481722657,
      "loss": 4.8014,
      "mean_token_accuracy": 0.23650410026311874,
      "num_tokens": 114395139.0,
      "step": 55125
    },
    {
      "entropy": 5.725746393203735,
      "epoch": 5.0009071117561685,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027814494159095996,
      "loss": 4.7817,
      "mean_token_accuracy": 0.23808474838733673,
      "num_tokens": 114405378.0,
      "step": 55130
    },
    {
      "entropy": 5.758464002609253,
      "epoch": 5.001360667634253,
      "grad_norm": 1.375,
      "learning_rate": 0.00027811258829965515,
      "loss": 4.8712,
      "mean_token_accuracy": 0.23315038979053498,
      "num_tokens": 114415920.0,
      "step": 55135
    },
    {
      "entropy": 5.792632007598877,
      "epoch": 5.001814223512337,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002780802349439812,
      "loss": 4.9368,
      "mean_token_accuracy": 0.22020885199308396,
      "num_tokens": 114426177.0,
      "step": 55140
    },
    {
      "entropy": 5.778587055206299,
      "epoch": 5.002267779390421,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027804788152460723,
      "loss": 4.8439,
      "mean_token_accuracy": 0.2314525157213211,
      "num_tokens": 114436221.0,
      "step": 55145
    },
    {
      "entropy": 5.63762526512146,
      "epoch": 5.002721335268505,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002780155280422023,
      "loss": 4.7268,
      "mean_token_accuracy": 0.24252526313066483,
      "num_tokens": 114445828.0,
      "step": 55150
    },
    {
      "entropy": 5.750668716430664,
      "epoch": 5.003174891146589,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027798317449743553,
      "loss": 4.8672,
      "mean_token_accuracy": 0.22933160066604613,
      "num_tokens": 114454792.0,
      "step": 55155
    },
    {
      "entropy": 5.72804069519043,
      "epoch": 5.003628447024673,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002779508208909759,
      "loss": 4.8717,
      "mean_token_accuracy": 0.22909872978925705,
      "num_tokens": 114465357.0,
      "step": 55160
    },
    {
      "entropy": 5.754390478134155,
      "epoch": 5.0040820029027575,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002779184672234925,
      "loss": 4.8225,
      "mean_token_accuracy": 0.23363883793354034,
      "num_tokens": 114475947.0,
      "step": 55165
    },
    {
      "entropy": 5.735841560363769,
      "epoch": 5.004535558780842,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002778861134956545,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2320389121770859,
      "num_tokens": 114486335.0,
      "step": 55170
    },
    {
      "entropy": 5.80067400932312,
      "epoch": 5.004989114658926,
      "grad_norm": 1.4375,
      "learning_rate": 0.000277853759708131,
      "loss": 4.8898,
      "mean_token_accuracy": 0.23231588751077653,
      "num_tokens": 114495959.0,
      "step": 55175
    },
    {
      "entropy": 5.7826985836029055,
      "epoch": 5.00544267053701,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002778214058615909,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2241130217909813,
      "num_tokens": 114506328.0,
      "step": 55180
    },
    {
      "entropy": 5.719534397125244,
      "epoch": 5.005896226415095,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002777890519567035,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23599251210689545,
      "num_tokens": 114515368.0,
      "step": 55185
    },
    {
      "entropy": 5.6812536239624025,
      "epoch": 5.006349782293179,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027775669799413774,
      "loss": 4.8666,
      "mean_token_accuracy": 0.2277531683444977,
      "num_tokens": 114525894.0,
      "step": 55190
    },
    {
      "entropy": 5.7234029293060305,
      "epoch": 5.006803338171263,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027772434397456284,
      "loss": 4.936,
      "mean_token_accuracy": 0.22524888664484025,
      "num_tokens": 114535739.0,
      "step": 55195
    },
    {
      "entropy": 5.6939009666442875,
      "epoch": 5.0072568940493465,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027769198989864764,
      "loss": 4.7793,
      "mean_token_accuracy": 0.23808362632989882,
      "num_tokens": 114546519.0,
      "step": 55200
    },
    {
      "entropy": 5.757656049728394,
      "epoch": 5.007710449927431,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002776596357670616,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23847619742155074,
      "num_tokens": 114556523.0,
      "step": 55205
    },
    {
      "entropy": 5.808803939819336,
      "epoch": 5.008164005805515,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027762728158047353,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22562931180000306,
      "num_tokens": 114566255.0,
      "step": 55210
    },
    {
      "entropy": 5.741128873825073,
      "epoch": 5.008617561683599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027759492733955257,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23354620784521102,
      "num_tokens": 114576686.0,
      "step": 55215
    },
    {
      "entropy": 5.704391670227051,
      "epoch": 5.0090711175616836,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002775625730449679,
      "loss": 4.871,
      "mean_token_accuracy": 0.2223159447312355,
      "num_tokens": 114586606.0,
      "step": 55220
    },
    {
      "entropy": 5.7055024147033695,
      "epoch": 5.009524673439768,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027753021869738855,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2266133427619934,
      "num_tokens": 114598450.0,
      "step": 55225
    },
    {
      "entropy": 5.8288373947143555,
      "epoch": 5.009978229317852,
      "grad_norm": 1.421875,
      "learning_rate": 0.00027749786429748366,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2201400026679039,
      "num_tokens": 114608761.0,
      "step": 55230
    },
    {
      "entropy": 5.782835912704468,
      "epoch": 5.010431785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002774655098459223,
      "loss": 4.8481,
      "mean_token_accuracy": 0.22627829164266586,
      "num_tokens": 114619100.0,
      "step": 55235
    },
    {
      "entropy": 5.8146604061126705,
      "epoch": 5.010885341074021,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002774331553433735,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22782895267009734,
      "num_tokens": 114629594.0,
      "step": 55240
    },
    {
      "entropy": 5.703615379333496,
      "epoch": 5.011338896952105,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002774008007905065,
      "loss": 4.8011,
      "mean_token_accuracy": 0.22803122997283937,
      "num_tokens": 114639090.0,
      "step": 55245
    },
    {
      "entropy": 5.739463424682617,
      "epoch": 5.011792452830188,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027736844618799026,
      "loss": 4.8303,
      "mean_token_accuracy": 0.23195790201425553,
      "num_tokens": 114648707.0,
      "step": 55250
    },
    {
      "entropy": 5.706911611557007,
      "epoch": 5.0122460087082725,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002773360915364941,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2259377047419548,
      "num_tokens": 114659534.0,
      "step": 55255
    },
    {
      "entropy": 5.729710578918457,
      "epoch": 5.012699564586357,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002773037368366868,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23145507127046586,
      "num_tokens": 114670094.0,
      "step": 55260
    },
    {
      "entropy": 5.731448459625244,
      "epoch": 5.013153120464441,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002772713820892377,
      "loss": 4.884,
      "mean_token_accuracy": 0.23443845957517623,
      "num_tokens": 114680807.0,
      "step": 55265
    },
    {
      "entropy": 5.754070568084717,
      "epoch": 5.013606676342525,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002772390272948159,
      "loss": 4.8506,
      "mean_token_accuracy": 0.23046160936355592,
      "num_tokens": 114691334.0,
      "step": 55270
    },
    {
      "entropy": 5.705181837081909,
      "epoch": 5.01406023222061,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027720667245409035,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2182374879717827,
      "num_tokens": 114702519.0,
      "step": 55275
    },
    {
      "entropy": 5.717446136474609,
      "epoch": 5.014513788098694,
      "grad_norm": 1.375,
      "learning_rate": 0.00027717431756773024,
      "loss": 4.8374,
      "mean_token_accuracy": 0.2336589589715004,
      "num_tokens": 114712893.0,
      "step": 55280
    },
    {
      "entropy": 5.772413539886474,
      "epoch": 5.014967343976778,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002771419626364048,
      "loss": 4.9197,
      "mean_token_accuracy": 0.21459054946899414,
      "num_tokens": 114722101.0,
      "step": 55285
    },
    {
      "entropy": 5.782913541793823,
      "epoch": 5.015420899854862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027710960766078296,
      "loss": 4.9011,
      "mean_token_accuracy": 0.2233288198709488,
      "num_tokens": 114733184.0,
      "step": 55290
    },
    {
      "entropy": 5.730598068237304,
      "epoch": 5.015874455732947,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027707725264153386,
      "loss": 4.7428,
      "mean_token_accuracy": 0.23344592452049256,
      "num_tokens": 114743553.0,
      "step": 55295
    },
    {
      "entropy": 5.761511754989624,
      "epoch": 5.01632801161103,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027704489757932667,
      "loss": 4.8556,
      "mean_token_accuracy": 0.2341866448521614,
      "num_tokens": 114754655.0,
      "step": 55300
    },
    {
      "entropy": 5.77126407623291,
      "epoch": 5.016781567489114,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002770125424748305,
      "loss": 4.9222,
      "mean_token_accuracy": 0.21875771284103393,
      "num_tokens": 114764918.0,
      "step": 55305
    },
    {
      "entropy": 5.756645154953003,
      "epoch": 5.017235123367199,
      "grad_norm": 1.734375,
      "learning_rate": 0.0002769801873287145,
      "loss": 4.8504,
      "mean_token_accuracy": 0.22734280675649643,
      "num_tokens": 114775968.0,
      "step": 55310
    },
    {
      "entropy": 5.759877252578735,
      "epoch": 5.017688679245283,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027694783214164757,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22943220138549805,
      "num_tokens": 114786363.0,
      "step": 55315
    },
    {
      "entropy": 5.785551643371582,
      "epoch": 5.018142235123367,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002769154769142991,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22816265672445296,
      "num_tokens": 114797034.0,
      "step": 55320
    },
    {
      "entropy": 5.72124490737915,
      "epoch": 5.018595791001451,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002768831216473381,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23327614217996598,
      "num_tokens": 114806493.0,
      "step": 55325
    },
    {
      "entropy": 5.7297886371612545,
      "epoch": 5.019049346879536,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027685076634143354,
      "loss": 4.9234,
      "mean_token_accuracy": 0.21945569664239883,
      "num_tokens": 114816986.0,
      "step": 55330
    },
    {
      "entropy": 5.717803144454956,
      "epoch": 5.01950290275762,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002768184109972547,
      "loss": 4.861,
      "mean_token_accuracy": 0.22545979171991348,
      "num_tokens": 114827592.0,
      "step": 55335
    },
    {
      "entropy": 5.757610940933228,
      "epoch": 5.019956458635704,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027678605561547073,
      "loss": 4.8801,
      "mean_token_accuracy": 0.22836952656507492,
      "num_tokens": 114839080.0,
      "step": 55340
    },
    {
      "entropy": 5.787270450592041,
      "epoch": 5.020410014513788,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002767537001967506,
      "loss": 4.8386,
      "mean_token_accuracy": 0.23286975622177125,
      "num_tokens": 114848816.0,
      "step": 55345
    },
    {
      "entropy": 5.745906162261963,
      "epoch": 5.020863570391873,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027672134474176347,
      "loss": 4.8608,
      "mean_token_accuracy": 0.23199104517698288,
      "num_tokens": 114858781.0,
      "step": 55350
    },
    {
      "entropy": 5.708705949783325,
      "epoch": 5.021317126269956,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027668898925117856,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2270657643675804,
      "num_tokens": 114869257.0,
      "step": 55355
    },
    {
      "entropy": 5.6933221340179445,
      "epoch": 5.02177068214804,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002766566337256649,
      "loss": 4.8289,
      "mean_token_accuracy": 0.24605948477983475,
      "num_tokens": 114879878.0,
      "step": 55360
    },
    {
      "entropy": 5.769289207458496,
      "epoch": 5.022224238026125,
      "grad_norm": 1.515625,
      "learning_rate": 0.00027662427816589163,
      "loss": 4.9203,
      "mean_token_accuracy": 0.22494234144687653,
      "num_tokens": 114890253.0,
      "step": 55365
    },
    {
      "entropy": 5.779898023605346,
      "epoch": 5.022677793904209,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002765919225725279,
      "loss": 4.8754,
      "mean_token_accuracy": 0.22741008996963502,
      "num_tokens": 114900715.0,
      "step": 55370
    },
    {
      "entropy": 5.727593898773193,
      "epoch": 5.023131349782293,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002765595669462428,
      "loss": 4.8341,
      "mean_token_accuracy": 0.2265070840716362,
      "num_tokens": 114910663.0,
      "step": 55375
    },
    {
      "entropy": 5.771023464202881,
      "epoch": 5.023584905660377,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002765272112877054,
      "loss": 4.9575,
      "mean_token_accuracy": 0.22001923471689225,
      "num_tokens": 114920844.0,
      "step": 55380
    },
    {
      "entropy": 5.739112043380738,
      "epoch": 5.024038461538462,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000276494855597585,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22862472683191298,
      "num_tokens": 114931701.0,
      "step": 55385
    },
    {
      "entropy": 5.7420830726623535,
      "epoch": 5.024492017416546,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027646249987655047,
      "loss": 4.8741,
      "mean_token_accuracy": 0.230885049700737,
      "num_tokens": 114941764.0,
      "step": 55390
    },
    {
      "entropy": 5.721848726272583,
      "epoch": 5.02494557329463,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002764301441252712,
      "loss": 4.8073,
      "mean_token_accuracy": 0.23492862284183502,
      "num_tokens": 114952126.0,
      "step": 55395
    },
    {
      "entropy": 5.6930078029632565,
      "epoch": 5.0253991291727145,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027639778834441606,
      "loss": 4.814,
      "mean_token_accuracy": 0.23971620947122574,
      "num_tokens": 114961952.0,
      "step": 55400
    },
    {
      "entropy": 5.7890575408935545,
      "epoch": 5.025852685050798,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002763654325346544,
      "loss": 4.9075,
      "mean_token_accuracy": 0.22643909454345704,
      "num_tokens": 114971338.0,
      "step": 55405
    },
    {
      "entropy": 5.755130338668823,
      "epoch": 5.026306240928882,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027633307669665524,
      "loss": 4.9027,
      "mean_token_accuracy": 0.23183279484510422,
      "num_tokens": 114981792.0,
      "step": 55410
    },
    {
      "entropy": 5.717730236053467,
      "epoch": 5.026759796806966,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002763007208310876,
      "loss": 4.7663,
      "mean_token_accuracy": 0.23591516762971879,
      "num_tokens": 114992619.0,
      "step": 55415
    },
    {
      "entropy": 5.721197032928467,
      "epoch": 5.027213352685051,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027626836493862086,
      "loss": 4.816,
      "mean_token_accuracy": 0.23718929886817933,
      "num_tokens": 115003551.0,
      "step": 55420
    },
    {
      "entropy": 5.840988826751709,
      "epoch": 5.027666908563135,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027623600901992393,
      "loss": 4.9172,
      "mean_token_accuracy": 0.2186077803373337,
      "num_tokens": 115013155.0,
      "step": 55425
    },
    {
      "entropy": 5.788182353973388,
      "epoch": 5.028120464441219,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027620365307566607,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2214490458369255,
      "num_tokens": 115023673.0,
      "step": 55430
    },
    {
      "entropy": 5.764286613464355,
      "epoch": 5.0285740203193035,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002761712971065163,
      "loss": 4.9062,
      "mean_token_accuracy": 0.23025648593902587,
      "num_tokens": 115033597.0,
      "step": 55435
    },
    {
      "entropy": 5.726695013046265,
      "epoch": 5.029027576197388,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027613894111314384,
      "loss": 4.8022,
      "mean_token_accuracy": 0.23617892414331437,
      "num_tokens": 115043732.0,
      "step": 55440
    },
    {
      "entropy": 5.743031358718872,
      "epoch": 5.029481132075472,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027610658509621776,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23308717757463454,
      "num_tokens": 115054124.0,
      "step": 55445
    },
    {
      "entropy": 5.70295672416687,
      "epoch": 5.029934687953556,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002760742290564072,
      "loss": 4.7409,
      "mean_token_accuracy": 0.2472650796175003,
      "num_tokens": 115064513.0,
      "step": 55450
    },
    {
      "entropy": 5.744663858413697,
      "epoch": 5.03038824383164,
      "grad_norm": 1.59375,
      "learning_rate": 0.00027604187299438146,
      "loss": 4.9386,
      "mean_token_accuracy": 0.22697410881519317,
      "num_tokens": 115074060.0,
      "step": 55455
    },
    {
      "entropy": 5.793066692352295,
      "epoch": 5.030841799709724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00027600951691080937,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22257995158433913,
      "num_tokens": 115084615.0,
      "step": 55460
    },
    {
      "entropy": 5.767037200927734,
      "epoch": 5.031295355587808,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027597716080636026,
      "loss": 4.7699,
      "mean_token_accuracy": 0.234649957716465,
      "num_tokens": 115095899.0,
      "step": 55465
    },
    {
      "entropy": 5.695185279846191,
      "epoch": 5.031748911465892,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002759448046817033,
      "loss": 4.8104,
      "mean_token_accuracy": 0.22820169776678084,
      "num_tokens": 115107121.0,
      "step": 55470
    },
    {
      "entropy": 5.7642394542694095,
      "epoch": 5.032202467343977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002759124485375074,
      "loss": 4.914,
      "mean_token_accuracy": 0.2279193177819252,
      "num_tokens": 115117842.0,
      "step": 55475
    },
    {
      "entropy": 5.71409969329834,
      "epoch": 5.032656023222061,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002758800923744419,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23642177134752274,
      "num_tokens": 115127268.0,
      "step": 55480
    },
    {
      "entropy": 5.757118606567383,
      "epoch": 5.033109579100145,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027584773619317583,
      "loss": 4.7932,
      "mean_token_accuracy": 0.2366337701678276,
      "num_tokens": 115137102.0,
      "step": 55485
    },
    {
      "entropy": 5.770935106277466,
      "epoch": 5.0335631349782295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002758153799943784,
      "loss": 4.91,
      "mean_token_accuracy": 0.22573933601379395,
      "num_tokens": 115147822.0,
      "step": 55490
    },
    {
      "entropy": 5.866298484802246,
      "epoch": 5.034016690856314,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027578302377871867,
      "loss": 5.0584,
      "mean_token_accuracy": 0.21470645368099212,
      "num_tokens": 115158163.0,
      "step": 55495
    },
    {
      "entropy": 5.673317861557007,
      "epoch": 5.034470246734398,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027575066754686583,
      "loss": 4.842,
      "mean_token_accuracy": 0.22446615397930145,
      "num_tokens": 115169224.0,
      "step": 55500
    },
    {
      "entropy": 5.7015145778656,
      "epoch": 5.034923802612482,
      "grad_norm": 1.375,
      "learning_rate": 0.00027571831129948895,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23552487939596176,
      "num_tokens": 115179357.0,
      "step": 55505
    },
    {
      "entropy": 5.7792644023895265,
      "epoch": 5.035377358490566,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002756859550372573,
      "loss": 4.8656,
      "mean_token_accuracy": 0.22938445061445237,
      "num_tokens": 115190245.0,
      "step": 55510
    },
    {
      "entropy": 5.805054330825806,
      "epoch": 5.03583091436865,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002756535987608398,
      "loss": 4.9077,
      "mean_token_accuracy": 0.2248205080628395,
      "num_tokens": 115201106.0,
      "step": 55515
    },
    {
      "entropy": 5.7779792785644535,
      "epoch": 5.036284470246734,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002756212424709058,
      "loss": 4.8573,
      "mean_token_accuracy": 0.23268459886312484,
      "num_tokens": 115211822.0,
      "step": 55520
    },
    {
      "entropy": 5.753941440582276,
      "epoch": 5.0367380261248185,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027558888616812435,
      "loss": 4.8353,
      "mean_token_accuracy": 0.2354719504714012,
      "num_tokens": 115222662.0,
      "step": 55525
    },
    {
      "entropy": 5.784066915512085,
      "epoch": 5.037191582002903,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002755565298531645,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22625831812620162,
      "num_tokens": 115232534.0,
      "step": 55530
    },
    {
      "entropy": 5.794498157501221,
      "epoch": 5.037645137880987,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002755241735266956,
      "loss": 4.9339,
      "mean_token_accuracy": 0.2228977307677269,
      "num_tokens": 115243669.0,
      "step": 55535
    },
    {
      "entropy": 5.766648054122925,
      "epoch": 5.038098693759071,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002754918171893865,
      "loss": 4.8064,
      "mean_token_accuracy": 0.2409607321023941,
      "num_tokens": 115253482.0,
      "step": 55540
    },
    {
      "entropy": 5.732803916931152,
      "epoch": 5.0385522496371555,
      "grad_norm": 1.546875,
      "learning_rate": 0.00027545946084190666,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22112515419721604,
      "num_tokens": 115263037.0,
      "step": 55545
    },
    {
      "entropy": 5.788962459564209,
      "epoch": 5.03900580551524,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002754271044849249,
      "loss": 4.9133,
      "mean_token_accuracy": 0.2294054999947548,
      "num_tokens": 115273135.0,
      "step": 55550
    },
    {
      "entropy": 5.781568813323974,
      "epoch": 5.039459361393324,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027539474811911065,
      "loss": 4.88,
      "mean_token_accuracy": 0.22672174870967865,
      "num_tokens": 115283482.0,
      "step": 55555
    },
    {
      "entropy": 5.7130285739898685,
      "epoch": 5.0399129172714074,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002753623917451329,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2360398530960083,
      "num_tokens": 115293855.0,
      "step": 55560
    },
    {
      "entropy": 5.739299488067627,
      "epoch": 5.040366473149492,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002753300353636607,
      "loss": 4.8898,
      "mean_token_accuracy": 0.23185523301362992,
      "num_tokens": 115304162.0,
      "step": 55565
    },
    {
      "entropy": 5.792537212371826,
      "epoch": 5.040820029027576,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027529767897536327,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22389875948429108,
      "num_tokens": 115315108.0,
      "step": 55570
    },
    {
      "entropy": 5.749419069290161,
      "epoch": 5.04127358490566,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027526532258090986,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2237188011407852,
      "num_tokens": 115325434.0,
      "step": 55575
    },
    {
      "entropy": 5.759827756881714,
      "epoch": 5.0417271407837445,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027523296618096947,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22254514545202256,
      "num_tokens": 115335410.0,
      "step": 55580
    },
    {
      "entropy": 5.745727682113648,
      "epoch": 5.042180696661829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027520060977621126,
      "loss": 4.7906,
      "mean_token_accuracy": 0.23710468262434006,
      "num_tokens": 115345838.0,
      "step": 55585
    },
    {
      "entropy": 5.692580127716065,
      "epoch": 5.042634252539913,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002751682533673044,
      "loss": 4.801,
      "mean_token_accuracy": 0.23808070719242097,
      "num_tokens": 115355988.0,
      "step": 55590
    },
    {
      "entropy": 5.662794733047486,
      "epoch": 5.043087808417997,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002751358969549181,
      "loss": 4.8175,
      "mean_token_accuracy": 0.22523844242095947,
      "num_tokens": 115366942.0,
      "step": 55595
    },
    {
      "entropy": 5.744259929656982,
      "epoch": 5.043541364296082,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027510354053972123,
      "loss": 4.7675,
      "mean_token_accuracy": 0.25168678611516954,
      "num_tokens": 115376948.0,
      "step": 55600
    },
    {
      "entropy": 5.7882836818695065,
      "epoch": 5.043994920174166,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027507118412238324,
      "loss": 4.9051,
      "mean_token_accuracy": 0.23352132439613343,
      "num_tokens": 115387831.0,
      "step": 55605
    },
    {
      "entropy": 5.784084033966065,
      "epoch": 5.044448476052249,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002750388277035732,
      "loss": 4.8593,
      "mean_token_accuracy": 0.2322611078619957,
      "num_tokens": 115397511.0,
      "step": 55610
    },
    {
      "entropy": 5.697371196746826,
      "epoch": 5.0449020319303335,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002750064712839601,
      "loss": 4.8235,
      "mean_token_accuracy": 0.23336742222309112,
      "num_tokens": 115408049.0,
      "step": 55615
    },
    {
      "entropy": 5.744609403610229,
      "epoch": 5.045355587808418,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027497411486421323,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23034833371639252,
      "num_tokens": 115418811.0,
      "step": 55620
    },
    {
      "entropy": 5.760235071182251,
      "epoch": 5.045809143686502,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027494175844500164,
      "loss": 4.8079,
      "mean_token_accuracy": 0.23163919299840927,
      "num_tokens": 115428324.0,
      "step": 55625
    },
    {
      "entropy": 5.745297193527222,
      "epoch": 5.046262699564586,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002749094020269945,
      "loss": 4.82,
      "mean_token_accuracy": 0.232093209028244,
      "num_tokens": 115437891.0,
      "step": 55630
    },
    {
      "entropy": 5.724225044250488,
      "epoch": 5.0467162554426706,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000274877045610861,
      "loss": 4.8486,
      "mean_token_accuracy": 0.22838467359542847,
      "num_tokens": 115448100.0,
      "step": 55635
    },
    {
      "entropy": 5.71543607711792,
      "epoch": 5.047169811320755,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002748446891972701,
      "loss": 4.8373,
      "mean_token_accuracy": 0.23660587668418884,
      "num_tokens": 115458530.0,
      "step": 55640
    },
    {
      "entropy": 5.733639287948608,
      "epoch": 5.047623367198839,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027481233278689124,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2300538793206215,
      "num_tokens": 115469222.0,
      "step": 55645
    },
    {
      "entropy": 5.788974618911743,
      "epoch": 5.048076923076923,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002747799763803932,
      "loss": 4.9041,
      "mean_token_accuracy": 0.22262870520353317,
      "num_tokens": 115478704.0,
      "step": 55650
    },
    {
      "entropy": 5.750610733032227,
      "epoch": 5.048530478955008,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027474761997844536,
      "loss": 4.8825,
      "mean_token_accuracy": 0.2257686361670494,
      "num_tokens": 115488430.0,
      "step": 55655
    },
    {
      "entropy": 5.744847917556763,
      "epoch": 5.048984034833092,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002747152635817169,
      "loss": 4.888,
      "mean_token_accuracy": 0.22554240077733995,
      "num_tokens": 115499010.0,
      "step": 55660
    },
    {
      "entropy": 5.735460186004639,
      "epoch": 5.049437590711175,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002746829071908768,
      "loss": 4.8416,
      "mean_token_accuracy": 0.23693214505910873,
      "num_tokens": 115508278.0,
      "step": 55665
    },
    {
      "entropy": 5.735982847213745,
      "epoch": 5.0498911465892595,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002746505508065942,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23376896977424622,
      "num_tokens": 115517977.0,
      "step": 55670
    },
    {
      "entropy": 5.728127145767212,
      "epoch": 5.050344702467344,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027461819442953846,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2300807535648346,
      "num_tokens": 115528533.0,
      "step": 55675
    },
    {
      "entropy": 5.728024959564209,
      "epoch": 5.050798258345428,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002745858380603784,
      "loss": 4.8516,
      "mean_token_accuracy": 0.23299282938241958,
      "num_tokens": 115537930.0,
      "step": 55680
    },
    {
      "entropy": 5.784687519073486,
      "epoch": 5.051251814223512,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002745534816997834,
      "loss": 4.9259,
      "mean_token_accuracy": 0.21683653742074965,
      "num_tokens": 115547436.0,
      "step": 55685
    },
    {
      "entropy": 5.712627410888672,
      "epoch": 5.051705370101597,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027452112534842246,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23708096593618394,
      "num_tokens": 115559313.0,
      "step": 55690
    },
    {
      "entropy": 5.808770799636841,
      "epoch": 5.052158925979681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002744887690069648,
      "loss": 4.8723,
      "mean_token_accuracy": 0.2316515862941742,
      "num_tokens": 115568773.0,
      "step": 55695
    },
    {
      "entropy": 5.75312237739563,
      "epoch": 5.052612481857765,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027445641267607947,
      "loss": 4.8899,
      "mean_token_accuracy": 0.22664265483617782,
      "num_tokens": 115579402.0,
      "step": 55700
    },
    {
      "entropy": 5.714404296875,
      "epoch": 5.053066037735849,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027442405635643574,
      "loss": 4.8997,
      "mean_token_accuracy": 0.23299072682857513,
      "num_tokens": 115588747.0,
      "step": 55705
    },
    {
      "entropy": 5.720153331756592,
      "epoch": 5.053519593613934,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002743917000487027,
      "loss": 4.9333,
      "mean_token_accuracy": 0.22674206048250198,
      "num_tokens": 115599927.0,
      "step": 55710
    },
    {
      "entropy": 5.80821123123169,
      "epoch": 5.053973149492017,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027435934375354934,
      "loss": 4.9445,
      "mean_token_accuracy": 0.22438188046216964,
      "num_tokens": 115610120.0,
      "step": 55715
    },
    {
      "entropy": 5.750439167022705,
      "epoch": 5.054426705370101,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027432698747164496,
      "loss": 4.8704,
      "mean_token_accuracy": 0.22260251194238662,
      "num_tokens": 115622039.0,
      "step": 55720
    },
    {
      "entropy": 5.7174904346466064,
      "epoch": 5.054880261248186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027429463120365874,
      "loss": 4.8281,
      "mean_token_accuracy": 0.22496728599071503,
      "num_tokens": 115633527.0,
      "step": 55725
    },
    {
      "entropy": 5.722726583480835,
      "epoch": 5.05533381712627,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002742622749502596,
      "loss": 4.8547,
      "mean_token_accuracy": 0.22711869478225707,
      "num_tokens": 115644363.0,
      "step": 55730
    },
    {
      "entropy": 5.816569995880127,
      "epoch": 5.055787373004354,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002742299187121168,
      "loss": 4.9428,
      "mean_token_accuracy": 0.2142679363489151,
      "num_tokens": 115654313.0,
      "step": 55735
    },
    {
      "entropy": 5.763773250579834,
      "epoch": 5.056240928882438,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027419756248989955,
      "loss": 4.784,
      "mean_token_accuracy": 0.2376966282725334,
      "num_tokens": 115664724.0,
      "step": 55740
    },
    {
      "entropy": 5.734986162185669,
      "epoch": 5.056694484760523,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002741652062842769,
      "loss": 4.8282,
      "mean_token_accuracy": 0.23053455650806426,
      "num_tokens": 115674037.0,
      "step": 55745
    },
    {
      "entropy": 5.746109247207642,
      "epoch": 5.057148040638607,
      "grad_norm": 1.359375,
      "learning_rate": 0.000274132850095918,
      "loss": 4.8229,
      "mean_token_accuracy": 0.2292885035276413,
      "num_tokens": 115684431.0,
      "step": 55750
    },
    {
      "entropy": 5.735474252700806,
      "epoch": 5.057601596516691,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027410049392549197,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23018597960472106,
      "num_tokens": 115695293.0,
      "step": 55755
    },
    {
      "entropy": 5.723589038848877,
      "epoch": 5.058055152394775,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027406813777366794,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23303620368242264,
      "num_tokens": 115705906.0,
      "step": 55760
    },
    {
      "entropy": 5.751614809036255,
      "epoch": 5.058508708272859,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027403578164111517,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23822540640830994,
      "num_tokens": 115716157.0,
      "step": 55765
    },
    {
      "entropy": 5.842559337615967,
      "epoch": 5.058962264150943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002740034255285025,
      "loss": 4.9229,
      "mean_token_accuracy": 0.22663590162992478,
      "num_tokens": 115727691.0,
      "step": 55770
    },
    {
      "entropy": 5.7364896774292,
      "epoch": 5.059415820029027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002739710694364994,
      "loss": 4.8255,
      "mean_token_accuracy": 0.22573816776275635,
      "num_tokens": 115738984.0,
      "step": 55775
    },
    {
      "entropy": 5.7207153797149655,
      "epoch": 5.059869375907112,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027393871336577476,
      "loss": 4.816,
      "mean_token_accuracy": 0.23041386157274246,
      "num_tokens": 115749040.0,
      "step": 55780
    },
    {
      "entropy": 5.7692817687988285,
      "epoch": 5.060322931785196,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027390635731699774,
      "loss": 4.9471,
      "mean_token_accuracy": 0.2152276709675789,
      "num_tokens": 115760125.0,
      "step": 55785
    },
    {
      "entropy": 5.7422727108001705,
      "epoch": 5.06077648766328,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002738740012908376,
      "loss": 4.8752,
      "mean_token_accuracy": 0.23054446876049042,
      "num_tokens": 115770780.0,
      "step": 55790
    },
    {
      "entropy": 5.770628261566162,
      "epoch": 5.061230043541364,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027384164528796345,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23347513228654862,
      "num_tokens": 115782051.0,
      "step": 55795
    },
    {
      "entropy": 5.678454828262329,
      "epoch": 5.061683599419449,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002738092893090443,
      "loss": 4.781,
      "mean_token_accuracy": 0.23388156145811081,
      "num_tokens": 115791692.0,
      "step": 55800
    },
    {
      "entropy": 5.743256998062134,
      "epoch": 5.062137155297533,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002737769333547494,
      "loss": 4.8112,
      "mean_token_accuracy": 0.24206740409135818,
      "num_tokens": 115802738.0,
      "step": 55805
    },
    {
      "entropy": 5.779575490951538,
      "epoch": 5.062590711175617,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002737445774257478,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2250595435500145,
      "num_tokens": 115813073.0,
      "step": 55810
    },
    {
      "entropy": 5.812508249282837,
      "epoch": 5.063044267053701,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002737122215227087,
      "loss": 4.9162,
      "mean_token_accuracy": 0.22151711881160735,
      "num_tokens": 115823976.0,
      "step": 55815
    },
    {
      "entropy": 5.750986862182617,
      "epoch": 5.063497822931785,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002736798656463011,
      "loss": 4.8218,
      "mean_token_accuracy": 0.23006895780563355,
      "num_tokens": 115833332.0,
      "step": 55820
    },
    {
      "entropy": 5.728865242004394,
      "epoch": 5.063951378809869,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027364750979719433,
      "loss": 4.826,
      "mean_token_accuracy": 0.2320621594786644,
      "num_tokens": 115844053.0,
      "step": 55825
    },
    {
      "entropy": 5.732742643356323,
      "epoch": 5.064404934687953,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002736151539760573,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23828265070915222,
      "num_tokens": 115854860.0,
      "step": 55830
    },
    {
      "entropy": 5.746911334991455,
      "epoch": 5.064858490566038,
      "grad_norm": 1.59375,
      "learning_rate": 0.00027358279818355926,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23527501970529557,
      "num_tokens": 115865523.0,
      "step": 55835
    },
    {
      "entropy": 5.7386455059051515,
      "epoch": 5.065312046444122,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027355044242036934,
      "loss": 4.8642,
      "mean_token_accuracy": 0.23048104494810104,
      "num_tokens": 115876033.0,
      "step": 55840
    },
    {
      "entropy": 5.773918724060058,
      "epoch": 5.065765602322206,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002735180866871567,
      "loss": 4.902,
      "mean_token_accuracy": 0.2334411308169365,
      "num_tokens": 115886575.0,
      "step": 55845
    },
    {
      "entropy": 5.7499024868011475,
      "epoch": 5.0662191582002905,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002734857309845903,
      "loss": 4.8806,
      "mean_token_accuracy": 0.22212490141391755,
      "num_tokens": 115897628.0,
      "step": 55850
    },
    {
      "entropy": 5.729778242111206,
      "epoch": 5.066672714078375,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002734533753133394,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23299334645271302,
      "num_tokens": 115908472.0,
      "step": 55855
    },
    {
      "entropy": 5.720535707473755,
      "epoch": 5.067126269956459,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027342101967407306,
      "loss": 4.7747,
      "mean_token_accuracy": 0.23420990109443665,
      "num_tokens": 115918798.0,
      "step": 55860
    },
    {
      "entropy": 5.718428754806519,
      "epoch": 5.067579825834543,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002733886640674605,
      "loss": 4.7936,
      "mean_token_accuracy": 0.2346283122897148,
      "num_tokens": 115928825.0,
      "step": 55865
    },
    {
      "entropy": 5.7296537399292,
      "epoch": 5.068033381712627,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002733563084941707,
      "loss": 4.8368,
      "mean_token_accuracy": 0.23671133518218995,
      "num_tokens": 115939315.0,
      "step": 55870
    },
    {
      "entropy": 5.723471164703369,
      "epoch": 5.068486937590711,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027332395295487296,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2291862666606903,
      "num_tokens": 115949870.0,
      "step": 55875
    },
    {
      "entropy": 5.825713729858398,
      "epoch": 5.068940493468795,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027329159745023625,
      "loss": 5.0007,
      "mean_token_accuracy": 0.2145875498652458,
      "num_tokens": 115961108.0,
      "step": 55880
    },
    {
      "entropy": 5.734103012084961,
      "epoch": 5.069394049346879,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027325924198092974,
      "loss": 4.8448,
      "mean_token_accuracy": 0.23232749253511428,
      "num_tokens": 115972032.0,
      "step": 55885
    },
    {
      "entropy": 5.747336673736572,
      "epoch": 5.069847605224964,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002732268865476226,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23231214731931688,
      "num_tokens": 115982819.0,
      "step": 55890
    },
    {
      "entropy": 5.7494298934936525,
      "epoch": 5.070301161103048,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002731945311509838,
      "loss": 4.861,
      "mean_token_accuracy": 0.23112165182828903,
      "num_tokens": 115992856.0,
      "step": 55895
    },
    {
      "entropy": 5.760613346099854,
      "epoch": 5.070754716981132,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002731621757916826,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22861369401216508,
      "num_tokens": 116004122.0,
      "step": 55900
    },
    {
      "entropy": 5.720883893966675,
      "epoch": 5.0712082728592165,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000273129820470388,
      "loss": 4.795,
      "mean_token_accuracy": 0.2385830655694008,
      "num_tokens": 116014238.0,
      "step": 55905
    },
    {
      "entropy": 5.665817642211914,
      "epoch": 5.071661828737301,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027309746518776936,
      "loss": 4.7072,
      "mean_token_accuracy": 0.24023172557353972,
      "num_tokens": 116024310.0,
      "step": 55910
    },
    {
      "entropy": 5.76065149307251,
      "epoch": 5.072115384615385,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002730651099444955,
      "loss": 4.8337,
      "mean_token_accuracy": 0.23565730601549148,
      "num_tokens": 116034648.0,
      "step": 55915
    },
    {
      "entropy": 5.74302110671997,
      "epoch": 5.072568940493468,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002730327547412357,
      "loss": 4.8626,
      "mean_token_accuracy": 0.23124008178710936,
      "num_tokens": 116045178.0,
      "step": 55920
    },
    {
      "entropy": 5.723499870300293,
      "epoch": 5.073022496371553,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027300039957865903,
      "loss": 4.8865,
      "mean_token_accuracy": 0.2327123761177063,
      "num_tokens": 116056429.0,
      "step": 55925
    },
    {
      "entropy": 5.690951824188232,
      "epoch": 5.073476052249637,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002729680444574346,
      "loss": 4.834,
      "mean_token_accuracy": 0.23467811048030854,
      "num_tokens": 116066151.0,
      "step": 55930
    },
    {
      "entropy": 5.761924362182617,
      "epoch": 5.073929608127721,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027293568937823156,
      "loss": 4.8535,
      "mean_token_accuracy": 0.22537243664264678,
      "num_tokens": 116075330.0,
      "step": 55935
    },
    {
      "entropy": 5.773995399475098,
      "epoch": 5.0743831640058055,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002729033343417189,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22883452773094176,
      "num_tokens": 116086108.0,
      "step": 55940
    },
    {
      "entropy": 5.733337783813477,
      "epoch": 5.07483671988389,
      "grad_norm": 1.40625,
      "learning_rate": 0.00027287097934856593,
      "loss": 4.8554,
      "mean_token_accuracy": 0.22132653445005418,
      "num_tokens": 116097046.0,
      "step": 55945
    },
    {
      "entropy": 5.701938199996948,
      "epoch": 5.075290275761974,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027283862439944157,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23305839598178862,
      "num_tokens": 116108048.0,
      "step": 55950
    },
    {
      "entropy": 5.769516086578369,
      "epoch": 5.075743831640058,
      "grad_norm": 1.515625,
      "learning_rate": 0.000272806269495015,
      "loss": 4.8949,
      "mean_token_accuracy": 0.2336965024471283,
      "num_tokens": 116117417.0,
      "step": 55955
    },
    {
      "entropy": 5.85616774559021,
      "epoch": 5.0761973875181425,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00027277391463595543,
      "loss": 5.0027,
      "mean_token_accuracy": 0.21369536817073823,
      "num_tokens": 116128462.0,
      "step": 55960
    },
    {
      "entropy": 5.8750985622406,
      "epoch": 5.076650943396227,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002727415598229318,
      "loss": 4.956,
      "mean_token_accuracy": 0.21992797553539276,
      "num_tokens": 116139908.0,
      "step": 55965
    },
    {
      "entropy": 5.804120826721191,
      "epoch": 5.07710449927431,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002727092050566133,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23008720427751542,
      "num_tokens": 116149310.0,
      "step": 55970
    },
    {
      "entropy": 5.7399474620819095,
      "epoch": 5.0775580551523944,
      "grad_norm": 1.296875,
      "learning_rate": 0.000272676850337669,
      "loss": 4.839,
      "mean_token_accuracy": 0.23262473493814467,
      "num_tokens": 116160010.0,
      "step": 55975
    },
    {
      "entropy": 5.726359558105469,
      "epoch": 5.078011611030479,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002726444956667681,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22757839560508727,
      "num_tokens": 116170042.0,
      "step": 55980
    },
    {
      "entropy": 5.7398316860198975,
      "epoch": 5.078465166908563,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002726121410445796,
      "loss": 4.8626,
      "mean_token_accuracy": 0.2310586079955101,
      "num_tokens": 116180842.0,
      "step": 55985
    },
    {
      "entropy": 5.7975140571594235,
      "epoch": 5.078918722786647,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002725797864717726,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22725329399108887,
      "num_tokens": 116191496.0,
      "step": 55990
    },
    {
      "entropy": 5.715862131118774,
      "epoch": 5.0793722786647315,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002725474319490162,
      "loss": 4.8115,
      "mean_token_accuracy": 0.2371105819940567,
      "num_tokens": 116201451.0,
      "step": 55995
    },
    {
      "entropy": 5.7311443328857425,
      "epoch": 5.079825834542816,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002725150774769796,
      "loss": 4.7761,
      "mean_token_accuracy": 0.22979567050933838,
      "num_tokens": 116211327.0,
      "step": 56000
    },
    {
      "entropy": 5.791390895843506,
      "epoch": 5.0802793904209,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002724827230563319,
      "loss": 4.8919,
      "mean_token_accuracy": 0.23156944960355758,
      "num_tokens": 116221676.0,
      "step": 56005
    },
    {
      "entropy": 5.740667200088501,
      "epoch": 5.080732946298984,
      "grad_norm": 1.4375,
      "learning_rate": 0.00027245036868774205,
      "loss": 4.8613,
      "mean_token_accuracy": 0.22652396261692048,
      "num_tokens": 116232778.0,
      "step": 56010
    },
    {
      "entropy": 5.719668531417847,
      "epoch": 5.081186502177069,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002724180143718792,
      "loss": 4.76,
      "mean_token_accuracy": 0.23593908697366714,
      "num_tokens": 116242198.0,
      "step": 56015
    },
    {
      "entropy": 5.691205453872681,
      "epoch": 5.081640058055152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027238566010941243,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23455014675855637,
      "num_tokens": 116252171.0,
      "step": 56020
    },
    {
      "entropy": 5.664731979370117,
      "epoch": 5.082093613933236,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000272353305901011,
      "loss": 4.7253,
      "mean_token_accuracy": 0.23733261674642564,
      "num_tokens": 116263014.0,
      "step": 56025
    },
    {
      "entropy": 5.712905311584473,
      "epoch": 5.0825471698113205,
      "grad_norm": 1.25,
      "learning_rate": 0.0002723209517473438,
      "loss": 4.8434,
      "mean_token_accuracy": 0.22905875146389007,
      "num_tokens": 116273126.0,
      "step": 56030
    },
    {
      "entropy": 5.747695779800415,
      "epoch": 5.083000725689405,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027228859764908,
      "loss": 4.8534,
      "mean_token_accuracy": 0.23206622749567032,
      "num_tokens": 116283174.0,
      "step": 56035
    },
    {
      "entropy": 5.743677377700806,
      "epoch": 5.083454281567489,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00027225624360688864,
      "loss": 4.8741,
      "mean_token_accuracy": 0.22948177605867387,
      "num_tokens": 116293015.0,
      "step": 56040
    },
    {
      "entropy": 5.754269552230835,
      "epoch": 5.083907837445573,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000272223889621439,
      "loss": 4.9171,
      "mean_token_accuracy": 0.22636842876672744,
      "num_tokens": 116304578.0,
      "step": 56045
    },
    {
      "entropy": 5.751530981063842,
      "epoch": 5.0843613933236576,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002721915356933999,
      "loss": 4.8621,
      "mean_token_accuracy": 0.2308286413550377,
      "num_tokens": 116314666.0,
      "step": 56050
    },
    {
      "entropy": 5.762468957901001,
      "epoch": 5.084814949201742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027215918182344064,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23612837344408036,
      "num_tokens": 116324972.0,
      "step": 56055
    },
    {
      "entropy": 5.740555667877198,
      "epoch": 5.085268505079826,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002721268280122302,
      "loss": 4.8335,
      "mean_token_accuracy": 0.2269410490989685,
      "num_tokens": 116335653.0,
      "step": 56060
    },
    {
      "entropy": 5.833990144729614,
      "epoch": 5.08572206095791,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002720944742604377,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22502049505710603,
      "num_tokens": 116345522.0,
      "step": 56065
    },
    {
      "entropy": 5.733820962905884,
      "epoch": 5.086175616835995,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00027206212056873213,
      "loss": 4.8486,
      "mean_token_accuracy": 0.23043663501739503,
      "num_tokens": 116354889.0,
      "step": 56070
    },
    {
      "entropy": 5.6552086353302,
      "epoch": 5.086629172714078,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002720297669377828,
      "loss": 4.7781,
      "mean_token_accuracy": 0.24026160389184953,
      "num_tokens": 116364619.0,
      "step": 56075
    },
    {
      "entropy": 5.660520601272583,
      "epoch": 5.087082728592162,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027199741336825857,
      "loss": 4.79,
      "mean_token_accuracy": 0.23506757318973542,
      "num_tokens": 116374346.0,
      "step": 56080
    },
    {
      "entropy": 5.754910802841186,
      "epoch": 5.0875362844702465,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002719650598608285,
      "loss": 4.8967,
      "mean_token_accuracy": 0.23466889858245848,
      "num_tokens": 116385780.0,
      "step": 56085
    },
    {
      "entropy": 5.813379716873169,
      "epoch": 5.087989840348331,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002719327064161619,
      "loss": 4.8363,
      "mean_token_accuracy": 0.22695365101099013,
      "num_tokens": 116395232.0,
      "step": 56090
    },
    {
      "entropy": 5.737285995483399,
      "epoch": 5.088443396226415,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002719003530349277,
      "loss": 4.8283,
      "mean_token_accuracy": 0.2343090444803238,
      "num_tokens": 116404478.0,
      "step": 56095
    },
    {
      "entropy": 5.727014446258545,
      "epoch": 5.088896952104499,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002718679997177949,
      "loss": 4.7759,
      "mean_token_accuracy": 0.24196468591690062,
      "num_tokens": 116414493.0,
      "step": 56100
    },
    {
      "entropy": 5.791533994674682,
      "epoch": 5.089350507982584,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00027183564646543273,
      "loss": 4.9083,
      "mean_token_accuracy": 0.21954996436834334,
      "num_tokens": 116424722.0,
      "step": 56105
    },
    {
      "entropy": 5.662182950973511,
      "epoch": 5.089804063860668,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027180329327851024,
      "loss": 4.7687,
      "mean_token_accuracy": 0.23162221163511276,
      "num_tokens": 116435639.0,
      "step": 56110
    },
    {
      "entropy": 5.65686674118042,
      "epoch": 5.090257619738752,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027177094015769644,
      "loss": 4.7626,
      "mean_token_accuracy": 0.22824383080005645,
      "num_tokens": 116444908.0,
      "step": 56115
    },
    {
      "entropy": 5.736851835250855,
      "epoch": 5.090711175616836,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027173858710366033,
      "loss": 4.8028,
      "mean_token_accuracy": 0.23687777668237686,
      "num_tokens": 116454965.0,
      "step": 56120
    },
    {
      "entropy": 5.7180845737457275,
      "epoch": 5.09116473149492,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027170623411707114,
      "loss": 4.793,
      "mean_token_accuracy": 0.23011509329080582,
      "num_tokens": 116465861.0,
      "step": 56125
    },
    {
      "entropy": 5.747059059143067,
      "epoch": 5.091618287373004,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027167388119859787,
      "loss": 4.9401,
      "mean_token_accuracy": 0.22339320033788682,
      "num_tokens": 116476367.0,
      "step": 56130
    },
    {
      "entropy": 5.725572347640991,
      "epoch": 5.092071843251088,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027164152834890953,
      "loss": 4.8504,
      "mean_token_accuracy": 0.231588913500309,
      "num_tokens": 116486492.0,
      "step": 56135
    },
    {
      "entropy": 5.713303327560425,
      "epoch": 5.092525399129173,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027160917556867534,
      "loss": 4.804,
      "mean_token_accuracy": 0.2328809529542923,
      "num_tokens": 116496280.0,
      "step": 56140
    },
    {
      "entropy": 5.812090730667114,
      "epoch": 5.092978955007257,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002715768228585642,
      "loss": 4.9211,
      "mean_token_accuracy": 0.21702938824892043,
      "num_tokens": 116505942.0,
      "step": 56145
    },
    {
      "entropy": 5.792029523849488,
      "epoch": 5.093432510885341,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002715444702192453,
      "loss": 4.8284,
      "mean_token_accuracy": 0.22526985555887222,
      "num_tokens": 116515749.0,
      "step": 56150
    },
    {
      "entropy": 5.728026294708252,
      "epoch": 5.093886066763425,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002715121176513875,
      "loss": 4.8345,
      "mean_token_accuracy": 0.23171899914741517,
      "num_tokens": 116525569.0,
      "step": 56155
    },
    {
      "entropy": 5.816686582565308,
      "epoch": 5.09433962264151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002714797651556601,
      "loss": 4.9317,
      "mean_token_accuracy": 0.2337749183177948,
      "num_tokens": 116535837.0,
      "step": 56160
    },
    {
      "entropy": 5.718699789047241,
      "epoch": 5.094793178519594,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027144741273273207,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2306734263896942,
      "num_tokens": 116546299.0,
      "step": 56165
    },
    {
      "entropy": 5.780652475357056,
      "epoch": 5.095246734397678,
      "grad_norm": 1.359375,
      "learning_rate": 0.00027141506038327233,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23437633663415908,
      "num_tokens": 116556725.0,
      "step": 56170
    },
    {
      "entropy": 5.731170225143432,
      "epoch": 5.0957002902757615,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002713827081079502,
      "loss": 4.812,
      "mean_token_accuracy": 0.22982341945171356,
      "num_tokens": 116566853.0,
      "step": 56175
    },
    {
      "entropy": 5.720853328704834,
      "epoch": 5.096153846153846,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002713503559074345,
      "loss": 4.8351,
      "mean_token_accuracy": 0.22717358469963073,
      "num_tokens": 116578347.0,
      "step": 56180
    },
    {
      "entropy": 5.772895860671997,
      "epoch": 5.09660740203193,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027131800378239437,
      "loss": 4.92,
      "mean_token_accuracy": 0.2190944790840149,
      "num_tokens": 116588836.0,
      "step": 56185
    },
    {
      "entropy": 5.744914245605469,
      "epoch": 5.097060957910014,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002712856517334989,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23578239381313323,
      "num_tokens": 116599463.0,
      "step": 56190
    },
    {
      "entropy": 5.779918575286866,
      "epoch": 5.097514513788099,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027125329976141713,
      "loss": 4.8757,
      "mean_token_accuracy": 0.2226291298866272,
      "num_tokens": 116610329.0,
      "step": 56195
    },
    {
      "entropy": 5.7832050800323485,
      "epoch": 5.097968069666183,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000271220947866818,
      "loss": 4.8409,
      "mean_token_accuracy": 0.23854517787694932,
      "num_tokens": 116619954.0,
      "step": 56200
    },
    {
      "entropy": 5.761743259429932,
      "epoch": 5.098421625544267,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002711885960503706,
      "loss": 4.8867,
      "mean_token_accuracy": 0.23233230262994767,
      "num_tokens": 116629347.0,
      "step": 56205
    },
    {
      "entropy": 5.764854383468628,
      "epoch": 5.098875181422351,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002711562443127441,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23112774193286895,
      "num_tokens": 116639674.0,
      "step": 56210
    },
    {
      "entropy": 5.72292423248291,
      "epoch": 5.099328737300436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002711238926546074,
      "loss": 4.8138,
      "mean_token_accuracy": 0.22470631450414658,
      "num_tokens": 116650889.0,
      "step": 56215
    },
    {
      "entropy": 5.638237333297729,
      "epoch": 5.09978229317852,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00027109154107662957,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22451765090227127,
      "num_tokens": 116661213.0,
      "step": 56220
    },
    {
      "entropy": 5.688112449645996,
      "epoch": 5.100235849056604,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002710591895794797,
      "loss": 4.7877,
      "mean_token_accuracy": 0.23098811954259874,
      "num_tokens": 116672064.0,
      "step": 56225
    },
    {
      "entropy": 5.7854715347290036,
      "epoch": 5.100689404934688,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002710268381638268,
      "loss": 4.9078,
      "mean_token_accuracy": 0.22141115218400956,
      "num_tokens": 116681616.0,
      "step": 56230
    },
    {
      "entropy": 5.7467268943786625,
      "epoch": 5.101142960812772,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002709944868303399,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23258928060531617,
      "num_tokens": 116691427.0,
      "step": 56235
    },
    {
      "entropy": 5.712050056457519,
      "epoch": 5.101596516690856,
      "grad_norm": 1.40625,
      "learning_rate": 0.000270962135579688,
      "loss": 4.8324,
      "mean_token_accuracy": 0.22627830803394317,
      "num_tokens": 116701326.0,
      "step": 56240
    },
    {
      "entropy": 5.72963514328003,
      "epoch": 5.10205007256894,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002709297844125402,
      "loss": 4.8503,
      "mean_token_accuracy": 0.23400108367204667,
      "num_tokens": 116712353.0,
      "step": 56245
    },
    {
      "entropy": 5.768566274642945,
      "epoch": 5.102503628447025,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002708974333295655,
      "loss": 4.8692,
      "mean_token_accuracy": 0.23433122485876084,
      "num_tokens": 116721906.0,
      "step": 56250
    },
    {
      "entropy": 5.7668030738830565,
      "epoch": 5.102957184325109,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027086508233143287,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22917971163988113,
      "num_tokens": 116732684.0,
      "step": 56255
    },
    {
      "entropy": 5.696315431594849,
      "epoch": 5.103410740203193,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002708327314188114,
      "loss": 4.8576,
      "mean_token_accuracy": 0.22901872545480728,
      "num_tokens": 116743071.0,
      "step": 56260
    },
    {
      "entropy": 5.743998670578003,
      "epoch": 5.1038642960812775,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002708003805923702,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22665836811065673,
      "num_tokens": 116752756.0,
      "step": 56265
    },
    {
      "entropy": 5.719941329956055,
      "epoch": 5.104317851959362,
      "grad_norm": 1.453125,
      "learning_rate": 0.00027076802985277813,
      "loss": 4.8269,
      "mean_token_accuracy": 0.22865709364414216,
      "num_tokens": 116762878.0,
      "step": 56270
    },
    {
      "entropy": 5.764460468292237,
      "epoch": 5.104771407837446,
      "grad_norm": 1.375,
      "learning_rate": 0.0002707356792007042,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2338831678032875,
      "num_tokens": 116772953.0,
      "step": 56275
    },
    {
      "entropy": 5.828871202468872,
      "epoch": 5.105224963715529,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027070332863681764,
      "loss": 4.8854,
      "mean_token_accuracy": 0.22375017553567886,
      "num_tokens": 116783530.0,
      "step": 56280
    },
    {
      "entropy": 5.741694164276123,
      "epoch": 5.105678519593614,
      "grad_norm": 1.5625,
      "learning_rate": 0.00027067097816178734,
      "loss": 4.833,
      "mean_token_accuracy": 0.22903510928153992,
      "num_tokens": 116792740.0,
      "step": 56285
    },
    {
      "entropy": 5.719997072219849,
      "epoch": 5.106132075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002706386277762822,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23119454681873322,
      "num_tokens": 116802846.0,
      "step": 56290
    },
    {
      "entropy": 5.785612010955811,
      "epoch": 5.106585631349782,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002706062774809714,
      "loss": 4.8897,
      "mean_token_accuracy": 0.22836748212575914,
      "num_tokens": 116812787.0,
      "step": 56295
    },
    {
      "entropy": 5.796293830871582,
      "epoch": 5.107039187227866,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000270573927276524,
      "loss": 4.9052,
      "mean_token_accuracy": 0.2201707199215889,
      "num_tokens": 116823840.0,
      "step": 56300
    },
    {
      "entropy": 5.738672494888306,
      "epoch": 5.107492743105951,
      "grad_norm": 1.25,
      "learning_rate": 0.0002705415771636088,
      "loss": 4.8113,
      "mean_token_accuracy": 0.23374902456998825,
      "num_tokens": 116833971.0,
      "step": 56305
    },
    {
      "entropy": 5.715430212020874,
      "epoch": 5.107946298984035,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027050922714289493,
      "loss": 4.791,
      "mean_token_accuracy": 0.2261875033378601,
      "num_tokens": 116844292.0,
      "step": 56310
    },
    {
      "entropy": 5.722376108169556,
      "epoch": 5.108399854862119,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00027047687721505143,
      "loss": 4.8645,
      "mean_token_accuracy": 0.22393683791160585,
      "num_tokens": 116854399.0,
      "step": 56315
    },
    {
      "entropy": 5.741343593597412,
      "epoch": 5.1088534107402035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027044452738074726,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2304689407348633,
      "num_tokens": 116865418.0,
      "step": 56320
    },
    {
      "entropy": 5.6822999000549315,
      "epoch": 5.109306966618288,
      "grad_norm": 1.375,
      "learning_rate": 0.00027041217764065136,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23872536718845366,
      "num_tokens": 116875225.0,
      "step": 56325
    },
    {
      "entropy": 5.790560388565064,
      "epoch": 5.109760522496371,
      "grad_norm": 1.390625,
      "learning_rate": 0.00027037982799543286,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2257324457168579,
      "num_tokens": 116885062.0,
      "step": 56330
    },
    {
      "entropy": 5.76598219871521,
      "epoch": 5.110214078374455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002703474784457607,
      "loss": 4.9008,
      "mean_token_accuracy": 0.223402801156044,
      "num_tokens": 116894705.0,
      "step": 56335
    },
    {
      "entropy": 5.7281969547271725,
      "epoch": 5.11066763425254,
      "grad_norm": 1.375,
      "learning_rate": 0.00027031512899230374,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2201962798833847,
      "num_tokens": 116904441.0,
      "step": 56340
    },
    {
      "entropy": 5.706724739074707,
      "epoch": 5.111121190130624,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002702827796357313,
      "loss": 4.8012,
      "mean_token_accuracy": 0.2341700628399849,
      "num_tokens": 116915008.0,
      "step": 56345
    },
    {
      "entropy": 5.768075370788575,
      "epoch": 5.111574746008708,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002702504303767121,
      "loss": 4.9194,
      "mean_token_accuracy": 0.2217234268784523,
      "num_tokens": 116926035.0,
      "step": 56350
    },
    {
      "entropy": 5.815449285507202,
      "epoch": 5.1120283018867925,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002702180812159152,
      "loss": 4.8989,
      "mean_token_accuracy": 0.22918594181537627,
      "num_tokens": 116936263.0,
      "step": 56355
    },
    {
      "entropy": 5.745538330078125,
      "epoch": 5.112481857764877,
      "grad_norm": 1.328125,
      "learning_rate": 0.00027018573215400955,
      "loss": 4.7208,
      "mean_token_accuracy": 0.2379728749394417,
      "num_tokens": 116947080.0,
      "step": 56360
    },
    {
      "entropy": 5.771700286865235,
      "epoch": 5.112935413642961,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027015338319166424,
      "loss": 4.7873,
      "mean_token_accuracy": 0.22942231148481368,
      "num_tokens": 116958224.0,
      "step": 56365
    },
    {
      "entropy": 5.7130656242370605,
      "epoch": 5.113388969521045,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002701210343295482,
      "loss": 4.8959,
      "mean_token_accuracy": 0.2257252037525177,
      "num_tokens": 116968385.0,
      "step": 56370
    },
    {
      "entropy": 5.752263164520263,
      "epoch": 5.1138425253991295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00027008868556833044,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22539214342832564,
      "num_tokens": 116979309.0,
      "step": 56375
    },
    {
      "entropy": 5.792833995819092,
      "epoch": 5.114296081277214,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00027005633690867985,
      "loss": 4.8979,
      "mean_token_accuracy": 0.23179445862770082,
      "num_tokens": 116990739.0,
      "step": 56380
    },
    {
      "entropy": 5.796722412109375,
      "epoch": 5.114749637155297,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002700239883512655,
      "loss": 4.8717,
      "mean_token_accuracy": 0.23292355835437775,
      "num_tokens": 117000820.0,
      "step": 56385
    },
    {
      "entropy": 5.699724435806274,
      "epoch": 5.1152031930333814,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002699916398967563,
      "loss": 4.8021,
      "mean_token_accuracy": 0.23809556066989898,
      "num_tokens": 117011319.0,
      "step": 56390
    },
    {
      "entropy": 5.7361408233642575,
      "epoch": 5.115656748911466,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002699592915458214,
      "loss": 4.8238,
      "mean_token_accuracy": 0.22901203483343124,
      "num_tokens": 117021480.0,
      "step": 56395
    },
    {
      "entropy": 5.788043355941772,
      "epoch": 5.11611030478955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026992694329912954,
      "loss": 4.8933,
      "mean_token_accuracy": 0.22485929429531099,
      "num_tokens": 117033237.0,
      "step": 56400
    },
    {
      "entropy": 5.752782773971558,
      "epoch": 5.116563860667634,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026989459515734974,
      "loss": 4.817,
      "mean_token_accuracy": 0.23556259125471116,
      "num_tokens": 117044237.0,
      "step": 56405
    },
    {
      "entropy": 5.788653421401977,
      "epoch": 5.1170174165457185,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026986224712115107,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2310333162546158,
      "num_tokens": 117054783.0,
      "step": 56410
    },
    {
      "entropy": 5.764601516723633,
      "epoch": 5.117470972423803,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002698298991912024,
      "loss": 4.8696,
      "mean_token_accuracy": 0.23246980607509612,
      "num_tokens": 117064865.0,
      "step": 56415
    },
    {
      "entropy": 5.771964025497437,
      "epoch": 5.117924528301887,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002697975513681728,
      "loss": 4.861,
      "mean_token_accuracy": 0.2277739927172661,
      "num_tokens": 117074968.0,
      "step": 56420
    },
    {
      "entropy": 5.745460939407349,
      "epoch": 5.118378084179971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002697652036527311,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22909265458583833,
      "num_tokens": 117086574.0,
      "step": 56425
    },
    {
      "entropy": 5.684934329986572,
      "epoch": 5.118831640058056,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026973285604554647,
      "loss": 4.7037,
      "mean_token_accuracy": 0.24084874540567397,
      "num_tokens": 117097212.0,
      "step": 56430
    },
    {
      "entropy": 5.73241810798645,
      "epoch": 5.119285195936139,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026970050854728756,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2393292024731636,
      "num_tokens": 117107544.0,
      "step": 56435
    },
    {
      "entropy": 5.772846984863281,
      "epoch": 5.119738751814223,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026966816115862354,
      "loss": 4.9238,
      "mean_token_accuracy": 0.22044626772403716,
      "num_tokens": 117118300.0,
      "step": 56440
    },
    {
      "entropy": 5.7602863788604735,
      "epoch": 5.1201923076923075,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002696358138802233,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23991342186927794,
      "num_tokens": 117128702.0,
      "step": 56445
    },
    {
      "entropy": 5.664323949813843,
      "epoch": 5.120645863570392,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026960346671275583,
      "loss": 4.7521,
      "mean_token_accuracy": 0.23928026705980301,
      "num_tokens": 117139485.0,
      "step": 56450
    },
    {
      "entropy": 5.8155464172363285,
      "epoch": 5.121099419448476,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026957111965688994,
      "loss": 4.9101,
      "mean_token_accuracy": 0.2264775738120079,
      "num_tokens": 117149895.0,
      "step": 56455
    },
    {
      "entropy": 5.808515644073486,
      "epoch": 5.12155297532656,
      "grad_norm": 1.375,
      "learning_rate": 0.00026953877271329476,
      "loss": 4.934,
      "mean_token_accuracy": 0.22286674231290818,
      "num_tokens": 117160901.0,
      "step": 56460
    },
    {
      "entropy": 5.698965406417846,
      "epoch": 5.1220065312046446,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026950642588263914,
      "loss": 4.7332,
      "mean_token_accuracy": 0.24170496612787246,
      "num_tokens": 117170820.0,
      "step": 56465
    },
    {
      "entropy": 5.695975017547608,
      "epoch": 5.122460087082729,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026947407916559196,
      "loss": 4.8141,
      "mean_token_accuracy": 0.23510272800922394,
      "num_tokens": 117181581.0,
      "step": 56470
    },
    {
      "entropy": 5.6778388023376465,
      "epoch": 5.122913642960813,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026944173256282234,
      "loss": 4.7885,
      "mean_token_accuracy": 0.23263873904943466,
      "num_tokens": 117190817.0,
      "step": 56475
    },
    {
      "entropy": 5.766508150100708,
      "epoch": 5.123367198838897,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026940938607499903,
      "loss": 4.881,
      "mean_token_accuracy": 0.23058799058198928,
      "num_tokens": 117201459.0,
      "step": 56480
    },
    {
      "entropy": 5.687958240509033,
      "epoch": 5.123820754716981,
      "grad_norm": 1.375,
      "learning_rate": 0.0002693770397027911,
      "loss": 4.7872,
      "mean_token_accuracy": 0.22846392244100572,
      "num_tokens": 117211941.0,
      "step": 56485
    },
    {
      "entropy": 5.747251176834107,
      "epoch": 5.124274310595065,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026934469344686734,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22158057391643524,
      "num_tokens": 117222010.0,
      "step": 56490
    },
    {
      "entropy": 5.747307872772216,
      "epoch": 5.124727866473149,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002693123473078968,
      "loss": 4.8505,
      "mean_token_accuracy": 0.21942156553268433,
      "num_tokens": 117231453.0,
      "step": 56495
    },
    {
      "entropy": 5.7431321144104,
      "epoch": 5.1251814223512335,
      "grad_norm": 1.515625,
      "learning_rate": 0.00026928000128654845,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23358263671398163,
      "num_tokens": 117241348.0,
      "step": 56500
    },
    {
      "entropy": 5.7044285297393795,
      "epoch": 5.125634978229318,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000269247655383491,
      "loss": 4.7893,
      "mean_token_accuracy": 0.23211827427148818,
      "num_tokens": 117251472.0,
      "step": 56505
    },
    {
      "entropy": 5.703810310363769,
      "epoch": 5.126088534107402,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026921530959939354,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23167288154363633,
      "num_tokens": 117262007.0,
      "step": 56510
    },
    {
      "entropy": 5.709354639053345,
      "epoch": 5.126542089985486,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000269182963934925,
      "loss": 4.8279,
      "mean_token_accuracy": 0.22679827213287354,
      "num_tokens": 117272573.0,
      "step": 56515
    },
    {
      "entropy": 5.769313764572144,
      "epoch": 5.126995645863571,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002691506183907542,
      "loss": 4.9299,
      "mean_token_accuracy": 0.22372591197490693,
      "num_tokens": 117283832.0,
      "step": 56520
    },
    {
      "entropy": 5.783345174789429,
      "epoch": 5.127449201741655,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026911827296755014,
      "loss": 4.9079,
      "mean_token_accuracy": 0.22762876451015474,
      "num_tokens": 117295287.0,
      "step": 56525
    },
    {
      "entropy": 5.733203315734864,
      "epoch": 5.127902757619739,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002690859276659817,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23450883477926254,
      "num_tokens": 117305585.0,
      "step": 56530
    },
    {
      "entropy": 5.776085376739502,
      "epoch": 5.1283563134978225,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002690535824867178,
      "loss": 4.8827,
      "mean_token_accuracy": 0.23009970188140869,
      "num_tokens": 117316149.0,
      "step": 56535
    },
    {
      "entropy": 5.751663589477539,
      "epoch": 5.128809869375907,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002690212374304273,
      "loss": 4.8364,
      "mean_token_accuracy": 0.2366613283753395,
      "num_tokens": 117325777.0,
      "step": 56540
    },
    {
      "entropy": 5.800431394577027,
      "epoch": 5.129263425253991,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002689888924977791,
      "loss": 4.9509,
      "mean_token_accuracy": 0.22412606775760652,
      "num_tokens": 117336303.0,
      "step": 56545
    },
    {
      "entropy": 5.726544380187988,
      "epoch": 5.129716981132075,
      "grad_norm": 1.375,
      "learning_rate": 0.0002689565476894422,
      "loss": 4.8001,
      "mean_token_accuracy": 0.23580704927444457,
      "num_tokens": 117347169.0,
      "step": 56550
    },
    {
      "entropy": 5.756903123855591,
      "epoch": 5.13017053701016,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00026892420300608533,
      "loss": 4.8269,
      "mean_token_accuracy": 0.2370362773537636,
      "num_tokens": 117357461.0,
      "step": 56555
    },
    {
      "entropy": 5.695665407180786,
      "epoch": 5.130624092888244,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026889185844837754,
      "loss": 4.8795,
      "mean_token_accuracy": 0.2294464722275734,
      "num_tokens": 117367915.0,
      "step": 56560
    },
    {
      "entropy": 5.7451213836669925,
      "epoch": 5.131077648766328,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002688595140169877,
      "loss": 4.9001,
      "mean_token_accuracy": 0.22345434576272966,
      "num_tokens": 117378352.0,
      "step": 56565
    },
    {
      "entropy": 5.763147401809692,
      "epoch": 5.131531204644412,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002688271697125847,
      "loss": 4.8676,
      "mean_token_accuracy": 0.2246803253889084,
      "num_tokens": 117389074.0,
      "step": 56570
    },
    {
      "entropy": 5.829005908966065,
      "epoch": 5.131984760522497,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002687948255358373,
      "loss": 4.8464,
      "mean_token_accuracy": 0.23010637760162353,
      "num_tokens": 117399643.0,
      "step": 56575
    },
    {
      "entropy": 5.745291042327881,
      "epoch": 5.132438316400581,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026876248148741464,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23428250402212142,
      "num_tokens": 117410015.0,
      "step": 56580
    },
    {
      "entropy": 5.724801158905029,
      "epoch": 5.132891872278664,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026873013756798533,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23498741835355758,
      "num_tokens": 117419517.0,
      "step": 56585
    },
    {
      "entropy": 5.7049243450164795,
      "epoch": 5.1333454281567485,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026869779377821837,
      "loss": 4.8011,
      "mean_token_accuracy": 0.23183590769767762,
      "num_tokens": 117430450.0,
      "step": 56590
    },
    {
      "entropy": 5.725224828720092,
      "epoch": 5.133798984034833,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002686654501187827,
      "loss": 4.8959,
      "mean_token_accuracy": 0.22775086164474487,
      "num_tokens": 117440403.0,
      "step": 56595
    },
    {
      "entropy": 5.7116883277893065,
      "epoch": 5.134252539912917,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002686331065903471,
      "loss": 4.8814,
      "mean_token_accuracy": 0.22747863531112672,
      "num_tokens": 117451014.0,
      "step": 56600
    },
    {
      "entropy": 5.742909240722656,
      "epoch": 5.134706095791001,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002686007631935805,
      "loss": 4.8663,
      "mean_token_accuracy": 0.22642363607883453,
      "num_tokens": 117461238.0,
      "step": 56605
    },
    {
      "entropy": 5.758097362518311,
      "epoch": 5.135159651669086,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002685684199291518,
      "loss": 4.8488,
      "mean_token_accuracy": 0.2287938579916954,
      "num_tokens": 117472298.0,
      "step": 56610
    },
    {
      "entropy": 5.805613803863525,
      "epoch": 5.13561320754717,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026853607679772973,
      "loss": 4.9109,
      "mean_token_accuracy": 0.2303944334387779,
      "num_tokens": 117483010.0,
      "step": 56615
    },
    {
      "entropy": 5.6734068393707275,
      "epoch": 5.136066763425254,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026850373379998325,
      "loss": 4.762,
      "mean_token_accuracy": 0.24263576418161392,
      "num_tokens": 117494325.0,
      "step": 56620
    },
    {
      "entropy": 5.728046464920044,
      "epoch": 5.136520319303338,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026847139093658123,
      "loss": 4.9241,
      "mean_token_accuracy": 0.22531863152980805,
      "num_tokens": 117505426.0,
      "step": 56625
    },
    {
      "entropy": 5.751479005813598,
      "epoch": 5.136973875181423,
      "grad_norm": 1.53125,
      "learning_rate": 0.00026843904820819247,
      "loss": 4.8474,
      "mean_token_accuracy": 0.22326727360486984,
      "num_tokens": 117515920.0,
      "step": 56630
    },
    {
      "entropy": 5.730240201950073,
      "epoch": 5.137427431059507,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026840670561548597,
      "loss": 4.8253,
      "mean_token_accuracy": 0.22735015749931337,
      "num_tokens": 117525454.0,
      "step": 56635
    },
    {
      "entropy": 5.709705257415772,
      "epoch": 5.13788098693759,
      "grad_norm": 1.375,
      "learning_rate": 0.0002683743631591303,
      "loss": 4.8048,
      "mean_token_accuracy": 0.23719432801008225,
      "num_tokens": 117536234.0,
      "step": 56640
    },
    {
      "entropy": 5.746394252777099,
      "epoch": 5.138334542815675,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026834202083979465,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2287971630692482,
      "num_tokens": 117546639.0,
      "step": 56645
    },
    {
      "entropy": 5.711922693252563,
      "epoch": 5.138788098693759,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026830967865814763,
      "loss": 4.832,
      "mean_token_accuracy": 0.2290983259677887,
      "num_tokens": 117557104.0,
      "step": 56650
    },
    {
      "entropy": 5.801455116271972,
      "epoch": 5.139241654571843,
      "grad_norm": 1.5625,
      "learning_rate": 0.00026827733661485825,
      "loss": 4.9047,
      "mean_token_accuracy": 0.236823570728302,
      "num_tokens": 117568909.0,
      "step": 56655
    },
    {
      "entropy": 5.757074975967408,
      "epoch": 5.139695210449927,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026824499471059514,
      "loss": 4.8614,
      "mean_token_accuracy": 0.22862014919519424,
      "num_tokens": 117579987.0,
      "step": 56660
    },
    {
      "entropy": 5.6729652881622314,
      "epoch": 5.140148766328012,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026821265294602736,
      "loss": 4.8612,
      "mean_token_accuracy": 0.22393427789211273,
      "num_tokens": 117591152.0,
      "step": 56665
    },
    {
      "entropy": 5.711812257766724,
      "epoch": 5.140602322206096,
      "grad_norm": 1.375,
      "learning_rate": 0.0002681803113218236,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23745160847902297,
      "num_tokens": 117601219.0,
      "step": 56670
    },
    {
      "entropy": 5.798388528823852,
      "epoch": 5.14105587808418,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002681479698386527,
      "loss": 4.9034,
      "mean_token_accuracy": 0.2270921289920807,
      "num_tokens": 117611670.0,
      "step": 56675
    },
    {
      "entropy": 5.772648191452026,
      "epoch": 5.1415094339622645,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002681156284971836,
      "loss": 4.8502,
      "mean_token_accuracy": 0.22712677717208862,
      "num_tokens": 117622016.0,
      "step": 56680
    },
    {
      "entropy": 5.753845262527466,
      "epoch": 5.141962989840349,
      "grad_norm": 1.390625,
      "learning_rate": 0.000268083287298085,
      "loss": 4.8401,
      "mean_token_accuracy": 0.23439722657203674,
      "num_tokens": 117631891.0,
      "step": 56685
    },
    {
      "entropy": 5.685210466384888,
      "epoch": 5.142416545718432,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002680509462420258,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23977771997451783,
      "num_tokens": 117642333.0,
      "step": 56690
    },
    {
      "entropy": 5.68830509185791,
      "epoch": 5.142870101596516,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026801860532967486,
      "loss": 4.8166,
      "mean_token_accuracy": 0.23750678598880767,
      "num_tokens": 117652761.0,
      "step": 56695
    },
    {
      "entropy": 5.752773237228394,
      "epoch": 5.143323657474601,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002679862645617009,
      "loss": 4.8255,
      "mean_token_accuracy": 0.24009413719177247,
      "num_tokens": 117661660.0,
      "step": 56700
    },
    {
      "entropy": 5.76942286491394,
      "epoch": 5.143777213352685,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002679539239387728,
      "loss": 4.8599,
      "mean_token_accuracy": 0.2241100549697876,
      "num_tokens": 117672083.0,
      "step": 56705
    },
    {
      "entropy": 5.725826692581177,
      "epoch": 5.144230769230769,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026792158346155924,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23441110253334047,
      "num_tokens": 117682923.0,
      "step": 56710
    },
    {
      "entropy": 5.677290916442871,
      "epoch": 5.144684325108853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026788924313072925,
      "loss": 4.7927,
      "mean_token_accuracy": 0.2322608634829521,
      "num_tokens": 117694409.0,
      "step": 56715
    },
    {
      "entropy": 5.742866134643554,
      "epoch": 5.145137880986938,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026785690294695155,
      "loss": 4.8556,
      "mean_token_accuracy": 0.23190476447343827,
      "num_tokens": 117704185.0,
      "step": 56720
    },
    {
      "entropy": 5.775363636016846,
      "epoch": 5.145591436865022,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026782456291089477,
      "loss": 4.8724,
      "mean_token_accuracy": 0.22171459794044496,
      "num_tokens": 117715067.0,
      "step": 56725
    },
    {
      "entropy": 5.673662757873535,
      "epoch": 5.146044992743106,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026779222302322796,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2344796285033226,
      "num_tokens": 117724258.0,
      "step": 56730
    },
    {
      "entropy": 5.771258211135864,
      "epoch": 5.1464985486211905,
      "grad_norm": 1.578125,
      "learning_rate": 0.0002677598832846198,
      "loss": 4.913,
      "mean_token_accuracy": 0.23364405035972596,
      "num_tokens": 117733334.0,
      "step": 56735
    },
    {
      "entropy": 5.7384096622467045,
      "epoch": 5.146952104499274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026772754369573914,
      "loss": 4.865,
      "mean_token_accuracy": 0.23696191161870955,
      "num_tokens": 117744482.0,
      "step": 56740
    },
    {
      "entropy": 5.786775588989258,
      "epoch": 5.147405660377358,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026769520425725464,
      "loss": 4.9202,
      "mean_token_accuracy": 0.23056211173534394,
      "num_tokens": 117755313.0,
      "step": 56745
    },
    {
      "entropy": 5.715366268157959,
      "epoch": 5.147859216255442,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026766286496983523,
      "loss": 4.8641,
      "mean_token_accuracy": 0.23353154957294464,
      "num_tokens": 117765892.0,
      "step": 56750
    },
    {
      "entropy": 5.687646675109863,
      "epoch": 5.148312772133527,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026763052583414964,
      "loss": 4.7933,
      "mean_token_accuracy": 0.2361868366599083,
      "num_tokens": 117776132.0,
      "step": 56755
    },
    {
      "entropy": 5.772993421554565,
      "epoch": 5.148766328011611,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002675981868508666,
      "loss": 4.8846,
      "mean_token_accuracy": 0.23042945861816405,
      "num_tokens": 117787194.0,
      "step": 56760
    },
    {
      "entropy": 5.726152944564819,
      "epoch": 5.149219883889695,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026756584802065504,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22713191956281661,
      "num_tokens": 117797991.0,
      "step": 56765
    },
    {
      "entropy": 5.806693649291992,
      "epoch": 5.1496734397677795,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002675335093441836,
      "loss": 4.9605,
      "mean_token_accuracy": 0.22439916729927062,
      "num_tokens": 117808114.0,
      "step": 56770
    },
    {
      "entropy": 5.737823915481568,
      "epoch": 5.150126995645864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002675011708221211,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23177096992731094,
      "num_tokens": 117818413.0,
      "step": 56775
    },
    {
      "entropy": 5.696425247192383,
      "epoch": 5.150580551523948,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026746883245513627,
      "loss": 4.7885,
      "mean_token_accuracy": 0.22587960511445998,
      "num_tokens": 117828926.0,
      "step": 56780
    },
    {
      "entropy": 5.723339319229126,
      "epoch": 5.151034107402032,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002674364942438979,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2311686471104622,
      "num_tokens": 117841051.0,
      "step": 56785
    },
    {
      "entropy": 5.679998779296875,
      "epoch": 5.1514876632801165,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026740415618907475,
      "loss": 4.8231,
      "mean_token_accuracy": 0.22973985075950623,
      "num_tokens": 117852010.0,
      "step": 56790
    },
    {
      "entropy": 5.768420696258545,
      "epoch": 5.1519412191582,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002673718182913356,
      "loss": 4.8406,
      "mean_token_accuracy": 0.22589776366949083,
      "num_tokens": 117860873.0,
      "step": 56795
    },
    {
      "entropy": 5.738710975646972,
      "epoch": 5.152394775036284,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002673394805513492,
      "loss": 4.8937,
      "mean_token_accuracy": 0.22962574660778046,
      "num_tokens": 117871417.0,
      "step": 56800
    },
    {
      "entropy": 5.726481485366821,
      "epoch": 5.1528483309143684,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026730714296978433,
      "loss": 4.8282,
      "mean_token_accuracy": 0.2291535660624504,
      "num_tokens": 117881115.0,
      "step": 56805
    },
    {
      "entropy": 5.754043817520142,
      "epoch": 5.153301886792453,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002672748055473096,
      "loss": 4.8454,
      "mean_token_accuracy": 0.2259107083082199,
      "num_tokens": 117891341.0,
      "step": 56810
    },
    {
      "entropy": 5.70675630569458,
      "epoch": 5.153755442670537,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000267242468284594,
      "loss": 4.8203,
      "mean_token_accuracy": 0.2302796095609665,
      "num_tokens": 117902304.0,
      "step": 56815
    },
    {
      "entropy": 5.663811731338501,
      "epoch": 5.154208998548621,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026721013118230605,
      "loss": 4.811,
      "mean_token_accuracy": 0.23311594426631926,
      "num_tokens": 117913367.0,
      "step": 56820
    },
    {
      "entropy": 5.7053954124450685,
      "epoch": 5.1546625544267055,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026717779424111456,
      "loss": 4.777,
      "mean_token_accuracy": 0.23465617448091508,
      "num_tokens": 117922750.0,
      "step": 56825
    },
    {
      "entropy": 5.739103221893311,
      "epoch": 5.15511611030479,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026714545746168826,
      "loss": 4.8277,
      "mean_token_accuracy": 0.22893625050783156,
      "num_tokens": 117933950.0,
      "step": 56830
    },
    {
      "entropy": 5.716288709640503,
      "epoch": 5.155569666182874,
      "grad_norm": 1.296875,
      "learning_rate": 0.000267113120844696,
      "loss": 4.8634,
      "mean_token_accuracy": 0.22756631225347518,
      "num_tokens": 117944985.0,
      "step": 56835
    },
    {
      "entropy": 5.64690933227539,
      "epoch": 5.156023222060958,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002670807843908064,
      "loss": 4.7263,
      "mean_token_accuracy": 0.23804692029953003,
      "num_tokens": 117954560.0,
      "step": 56840
    },
    {
      "entropy": 5.69354281425476,
      "epoch": 5.156476777939042,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026704844810068805,
      "loss": 4.8056,
      "mean_token_accuracy": 0.2363026723265648,
      "num_tokens": 117965690.0,
      "step": 56845
    },
    {
      "entropy": 5.74739933013916,
      "epoch": 5.156930333817126,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002670161119750099,
      "loss": 4.8208,
      "mean_token_accuracy": 0.2316965401172638,
      "num_tokens": 117975856.0,
      "step": 56850
    },
    {
      "entropy": 5.744196939468384,
      "epoch": 5.15738388969521,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002669837760144407,
      "loss": 4.8075,
      "mean_token_accuracy": 0.23128207474946977,
      "num_tokens": 117985673.0,
      "step": 56855
    },
    {
      "entropy": 5.707855129241944,
      "epoch": 5.1578374455732945,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026695144021964895,
      "loss": 4.7896,
      "mean_token_accuracy": 0.2355269655585289,
      "num_tokens": 117995866.0,
      "step": 56860
    },
    {
      "entropy": 5.687462186813354,
      "epoch": 5.158291001451379,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026691910459130346,
      "loss": 4.86,
      "mean_token_accuracy": 0.22219104170799256,
      "num_tokens": 118006756.0,
      "step": 56865
    },
    {
      "entropy": 5.748944425582886,
      "epoch": 5.158744557329463,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00026688676913007303,
      "loss": 4.9259,
      "mean_token_accuracy": 0.2304922327399254,
      "num_tokens": 118016506.0,
      "step": 56870
    },
    {
      "entropy": 5.7571605205535885,
      "epoch": 5.159198113207547,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026685443383662615,
      "loss": 4.836,
      "mean_token_accuracy": 0.22703455090522767,
      "num_tokens": 118026681.0,
      "step": 56875
    },
    {
      "entropy": 5.718334436416626,
      "epoch": 5.1596516690856316,
      "grad_norm": 1.375,
      "learning_rate": 0.00026682209871163175,
      "loss": 4.838,
      "mean_token_accuracy": 0.2318871349096298,
      "num_tokens": 118036376.0,
      "step": 56880
    },
    {
      "entropy": 5.775039720535278,
      "epoch": 5.160105224963716,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002667897637557584,
      "loss": 4.8806,
      "mean_token_accuracy": 0.23184294253587723,
      "num_tokens": 118047029.0,
      "step": 56885
    },
    {
      "entropy": 5.768552684783936,
      "epoch": 5.1605587808418,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002667574289696748,
      "loss": 4.8248,
      "mean_token_accuracy": 0.2277754381299019,
      "num_tokens": 118056887.0,
      "step": 56890
    },
    {
      "entropy": 5.779972219467163,
      "epoch": 5.1610123367198835,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026672509435404974,
      "loss": 4.8091,
      "mean_token_accuracy": 0.22905089408159257,
      "num_tokens": 118067036.0,
      "step": 56895
    },
    {
      "entropy": 5.725745296478271,
      "epoch": 5.161465892597968,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026669275990955175,
      "loss": 4.8644,
      "mean_token_accuracy": 0.2325099989771843,
      "num_tokens": 118077439.0,
      "step": 56900
    },
    {
      "entropy": 5.74407639503479,
      "epoch": 5.161919448476052,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026666042563684966,
      "loss": 4.8895,
      "mean_token_accuracy": 0.2206459328532219,
      "num_tokens": 118089009.0,
      "step": 56905
    },
    {
      "entropy": 5.724738311767578,
      "epoch": 5.162373004354136,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002666280915366121,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22594888806343078,
      "num_tokens": 118098820.0,
      "step": 56910
    },
    {
      "entropy": 5.743892574310303,
      "epoch": 5.1628265602322205,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026659575760950774,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23526585549116136,
      "num_tokens": 118108896.0,
      "step": 56915
    },
    {
      "entropy": 5.80528712272644,
      "epoch": 5.163280116110305,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002665634238562052,
      "loss": 4.9591,
      "mean_token_accuracy": 0.2277496337890625,
      "num_tokens": 118118481.0,
      "step": 56920
    },
    {
      "entropy": 5.808727741241455,
      "epoch": 5.163733671988389,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002665310902773733,
      "loss": 4.9164,
      "mean_token_accuracy": 0.2302661195397377,
      "num_tokens": 118128236.0,
      "step": 56925
    },
    {
      "entropy": 5.764145994186402,
      "epoch": 5.164187227866473,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026649875687368053,
      "loss": 4.8242,
      "mean_token_accuracy": 0.23507175147533416,
      "num_tokens": 118138967.0,
      "step": 56930
    },
    {
      "entropy": 5.788683223724365,
      "epoch": 5.164640783744558,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026646642364579567,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2187125340104103,
      "num_tokens": 118148920.0,
      "step": 56935
    },
    {
      "entropy": 5.740116930007934,
      "epoch": 5.165094339622642,
      "grad_norm": 1.375,
      "learning_rate": 0.00026643409059438726,
      "loss": 4.8102,
      "mean_token_accuracy": 0.22826606184244155,
      "num_tokens": 118158184.0,
      "step": 56940
    },
    {
      "entropy": 5.729327583312989,
      "epoch": 5.165547895500726,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026640175772012415,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2303391620516777,
      "num_tokens": 118168785.0,
      "step": 56945
    },
    {
      "entropy": 5.736867761611938,
      "epoch": 5.1660014513788095,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026636942502367486,
      "loss": 4.9264,
      "mean_token_accuracy": 0.220053930580616,
      "num_tokens": 118179713.0,
      "step": 56950
    },
    {
      "entropy": 5.735652303695678,
      "epoch": 5.166455007256894,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000266337092505708,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2246882975101471,
      "num_tokens": 118190226.0,
      "step": 56955
    },
    {
      "entropy": 5.7391283988952635,
      "epoch": 5.166908563134978,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002663047601668923,
      "loss": 4.8132,
      "mean_token_accuracy": 0.230804842710495,
      "num_tokens": 118200649.0,
      "step": 56960
    },
    {
      "entropy": 5.783419895172119,
      "epoch": 5.167362119013062,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002662724280078964,
      "loss": 4.9322,
      "mean_token_accuracy": 0.22314283847808838,
      "num_tokens": 118211249.0,
      "step": 56965
    },
    {
      "entropy": 5.842233562469483,
      "epoch": 5.167815674891147,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002662400960293889,
      "loss": 4.951,
      "mean_token_accuracy": 0.22421740144491195,
      "num_tokens": 118221734.0,
      "step": 56970
    },
    {
      "entropy": 5.772772741317749,
      "epoch": 5.168269230769231,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026620776423203846,
      "loss": 4.8651,
      "mean_token_accuracy": 0.23214904665946962,
      "num_tokens": 118232371.0,
      "step": 56975
    },
    {
      "entropy": 5.812578630447388,
      "epoch": 5.168722786647315,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026617543261651374,
      "loss": 4.9337,
      "mean_token_accuracy": 0.22548143118619918,
      "num_tokens": 118242830.0,
      "step": 56980
    },
    {
      "entropy": 5.726014232635498,
      "epoch": 5.169176342525399,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002661431011834833,
      "loss": 4.7944,
      "mean_token_accuracy": 0.2304333597421646,
      "num_tokens": 118253415.0,
      "step": 56985
    },
    {
      "entropy": 5.7991691589355465,
      "epoch": 5.169629898403484,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026611076993361574,
      "loss": 4.8924,
      "mean_token_accuracy": 0.2251475378870964,
      "num_tokens": 118263744.0,
      "step": 56990
    },
    {
      "entropy": 5.743007564544678,
      "epoch": 5.170083454281568,
      "grad_norm": 1.375,
      "learning_rate": 0.00026607843886757964,
      "loss": 4.8396,
      "mean_token_accuracy": 0.23004789799451827,
      "num_tokens": 118274563.0,
      "step": 56995
    },
    {
      "entropy": 5.726328039169312,
      "epoch": 5.170537010159651,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026604610798604386,
      "loss": 4.9608,
      "mean_token_accuracy": 0.22118237763643264,
      "num_tokens": 118285152.0,
      "step": 57000
    },
    {
      "epoch": 5.170537010159651,
      "eval_entropy": 5.408167106749246,
      "eval_loss": 4.9551825523376465,
      "eval_mean_token_accuracy": 0.23214119988160706,
      "eval_num_tokens": 118285152.0,
      "eval_runtime": 20.5264,
      "eval_samples_per_second": 1722.659,
      "eval_steps_per_second": 215.332,
      "step": 57000
    },
    {
      "entropy": 5.699273157119751,
      "epoch": 5.1709905660377355,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002660137772896768,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22869240790605544,
      "num_tokens": 118295783.0,
      "step": 57005
    },
    {
      "entropy": 5.772755861282349,
      "epoch": 5.17144412191582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026598144677914703,
      "loss": 4.8327,
      "mean_token_accuracy": 0.23387792259454726,
      "num_tokens": 118305432.0,
      "step": 57010
    },
    {
      "entropy": 5.8329394340515135,
      "epoch": 5.171897677793904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002659491164551233,
      "loss": 4.9658,
      "mean_token_accuracy": 0.2151197835803032,
      "num_tokens": 118316943.0,
      "step": 57015
    },
    {
      "entropy": 5.723055696487426,
      "epoch": 5.172351233671988,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002659167863182742,
      "loss": 4.8748,
      "mean_token_accuracy": 0.22692510783672332,
      "num_tokens": 118326915.0,
      "step": 57020
    },
    {
      "entropy": 5.69899435043335,
      "epoch": 5.172804789550073,
      "grad_norm": 1.5,
      "learning_rate": 0.00026588445636926815,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22970376610755922,
      "num_tokens": 118336925.0,
      "step": 57025
    },
    {
      "entropy": 5.746276760101319,
      "epoch": 5.173258345428157,
      "grad_norm": 1.3125,
      "learning_rate": 0.000265852126608774,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2275429755449295,
      "num_tokens": 118347681.0,
      "step": 57030
    },
    {
      "entropy": 5.779839086532593,
      "epoch": 5.173711901306241,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026581979703746016,
      "loss": 4.8647,
      "mean_token_accuracy": 0.22831450551748275,
      "num_tokens": 118359429.0,
      "step": 57035
    },
    {
      "entropy": 5.791433906555175,
      "epoch": 5.174165457184325,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002657874676559953,
      "loss": 4.8781,
      "mean_token_accuracy": 0.2253456309437752,
      "num_tokens": 118369800.0,
      "step": 57040
    },
    {
      "entropy": 5.717255115509033,
      "epoch": 5.17461901306241,
      "grad_norm": 1.375,
      "learning_rate": 0.00026575513846504786,
      "loss": 4.8577,
      "mean_token_accuracy": 0.2237341359257698,
      "num_tokens": 118381822.0,
      "step": 57045
    },
    {
      "entropy": 5.840226793289185,
      "epoch": 5.175072568940493,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026572280946528655,
      "loss": 4.9746,
      "mean_token_accuracy": 0.22571128159761428,
      "num_tokens": 118393157.0,
      "step": 57050
    },
    {
      "entropy": 5.739333724975586,
      "epoch": 5.175526124818577,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002656904806573799,
      "loss": 4.8867,
      "mean_token_accuracy": 0.22796417474746705,
      "num_tokens": 118403538.0,
      "step": 57055
    },
    {
      "entropy": 5.757237720489502,
      "epoch": 5.175979680696662,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026565815204199645,
      "loss": 4.7849,
      "mean_token_accuracy": 0.2353193372488022,
      "num_tokens": 118413756.0,
      "step": 57060
    },
    {
      "entropy": 5.771017980575562,
      "epoch": 5.176433236574746,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026562582361980486,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23424521833658218,
      "num_tokens": 118423967.0,
      "step": 57065
    },
    {
      "entropy": 5.707622337341308,
      "epoch": 5.17688679245283,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002655934953914736,
      "loss": 4.8117,
      "mean_token_accuracy": 0.22941123098134994,
      "num_tokens": 118434004.0,
      "step": 57070
    },
    {
      "entropy": 5.810789680480957,
      "epoch": 5.177340348330914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002655611673576713,
      "loss": 4.974,
      "mean_token_accuracy": 0.22643265426158904,
      "num_tokens": 118445083.0,
      "step": 57075
    },
    {
      "entropy": 5.743209600448608,
      "epoch": 5.177793904208999,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026552883951906636,
      "loss": 4.7812,
      "mean_token_accuracy": 0.24726655036211015,
      "num_tokens": 118454925.0,
      "step": 57080
    },
    {
      "entropy": 5.692591190338135,
      "epoch": 5.178247460087083,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002654965118763275,
      "loss": 4.8167,
      "mean_token_accuracy": 0.2338818371295929,
      "num_tokens": 118465562.0,
      "step": 57085
    },
    {
      "entropy": 5.7220207214355465,
      "epoch": 5.178701015965167,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002654641844301232,
      "loss": 4.8755,
      "mean_token_accuracy": 0.2321154937148094,
      "num_tokens": 118476884.0,
      "step": 57090
    },
    {
      "entropy": 5.761951494216919,
      "epoch": 5.1791545718432515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026543185718112193,
      "loss": 4.8917,
      "mean_token_accuracy": 0.222500242292881,
      "num_tokens": 118486874.0,
      "step": 57095
    },
    {
      "entropy": 5.818810367584229,
      "epoch": 5.179608127721336,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002653995301299924,
      "loss": 4.8918,
      "mean_token_accuracy": 0.23083569705486298,
      "num_tokens": 118496427.0,
      "step": 57100
    },
    {
      "entropy": 5.738851833343506,
      "epoch": 5.180061683599419,
      "grad_norm": 1.421875,
      "learning_rate": 0.000265367203277403,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2364002451300621,
      "num_tokens": 118506809.0,
      "step": 57105
    },
    {
      "entropy": 5.831502532958984,
      "epoch": 5.180515239477503,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002653348766240222,
      "loss": 4.9389,
      "mean_token_accuracy": 0.22381169646978377,
      "num_tokens": 118516525.0,
      "step": 57110
    },
    {
      "entropy": 5.647779512405395,
      "epoch": 5.180968795355588,
      "grad_norm": 1.375,
      "learning_rate": 0.00026530255017051876,
      "loss": 4.744,
      "mean_token_accuracy": 0.24050375819206238,
      "num_tokens": 118527508.0,
      "step": 57115
    },
    {
      "entropy": 5.76877064704895,
      "epoch": 5.181422351233672,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026527022391756094,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22801747620105745,
      "num_tokens": 118538038.0,
      "step": 57120
    },
    {
      "entropy": 5.7215794086456295,
      "epoch": 5.181875907111756,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026523789786581743,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23753680884838105,
      "num_tokens": 118548521.0,
      "step": 57125
    },
    {
      "entropy": 5.7134147644042965,
      "epoch": 5.18232946298984,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026520557201595665,
      "loss": 4.7839,
      "mean_token_accuracy": 0.23182654082775117,
      "num_tokens": 118558703.0,
      "step": 57130
    },
    {
      "entropy": 5.817881965637207,
      "epoch": 5.182783018867925,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002651732463686471,
      "loss": 4.926,
      "mean_token_accuracy": 0.22461117058992386,
      "num_tokens": 118568746.0,
      "step": 57135
    },
    {
      "entropy": 5.798840856552124,
      "epoch": 5.183236574746009,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002651409209245574,
      "loss": 4.841,
      "mean_token_accuracy": 0.22833438962697983,
      "num_tokens": 118578822.0,
      "step": 57140
    },
    {
      "entropy": 5.787361192703247,
      "epoch": 5.183690130624093,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002651085956843559,
      "loss": 4.8923,
      "mean_token_accuracy": 0.2305052697658539,
      "num_tokens": 118588363.0,
      "step": 57145
    },
    {
      "entropy": 5.833042812347412,
      "epoch": 5.1841436865021775,
      "grad_norm": 1.375,
      "learning_rate": 0.00026507627064871123,
      "loss": 4.8799,
      "mean_token_accuracy": 0.23147900700569152,
      "num_tokens": 118597869.0,
      "step": 57150
    },
    {
      "entropy": 5.746020412445068,
      "epoch": 5.184597242380261,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002650439458182917,
      "loss": 4.8,
      "mean_token_accuracy": 0.235142520070076,
      "num_tokens": 118608196.0,
      "step": 57155
    },
    {
      "entropy": 5.718560266494751,
      "epoch": 5.185050798258345,
      "grad_norm": 1.46875,
      "learning_rate": 0.000265011621193766,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2258942738175392,
      "num_tokens": 118617987.0,
      "step": 57160
    },
    {
      "entropy": 5.7095552444458,
      "epoch": 5.185504354136429,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026497929677580254,
      "loss": 4.9249,
      "mean_token_accuracy": 0.2164493218064308,
      "num_tokens": 118629740.0,
      "step": 57165
    },
    {
      "entropy": 5.781260538101196,
      "epoch": 5.185957910014514,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002649469725650697,
      "loss": 4.8601,
      "mean_token_accuracy": 0.22853533178567886,
      "num_tokens": 118640752.0,
      "step": 57170
    },
    {
      "entropy": 5.771016550064087,
      "epoch": 5.186411465892598,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026491464856223613,
      "loss": 4.9004,
      "mean_token_accuracy": 0.22349375933408738,
      "num_tokens": 118650064.0,
      "step": 57175
    },
    {
      "entropy": 5.681456518173218,
      "epoch": 5.186865021770682,
      "grad_norm": 1.375,
      "learning_rate": 0.0002648823247679701,
      "loss": 4.8188,
      "mean_token_accuracy": 0.22879400551319123,
      "num_tokens": 118660332.0,
      "step": 57180
    },
    {
      "entropy": 5.6983567237854,
      "epoch": 5.1873185776487665,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026485000118294015,
      "loss": 4.8398,
      "mean_token_accuracy": 0.22852174639701844,
      "num_tokens": 118671694.0,
      "step": 57185
    },
    {
      "entropy": 5.8092203617095945,
      "epoch": 5.187772133526851,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026481767780781485,
      "loss": 4.9231,
      "mean_token_accuracy": 0.22420667260885238,
      "num_tokens": 118682103.0,
      "step": 57190
    },
    {
      "entropy": 5.768647956848144,
      "epoch": 5.188225689404935,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002647853546432626,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2284164845943451,
      "num_tokens": 118691995.0,
      "step": 57195
    },
    {
      "entropy": 5.727576446533203,
      "epoch": 5.188679245283019,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026475303168995167,
      "loss": 4.8916,
      "mean_token_accuracy": 0.2286909118294716,
      "num_tokens": 118702587.0,
      "step": 57200
    },
    {
      "entropy": 5.743663787841797,
      "epoch": 5.189132801161103,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026472070894855077,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2262975126504898,
      "num_tokens": 118712032.0,
      "step": 57205
    },
    {
      "entropy": 5.700171852111817,
      "epoch": 5.189586357039187,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002646883864197282,
      "loss": 4.8225,
      "mean_token_accuracy": 0.23829605132341386,
      "num_tokens": 118723266.0,
      "step": 57210
    },
    {
      "entropy": 5.76467866897583,
      "epoch": 5.190039912917271,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002646560641041524,
      "loss": 4.8532,
      "mean_token_accuracy": 0.2287070035934448,
      "num_tokens": 118733025.0,
      "step": 57215
    },
    {
      "entropy": 5.73843240737915,
      "epoch": 5.1904934687953554,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026462374200249186,
      "loss": 4.9065,
      "mean_token_accuracy": 0.2284810796380043,
      "num_tokens": 118742690.0,
      "step": 57220
    },
    {
      "entropy": 5.748357772827148,
      "epoch": 5.19094702467344,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000264591420115415,
      "loss": 4.858,
      "mean_token_accuracy": 0.23269064724445343,
      "num_tokens": 118753032.0,
      "step": 57225
    },
    {
      "entropy": 5.754363298416138,
      "epoch": 5.191400580551524,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002645590984435902,
      "loss": 4.8165,
      "mean_token_accuracy": 0.23426015675067902,
      "num_tokens": 118763764.0,
      "step": 57230
    },
    {
      "entropy": 5.820304489135742,
      "epoch": 5.191854136429608,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002645267769876859,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22519617676734924,
      "num_tokens": 118773236.0,
      "step": 57235
    },
    {
      "entropy": 5.710571241378784,
      "epoch": 5.1923076923076925,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002644944557483705,
      "loss": 4.8352,
      "mean_token_accuracy": 0.230201917886734,
      "num_tokens": 118783239.0,
      "step": 57240
    },
    {
      "entropy": 5.727662181854248,
      "epoch": 5.192761248185777,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002644621347263125,
      "loss": 4.7754,
      "mean_token_accuracy": 0.23719450533390046,
      "num_tokens": 118792514.0,
      "step": 57245
    },
    {
      "entropy": 5.673587322235107,
      "epoch": 5.193214804063861,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026442981392218014,
      "loss": 4.7346,
      "mean_token_accuracy": 0.23845145404338836,
      "num_tokens": 118802673.0,
      "step": 57250
    },
    {
      "entropy": 5.720385980606079,
      "epoch": 5.193668359941944,
      "grad_norm": 1.453125,
      "learning_rate": 0.000264397493336642,
      "loss": 4.8574,
      "mean_token_accuracy": 0.22463958859443664,
      "num_tokens": 118812917.0,
      "step": 57255
    },
    {
      "entropy": 5.697715520858765,
      "epoch": 5.194121915820029,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002643651729703664,
      "loss": 4.765,
      "mean_token_accuracy": 0.23736517280340194,
      "num_tokens": 118822907.0,
      "step": 57260
    },
    {
      "entropy": 5.80339412689209,
      "epoch": 5.194575471698113,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002643328528240216,
      "loss": 4.8825,
      "mean_token_accuracy": 0.22684013843536377,
      "num_tokens": 118832072.0,
      "step": 57265
    },
    {
      "entropy": 5.698293209075928,
      "epoch": 5.195029027576197,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002643005328982763,
      "loss": 4.8154,
      "mean_token_accuracy": 0.23785149455070495,
      "num_tokens": 118842499.0,
      "step": 57270
    },
    {
      "entropy": 5.766353178024292,
      "epoch": 5.1954825834542815,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026426821319379867,
      "loss": 4.8657,
      "mean_token_accuracy": 0.22231709212064743,
      "num_tokens": 118853266.0,
      "step": 57275
    },
    {
      "entropy": 5.761991167068482,
      "epoch": 5.195936139332366,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026423589371125705,
      "loss": 4.9037,
      "mean_token_accuracy": 0.2221872851252556,
      "num_tokens": 118864052.0,
      "step": 57280
    },
    {
      "entropy": 5.7628014087677,
      "epoch": 5.19638969521045,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026420357445132,
      "loss": 4.8667,
      "mean_token_accuracy": 0.22605955898761748,
      "num_tokens": 118874301.0,
      "step": 57285
    },
    {
      "entropy": 5.776657819747925,
      "epoch": 5.196843251088534,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026417125541465574,
      "loss": 4.7869,
      "mean_token_accuracy": 0.22519824504852295,
      "num_tokens": 118884387.0,
      "step": 57290
    },
    {
      "entropy": 5.738284492492676,
      "epoch": 5.1972968069666186,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026413893660193265,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23604946285486222,
      "num_tokens": 118894612.0,
      "step": 57295
    },
    {
      "entropy": 5.7629786968231205,
      "epoch": 5.197750362844703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002641066180138191,
      "loss": 4.8841,
      "mean_token_accuracy": 0.21875648498535155,
      "num_tokens": 118904185.0,
      "step": 57300
    },
    {
      "entropy": 5.848327112197876,
      "epoch": 5.198203918722786,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026407429965098354,
      "loss": 4.9802,
      "mean_token_accuracy": 0.22274287939071655,
      "num_tokens": 118915013.0,
      "step": 57305
    },
    {
      "entropy": 5.79775037765503,
      "epoch": 5.1986574746008705,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026404198151409414,
      "loss": 4.886,
      "mean_token_accuracy": 0.23813897967338563,
      "num_tokens": 118926520.0,
      "step": 57310
    },
    {
      "entropy": 5.6583483695983885,
      "epoch": 5.199111030478955,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026400966360381946,
      "loss": 4.7885,
      "mean_token_accuracy": 0.2334123209118843,
      "num_tokens": 118937601.0,
      "step": 57315
    },
    {
      "entropy": 5.716318559646607,
      "epoch": 5.199564586357039,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026397734592082766,
      "loss": 4.8636,
      "mean_token_accuracy": 0.22856749594211578,
      "num_tokens": 118948030.0,
      "step": 57320
    },
    {
      "entropy": 5.727548503875733,
      "epoch": 5.200018142235123,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026394502846578715,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2342899590730667,
      "num_tokens": 118958670.0,
      "step": 57325
    },
    {
      "entropy": 5.833939123153686,
      "epoch": 5.2004716981132075,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026391271123936633,
      "loss": 4.9635,
      "mean_token_accuracy": 0.22898880690336226,
      "num_tokens": 118969211.0,
      "step": 57330
    },
    {
      "entropy": 5.825479459762573,
      "epoch": 5.200925253991292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002638803942422335,
      "loss": 4.9294,
      "mean_token_accuracy": 0.2235774889588356,
      "num_tokens": 118980837.0,
      "step": 57335
    },
    {
      "entropy": 5.750619602203369,
      "epoch": 5.201378809869376,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00026384807747505687,
      "loss": 4.8539,
      "mean_token_accuracy": 0.2277578353881836,
      "num_tokens": 118991266.0,
      "step": 57340
    },
    {
      "entropy": 5.833273410797119,
      "epoch": 5.20183236574746,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026381576093850484,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2252048820257187,
      "num_tokens": 119001459.0,
      "step": 57345
    },
    {
      "entropy": 5.808139610290527,
      "epoch": 5.202285921625545,
      "grad_norm": 1.5,
      "learning_rate": 0.0002637834446332457,
      "loss": 4.9093,
      "mean_token_accuracy": 0.23210416436195375,
      "num_tokens": 119012010.0,
      "step": 57350
    },
    {
      "entropy": 5.705382013320923,
      "epoch": 5.202739477503629,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026375112855994785,
      "loss": 4.8007,
      "mean_token_accuracy": 0.234898541867733,
      "num_tokens": 119022371.0,
      "step": 57355
    },
    {
      "entropy": 5.808044195175171,
      "epoch": 5.203193033381712,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002637188127192795,
      "loss": 4.9291,
      "mean_token_accuracy": 0.22888227850198745,
      "num_tokens": 119032457.0,
      "step": 57360
    },
    {
      "entropy": 5.700084829330445,
      "epoch": 5.2036465892597965,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000263686497111909,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22429586052894593,
      "num_tokens": 119044133.0,
      "step": 57365
    },
    {
      "entropy": 5.713538408279419,
      "epoch": 5.204100145137881,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026365418173850466,
      "loss": 4.7463,
      "mean_token_accuracy": 0.23495604693889618,
      "num_tokens": 119054412.0,
      "step": 57370
    },
    {
      "entropy": 5.704257678985596,
      "epoch": 5.204553701015965,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026362186659973464,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23962566405534744,
      "num_tokens": 119064950.0,
      "step": 57375
    },
    {
      "entropy": 5.765338039398193,
      "epoch": 5.205007256894049,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002635895516962673,
      "loss": 4.9509,
      "mean_token_accuracy": 0.2198440060019493,
      "num_tokens": 119074945.0,
      "step": 57380
    },
    {
      "entropy": 5.752412414550781,
      "epoch": 5.205460812772134,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026355723702877105,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22869546264410018,
      "num_tokens": 119086193.0,
      "step": 57385
    },
    {
      "entropy": 5.739171552658081,
      "epoch": 5.205914368650218,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000263524922597914,
      "loss": 4.8471,
      "mean_token_accuracy": 0.24211318790912628,
      "num_tokens": 119096697.0,
      "step": 57390
    },
    {
      "entropy": 5.737106370925903,
      "epoch": 5.206367924528302,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002634926084043645,
      "loss": 4.8373,
      "mean_token_accuracy": 0.2260235592722893,
      "num_tokens": 119107833.0,
      "step": 57395
    },
    {
      "entropy": 5.777524471282959,
      "epoch": 5.206821480406386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026346029444879074,
      "loss": 4.8395,
      "mean_token_accuracy": 0.23336706608533858,
      "num_tokens": 119117602.0,
      "step": 57400
    },
    {
      "entropy": 5.7668829441070555,
      "epoch": 5.207275036284471,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026342798073186106,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23122793436050415,
      "num_tokens": 119128546.0,
      "step": 57405
    },
    {
      "entropy": 5.758804130554199,
      "epoch": 5.207728592162554,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026339566725424365,
      "loss": 4.8573,
      "mean_token_accuracy": 0.23122089803218843,
      "num_tokens": 119138079.0,
      "step": 57410
    },
    {
      "entropy": 5.741027164459228,
      "epoch": 5.208182148040638,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026336335401660685,
      "loss": 4.8167,
      "mean_token_accuracy": 0.22768481373786925,
      "num_tokens": 119148239.0,
      "step": 57415
    },
    {
      "entropy": 5.750075101852417,
      "epoch": 5.2086357039187225,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026333104101961886,
      "loss": 4.8341,
      "mean_token_accuracy": 0.22955388724803924,
      "num_tokens": 119159266.0,
      "step": 57420
    },
    {
      "entropy": 5.74737663269043,
      "epoch": 5.209089259796807,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002632987282639479,
      "loss": 4.8995,
      "mean_token_accuracy": 0.21929021775722504,
      "num_tokens": 119168766.0,
      "step": 57425
    },
    {
      "entropy": 5.698893308639526,
      "epoch": 5.209542815674891,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002632664157502622,
      "loss": 4.7624,
      "mean_token_accuracy": 0.23811435103416442,
      "num_tokens": 119179529.0,
      "step": 57430
    },
    {
      "entropy": 5.724727725982666,
      "epoch": 5.209996371552975,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026323410347923,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2289437934756279,
      "num_tokens": 119189796.0,
      "step": 57435
    },
    {
      "entropy": 5.79012541770935,
      "epoch": 5.21044992743106,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00026320179145151957,
      "loss": 4.903,
      "mean_token_accuracy": 0.23546014875173568,
      "num_tokens": 119199968.0,
      "step": 57440
    },
    {
      "entropy": 5.774277448654175,
      "epoch": 5.210903483309144,
      "grad_norm": 1.53125,
      "learning_rate": 0.000263169479667799,
      "loss": 4.9356,
      "mean_token_accuracy": 0.22711038440465928,
      "num_tokens": 119210179.0,
      "step": 57445
    },
    {
      "entropy": 5.767448568344117,
      "epoch": 5.211357039187228,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002631371681287368,
      "loss": 4.8967,
      "mean_token_accuracy": 0.2311580702662468,
      "num_tokens": 119222011.0,
      "step": 57450
    },
    {
      "entropy": 5.7347925186157225,
      "epoch": 5.211810595065312,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00026310485683500085,
      "loss": 4.86,
      "mean_token_accuracy": 0.23820654451847076,
      "num_tokens": 119232166.0,
      "step": 57455
    },
    {
      "entropy": 5.843944931030274,
      "epoch": 5.212264150943396,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002630725457872595,
      "loss": 4.9689,
      "mean_token_accuracy": 0.22354728281497954,
      "num_tokens": 119241473.0,
      "step": 57460
    },
    {
      "entropy": 5.807803773880005,
      "epoch": 5.21271770682148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026304023498618093,
      "loss": 4.9064,
      "mean_token_accuracy": 0.23454933017492294,
      "num_tokens": 119252092.0,
      "step": 57465
    },
    {
      "entropy": 5.70580153465271,
      "epoch": 5.213171262699564,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026300792443243343,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23363620936870574,
      "num_tokens": 119262473.0,
      "step": 57470
    },
    {
      "entropy": 5.769939136505127,
      "epoch": 5.213624818577649,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002629756141266851,
      "loss": 4.8467,
      "mean_token_accuracy": 0.2307567834854126,
      "num_tokens": 119271639.0,
      "step": 57475
    },
    {
      "entropy": 5.745132541656494,
      "epoch": 5.214078374455733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002629433040696041,
      "loss": 4.8497,
      "mean_token_accuracy": 0.22785588204860688,
      "num_tokens": 119282463.0,
      "step": 57480
    },
    {
      "entropy": 5.802043390274048,
      "epoch": 5.214531930333817,
      "grad_norm": 1.4375,
      "learning_rate": 0.00026291099426185865,
      "loss": 4.8755,
      "mean_token_accuracy": 0.23273638188838958,
      "num_tokens": 119292472.0,
      "step": 57485
    },
    {
      "entropy": 5.734532022476197,
      "epoch": 5.214985486211901,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002628786847041169,
      "loss": 4.8192,
      "mean_token_accuracy": 0.2275092586874962,
      "num_tokens": 119302447.0,
      "step": 57490
    },
    {
      "entropy": 5.728944778442383,
      "epoch": 5.215439042089986,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026284637539704706,
      "loss": 4.7806,
      "mean_token_accuracy": 0.2311622306704521,
      "num_tokens": 119312864.0,
      "step": 57495
    },
    {
      "entropy": 5.820295238494873,
      "epoch": 5.21589259796807,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002628140663413172,
      "loss": 4.9379,
      "mean_token_accuracy": 0.2258877381682396,
      "num_tokens": 119322866.0,
      "step": 57500
    },
    {
      "entropy": 5.758116149902344,
      "epoch": 5.216346153846154,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002627817575375957,
      "loss": 4.7928,
      "mean_token_accuracy": 0.23607232570648193,
      "num_tokens": 119333185.0,
      "step": 57505
    },
    {
      "entropy": 5.710790300369263,
      "epoch": 5.2167997097242385,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00026274944898655047,
      "loss": 4.7829,
      "mean_token_accuracy": 0.2438657894730568,
      "num_tokens": 119342955.0,
      "step": 57510
    },
    {
      "entropy": 5.720876598358155,
      "epoch": 5.217253265602322,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026271714068884977,
      "loss": 4.9151,
      "mean_token_accuracy": 0.22879286408424376,
      "num_tokens": 119353346.0,
      "step": 57515
    },
    {
      "entropy": 5.756113958358765,
      "epoch": 5.217706821480406,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026268483264516176,
      "loss": 4.8705,
      "mean_token_accuracy": 0.230208383500576,
      "num_tokens": 119364300.0,
      "step": 57520
    },
    {
      "entropy": 5.776953268051147,
      "epoch": 5.21816037735849,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00026265252485615446,
      "loss": 4.8648,
      "mean_token_accuracy": 0.23403764963150026,
      "num_tokens": 119374742.0,
      "step": 57525
    },
    {
      "entropy": 5.747763681411743,
      "epoch": 5.218613933236575,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002626202173224962,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22140754908323287,
      "num_tokens": 119384948.0,
      "step": 57530
    },
    {
      "entropy": 5.753099822998047,
      "epoch": 5.219067489114659,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026258791004485496,
      "loss": 4.8719,
      "mean_token_accuracy": 0.23012969493865967,
      "num_tokens": 119395883.0,
      "step": 57535
    },
    {
      "entropy": 5.784112548828125,
      "epoch": 5.219521044992743,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026255560302389886,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22931280434131623,
      "num_tokens": 119406285.0,
      "step": 57540
    },
    {
      "entropy": 5.730083990097046,
      "epoch": 5.219974600870827,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00026252329626029616,
      "loss": 4.8379,
      "mean_token_accuracy": 0.2368006944656372,
      "num_tokens": 119417205.0,
      "step": 57545
    },
    {
      "entropy": 5.777508211135864,
      "epoch": 5.220428156748912,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026249098975471475,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23525601327419282,
      "num_tokens": 119427619.0,
      "step": 57550
    },
    {
      "entropy": 5.752841949462891,
      "epoch": 5.220881712626996,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002624586835078229,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23241974860429765,
      "num_tokens": 119437020.0,
      "step": 57555
    },
    {
      "entropy": 5.802765512466431,
      "epoch": 5.22133526850508,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026242637752028875,
      "loss": 4.8945,
      "mean_token_accuracy": 0.22904841899871825,
      "num_tokens": 119447111.0,
      "step": 57560
    },
    {
      "entropy": 5.755113124847412,
      "epoch": 5.221788824383164,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026239407179278027,
      "loss": 4.8332,
      "mean_token_accuracy": 0.24091894477605819,
      "num_tokens": 119458634.0,
      "step": 57565
    },
    {
      "entropy": 5.78506555557251,
      "epoch": 5.222242380261248,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026236176632596554,
      "loss": 4.8477,
      "mean_token_accuracy": 0.22980028092861177,
      "num_tokens": 119468193.0,
      "step": 57570
    },
    {
      "entropy": 5.709089756011963,
      "epoch": 5.222695936139332,
      "grad_norm": 1.390625,
      "learning_rate": 0.00026232946112051276,
      "loss": 4.846,
      "mean_token_accuracy": 0.22720832079648973,
      "num_tokens": 119477755.0,
      "step": 57575
    },
    {
      "entropy": 5.723527336120606,
      "epoch": 5.223149492017416,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026229715617708996,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23408553153276443,
      "num_tokens": 119487943.0,
      "step": 57580
    },
    {
      "entropy": 5.772112464904785,
      "epoch": 5.223603047895501,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002622648514963652,
      "loss": 4.8609,
      "mean_token_accuracy": 0.22106145322322845,
      "num_tokens": 119498731.0,
      "step": 57585
    },
    {
      "entropy": 5.77610330581665,
      "epoch": 5.224056603773585,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026223254707900655,
      "loss": 4.895,
      "mean_token_accuracy": 0.23711722195148469,
      "num_tokens": 119509991.0,
      "step": 57590
    },
    {
      "entropy": 5.800629043579102,
      "epoch": 5.224510159651669,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026220024292568204,
      "loss": 4.9112,
      "mean_token_accuracy": 0.2257620856165886,
      "num_tokens": 119520208.0,
      "step": 57595
    },
    {
      "entropy": 5.837824487686158,
      "epoch": 5.2249637155297535,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002621679390370598,
      "loss": 4.9402,
      "mean_token_accuracy": 0.22044828087091445,
      "num_tokens": 119530061.0,
      "step": 57600
    },
    {
      "entropy": 5.848437356948852,
      "epoch": 5.225417271407838,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026213563541380787,
      "loss": 4.8892,
      "mean_token_accuracy": 0.23429669439792633,
      "num_tokens": 119540050.0,
      "step": 57605
    },
    {
      "entropy": 5.82088794708252,
      "epoch": 5.225870827285922,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002621033320565943,
      "loss": 4.9762,
      "mean_token_accuracy": 0.21969554871320723,
      "num_tokens": 119550526.0,
      "step": 57610
    },
    {
      "entropy": 5.760790920257568,
      "epoch": 5.226324383164005,
      "grad_norm": 1.515625,
      "learning_rate": 0.000262071028966087,
      "loss": 4.908,
      "mean_token_accuracy": 0.23068923950195314,
      "num_tokens": 119560625.0,
      "step": 57615
    },
    {
      "entropy": 5.737749862670898,
      "epoch": 5.22677793904209,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002620387261429543,
      "loss": 4.7869,
      "mean_token_accuracy": 0.23474531769752502,
      "num_tokens": 119571125.0,
      "step": 57620
    },
    {
      "entropy": 5.765307331085205,
      "epoch": 5.227231494920174,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026200642358786385,
      "loss": 4.8441,
      "mean_token_accuracy": 0.23021037727594376,
      "num_tokens": 119581629.0,
      "step": 57625
    },
    {
      "entropy": 5.766329097747803,
      "epoch": 5.227685050798258,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000261974121301484,
      "loss": 4.8674,
      "mean_token_accuracy": 0.22657221406698227,
      "num_tokens": 119593421.0,
      "step": 57630
    },
    {
      "entropy": 5.752644634246826,
      "epoch": 5.2281386066763424,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026194181928448256,
      "loss": 4.7952,
      "mean_token_accuracy": 0.2392216518521309,
      "num_tokens": 119602446.0,
      "step": 57635
    },
    {
      "entropy": 5.719210386276245,
      "epoch": 5.228592162554427,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002619095175375277,
      "loss": 4.7753,
      "mean_token_accuracy": 0.23874304443597794,
      "num_tokens": 119613071.0,
      "step": 57640
    },
    {
      "entropy": 5.7693897724151615,
      "epoch": 5.229045718432511,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002618772160612873,
      "loss": 4.8519,
      "mean_token_accuracy": 0.23214143067598342,
      "num_tokens": 119623945.0,
      "step": 57645
    },
    {
      "entropy": 5.751653861999512,
      "epoch": 5.229499274310595,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002618449148564294,
      "loss": 4.8112,
      "mean_token_accuracy": 0.2355145424604416,
      "num_tokens": 119634491.0,
      "step": 57650
    },
    {
      "entropy": 5.844055700302124,
      "epoch": 5.2299528301886795,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002618126139236221,
      "loss": 5.0257,
      "mean_token_accuracy": 0.22125775068998338,
      "num_tokens": 119645076.0,
      "step": 57655
    },
    {
      "entropy": 5.719178485870361,
      "epoch": 5.230406386066764,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026178031326353325,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23083217591047286,
      "num_tokens": 119655030.0,
      "step": 57660
    },
    {
      "entropy": 5.747895097732544,
      "epoch": 5.230859941944848,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002617480128768309,
      "loss": 4.827,
      "mean_token_accuracy": 0.23054133206605912,
      "num_tokens": 119664914.0,
      "step": 57665
    },
    {
      "entropy": 5.6890716552734375,
      "epoch": 5.231313497822931,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026171571276418295,
      "loss": 4.8522,
      "mean_token_accuracy": 0.23640925735235213,
      "num_tokens": 119676281.0,
      "step": 57670
    },
    {
      "entropy": 5.740273189544678,
      "epoch": 5.231767053701016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002616834129262575,
      "loss": 4.8178,
      "mean_token_accuracy": 0.23214916586875917,
      "num_tokens": 119686988.0,
      "step": 57675
    },
    {
      "entropy": 5.692128038406372,
      "epoch": 5.2322206095791,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026165111336372255,
      "loss": 4.7313,
      "mean_token_accuracy": 0.24059873819351196,
      "num_tokens": 119699327.0,
      "step": 57680
    },
    {
      "entropy": 5.725168657302857,
      "epoch": 5.232674165457184,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026161881407724586,
      "loss": 4.8702,
      "mean_token_accuracy": 0.2288939267396927,
      "num_tokens": 119709527.0,
      "step": 57685
    },
    {
      "entropy": 5.6914409637451175,
      "epoch": 5.2331277213352685,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026158651506749556,
      "loss": 4.843,
      "mean_token_accuracy": 0.23448372334241868,
      "num_tokens": 119720237.0,
      "step": 57690
    },
    {
      "entropy": 5.728109312057495,
      "epoch": 5.233581277213353,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026155421633513947,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22622449398040773,
      "num_tokens": 119731091.0,
      "step": 57695
    },
    {
      "entropy": 5.756864881515503,
      "epoch": 5.234034833091437,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026152191788084573,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23442850559949874,
      "num_tokens": 119740986.0,
      "step": 57700
    },
    {
      "entropy": 5.723059511184692,
      "epoch": 5.234488388969521,
      "grad_norm": 1.34375,
      "learning_rate": 0.00026148961970528205,
      "loss": 4.8441,
      "mean_token_accuracy": 0.2307480737566948,
      "num_tokens": 119751045.0,
      "step": 57705
    },
    {
      "entropy": 5.745478105545044,
      "epoch": 5.2349419448476056,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00026145732180911667,
      "loss": 4.7973,
      "mean_token_accuracy": 0.23088393956422806,
      "num_tokens": 119761404.0,
      "step": 57710
    },
    {
      "entropy": 5.771057176589966,
      "epoch": 5.23539550072569,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002614250241930172,
      "loss": 4.8481,
      "mean_token_accuracy": 0.23310334235429764,
      "num_tokens": 119771827.0,
      "step": 57715
    },
    {
      "entropy": 5.795271730422973,
      "epoch": 5.235849056603773,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026139272685765164,
      "loss": 4.9007,
      "mean_token_accuracy": 0.2295002296566963,
      "num_tokens": 119782688.0,
      "step": 57720
    },
    {
      "entropy": 5.804090976715088,
      "epoch": 5.2363026124818575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000261360429803688,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22740901559591292,
      "num_tokens": 119792498.0,
      "step": 57725
    },
    {
      "entropy": 5.7012382507324215,
      "epoch": 5.236756168359942,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002613281330317941,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23366560637950898,
      "num_tokens": 119803855.0,
      "step": 57730
    },
    {
      "entropy": 5.686465215682984,
      "epoch": 5.237209724238026,
      "grad_norm": 1.453125,
      "learning_rate": 0.00026129583654263793,
      "loss": 4.7895,
      "mean_token_accuracy": 0.23249991089105607,
      "num_tokens": 119813280.0,
      "step": 57735
    },
    {
      "entropy": 5.755679988861084,
      "epoch": 5.23766328011611,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026126354033688745,
      "loss": 4.8956,
      "mean_token_accuracy": 0.22564112544059753,
      "num_tokens": 119823966.0,
      "step": 57740
    },
    {
      "entropy": 5.851648998260498,
      "epoch": 5.2381168359941945,
      "grad_norm": 1.5,
      "learning_rate": 0.0002612312444152103,
      "loss": 4.9656,
      "mean_token_accuracy": 0.2281494230031967,
      "num_tokens": 119834134.0,
      "step": 57745
    },
    {
      "entropy": 5.734207010269165,
      "epoch": 5.238570391872279,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002611989487782746,
      "loss": 4.7719,
      "mean_token_accuracy": 0.23588687479496,
      "num_tokens": 119844134.0,
      "step": 57750
    },
    {
      "entropy": 5.775241088867188,
      "epoch": 5.239023947750363,
      "grad_norm": 1.359375,
      "learning_rate": 0.00026116665342674816,
      "loss": 4.9365,
      "mean_token_accuracy": 0.22712356448173524,
      "num_tokens": 119855029.0,
      "step": 57755
    },
    {
      "entropy": 5.741457128524781,
      "epoch": 5.239477503628447,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026113435836129886,
      "loss": 4.8299,
      "mean_token_accuracy": 0.22763761579990388,
      "num_tokens": 119865453.0,
      "step": 57760
    },
    {
      "entropy": 5.763896799087524,
      "epoch": 5.239931059506532,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026110206358259447,
      "loss": 4.8555,
      "mean_token_accuracy": 0.23249513506889344,
      "num_tokens": 119876866.0,
      "step": 57765
    },
    {
      "entropy": 5.748286437988281,
      "epoch": 5.240384615384615,
      "grad_norm": 1.328125,
      "learning_rate": 0.000261069769091303,
      "loss": 4.8164,
      "mean_token_accuracy": 0.2317485600709915,
      "num_tokens": 119886944.0,
      "step": 57770
    },
    {
      "entropy": 5.8179600715637205,
      "epoch": 5.240838171262699,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002610374748880923,
      "loss": 4.9455,
      "mean_token_accuracy": 0.226500204205513,
      "num_tokens": 119896599.0,
      "step": 57775
    },
    {
      "entropy": 5.777005529403686,
      "epoch": 5.2412917271407835,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00026100518097363015,
      "loss": 4.9527,
      "mean_token_accuracy": 0.2265087068080902,
      "num_tokens": 119907360.0,
      "step": 57780
    },
    {
      "entropy": 5.694047307968139,
      "epoch": 5.241745283018868,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00026097288734858436,
      "loss": 4.8139,
      "mean_token_accuracy": 0.2357266962528229,
      "num_tokens": 119917269.0,
      "step": 57785
    },
    {
      "entropy": 5.687475967407226,
      "epoch": 5.242198838896952,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000260940594013623,
      "loss": 4.7878,
      "mean_token_accuracy": 0.23458537459373474,
      "num_tokens": 119927234.0,
      "step": 57790
    },
    {
      "entropy": 5.750402545928955,
      "epoch": 5.242652394775036,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002609083009694136,
      "loss": 4.8406,
      "mean_token_accuracy": 0.23324733823537827,
      "num_tokens": 119937208.0,
      "step": 57795
    },
    {
      "entropy": 5.767661809921265,
      "epoch": 5.243105950653121,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026087600821662416,
      "loss": 4.8804,
      "mean_token_accuracy": 0.22772018611431122,
      "num_tokens": 119948385.0,
      "step": 57800
    },
    {
      "entropy": 5.708481168746948,
      "epoch": 5.243559506531205,
      "grad_norm": 1.375,
      "learning_rate": 0.0002608437157559225,
      "loss": 4.8114,
      "mean_token_accuracy": 0.23274375051259993,
      "num_tokens": 119958155.0,
      "step": 57805
    },
    {
      "entropy": 5.794917488098145,
      "epoch": 5.244013062409289,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026081142358797633,
      "loss": 4.8767,
      "mean_token_accuracy": 0.22788176387548448,
      "num_tokens": 119968822.0,
      "step": 57810
    },
    {
      "entropy": 5.7087390422821045,
      "epoch": 5.244466618287373,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026077913171345356,
      "loss": 4.812,
      "mean_token_accuracy": 0.23621733486652374,
      "num_tokens": 119979363.0,
      "step": 57815
    },
    {
      "entropy": 5.747918033599854,
      "epoch": 5.244920174165458,
      "grad_norm": 1.484375,
      "learning_rate": 0.00026074684013302203,
      "loss": 4.8752,
      "mean_token_accuracy": 0.23256305754184722,
      "num_tokens": 119989766.0,
      "step": 57820
    },
    {
      "entropy": 5.777681875228882,
      "epoch": 5.245373730043541,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002607145488473494,
      "loss": 4.936,
      "mean_token_accuracy": 0.22497897148132323,
      "num_tokens": 120000836.0,
      "step": 57825
    },
    {
      "entropy": 5.786275720596313,
      "epoch": 5.245827285921625,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00026068225785710354,
      "loss": 4.8735,
      "mean_token_accuracy": 0.22752296626567842,
      "num_tokens": 120010747.0,
      "step": 57830
    },
    {
      "entropy": 5.739496278762817,
      "epoch": 5.2462808417997095,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002606499671629523,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23155307918787002,
      "num_tokens": 120020931.0,
      "step": 57835
    },
    {
      "entropy": 5.709837484359741,
      "epoch": 5.246734397677794,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002606176767655634,
      "loss": 4.871,
      "mean_token_accuracy": 0.22401734590530395,
      "num_tokens": 120031631.0,
      "step": 57840
    },
    {
      "entropy": 5.7967095375061035,
      "epoch": 5.247187953555878,
      "grad_norm": 1.5,
      "learning_rate": 0.0002605853866656046,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22854049056768416,
      "num_tokens": 120041861.0,
      "step": 57845
    },
    {
      "entropy": 5.708958721160888,
      "epoch": 5.247641509433962,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002605530968637436,
      "loss": 4.7994,
      "mean_token_accuracy": 0.23366958498954774,
      "num_tokens": 120052007.0,
      "step": 57850
    },
    {
      "entropy": 5.754988956451416,
      "epoch": 5.248095065312047,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002605208073606483,
      "loss": 4.8671,
      "mean_token_accuracy": 0.23134178221225737,
      "num_tokens": 120062001.0,
      "step": 57855
    },
    {
      "entropy": 5.680440044403076,
      "epoch": 5.248548621190131,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002604885181569864,
      "loss": 4.8237,
      "mean_token_accuracy": 0.2291140601038933,
      "num_tokens": 120072709.0,
      "step": 57860
    },
    {
      "entropy": 5.736977863311767,
      "epoch": 5.249002177068215,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002604562292534256,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2284591466188431,
      "num_tokens": 120083659.0,
      "step": 57865
    },
    {
      "entropy": 5.801371669769287,
      "epoch": 5.249455732946299,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002604239406506338,
      "loss": 4.8958,
      "mean_token_accuracy": 0.22847157567739487,
      "num_tokens": 120093872.0,
      "step": 57870
    },
    {
      "entropy": 5.738225030899048,
      "epoch": 5.249909288824383,
      "grad_norm": 1.375,
      "learning_rate": 0.00026039165234927855,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23141127973794937,
      "num_tokens": 120103533.0,
      "step": 57875
    },
    {
      "entropy": 5.782400369644165,
      "epoch": 5.250362844702467,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002603593643500276,
      "loss": 4.859,
      "mean_token_accuracy": 0.23528838902711868,
      "num_tokens": 120113006.0,
      "step": 57880
    },
    {
      "entropy": 5.75044264793396,
      "epoch": 5.250816400580551,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002603270766535488,
      "loss": 4.9019,
      "mean_token_accuracy": 0.2290702983736992,
      "num_tokens": 120124449.0,
      "step": 57885
    },
    {
      "entropy": 5.763515520095825,
      "epoch": 5.251269956458636,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00026029478926050976,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23360873758792877,
      "num_tokens": 120135100.0,
      "step": 57890
    },
    {
      "entropy": 5.812651205062866,
      "epoch": 5.25172351233672,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00026026250217157823,
      "loss": 4.8895,
      "mean_token_accuracy": 0.2258877143263817,
      "num_tokens": 120146276.0,
      "step": 57895
    },
    {
      "entropy": 5.77949538230896,
      "epoch": 5.252177068214804,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00026023021538742184,
      "loss": 4.8722,
      "mean_token_accuracy": 0.2367260903120041,
      "num_tokens": 120157028.0,
      "step": 57900
    },
    {
      "entropy": 5.7693733215332035,
      "epoch": 5.252630624092888,
      "grad_norm": 1.421875,
      "learning_rate": 0.00026019792890870846,
      "loss": 4.7736,
      "mean_token_accuracy": 0.2430350124835968,
      "num_tokens": 120167027.0,
      "step": 57905
    },
    {
      "entropy": 5.85188136100769,
      "epoch": 5.253084179970973,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026016564273610563,
      "loss": 4.9587,
      "mean_token_accuracy": 0.21873564124107361,
      "num_tokens": 120178060.0,
      "step": 57910
    },
    {
      "entropy": 5.72274169921875,
      "epoch": 5.253537735849057,
      "grad_norm": 1.40625,
      "learning_rate": 0.00026013335687028117,
      "loss": 4.8696,
      "mean_token_accuracy": 0.22924137562513353,
      "num_tokens": 120187869.0,
      "step": 57915
    },
    {
      "entropy": 5.75838885307312,
      "epoch": 5.253991291727141,
      "grad_norm": 1.46875,
      "learning_rate": 0.00026010107131190256,
      "loss": 4.8633,
      "mean_token_accuracy": 0.2346254125237465,
      "num_tokens": 120197701.0,
      "step": 57920
    },
    {
      "entropy": 5.658698797225952,
      "epoch": 5.254444847605225,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002600687860616376,
      "loss": 4.7545,
      "mean_token_accuracy": 0.23532989174127578,
      "num_tokens": 120208978.0,
      "step": 57925
    },
    {
      "entropy": 5.68812894821167,
      "epoch": 5.254898403483309,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002600365011201541,
      "loss": 4.8344,
      "mean_token_accuracy": 0.2370302602648735,
      "num_tokens": 120219021.0,
      "step": 57930
    },
    {
      "entropy": 5.726566934585572,
      "epoch": 5.255351959361393,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002600042164881194,
      "loss": 4.9236,
      "mean_token_accuracy": 0.21646389365196228,
      "num_tokens": 120229357.0,
      "step": 57935
    },
    {
      "entropy": 5.677946138381958,
      "epoch": 5.255805515239477,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002599719321662014,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2360778495669365,
      "num_tokens": 120239975.0,
      "step": 57940
    },
    {
      "entropy": 5.730272960662842,
      "epoch": 5.256259071117562,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002599396481550677,
      "loss": 4.7978,
      "mean_token_accuracy": 0.23611088693141938,
      "num_tokens": 120250764.0,
      "step": 57945
    },
    {
      "entropy": 5.734555864334107,
      "epoch": 5.256712626995646,
      "grad_norm": 1.375,
      "learning_rate": 0.0002599073644553858,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23086672127246857,
      "num_tokens": 120260503.0,
      "step": 57950
    },
    {
      "entropy": 5.760750436782837,
      "epoch": 5.25716618287373,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002598750810678236,
      "loss": 4.8188,
      "mean_token_accuracy": 0.23464962393045424,
      "num_tokens": 120270604.0,
      "step": 57955
    },
    {
      "entropy": 5.813714265823364,
      "epoch": 5.257619738751814,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002598427979930485,
      "loss": 4.8993,
      "mean_token_accuracy": 0.2259931355714798,
      "num_tokens": 120281601.0,
      "step": 57960
    },
    {
      "entropy": 5.776846170425415,
      "epoch": 5.258073294629899,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025981051523172823,
      "loss": 4.9253,
      "mean_token_accuracy": 0.227910815179348,
      "num_tokens": 120292431.0,
      "step": 57965
    },
    {
      "entropy": 5.821209669113159,
      "epoch": 5.258526850507983,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002597782327845303,
      "loss": 4.9341,
      "mean_token_accuracy": 0.22753820121288298,
      "num_tokens": 120303063.0,
      "step": 57970
    },
    {
      "entropy": 5.751330089569092,
      "epoch": 5.258980406386067,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002597459506521225,
      "loss": 4.8804,
      "mean_token_accuracy": 0.221867398917675,
      "num_tokens": 120313543.0,
      "step": 57975
    },
    {
      "entropy": 5.766824960708618,
      "epoch": 5.259433962264151,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025971366883517226,
      "loss": 4.8995,
      "mean_token_accuracy": 0.22445712685585023,
      "num_tokens": 120322409.0,
      "step": 57980
    },
    {
      "entropy": 5.733635282516479,
      "epoch": 5.259887518142235,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002596813873343472,
      "loss": 4.8588,
      "mean_token_accuracy": 0.23535745590925217,
      "num_tokens": 120332247.0,
      "step": 57985
    },
    {
      "entropy": 5.705030250549316,
      "epoch": 5.260341074020319,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000259649106150315,
      "loss": 4.8181,
      "mean_token_accuracy": 0.22976830452680588,
      "num_tokens": 120342224.0,
      "step": 57990
    },
    {
      "entropy": 5.8326653957366945,
      "epoch": 5.260794629898403,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025961682528374326,
      "loss": 4.9213,
      "mean_token_accuracy": 0.22413175255060197,
      "num_tokens": 120352093.0,
      "step": 57995
    },
    {
      "entropy": 5.841433906555176,
      "epoch": 5.261248185776488,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025958454473529944,
      "loss": 4.9659,
      "mean_token_accuracy": 0.2190907269716263,
      "num_tokens": 120363561.0,
      "step": 58000
    },
    {
      "entropy": 5.773711061477661,
      "epoch": 5.261701741654572,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025955226450565113,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22910771816968917,
      "num_tokens": 120374119.0,
      "step": 58005
    },
    {
      "entropy": 5.839850521087646,
      "epoch": 5.262155297532656,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025951998459546597,
      "loss": 4.985,
      "mean_token_accuracy": 0.21026467382907868,
      "num_tokens": 120384365.0,
      "step": 58010
    },
    {
      "entropy": 5.739672613143921,
      "epoch": 5.2626088534107405,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002594877050054115,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23670571744441987,
      "num_tokens": 120394341.0,
      "step": 58015
    },
    {
      "entropy": 5.709370088577271,
      "epoch": 5.263062409288825,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025945542573615517,
      "loss": 4.7637,
      "mean_token_accuracy": 0.2340611547231674,
      "num_tokens": 120404995.0,
      "step": 58020
    },
    {
      "entropy": 5.736138868331909,
      "epoch": 5.263515965166908,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025942314678836467,
      "loss": 4.9202,
      "mean_token_accuracy": 0.2237417832016945,
      "num_tokens": 120416443.0,
      "step": 58025
    },
    {
      "entropy": 5.737306308746338,
      "epoch": 5.263969521044992,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002593908681627074,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23620154559612275,
      "num_tokens": 120425915.0,
      "step": 58030
    },
    {
      "entropy": 5.846194458007813,
      "epoch": 5.264423076923077,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025935858985985097,
      "loss": 4.9831,
      "mean_token_accuracy": 0.21891015619039536,
      "num_tokens": 120436885.0,
      "step": 58035
    },
    {
      "entropy": 5.71836290359497,
      "epoch": 5.264876632801161,
      "grad_norm": 1.59375,
      "learning_rate": 0.0002593263118804629,
      "loss": 4.8096,
      "mean_token_accuracy": 0.23491533547639848,
      "num_tokens": 120447234.0,
      "step": 58040
    },
    {
      "entropy": 5.829079961776733,
      "epoch": 5.265330188679245,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002592940342252107,
      "loss": 4.8889,
      "mean_token_accuracy": 0.22820569425821305,
      "num_tokens": 120457385.0,
      "step": 58045
    },
    {
      "entropy": 5.753712558746338,
      "epoch": 5.2657837445573294,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025926175689476187,
      "loss": 4.7708,
      "mean_token_accuracy": 0.2336651623249054,
      "num_tokens": 120468495.0,
      "step": 58050
    },
    {
      "entropy": 5.7479650497436525,
      "epoch": 5.266237300435414,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002592294798897839,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2402576744556427,
      "num_tokens": 120478536.0,
      "step": 58055
    },
    {
      "entropy": 5.775179719924926,
      "epoch": 5.266690856313498,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025919720321094435,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2336572900414467,
      "num_tokens": 120489872.0,
      "step": 58060
    },
    {
      "entropy": 5.736604261398315,
      "epoch": 5.267144412191582,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002591649268589106,
      "loss": 4.8067,
      "mean_token_accuracy": 0.233882774412632,
      "num_tokens": 120499454.0,
      "step": 58065
    },
    {
      "entropy": 5.7551414489746096,
      "epoch": 5.2675979680696665,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025913265083435025,
      "loss": 4.8703,
      "mean_token_accuracy": 0.2319485828280449,
      "num_tokens": 120509974.0,
      "step": 58070
    },
    {
      "entropy": 5.746386194229126,
      "epoch": 5.268051523947751,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002591003751379307,
      "loss": 4.811,
      "mean_token_accuracy": 0.23060350865125656,
      "num_tokens": 120520310.0,
      "step": 58075
    },
    {
      "entropy": 5.766818332672119,
      "epoch": 5.268505079825834,
      "grad_norm": 1.375,
      "learning_rate": 0.0002590680997703195,
      "loss": 4.8252,
      "mean_token_accuracy": 0.22795485854148864,
      "num_tokens": 120530662.0,
      "step": 58080
    },
    {
      "entropy": 5.789261198043823,
      "epoch": 5.268958635703918,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025903582473218404,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22361414283514022,
      "num_tokens": 120541138.0,
      "step": 58085
    },
    {
      "entropy": 5.677810716629028,
      "epoch": 5.269412191582003,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025900355002419167,
      "loss": 4.7607,
      "mean_token_accuracy": 0.23790125995874406,
      "num_tokens": 120551758.0,
      "step": 58090
    },
    {
      "entropy": 5.697582530975342,
      "epoch": 5.269865747460087,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002589712756470101,
      "loss": 4.8342,
      "mean_token_accuracy": 0.22728884369134902,
      "num_tokens": 120562438.0,
      "step": 58095
    },
    {
      "entropy": 5.751409196853638,
      "epoch": 5.270319303338171,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002589390016013066,
      "loss": 4.8462,
      "mean_token_accuracy": 0.23252992033958436,
      "num_tokens": 120572032.0,
      "step": 58100
    },
    {
      "entropy": 5.833077096939087,
      "epoch": 5.2707728592162555,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002589067278877486,
      "loss": 4.8831,
      "mean_token_accuracy": 0.23223424702882767,
      "num_tokens": 120582415.0,
      "step": 58105
    },
    {
      "entropy": 5.778055334091187,
      "epoch": 5.27122641509434,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025887445450700357,
      "loss": 4.8639,
      "mean_token_accuracy": 0.23504395931959152,
      "num_tokens": 120593169.0,
      "step": 58110
    },
    {
      "entropy": 5.79623703956604,
      "epoch": 5.271679970972424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000258842181459739,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23230123072862624,
      "num_tokens": 120604472.0,
      "step": 58115
    },
    {
      "entropy": 5.764988946914673,
      "epoch": 5.272133526850508,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002588099087466222,
      "loss": 4.8388,
      "mean_token_accuracy": 0.2299970805644989,
      "num_tokens": 120614286.0,
      "step": 58120
    },
    {
      "entropy": 5.746915006637574,
      "epoch": 5.2725870827285926,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002587776363683206,
      "loss": 4.842,
      "mean_token_accuracy": 0.23240000605583191,
      "num_tokens": 120624130.0,
      "step": 58125
    },
    {
      "entropy": 5.754438638687134,
      "epoch": 5.273040638606676,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025874536432550163,
      "loss": 4.8891,
      "mean_token_accuracy": 0.23332862108945845,
      "num_tokens": 120634014.0,
      "step": 58130
    },
    {
      "entropy": 5.743218278884887,
      "epoch": 5.27349419448476,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002587130926188327,
      "loss": 4.865,
      "mean_token_accuracy": 0.2342361718416214,
      "num_tokens": 120645370.0,
      "step": 58135
    },
    {
      "entropy": 5.78173599243164,
      "epoch": 5.2739477503628445,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002586808212489811,
      "loss": 4.9248,
      "mean_token_accuracy": 0.21806923002004625,
      "num_tokens": 120655967.0,
      "step": 58140
    },
    {
      "entropy": 5.764715242385864,
      "epoch": 5.274401306240929,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025864855021661434,
      "loss": 4.853,
      "mean_token_accuracy": 0.22588689476251603,
      "num_tokens": 120666223.0,
      "step": 58145
    },
    {
      "entropy": 5.699695682525634,
      "epoch": 5.274854862119013,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002586162795223996,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23545179516077042,
      "num_tokens": 120676859.0,
      "step": 58150
    },
    {
      "entropy": 5.82137451171875,
      "epoch": 5.275308417997097,
      "grad_norm": 1.375,
      "learning_rate": 0.0002585840091670045,
      "loss": 4.9442,
      "mean_token_accuracy": 0.22889023274183273,
      "num_tokens": 120687930.0,
      "step": 58155
    },
    {
      "entropy": 5.733042860031128,
      "epoch": 5.2757619738751815,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025855173915109627,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23305374681949614,
      "num_tokens": 120698121.0,
      "step": 58160
    },
    {
      "entropy": 5.738875150680542,
      "epoch": 5.276215529753266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025851946947534214,
      "loss": 4.8259,
      "mean_token_accuracy": 0.22885158210992812,
      "num_tokens": 120708428.0,
      "step": 58165
    },
    {
      "entropy": 5.796575975418091,
      "epoch": 5.27666908563135,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025848720014040973,
      "loss": 4.9333,
      "mean_token_accuracy": 0.23108034133911132,
      "num_tokens": 120719726.0,
      "step": 58170
    },
    {
      "entropy": 5.740129899978638,
      "epoch": 5.277122641509434,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025845493114696615,
      "loss": 4.852,
      "mean_token_accuracy": 0.22875178456306458,
      "num_tokens": 120730698.0,
      "step": 58175
    },
    {
      "entropy": 5.785229587554932,
      "epoch": 5.277576197387518,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002584226624956789,
      "loss": 4.9094,
      "mean_token_accuracy": 0.2231430694460869,
      "num_tokens": 120741122.0,
      "step": 58180
    },
    {
      "entropy": 5.779577207565308,
      "epoch": 5.278029753265602,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002583903941872151,
      "loss": 4.865,
      "mean_token_accuracy": 0.22990071326494216,
      "num_tokens": 120751623.0,
      "step": 58185
    },
    {
      "entropy": 5.764954042434693,
      "epoch": 5.278483309143686,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002583581262222422,
      "loss": 4.8457,
      "mean_token_accuracy": 0.2291648805141449,
      "num_tokens": 120762203.0,
      "step": 58190
    },
    {
      "entropy": 5.720088243484497,
      "epoch": 5.2789368650217705,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025832585860142754,
      "loss": 4.8345,
      "mean_token_accuracy": 0.22912245094776154,
      "num_tokens": 120772732.0,
      "step": 58195
    },
    {
      "entropy": 5.760712671279907,
      "epoch": 5.279390420899855,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002582935913254383,
      "loss": 4.8149,
      "mean_token_accuracy": 0.2396260291337967,
      "num_tokens": 120782289.0,
      "step": 58200
    },
    {
      "entropy": 5.7803455829620365,
      "epoch": 5.279843976777939,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002582613243949419,
      "loss": 4.8938,
      "mean_token_accuracy": 0.2258755847811699,
      "num_tokens": 120793730.0,
      "step": 58205
    },
    {
      "entropy": 5.781848812103272,
      "epoch": 5.280297532656023,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025822905781060565,
      "loss": 4.92,
      "mean_token_accuracy": 0.2263464242219925,
      "num_tokens": 120804322.0,
      "step": 58210
    },
    {
      "entropy": 5.705565786361694,
      "epoch": 5.280751088534108,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025819679157309665,
      "loss": 4.8668,
      "mean_token_accuracy": 0.22718582302331924,
      "num_tokens": 120815095.0,
      "step": 58215
    },
    {
      "entropy": 5.741633558273316,
      "epoch": 5.281204644412192,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002581645256830823,
      "loss": 4.824,
      "mean_token_accuracy": 0.23060215264558792,
      "num_tokens": 120825941.0,
      "step": 58220
    },
    {
      "entropy": 5.71229100227356,
      "epoch": 5.281658200290276,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025813226014122985,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22922378182411193,
      "num_tokens": 120836512.0,
      "step": 58225
    },
    {
      "entropy": 5.847497892379761,
      "epoch": 5.28211175616836,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002580999949482066,
      "loss": 4.9003,
      "mean_token_accuracy": 0.23634578436613082,
      "num_tokens": 120846917.0,
      "step": 58230
    },
    {
      "entropy": 5.752352571487426,
      "epoch": 5.282565312046444,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002580677301046797,
      "loss": 4.8158,
      "mean_token_accuracy": 0.24143480509519577,
      "num_tokens": 120857084.0,
      "step": 58235
    },
    {
      "entropy": 5.73636269569397,
      "epoch": 5.283018867924528,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002580354656113165,
      "loss": 4.8346,
      "mean_token_accuracy": 0.22627011090517044,
      "num_tokens": 120867496.0,
      "step": 58240
    },
    {
      "entropy": 5.7112464427948,
      "epoch": 5.283472423802612,
      "grad_norm": 1.375,
      "learning_rate": 0.0002580032014687842,
      "loss": 4.8248,
      "mean_token_accuracy": 0.230662739276886,
      "num_tokens": 120877722.0,
      "step": 58245
    },
    {
      "entropy": 5.750569438934326,
      "epoch": 5.2839259796806965,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025797093767774995,
      "loss": 4.8923,
      "mean_token_accuracy": 0.23058352023363113,
      "num_tokens": 120888225.0,
      "step": 58250
    },
    {
      "entropy": 5.825892210006714,
      "epoch": 5.284379535558781,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002579386742388811,
      "loss": 4.9106,
      "mean_token_accuracy": 0.2212887540459633,
      "num_tokens": 120898664.0,
      "step": 58255
    },
    {
      "entropy": 5.727677869796753,
      "epoch": 5.284833091436865,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025790641115284486,
      "loss": 4.7781,
      "mean_token_accuracy": 0.23603951334953308,
      "num_tokens": 120909084.0,
      "step": 58260
    },
    {
      "entropy": 5.720179319381714,
      "epoch": 5.285286647314949,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002578741484203084,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23343392461538315,
      "num_tokens": 120919552.0,
      "step": 58265
    },
    {
      "entropy": 5.714810943603515,
      "epoch": 5.285740203193034,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002578418860419388,
      "loss": 4.839,
      "mean_token_accuracy": 0.22128835171461106,
      "num_tokens": 120929918.0,
      "step": 58270
    },
    {
      "entropy": 5.699560594558716,
      "epoch": 5.286193759071118,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002578096240184035,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23820021450519563,
      "num_tokens": 120940146.0,
      "step": 58275
    },
    {
      "entropy": 5.794398546218872,
      "epoch": 5.286647314949202,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002577773623503695,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2348590984940529,
      "num_tokens": 120950946.0,
      "step": 58280
    },
    {
      "entropy": 5.777624702453613,
      "epoch": 5.2871008708272855,
      "grad_norm": 1.375,
      "learning_rate": 0.000257745101038504,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22917619943618775,
      "num_tokens": 120961623.0,
      "step": 58285
    },
    {
      "entropy": 5.748729228973389,
      "epoch": 5.28755442670537,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002577128400834742,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23405245840549468,
      "num_tokens": 120971798.0,
      "step": 58290
    },
    {
      "entropy": 5.775893640518189,
      "epoch": 5.288007982583454,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025768057948594733,
      "loss": 4.9257,
      "mean_token_accuracy": 0.2224203661084175,
      "num_tokens": 120982042.0,
      "step": 58295
    },
    {
      "entropy": 5.741157102584839,
      "epoch": 5.288461538461538,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002576483192465905,
      "loss": 4.8533,
      "mean_token_accuracy": 0.22873178720474244,
      "num_tokens": 120991650.0,
      "step": 58300
    },
    {
      "entropy": 5.809757471084595,
      "epoch": 5.288915094339623,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025761605936607077,
      "loss": 4.8655,
      "mean_token_accuracy": 0.23241062909364701,
      "num_tokens": 121001660.0,
      "step": 58305
    },
    {
      "entropy": 5.778255558013916,
      "epoch": 5.289368650217707,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002575837998450554,
      "loss": 4.8607,
      "mean_token_accuracy": 0.23226837813854218,
      "num_tokens": 121011065.0,
      "step": 58310
    },
    {
      "entropy": 5.667554664611816,
      "epoch": 5.289822206095791,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025755154068421156,
      "loss": 4.7314,
      "mean_token_accuracy": 0.23774618804454803,
      "num_tokens": 121020520.0,
      "step": 58315
    },
    {
      "entropy": 5.737356090545655,
      "epoch": 5.290275761973875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002575192818842062,
      "loss": 4.8813,
      "mean_token_accuracy": 0.22211106419563292,
      "num_tokens": 121030525.0,
      "step": 58320
    },
    {
      "entropy": 5.717375659942627,
      "epoch": 5.29072931785196,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025748702344570657,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23344827741384505,
      "num_tokens": 121041285.0,
      "step": 58325
    },
    {
      "entropy": 5.78113489151001,
      "epoch": 5.291182873730044,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002574547653693798,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22623443454504014,
      "num_tokens": 121050787.0,
      "step": 58330
    },
    {
      "entropy": 5.820376396179199,
      "epoch": 5.291636429608127,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002574225076558929,
      "loss": 4.9686,
      "mean_token_accuracy": 0.21978258937597275,
      "num_tokens": 121061088.0,
      "step": 58335
    },
    {
      "entropy": 5.797939491271973,
      "epoch": 5.292089985486212,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025739025030591307,
      "loss": 4.923,
      "mean_token_accuracy": 0.22056026458740235,
      "num_tokens": 121071111.0,
      "step": 58340
    },
    {
      "entropy": 5.864497470855713,
      "epoch": 5.292543541364296,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002573579933201073,
      "loss": 4.9227,
      "mean_token_accuracy": 0.22373994141817094,
      "num_tokens": 121081339.0,
      "step": 58345
    },
    {
      "entropy": 5.704949522018433,
      "epoch": 5.29299709724238,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025732573669914286,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23256916403770447,
      "num_tokens": 121091830.0,
      "step": 58350
    },
    {
      "entropy": 5.6462428092956545,
      "epoch": 5.293450653120464,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002572934804436865,
      "loss": 4.7642,
      "mean_token_accuracy": 0.23830267786979675,
      "num_tokens": 121102727.0,
      "step": 58355
    },
    {
      "entropy": 5.764221096038819,
      "epoch": 5.293904208998549,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002572612245544056,
      "loss": 4.9323,
      "mean_token_accuracy": 0.22268909960985184,
      "num_tokens": 121112845.0,
      "step": 58360
    },
    {
      "entropy": 5.714921951293945,
      "epoch": 5.294357764876633,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025722896903196705,
      "loss": 4.7419,
      "mean_token_accuracy": 0.23399266302585603,
      "num_tokens": 121123456.0,
      "step": 58365
    },
    {
      "entropy": 5.703054618835449,
      "epoch": 5.294811320754717,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002571967138770379,
      "loss": 4.7352,
      "mean_token_accuracy": 0.23378469944000244,
      "num_tokens": 121133185.0,
      "step": 58370
    },
    {
      "entropy": 5.692541837692261,
      "epoch": 5.295264876632801,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025716445909028537,
      "loss": 4.8226,
      "mean_token_accuracy": 0.22771742194890976,
      "num_tokens": 121143962.0,
      "step": 58375
    },
    {
      "entropy": 5.729185390472412,
      "epoch": 5.295718432510886,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002571322046723763,
      "loss": 4.8173,
      "mean_token_accuracy": 0.2342734307050705,
      "num_tokens": 121153184.0,
      "step": 58380
    },
    {
      "entropy": 5.712761783599854,
      "epoch": 5.29617198838897,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002570999506239778,
      "loss": 4.8646,
      "mean_token_accuracy": 0.2249797537922859,
      "num_tokens": 121165189.0,
      "step": 58385
    },
    {
      "entropy": 5.786094808578492,
      "epoch": 5.296625544267053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025706769694575684,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2312637522816658,
      "num_tokens": 121174787.0,
      "step": 58390
    },
    {
      "entropy": 5.7396660327911375,
      "epoch": 5.297079100145138,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002570354436383805,
      "loss": 4.9471,
      "mean_token_accuracy": 0.22590580433607102,
      "num_tokens": 121185334.0,
      "step": 58395
    },
    {
      "entropy": 5.872246932983399,
      "epoch": 5.297532656023222,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002570031907025158,
      "loss": 4.9874,
      "mean_token_accuracy": 0.2158442035317421,
      "num_tokens": 121195928.0,
      "step": 58400
    },
    {
      "entropy": 5.840354871749878,
      "epoch": 5.297986211901306,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025697093813882967,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21571983993053437,
      "num_tokens": 121206369.0,
      "step": 58405
    },
    {
      "entropy": 5.808262300491333,
      "epoch": 5.29843976777939,
      "grad_norm": 1.375,
      "learning_rate": 0.0002569386859479891,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23361335545778275,
      "num_tokens": 121216573.0,
      "step": 58410
    },
    {
      "entropy": 5.718575286865234,
      "epoch": 5.298893323657475,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002569064341306612,
      "loss": 4.8141,
      "mean_token_accuracy": 0.23059422075748442,
      "num_tokens": 121226330.0,
      "step": 58415
    },
    {
      "entropy": 5.737021493911743,
      "epoch": 5.299346879535559,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025687418268751283,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2305547222495079,
      "num_tokens": 121236171.0,
      "step": 58420
    },
    {
      "entropy": 5.741543626785278,
      "epoch": 5.299800435413643,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025684193161921087,
      "loss": 4.864,
      "mean_token_accuracy": 0.22664555460214614,
      "num_tokens": 121246506.0,
      "step": 58425
    },
    {
      "entropy": 5.799118757247925,
      "epoch": 5.3002539912917275,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025680968092642255,
      "loss": 4.901,
      "mean_token_accuracy": 0.22325329780578612,
      "num_tokens": 121257080.0,
      "step": 58430
    },
    {
      "entropy": 5.8160576820373535,
      "epoch": 5.300707547169811,
      "grad_norm": 1.53125,
      "learning_rate": 0.00025677743060981454,
      "loss": 4.9186,
      "mean_token_accuracy": 0.23469147235155105,
      "num_tokens": 121266349.0,
      "step": 58435
    },
    {
      "entropy": 5.725163125991822,
      "epoch": 5.301161103047895,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002567451806700539,
      "loss": 4.8394,
      "mean_token_accuracy": 0.22615112513303756,
      "num_tokens": 121278410.0,
      "step": 58440
    },
    {
      "entropy": 5.749270963668823,
      "epoch": 5.301614658925979,
      "grad_norm": 1.375,
      "learning_rate": 0.0002567129311078077,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23388071358203888,
      "num_tokens": 121289334.0,
      "step": 58445
    },
    {
      "entropy": 5.767132568359375,
      "epoch": 5.302068214804064,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025668068192374264,
      "loss": 4.9161,
      "mean_token_accuracy": 0.22530641704797744,
      "num_tokens": 121299132.0,
      "step": 58450
    },
    {
      "entropy": 5.665415859222412,
      "epoch": 5.302521770682148,
      "grad_norm": 1.375,
      "learning_rate": 0.00025664843311852574,
      "loss": 4.7817,
      "mean_token_accuracy": 0.2316224068403244,
      "num_tokens": 121308905.0,
      "step": 58455
    },
    {
      "entropy": 5.7253954887390135,
      "epoch": 5.302975326560232,
      "grad_norm": 1.546875,
      "learning_rate": 0.000256616184692824,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23334518820047379,
      "num_tokens": 121318719.0,
      "step": 58460
    },
    {
      "entropy": 5.791615152359009,
      "epoch": 5.3034288824383164,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002565839366473042,
      "loss": 4.9476,
      "mean_token_accuracy": 0.2209972009062767,
      "num_tokens": 121328501.0,
      "step": 58465
    },
    {
      "entropy": 5.804558134078979,
      "epoch": 5.303882438316401,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025655168898263336,
      "loss": 4.8669,
      "mean_token_accuracy": 0.22649078220129013,
      "num_tokens": 121339050.0,
      "step": 58470
    },
    {
      "entropy": 5.756304693222046,
      "epoch": 5.304335994194485,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002565194416994782,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23225207924842833,
      "num_tokens": 121347699.0,
      "step": 58475
    },
    {
      "entropy": 5.725672817230224,
      "epoch": 5.304789550072569,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002564871947985057,
      "loss": 4.818,
      "mean_token_accuracy": 0.23687656670808793,
      "num_tokens": 121357623.0,
      "step": 58480
    },
    {
      "entropy": 5.67668662071228,
      "epoch": 5.3052431059506535,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002564549482803828,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23188210278749466,
      "num_tokens": 121367933.0,
      "step": 58485
    },
    {
      "entropy": 5.756233263015747,
      "epoch": 5.305696661828737,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025642270214577617,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22006010711193086,
      "num_tokens": 121377977.0,
      "step": 58490
    },
    {
      "entropy": 5.806051015853882,
      "epoch": 5.306150217706821,
      "grad_norm": 1.375,
      "learning_rate": 0.00025639045639535283,
      "loss": 4.9993,
      "mean_token_accuracy": 0.22080691158771515,
      "num_tokens": 121387866.0,
      "step": 58495
    },
    {
      "entropy": 5.734874534606933,
      "epoch": 5.306603773584905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002563582110297797,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2279014229774475,
      "num_tokens": 121397659.0,
      "step": 58500
    },
    {
      "entropy": 5.737406778335571,
      "epoch": 5.30705732946299,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025632596604972346,
      "loss": 4.843,
      "mean_token_accuracy": 0.23327345997095109,
      "num_tokens": 121407907.0,
      "step": 58505
    },
    {
      "entropy": 5.6882405281066895,
      "epoch": 5.307510885341074,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025629372145585095,
      "loss": 4.7466,
      "mean_token_accuracy": 0.24575672745704652,
      "num_tokens": 121417882.0,
      "step": 58510
    },
    {
      "entropy": 5.705540943145752,
      "epoch": 5.307964441219158,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002562614772488291,
      "loss": 4.8756,
      "mean_token_accuracy": 0.22735249549150466,
      "num_tokens": 121427522.0,
      "step": 58515
    },
    {
      "entropy": 5.747119283676147,
      "epoch": 5.3084179970972425,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002562292334293247,
      "loss": 4.8404,
      "mean_token_accuracy": 0.23718902319669724,
      "num_tokens": 121437796.0,
      "step": 58520
    },
    {
      "entropy": 5.755185985565186,
      "epoch": 5.308871552975327,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025619698999800446,
      "loss": 4.8482,
      "mean_token_accuracy": 0.23031722605228425,
      "num_tokens": 121448096.0,
      "step": 58525
    },
    {
      "entropy": 5.726437759399414,
      "epoch": 5.309325108853411,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025616474695553526,
      "loss": 4.7834,
      "mean_token_accuracy": 0.23277649879455567,
      "num_tokens": 121458017.0,
      "step": 58530
    },
    {
      "entropy": 5.785188770294189,
      "epoch": 5.309778664731495,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025613250430258397,
      "loss": 4.9222,
      "mean_token_accuracy": 0.22025909423828124,
      "num_tokens": 121468662.0,
      "step": 58535
    },
    {
      "entropy": 5.785566949844361,
      "epoch": 5.31023222060958,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025610026203981724,
      "loss": 4.8847,
      "mean_token_accuracy": 0.22891699373722077,
      "num_tokens": 121478988.0,
      "step": 58540
    },
    {
      "entropy": 5.738285398483276,
      "epoch": 5.310685776487663,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025606802016790194,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23565197438001634,
      "num_tokens": 121487737.0,
      "step": 58545
    },
    {
      "entropy": 5.771641206741333,
      "epoch": 5.311139332365747,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002560357786875048,
      "loss": 4.9032,
      "mean_token_accuracy": 0.23219513893127441,
      "num_tokens": 121498670.0,
      "step": 58550
    },
    {
      "entropy": 5.8035050392150875,
      "epoch": 5.3115928882438315,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002560035375992926,
      "loss": 4.8999,
      "mean_token_accuracy": 0.23067188858985901,
      "num_tokens": 121508294.0,
      "step": 58555
    },
    {
      "entropy": 5.766126251220703,
      "epoch": 5.312046444121916,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000255971296903932,
      "loss": 4.8876,
      "mean_token_accuracy": 0.23138874322175979,
      "num_tokens": 121518046.0,
      "step": 58560
    },
    {
      "entropy": 5.718022871017456,
      "epoch": 5.3125,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025593905660209004,
      "loss": 4.8204,
      "mean_token_accuracy": 0.23264157623052598,
      "num_tokens": 121527779.0,
      "step": 58565
    },
    {
      "entropy": 5.738268852233887,
      "epoch": 5.312953555878084,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025590681669443303,
      "loss": 4.872,
      "mean_token_accuracy": 0.22282867431640624,
      "num_tokens": 121538365.0,
      "step": 58570
    },
    {
      "entropy": 5.765929651260376,
      "epoch": 5.3134071117561685,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000255874577181628,
      "loss": 4.8539,
      "mean_token_accuracy": 0.22471943497657776,
      "num_tokens": 121548668.0,
      "step": 58575
    },
    {
      "entropy": 5.7709118843078615,
      "epoch": 5.313860667634253,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025584233806434165,
      "loss": 4.8805,
      "mean_token_accuracy": 0.22274496406316757,
      "num_tokens": 121559276.0,
      "step": 58580
    },
    {
      "entropy": 5.6963190078735355,
      "epoch": 5.314314223512337,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002558100993432405,
      "loss": 4.8313,
      "mean_token_accuracy": 0.24246373921632766,
      "num_tokens": 121569686.0,
      "step": 58585
    },
    {
      "entropy": 5.7888744354248045,
      "epoch": 5.31476777939042,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002557778610189914,
      "loss": 4.852,
      "mean_token_accuracy": 0.23456939607858657,
      "num_tokens": 121580443.0,
      "step": 58590
    },
    {
      "entropy": 5.7818115234375,
      "epoch": 5.315221335268505,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025574562309226115,
      "loss": 4.8797,
      "mean_token_accuracy": 0.22515979558229446,
      "num_tokens": 121592393.0,
      "step": 58595
    },
    {
      "entropy": 5.789609909057617,
      "epoch": 5.315674891146589,
      "grad_norm": 1.25,
      "learning_rate": 0.0002557133855637163,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22899097800254822,
      "num_tokens": 121603394.0,
      "step": 58600
    },
    {
      "entropy": 5.754934358596802,
      "epoch": 5.316128447024673,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002556811484340234,
      "loss": 4.8743,
      "mean_token_accuracy": 0.2197237566113472,
      "num_tokens": 121613719.0,
      "step": 58605
    },
    {
      "entropy": 5.728410720825195,
      "epoch": 5.3165820029027575,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002556489117038493,
      "loss": 4.8362,
      "mean_token_accuracy": 0.23416124284267426,
      "num_tokens": 121623897.0,
      "step": 58610
    },
    {
      "entropy": 5.7461730480194095,
      "epoch": 5.317035558780842,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002556166753738608,
      "loss": 4.834,
      "mean_token_accuracy": 0.2380073696374893,
      "num_tokens": 121633916.0,
      "step": 58615
    },
    {
      "entropy": 5.801687908172608,
      "epoch": 5.317489114658926,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025558443944472426,
      "loss": 4.9056,
      "mean_token_accuracy": 0.22618184089660645,
      "num_tokens": 121643828.0,
      "step": 58620
    },
    {
      "entropy": 5.670260286331176,
      "epoch": 5.31794267053701,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002555522039171065,
      "loss": 4.783,
      "mean_token_accuracy": 0.23175861239433287,
      "num_tokens": 121652956.0,
      "step": 58625
    },
    {
      "entropy": 5.687716484069824,
      "epoch": 5.318396226415095,
      "grad_norm": 1.5,
      "learning_rate": 0.00025551996879167417,
      "loss": 4.8367,
      "mean_token_accuracy": 0.23078349381685256,
      "num_tokens": 121663199.0,
      "step": 58630
    },
    {
      "entropy": 5.686079359054565,
      "epoch": 5.318849782293179,
      "grad_norm": 1.625,
      "learning_rate": 0.00025548773406909375,
      "loss": 4.8062,
      "mean_token_accuracy": 0.23465202301740645,
      "num_tokens": 121672731.0,
      "step": 58635
    },
    {
      "entropy": 5.825100898742676,
      "epoch": 5.319303338171263,
      "grad_norm": 1.28125,
      "learning_rate": 0.000255455499750032,
      "loss": 4.9563,
      "mean_token_accuracy": 0.21975841522216796,
      "num_tokens": 121683307.0,
      "step": 58640
    },
    {
      "entropy": 5.772555732727051,
      "epoch": 5.3197568940493465,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00025542326583515555,
      "loss": 4.905,
      "mean_token_accuracy": 0.2283428966999054,
      "num_tokens": 121693944.0,
      "step": 58645
    },
    {
      "entropy": 5.788543081283569,
      "epoch": 5.320210449927431,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002553910323251309,
      "loss": 4.8791,
      "mean_token_accuracy": 0.23728718608617783,
      "num_tokens": 121704930.0,
      "step": 58650
    },
    {
      "entropy": 5.788434791564941,
      "epoch": 5.320664005805515,
      "grad_norm": 1.375,
      "learning_rate": 0.00025535879922062465,
      "loss": 4.9039,
      "mean_token_accuracy": 0.22041577100753784,
      "num_tokens": 121715924.0,
      "step": 58655
    },
    {
      "entropy": 5.745467567443848,
      "epoch": 5.321117561683599,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002553265665223035,
      "loss": 4.8401,
      "mean_token_accuracy": 0.23918775767087935,
      "num_tokens": 121726658.0,
      "step": 58660
    },
    {
      "entropy": 5.802395153045654,
      "epoch": 5.3215711175616836,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025529433423083397,
      "loss": 4.9778,
      "mean_token_accuracy": 0.21960523426532746,
      "num_tokens": 121737019.0,
      "step": 58665
    },
    {
      "entropy": 5.817547941207886,
      "epoch": 5.322024673439768,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002552621023468826,
      "loss": 4.875,
      "mean_token_accuracy": 0.23022617548704147,
      "num_tokens": 121748013.0,
      "step": 58670
    },
    {
      "entropy": 5.719989776611328,
      "epoch": 5.322478229317852,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025522987087111604,
      "loss": 4.7797,
      "mean_token_accuracy": 0.23225608170032502,
      "num_tokens": 121758873.0,
      "step": 58675
    },
    {
      "entropy": 5.695825719833374,
      "epoch": 5.322931785195936,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025519763980420075,
      "loss": 4.7593,
      "mean_token_accuracy": 0.2404251828789711,
      "num_tokens": 121769565.0,
      "step": 58680
    },
    {
      "entropy": 5.727195739746094,
      "epoch": 5.323385341074021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025516540914680337,
      "loss": 4.8566,
      "mean_token_accuracy": 0.2276360273361206,
      "num_tokens": 121780259.0,
      "step": 58685
    },
    {
      "entropy": 5.680407524108887,
      "epoch": 5.323838896952105,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025513317889959024,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23580965846776963,
      "num_tokens": 121791608.0,
      "step": 58690
    },
    {
      "entropy": 5.679393672943116,
      "epoch": 5.324292452830189,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025510094906322815,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23284097462892533,
      "num_tokens": 121800954.0,
      "step": 58695
    },
    {
      "entropy": 5.68446455001831,
      "epoch": 5.3247460087082725,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002550687196383835,
      "loss": 4.7657,
      "mean_token_accuracy": 0.2319561406970024,
      "num_tokens": 121809780.0,
      "step": 58700
    },
    {
      "entropy": 5.697523927688598,
      "epoch": 5.325199564586357,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025503649062572276,
      "loss": 4.8524,
      "mean_token_accuracy": 0.22547666728496552,
      "num_tokens": 121819941.0,
      "step": 58705
    },
    {
      "entropy": 5.751297235488892,
      "epoch": 5.325653120464441,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002550042620259125,
      "loss": 4.9548,
      "mean_token_accuracy": 0.21555543541908265,
      "num_tokens": 121831176.0,
      "step": 58710
    },
    {
      "entropy": 5.774681472778321,
      "epoch": 5.326106676342525,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025497203383961927,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22789106369018555,
      "num_tokens": 121840832.0,
      "step": 58715
    },
    {
      "entropy": 5.745599746704102,
      "epoch": 5.32656023222061,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002549398060675094,
      "loss": 4.9081,
      "mean_token_accuracy": 0.21903801560401917,
      "num_tokens": 121851976.0,
      "step": 58720
    },
    {
      "entropy": 5.6736695766448975,
      "epoch": 5.327013788098694,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025490757871024944,
      "loss": 4.7727,
      "mean_token_accuracy": 0.23283199965953827,
      "num_tokens": 121862359.0,
      "step": 58725
    },
    {
      "entropy": 5.715384006500244,
      "epoch": 5.327467343976778,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025487535176850593,
      "loss": 4.7603,
      "mean_token_accuracy": 0.23497437983751296,
      "num_tokens": 121872306.0,
      "step": 58730
    },
    {
      "entropy": 5.778316831588745,
      "epoch": 5.327920899854862,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025484312524294523,
      "loss": 4.88,
      "mean_token_accuracy": 0.2280063211917877,
      "num_tokens": 121881711.0,
      "step": 58735
    },
    {
      "entropy": 5.725603008270264,
      "epoch": 5.328374455732947,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002548108991342338,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23454095125198365,
      "num_tokens": 121892781.0,
      "step": 58740
    },
    {
      "entropy": 5.746542167663574,
      "epoch": 5.32882801161103,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002547786734430382,
      "loss": 4.8885,
      "mean_token_accuracy": 0.2306187316775322,
      "num_tokens": 121902329.0,
      "step": 58745
    },
    {
      "entropy": 5.679496812820434,
      "epoch": 5.329281567489114,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002547464481700247,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23430726528167725,
      "num_tokens": 121913410.0,
      "step": 58750
    },
    {
      "entropy": 5.669157981872559,
      "epoch": 5.329735123367199,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002547142233158599,
      "loss": 4.7752,
      "mean_token_accuracy": 0.23324715495109558,
      "num_tokens": 121923959.0,
      "step": 58755
    },
    {
      "entropy": 5.704904985427857,
      "epoch": 5.330188679245283,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025468199888121015,
      "loss": 4.796,
      "mean_token_accuracy": 0.23502748906612397,
      "num_tokens": 121934988.0,
      "step": 58760
    },
    {
      "entropy": 5.7916093349456785,
      "epoch": 5.330642235123367,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002546497748667418,
      "loss": 4.9093,
      "mean_token_accuracy": 0.2223212391138077,
      "num_tokens": 121944848.0,
      "step": 58765
    },
    {
      "entropy": 5.798360443115234,
      "epoch": 5.331095791001451,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025461755127312123,
      "loss": 4.897,
      "mean_token_accuracy": 0.22752373069524764,
      "num_tokens": 121956085.0,
      "step": 58770
    },
    {
      "entropy": 5.851776266098023,
      "epoch": 5.331549346879536,
      "grad_norm": 1.390625,
      "learning_rate": 0.000254585328101015,
      "loss": 4.9786,
      "mean_token_accuracy": 0.2201611027121544,
      "num_tokens": 121967165.0,
      "step": 58775
    },
    {
      "entropy": 5.814404249191284,
      "epoch": 5.33200290275762,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002545531053510893,
      "loss": 4.9087,
      "mean_token_accuracy": 0.2285129025578499,
      "num_tokens": 121978951.0,
      "step": 58780
    },
    {
      "entropy": 5.748059272766113,
      "epoch": 5.332456458635704,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002545208830240107,
      "loss": 4.8538,
      "mean_token_accuracy": 0.22854154109954833,
      "num_tokens": 121989130.0,
      "step": 58785
    },
    {
      "entropy": 5.794121408462525,
      "epoch": 5.332910014513788,
      "grad_norm": 1.34375,
      "learning_rate": 0.00025448866112044526,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23381573259830474,
      "num_tokens": 121999097.0,
      "step": 58790
    },
    {
      "entropy": 5.816999435424805,
      "epoch": 5.333363570391873,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025445643964105974,
      "loss": 4.9624,
      "mean_token_accuracy": 0.21969965994358062,
      "num_tokens": 122009385.0,
      "step": 58795
    },
    {
      "entropy": 5.7185920715332035,
      "epoch": 5.333817126269956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002544242185865202,
      "loss": 4.8837,
      "mean_token_accuracy": 0.22890662550926208,
      "num_tokens": 122020331.0,
      "step": 58800
    },
    {
      "entropy": 5.759378337860108,
      "epoch": 5.33427068214804,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002543919979574931,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23183712363243103,
      "num_tokens": 122030968.0,
      "step": 58805
    },
    {
      "entropy": 5.774029016494751,
      "epoch": 5.334724238026125,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025435977775464466,
      "loss": 4.8398,
      "mean_token_accuracy": 0.2266295537352562,
      "num_tokens": 122040759.0,
      "step": 58810
    },
    {
      "entropy": 5.798574924468994,
      "epoch": 5.335177793904209,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002543275579786413,
      "loss": 4.9177,
      "mean_token_accuracy": 0.23353657871484756,
      "num_tokens": 122050986.0,
      "step": 58815
    },
    {
      "entropy": 5.753296995162964,
      "epoch": 5.335631349782293,
      "grad_norm": 1.515625,
      "learning_rate": 0.00025429533863014934,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22704880237579345,
      "num_tokens": 122060602.0,
      "step": 58820
    },
    {
      "entropy": 5.775974225997925,
      "epoch": 5.336084905660377,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000254263119709835,
      "loss": 4.9206,
      "mean_token_accuracy": 0.22651485353708267,
      "num_tokens": 122070863.0,
      "step": 58825
    },
    {
      "entropy": 5.734273052215576,
      "epoch": 5.336538461538462,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002542309012183647,
      "loss": 4.9308,
      "mean_token_accuracy": 0.22486617863178254,
      "num_tokens": 122080756.0,
      "step": 58830
    },
    {
      "entropy": 5.748002624511718,
      "epoch": 5.336992017416546,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002541986831564046,
      "loss": 4.8991,
      "mean_token_accuracy": 0.22841140031814575,
      "num_tokens": 122090695.0,
      "step": 58835
    },
    {
      "entropy": 5.7708676815032955,
      "epoch": 5.33744557329463,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025416646552462103,
      "loss": 4.8107,
      "mean_token_accuracy": 0.24028520584106444,
      "num_tokens": 122101269.0,
      "step": 58840
    },
    {
      "entropy": 5.757588624954224,
      "epoch": 5.3378991291727145,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002541342483236803,
      "loss": 4.8341,
      "mean_token_accuracy": 0.23167395144701003,
      "num_tokens": 122111966.0,
      "step": 58845
    },
    {
      "entropy": 5.820436000823975,
      "epoch": 5.338352685050798,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00025410203155424866,
      "loss": 4.8466,
      "mean_token_accuracy": 0.23750391900539397,
      "num_tokens": 122121936.0,
      "step": 58850
    },
    {
      "entropy": 5.763745832443237,
      "epoch": 5.338806240928882,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025406981521699227,
      "loss": 4.8649,
      "mean_token_accuracy": 0.22572024315595626,
      "num_tokens": 122131943.0,
      "step": 58855
    },
    {
      "entropy": 5.762346649169922,
      "epoch": 5.339259796806966,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025403759931257746,
      "loss": 4.8751,
      "mean_token_accuracy": 0.22685976177453995,
      "num_tokens": 122141613.0,
      "step": 58860
    },
    {
      "entropy": 5.771994113922119,
      "epoch": 5.339713352685051,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025400538384167044,
      "loss": 4.8896,
      "mean_token_accuracy": 0.22975657433271407,
      "num_tokens": 122153141.0,
      "step": 58865
    },
    {
      "entropy": 5.719660663604737,
      "epoch": 5.340166908563135,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002539731688049374,
      "loss": 4.8037,
      "mean_token_accuracy": 0.2292369857430458,
      "num_tokens": 122164556.0,
      "step": 58870
    },
    {
      "entropy": 5.832929706573486,
      "epoch": 5.340620464441219,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002539409542030446,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22560137659311294,
      "num_tokens": 122175504.0,
      "step": 58875
    },
    {
      "entropy": 5.703085660934448,
      "epoch": 5.3410740203193035,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025390874003665826,
      "loss": 4.8734,
      "mean_token_accuracy": 0.23496291637420655,
      "num_tokens": 122186346.0,
      "step": 58880
    },
    {
      "entropy": 5.749679803848267,
      "epoch": 5.341527576197388,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002538765263064446,
      "loss": 4.9225,
      "mean_token_accuracy": 0.22683555334806443,
      "num_tokens": 122196647.0,
      "step": 58885
    },
    {
      "entropy": 5.756049728393554,
      "epoch": 5.341981132075472,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002538443130130697,
      "loss": 4.8623,
      "mean_token_accuracy": 0.22601027190685272,
      "num_tokens": 122209041.0,
      "step": 58890
    },
    {
      "entropy": 5.777977085113525,
      "epoch": 5.342434687953556,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00025381210015719977,
      "loss": 4.8688,
      "mean_token_accuracy": 0.2308949872851372,
      "num_tokens": 122218330.0,
      "step": 58895
    },
    {
      "entropy": 5.742093944549561,
      "epoch": 5.34288824383164,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025377988773950103,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2271358385682106,
      "num_tokens": 122227930.0,
      "step": 58900
    },
    {
      "entropy": 5.775581407546997,
      "epoch": 5.343341799709724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025374767576063963,
      "loss": 4.887,
      "mean_token_accuracy": 0.2276812806725502,
      "num_tokens": 122238028.0,
      "step": 58905
    },
    {
      "entropy": 5.75932674407959,
      "epoch": 5.343795355587808,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025371546422128165,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23588544726371766,
      "num_tokens": 122249018.0,
      "step": 58910
    },
    {
      "entropy": 5.763100671768188,
      "epoch": 5.344248911465892,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025368325312209337,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2265843406319618,
      "num_tokens": 122258635.0,
      "step": 58915
    },
    {
      "entropy": 5.669419050216675,
      "epoch": 5.344702467343977,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002536510424637409,
      "loss": 4.7571,
      "mean_token_accuracy": 0.24175533503293992,
      "num_tokens": 122268370.0,
      "step": 58920
    },
    {
      "entropy": 5.679708766937256,
      "epoch": 5.345156023222061,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025361883224689027,
      "loss": 4.8028,
      "mean_token_accuracy": 0.23486155420541763,
      "num_tokens": 122279044.0,
      "step": 58925
    },
    {
      "entropy": 5.737201118469239,
      "epoch": 5.345609579100145,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025358662247220755,
      "loss": 4.8775,
      "mean_token_accuracy": 0.2275758296251297,
      "num_tokens": 122289539.0,
      "step": 58930
    },
    {
      "entropy": 5.748228693008423,
      "epoch": 5.3460631349782295,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025355441314035897,
      "loss": 4.7617,
      "mean_token_accuracy": 0.238874152302742,
      "num_tokens": 122300263.0,
      "step": 58935
    },
    {
      "entropy": 5.789642477035523,
      "epoch": 5.346516690856314,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002535222042520106,
      "loss": 4.872,
      "mean_token_accuracy": 0.22701238691806794,
      "num_tokens": 122310813.0,
      "step": 58940
    },
    {
      "entropy": 5.74608416557312,
      "epoch": 5.346970246734398,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002534899958078285,
      "loss": 4.904,
      "mean_token_accuracy": 0.22732759714126588,
      "num_tokens": 122321306.0,
      "step": 58945
    },
    {
      "entropy": 5.768684720993042,
      "epoch": 5.347423802612482,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025345778780847884,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23457985669374465,
      "num_tokens": 122331355.0,
      "step": 58950
    },
    {
      "entropy": 5.7464018821716305,
      "epoch": 5.347877358490566,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002534255802546276,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23574353605508805,
      "num_tokens": 122341707.0,
      "step": 58955
    },
    {
      "entropy": 5.809862518310547,
      "epoch": 5.34833091436865,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002533933731469408,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22833797186613083,
      "num_tokens": 122351680.0,
      "step": 58960
    },
    {
      "entropy": 5.781007194519043,
      "epoch": 5.348784470246734,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002533611664860846,
      "loss": 4.8597,
      "mean_token_accuracy": 0.2281623214483261,
      "num_tokens": 122361895.0,
      "step": 58965
    },
    {
      "entropy": 5.7567445755004885,
      "epoch": 5.3492380261248185,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025332896027272493,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2266845941543579,
      "num_tokens": 122372652.0,
      "step": 58970
    },
    {
      "entropy": 5.761467409133911,
      "epoch": 5.349691582002903,
      "grad_norm": 1.5,
      "learning_rate": 0.000253296754507528,
      "loss": 4.9051,
      "mean_token_accuracy": 0.21718395203351976,
      "num_tokens": 122383372.0,
      "step": 58975
    },
    {
      "entropy": 5.7597630500793455,
      "epoch": 5.350145137880987,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002532645491911596,
      "loss": 4.8909,
      "mean_token_accuracy": 0.22816707491874694,
      "num_tokens": 122392544.0,
      "step": 58980
    },
    {
      "entropy": 5.767343044281006,
      "epoch": 5.350598693759071,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025323234432428587,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2312971383333206,
      "num_tokens": 122402869.0,
      "step": 58985
    },
    {
      "entropy": 5.845137357711792,
      "epoch": 5.3510522496371555,
      "grad_norm": 1.375,
      "learning_rate": 0.00025320013990757297,
      "loss": 4.9867,
      "mean_token_accuracy": 0.21975106000900269,
      "num_tokens": 122413910.0,
      "step": 58990
    },
    {
      "entropy": 5.791248607635498,
      "epoch": 5.35150580551524,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025316793594168656,
      "loss": 4.8857,
      "mean_token_accuracy": 0.22185314446687698,
      "num_tokens": 122424322.0,
      "step": 58995
    },
    {
      "entropy": 5.799533939361572,
      "epoch": 5.351959361393324,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002531357324272929,
      "loss": 4.8588,
      "mean_token_accuracy": 0.22482138723134995,
      "num_tokens": 122434559.0,
      "step": 59000
    },
    {
      "entropy": 5.8421718120574955,
      "epoch": 5.3524129172714074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002531035293650578,
      "loss": 5.0022,
      "mean_token_accuracy": 0.2178020477294922,
      "num_tokens": 122445641.0,
      "step": 59005
    },
    {
      "entropy": 5.824236679077148,
      "epoch": 5.352866473149492,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002530713267556473,
      "loss": 4.9177,
      "mean_token_accuracy": 0.22611577212810516,
      "num_tokens": 122457524.0,
      "step": 59010
    },
    {
      "entropy": 5.788254642486573,
      "epoch": 5.353320029027576,
      "grad_norm": 1.375,
      "learning_rate": 0.0002530391245997274,
      "loss": 4.8654,
      "mean_token_accuracy": 0.2319887697696686,
      "num_tokens": 122468703.0,
      "step": 59015
    },
    {
      "entropy": 5.747129583358765,
      "epoch": 5.35377358490566,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002530069228979639,
      "loss": 4.8875,
      "mean_token_accuracy": 0.2304629296064377,
      "num_tokens": 122479187.0,
      "step": 59020
    },
    {
      "entropy": 5.769945192337036,
      "epoch": 5.3542271407837445,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025297472165102284,
      "loss": 4.8992,
      "mean_token_accuracy": 0.2147141292691231,
      "num_tokens": 122488893.0,
      "step": 59025
    },
    {
      "entropy": 5.679041719436645,
      "epoch": 5.354680696661829,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025294252085957024,
      "loss": 4.7833,
      "mean_token_accuracy": 0.22769387066364288,
      "num_tokens": 122499596.0,
      "step": 59030
    },
    {
      "entropy": 5.759934425354004,
      "epoch": 5.355134252539913,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002529103205242719,
      "loss": 4.8421,
      "mean_token_accuracy": 0.23154180198907853,
      "num_tokens": 122509219.0,
      "step": 59035
    },
    {
      "entropy": 5.802922153472901,
      "epoch": 5.355587808417997,
      "grad_norm": 1.453125,
      "learning_rate": 0.00025287812064579376,
      "loss": 4.9452,
      "mean_token_accuracy": 0.22006417512893678,
      "num_tokens": 122519118.0,
      "step": 59040
    },
    {
      "entropy": 5.776344680786133,
      "epoch": 5.356041364296082,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002528459212248017,
      "loss": 4.879,
      "mean_token_accuracy": 0.23235837370157242,
      "num_tokens": 122528641.0,
      "step": 59045
    },
    {
      "entropy": 5.721927499771118,
      "epoch": 5.356494920174166,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002528137222619616,
      "loss": 4.8055,
      "mean_token_accuracy": 0.2313826188445091,
      "num_tokens": 122539357.0,
      "step": 59050
    },
    {
      "entropy": 5.868796539306641,
      "epoch": 5.356948476052249,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025278152375793946,
      "loss": 4.9452,
      "mean_token_accuracy": 0.2274983063340187,
      "num_tokens": 122550218.0,
      "step": 59055
    },
    {
      "entropy": 5.834738159179688,
      "epoch": 5.3574020319303335,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025274932571340105,
      "loss": 4.9585,
      "mean_token_accuracy": 0.21775849014520646,
      "num_tokens": 122561196.0,
      "step": 59060
    },
    {
      "entropy": 5.792884588241577,
      "epoch": 5.357855587808418,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025271712812901224,
      "loss": 4.9097,
      "mean_token_accuracy": 0.22416260242462158,
      "num_tokens": 122572009.0,
      "step": 59065
    },
    {
      "entropy": 5.696451807022095,
      "epoch": 5.358309143686502,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025268493100543896,
      "loss": 4.7829,
      "mean_token_accuracy": 0.23100514560937882,
      "num_tokens": 122581677.0,
      "step": 59070
    },
    {
      "entropy": 5.677062225341797,
      "epoch": 5.358762699564586,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025265273434334684,
      "loss": 4.8255,
      "mean_token_accuracy": 0.2332460403442383,
      "num_tokens": 122591503.0,
      "step": 59075
    },
    {
      "entropy": 5.714112377166748,
      "epoch": 5.3592162554426706,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025262053814340197,
      "loss": 4.8181,
      "mean_token_accuracy": 0.22881806045770645,
      "num_tokens": 122602675.0,
      "step": 59080
    },
    {
      "entropy": 5.770518016815186,
      "epoch": 5.359669811320755,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002525883424062701,
      "loss": 4.8327,
      "mean_token_accuracy": 0.22530032843351364,
      "num_tokens": 122613832.0,
      "step": 59085
    },
    {
      "entropy": 5.7734050273895265,
      "epoch": 5.360123367198839,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025255614713261693,
      "loss": 4.9336,
      "mean_token_accuracy": 0.23250281065702438,
      "num_tokens": 122623910.0,
      "step": 59090
    },
    {
      "entropy": 5.811656045913696,
      "epoch": 5.360576923076923,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025252395232310844,
      "loss": 4.8854,
      "mean_token_accuracy": 0.2233816146850586,
      "num_tokens": 122634634.0,
      "step": 59095
    },
    {
      "entropy": 5.777563047409058,
      "epoch": 5.361030478955008,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002524917579784103,
      "loss": 4.8332,
      "mean_token_accuracy": 0.22835507690906526,
      "num_tokens": 122644555.0,
      "step": 59100
    },
    {
      "entropy": 5.719345808029175,
      "epoch": 5.361484034833092,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002524595640991884,
      "loss": 4.7942,
      "mean_token_accuracy": 0.2338922530412674,
      "num_tokens": 122654498.0,
      "step": 59105
    },
    {
      "entropy": 5.673157453536987,
      "epoch": 5.361937590711175,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002524273706861084,
      "loss": 4.7621,
      "mean_token_accuracy": 0.23317780196666718,
      "num_tokens": 122664540.0,
      "step": 59110
    },
    {
      "entropy": 5.744271945953369,
      "epoch": 5.3623911465892595,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002523951777398361,
      "loss": 4.8884,
      "mean_token_accuracy": 0.23324215710163115,
      "num_tokens": 122674372.0,
      "step": 59115
    },
    {
      "entropy": 5.820331764221192,
      "epoch": 5.362844702467344,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025236298526103737,
      "loss": 4.9564,
      "mean_token_accuracy": 0.219841668009758,
      "num_tokens": 122686187.0,
      "step": 59120
    },
    {
      "entropy": 5.724718618392944,
      "epoch": 5.363298258345428,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002523307932503778,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23051024973392487,
      "num_tokens": 122697347.0,
      "step": 59125
    },
    {
      "entropy": 5.83995509147644,
      "epoch": 5.363751814223512,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025229860170852326,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22435093075037002,
      "num_tokens": 122707164.0,
      "step": 59130
    },
    {
      "entropy": 5.7683202743530275,
      "epoch": 5.364205370101597,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002522664106361394,
      "loss": 4.8405,
      "mean_token_accuracy": 0.2247115418314934,
      "num_tokens": 122716275.0,
      "step": 59135
    },
    {
      "entropy": 5.70882773399353,
      "epoch": 5.364658925979681,
      "grad_norm": 1.4375,
      "learning_rate": 0.000252234220033892,
      "loss": 4.7888,
      "mean_token_accuracy": 0.23420851081609725,
      "num_tokens": 122726682.0,
      "step": 59140
    },
    {
      "entropy": 5.798706436157227,
      "epoch": 5.365112481857765,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002522020299024466,
      "loss": 4.9401,
      "mean_token_accuracy": 0.2182966500520706,
      "num_tokens": 122738037.0,
      "step": 59145
    },
    {
      "entropy": 5.684055614471435,
      "epoch": 5.365566037735849,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025216984024246914,
      "loss": 4.7707,
      "mean_token_accuracy": 0.22945379018783568,
      "num_tokens": 122748342.0,
      "step": 59150
    },
    {
      "entropy": 5.690022563934326,
      "epoch": 5.366019593613933,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025213765105462524,
      "loss": 4.732,
      "mean_token_accuracy": 0.2389059156179428,
      "num_tokens": 122759246.0,
      "step": 59155
    },
    {
      "entropy": 5.704335260391235,
      "epoch": 5.366473149492017,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002521054623395804,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23634961247444153,
      "num_tokens": 122769145.0,
      "step": 59160
    },
    {
      "entropy": 5.814648962020874,
      "epoch": 5.366926705370101,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002520732740980005,
      "loss": 5.0042,
      "mean_token_accuracy": 0.21866751462221146,
      "num_tokens": 122779765.0,
      "step": 59165
    },
    {
      "entropy": 5.775717735290527,
      "epoch": 5.367380261248186,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025204108633055116,
      "loss": 4.8576,
      "mean_token_accuracy": 0.2276033490896225,
      "num_tokens": 122790708.0,
      "step": 59170
    },
    {
      "entropy": 5.756658363342285,
      "epoch": 5.36783381712627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025200889903789796,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2325998604297638,
      "num_tokens": 122800616.0,
      "step": 59175
    },
    {
      "entropy": 5.7144246101379395,
      "epoch": 5.368287373004354,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002519767122207067,
      "loss": 4.8177,
      "mean_token_accuracy": 0.23168991655111312,
      "num_tokens": 122810385.0,
      "step": 59180
    },
    {
      "entropy": 5.7184583187103275,
      "epoch": 5.368740928882438,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025194452587964277,
      "loss": 4.8045,
      "mean_token_accuracy": 0.23189862966537475,
      "num_tokens": 122820375.0,
      "step": 59185
    },
    {
      "entropy": 5.820152282714844,
      "epoch": 5.369194484760523,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000251912340015372,
      "loss": 4.8659,
      "mean_token_accuracy": 0.22779636532068254,
      "num_tokens": 122831273.0,
      "step": 59190
    },
    {
      "entropy": 5.733159160614013,
      "epoch": 5.369648040638607,
      "grad_norm": 1.25,
      "learning_rate": 0.00025188015462855986,
      "loss": 4.8191,
      "mean_token_accuracy": 0.2355433464050293,
      "num_tokens": 122842049.0,
      "step": 59195
    },
    {
      "entropy": 5.795488023757935,
      "epoch": 5.370101596516691,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002518479697198721,
      "loss": 4.934,
      "mean_token_accuracy": 0.22555984556674957,
      "num_tokens": 122854049.0,
      "step": 59200
    },
    {
      "entropy": 5.816366243362427,
      "epoch": 5.370555152394775,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002518157852899742,
      "loss": 4.9024,
      "mean_token_accuracy": 0.2303333729505539,
      "num_tokens": 122864528.0,
      "step": 59205
    },
    {
      "entropy": 5.799755239486695,
      "epoch": 5.371008708272859,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00025178360133953166,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2293718382716179,
      "num_tokens": 122873975.0,
      "step": 59210
    },
    {
      "entropy": 5.8421965599060055,
      "epoch": 5.371462264150943,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025175141786921024,
      "loss": 4.9541,
      "mean_token_accuracy": 0.2208133414387703,
      "num_tokens": 122884742.0,
      "step": 59215
    },
    {
      "entropy": 5.73737735748291,
      "epoch": 5.371915820029027,
      "grad_norm": 1.546875,
      "learning_rate": 0.0002517192348796754,
      "loss": 4.8072,
      "mean_token_accuracy": 0.24060911685228348,
      "num_tokens": 122895300.0,
      "step": 59220
    },
    {
      "entropy": 5.7259602546691895,
      "epoch": 5.372369375907112,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002516870523715928,
      "loss": 4.8094,
      "mean_token_accuracy": 0.236881585419178,
      "num_tokens": 122905805.0,
      "step": 59225
    },
    {
      "entropy": 5.700653409957885,
      "epoch": 5.372822931785196,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002516548703456278,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23185192048549652,
      "num_tokens": 122916106.0,
      "step": 59230
    },
    {
      "entropy": 5.745480251312256,
      "epoch": 5.37327648766328,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000251622688802446,
      "loss": 4.8373,
      "mean_token_accuracy": 0.23055424988269807,
      "num_tokens": 122926074.0,
      "step": 59235
    },
    {
      "entropy": 5.779722404479981,
      "epoch": 5.373730043541364,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002515905077427131,
      "loss": 4.8511,
      "mean_token_accuracy": 0.22671821117401122,
      "num_tokens": 122936368.0,
      "step": 59240
    },
    {
      "entropy": 5.753577613830567,
      "epoch": 5.374183599419449,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00025155832716709434,
      "loss": 4.8266,
      "mean_token_accuracy": 0.2357180595397949,
      "num_tokens": 122946849.0,
      "step": 59245
    },
    {
      "entropy": 5.774622249603271,
      "epoch": 5.374637155297533,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025152614707625543,
      "loss": 4.8762,
      "mean_token_accuracy": 0.22692235559225082,
      "num_tokens": 122957093.0,
      "step": 59250
    },
    {
      "entropy": 5.7113570213317875,
      "epoch": 5.375090711175617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002514939674708617,
      "loss": 4.7813,
      "mean_token_accuracy": 0.24064603447914124,
      "num_tokens": 122967399.0,
      "step": 59255
    },
    {
      "entropy": 5.764472436904907,
      "epoch": 5.3755442670537015,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025146178835157876,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2239855244755745,
      "num_tokens": 122977891.0,
      "step": 59260
    },
    {
      "entropy": 5.792226886749267,
      "epoch": 5.375997822931785,
      "grad_norm": 1.390625,
      "learning_rate": 0.000251429609719072,
      "loss": 4.8975,
      "mean_token_accuracy": 0.2261684998869896,
      "num_tokens": 122988446.0,
      "step": 59265
    },
    {
      "entropy": 5.838032484054565,
      "epoch": 5.376451378809869,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000251397431574007,
      "loss": 4.8682,
      "mean_token_accuracy": 0.22752266079187394,
      "num_tokens": 122998021.0,
      "step": 59270
    },
    {
      "entropy": 5.70680079460144,
      "epoch": 5.376904934687953,
      "grad_norm": 1.375,
      "learning_rate": 0.00025136525391704904,
      "loss": 4.8564,
      "mean_token_accuracy": 0.22790154963731765,
      "num_tokens": 123008599.0,
      "step": 59275
    },
    {
      "entropy": 5.716510677337647,
      "epoch": 5.377358490566038,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002513330767488637,
      "loss": 4.8181,
      "mean_token_accuracy": 0.23232841342687607,
      "num_tokens": 123019586.0,
      "step": 59280
    },
    {
      "entropy": 5.714110517501831,
      "epoch": 5.377812046444122,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00025130090007011635,
      "loss": 4.8462,
      "mean_token_accuracy": 0.22632977664470671,
      "num_tokens": 123030222.0,
      "step": 59285
    },
    {
      "entropy": 5.707944536209107,
      "epoch": 5.378265602322206,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002512687238814724,
      "loss": 4.796,
      "mean_token_accuracy": 0.23199861347675324,
      "num_tokens": 123041035.0,
      "step": 59290
    },
    {
      "entropy": 5.759108304977417,
      "epoch": 5.3787191582002905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025123654818359737,
      "loss": 4.8905,
      "mean_token_accuracy": 0.22823053747415542,
      "num_tokens": 123052559.0,
      "step": 59295
    },
    {
      "entropy": 5.806287956237793,
      "epoch": 5.379172714078375,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002512043729771565,
      "loss": 4.9119,
      "mean_token_accuracy": 0.22670883685350418,
      "num_tokens": 123062106.0,
      "step": 59300
    },
    {
      "entropy": 5.759838342666626,
      "epoch": 5.379626269956459,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025117219826281526,
      "loss": 4.7909,
      "mean_token_accuracy": 0.23475249707698823,
      "num_tokens": 123071670.0,
      "step": 59305
    },
    {
      "entropy": 5.776859712600708,
      "epoch": 5.380079825834542,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025114002404123917,
      "loss": 4.7477,
      "mean_token_accuracy": 0.2469218045473099,
      "num_tokens": 123082023.0,
      "step": 59310
    },
    {
      "entropy": 5.741955375671386,
      "epoch": 5.380533381712627,
      "grad_norm": 1.484375,
      "learning_rate": 0.00025110785031309325,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22089990228414536,
      "num_tokens": 123092522.0,
      "step": 59315
    },
    {
      "entropy": 5.686279439926148,
      "epoch": 5.380986937590711,
      "grad_norm": 1.40625,
      "learning_rate": 0.00025107567707904324,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23210676163434982,
      "num_tokens": 123102537.0,
      "step": 59320
    },
    {
      "entropy": 5.768305540084839,
      "epoch": 5.381440493468795,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002510435043397542,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22749537229537964,
      "num_tokens": 123112544.0,
      "step": 59325
    },
    {
      "entropy": 5.816640043258667,
      "epoch": 5.381894049346879,
      "grad_norm": 1.375,
      "learning_rate": 0.0002510113320958917,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22902814596891402,
      "num_tokens": 123123089.0,
      "step": 59330
    },
    {
      "entropy": 5.760800313949585,
      "epoch": 5.382347605224964,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002509791603481209,
      "loss": 4.8534,
      "mean_token_accuracy": 0.2280653402209282,
      "num_tokens": 123133452.0,
      "step": 59335
    },
    {
      "entropy": 5.710411024093628,
      "epoch": 5.382801161103048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025094698909710716,
      "loss": 4.7915,
      "mean_token_accuracy": 0.2317988693714142,
      "num_tokens": 123143739.0,
      "step": 59340
    },
    {
      "entropy": 5.797215700149536,
      "epoch": 5.383254716981132,
      "grad_norm": 1.375,
      "learning_rate": 0.0002509148183435159,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23520738631486893,
      "num_tokens": 123154948.0,
      "step": 59345
    },
    {
      "entropy": 5.801603507995606,
      "epoch": 5.3837082728592165,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025088264808801225,
      "loss": 4.9282,
      "mean_token_accuracy": 0.2183308184146881,
      "num_tokens": 123165375.0,
      "step": 59350
    },
    {
      "entropy": 5.741341638565063,
      "epoch": 5.384161828737301,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002508504783312617,
      "loss": 4.8651,
      "mean_token_accuracy": 0.2297041893005371,
      "num_tokens": 123176129.0,
      "step": 59355
    },
    {
      "entropy": 5.794807052612304,
      "epoch": 5.384615384615385,
      "grad_norm": 1.421875,
      "learning_rate": 0.00025081830907392936,
      "loss": 4.8779,
      "mean_token_accuracy": 0.22951370477676392,
      "num_tokens": 123186499.0,
      "step": 59360
    },
    {
      "entropy": 5.722779989242554,
      "epoch": 5.385068940493468,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025078614031668044,
      "loss": 4.7786,
      "mean_token_accuracy": 0.23417637795209884,
      "num_tokens": 123197486.0,
      "step": 59365
    },
    {
      "entropy": 5.791292905807495,
      "epoch": 5.385522496371553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00025075397206018044,
      "loss": 4.8703,
      "mean_token_accuracy": 0.22691241651773453,
      "num_tokens": 123208107.0,
      "step": 59370
    },
    {
      "entropy": 5.741872692108155,
      "epoch": 5.385976052249637,
      "grad_norm": 1.25,
      "learning_rate": 0.00025072180430509445,
      "loss": 4.875,
      "mean_token_accuracy": 0.2236308768391609,
      "num_tokens": 123218993.0,
      "step": 59375
    },
    {
      "entropy": 5.780838298797607,
      "epoch": 5.386429608127721,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002506896370520877,
      "loss": 4.897,
      "mean_token_accuracy": 0.23006902635097504,
      "num_tokens": 123229728.0,
      "step": 59380
    },
    {
      "entropy": 5.800446224212647,
      "epoch": 5.3868831640058055,
      "grad_norm": 1.375,
      "learning_rate": 0.0002506574703018255,
      "loss": 4.8841,
      "mean_token_accuracy": 0.2361946776509285,
      "num_tokens": 123240009.0,
      "step": 59385
    },
    {
      "entropy": 5.7767445087432865,
      "epoch": 5.38733671988389,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025062530405497297,
      "loss": 4.8641,
      "mean_token_accuracy": 0.22707943618297577,
      "num_tokens": 123249827.0,
      "step": 59390
    },
    {
      "entropy": 5.679130506515503,
      "epoch": 5.387790275761974,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002505931383121954,
      "loss": 4.7862,
      "mean_token_accuracy": 0.24036111682653427,
      "num_tokens": 123260032.0,
      "step": 59395
    },
    {
      "entropy": 5.656583261489868,
      "epoch": 5.388243831640058,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00025056097307415786,
      "loss": 4.7735,
      "mean_token_accuracy": 0.23404638171195985,
      "num_tokens": 123269818.0,
      "step": 59400
    },
    {
      "entropy": 5.771254634857177,
      "epoch": 5.3886973875181425,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00025052880834152574,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22527411133050917,
      "num_tokens": 123280478.0,
      "step": 59405
    },
    {
      "entropy": 5.774083852767944,
      "epoch": 5.389150943396227,
      "grad_norm": 1.4375,
      "learning_rate": 0.000250496644114964,
      "loss": 4.901,
      "mean_token_accuracy": 0.21982172429561614,
      "num_tokens": 123291621.0,
      "step": 59410
    },
    {
      "entropy": 5.701043176651001,
      "epoch": 5.389604499274311,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002504644803951381,
      "loss": 4.7549,
      "mean_token_accuracy": 0.2397686243057251,
      "num_tokens": 123300906.0,
      "step": 59415
    },
    {
      "entropy": 5.755737018585205,
      "epoch": 5.3900580551523944,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002504323171827128,
      "loss": 4.8699,
      "mean_token_accuracy": 0.23358875662088394,
      "num_tokens": 123310961.0,
      "step": 59420
    },
    {
      "entropy": 5.724580192565918,
      "epoch": 5.390511611030479,
      "grad_norm": 1.375,
      "learning_rate": 0.0002504001544783535,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23735453486442565,
      "num_tokens": 123321570.0,
      "step": 59425
    },
    {
      "entropy": 5.768183040618896,
      "epoch": 5.390965166908563,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025036799228272526,
      "loss": 4.8452,
      "mean_token_accuracy": 0.22434639930725098,
      "num_tokens": 123331736.0,
      "step": 59430
    },
    {
      "entropy": 5.775891637802124,
      "epoch": 5.391418722786647,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002503358305964933,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2355358839035034,
      "num_tokens": 123341893.0,
      "step": 59435
    },
    {
      "entropy": 5.7740561962127686,
      "epoch": 5.3918722786647315,
      "grad_norm": 1.4375,
      "learning_rate": 0.00025030366942032257,
      "loss": 4.9053,
      "mean_token_accuracy": 0.22700002938508987,
      "num_tokens": 123351844.0,
      "step": 59440
    },
    {
      "entropy": 5.819478988647461,
      "epoch": 5.392325834542816,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002502715087548782,
      "loss": 4.8695,
      "mean_token_accuracy": 0.2254759892821312,
      "num_tokens": 123362376.0,
      "step": 59445
    },
    {
      "entropy": 5.794401979446411,
      "epoch": 5.3927793904209,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002502393486008254,
      "loss": 4.8147,
      "mean_token_accuracy": 0.2362274706363678,
      "num_tokens": 123372781.0,
      "step": 59450
    },
    {
      "entropy": 5.770057201385498,
      "epoch": 5.393232946298984,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002502071889588292,
      "loss": 4.9733,
      "mean_token_accuracy": 0.22696294635534286,
      "num_tokens": 123383047.0,
      "step": 59455
    },
    {
      "entropy": 5.674321508407592,
      "epoch": 5.393686502177069,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025017502982955457,
      "loss": 4.8098,
      "mean_token_accuracy": 0.23600797057151796,
      "num_tokens": 123393265.0,
      "step": 59460
    },
    {
      "entropy": 5.690494155883789,
      "epoch": 5.394140058055152,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002501428712136667,
      "loss": 4.77,
      "mean_token_accuracy": 0.23316708207130432,
      "num_tokens": 123403781.0,
      "step": 59465
    },
    {
      "entropy": 5.807736015319824,
      "epoch": 5.394593613933236,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002501107131118306,
      "loss": 4.827,
      "mean_token_accuracy": 0.231922709941864,
      "num_tokens": 123413260.0,
      "step": 59470
    },
    {
      "entropy": 5.780968952178955,
      "epoch": 5.3950471698113205,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025007855552471123,
      "loss": 4.8531,
      "mean_token_accuracy": 0.22516042739152908,
      "num_tokens": 123423578.0,
      "step": 59475
    },
    {
      "entropy": 5.6992710590362545,
      "epoch": 5.395500725689405,
      "grad_norm": 1.46875,
      "learning_rate": 0.00025004639845297377,
      "loss": 4.7696,
      "mean_token_accuracy": 0.23468492031097413,
      "num_tokens": 123434425.0,
      "step": 59480
    },
    {
      "entropy": 5.686392116546631,
      "epoch": 5.395954281567489,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00025001424189728314,
      "loss": 4.7378,
      "mean_token_accuracy": 0.24167473316192628,
      "num_tokens": 123444096.0,
      "step": 59485
    },
    {
      "entropy": 5.802724695205688,
      "epoch": 5.396407837445573,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024998208585830435,
      "loss": 4.9405,
      "mean_token_accuracy": 0.22735000401735306,
      "num_tokens": 123454655.0,
      "step": 59490
    },
    {
      "entropy": 5.729579257965088,
      "epoch": 5.3968613933236576,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024994993033670233,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23632412254810334,
      "num_tokens": 123465742.0,
      "step": 59495
    },
    {
      "entropy": 5.827843284606933,
      "epoch": 5.397314949201742,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002499177753331422,
      "loss": 4.9109,
      "mean_token_accuracy": 0.23050676435232162,
      "num_tokens": 123476110.0,
      "step": 59500
    },
    {
      "entropy": 5.730667924880981,
      "epoch": 5.397768505079826,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002498856208482889,
      "loss": 4.849,
      "mean_token_accuracy": 0.2238295704126358,
      "num_tokens": 123486572.0,
      "step": 59505
    },
    {
      "entropy": 5.711811113357544,
      "epoch": 5.39822206095791,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002498534668828072,
      "loss": 4.856,
      "mean_token_accuracy": 0.22224498689174652,
      "num_tokens": 123498096.0,
      "step": 59510
    },
    {
      "entropy": 5.7396753311157225,
      "epoch": 5.398675616835995,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002498213134373624,
      "loss": 4.8035,
      "mean_token_accuracy": 0.23594440668821334,
      "num_tokens": 123508290.0,
      "step": 59515
    },
    {
      "entropy": 5.816624593734741,
      "epoch": 5.399129172714078,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024978916051261914,
      "loss": 4.8473,
      "mean_token_accuracy": 0.2304805874824524,
      "num_tokens": 123518040.0,
      "step": 59520
    },
    {
      "entropy": 5.822852420806885,
      "epoch": 5.399582728592162,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002497570081092425,
      "loss": 4.9419,
      "mean_token_accuracy": 0.21893096417188646,
      "num_tokens": 123528799.0,
      "step": 59525
    },
    {
      "entropy": 5.746806287765503,
      "epoch": 5.4000362844702465,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002497248562278974,
      "loss": 4.8375,
      "mean_token_accuracy": 0.22960828691720964,
      "num_tokens": 123539114.0,
      "step": 59530
    },
    {
      "entropy": 5.767926597595215,
      "epoch": 5.400489840348331,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024969270486924867,
      "loss": 4.8142,
      "mean_token_accuracy": 0.2313730776309967,
      "num_tokens": 123549914.0,
      "step": 59535
    },
    {
      "entropy": 5.85749306678772,
      "epoch": 5.400943396226415,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024966055403396125,
      "loss": 4.9593,
      "mean_token_accuracy": 0.21858482509851457,
      "num_tokens": 123560067.0,
      "step": 59540
    },
    {
      "entropy": 5.725943899154663,
      "epoch": 5.401396952104499,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002496284037227,
      "loss": 4.7881,
      "mean_token_accuracy": 0.23375679403543473,
      "num_tokens": 123569756.0,
      "step": 59545
    },
    {
      "entropy": 5.779930305480957,
      "epoch": 5.401850507982584,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024959625393612987,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22468969225883484,
      "num_tokens": 123580482.0,
      "step": 59550
    },
    {
      "entropy": 5.7547149658203125,
      "epoch": 5.402304063860668,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002495641046749157,
      "loss": 4.8742,
      "mean_token_accuracy": 0.23132430762052536,
      "num_tokens": 123592212.0,
      "step": 59555
    },
    {
      "entropy": 5.71754298210144,
      "epoch": 5.402757619738752,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002495319559397223,
      "loss": 4.8361,
      "mean_token_accuracy": 0.2284839242696762,
      "num_tokens": 123602795.0,
      "step": 59560
    },
    {
      "entropy": 5.713863897323608,
      "epoch": 5.403211175616836,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002494998077312146,
      "loss": 4.8278,
      "mean_token_accuracy": 0.23258376121520996,
      "num_tokens": 123613856.0,
      "step": 59565
    },
    {
      "entropy": 5.71205325126648,
      "epoch": 5.40366473149492,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024946766005005727,
      "loss": 4.8427,
      "mean_token_accuracy": 0.22928122729063033,
      "num_tokens": 123623536.0,
      "step": 59570
    },
    {
      "entropy": 5.703930282592774,
      "epoch": 5.404118287373004,
      "grad_norm": 1.5,
      "learning_rate": 0.0002494355128969153,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23264421671628951,
      "num_tokens": 123634011.0,
      "step": 59575
    },
    {
      "entropy": 5.743545770645142,
      "epoch": 5.404571843251088,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00024940336627245343,
      "loss": 4.827,
      "mean_token_accuracy": 0.23455135077238082,
      "num_tokens": 123644064.0,
      "step": 59580
    },
    {
      "entropy": 5.694611024856568,
      "epoch": 5.405025399129173,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002493712201773365,
      "loss": 4.7723,
      "mean_token_accuracy": 0.23524772971868516,
      "num_tokens": 123654231.0,
      "step": 59585
    },
    {
      "entropy": 5.765783500671387,
      "epoch": 5.405478955007257,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002493390746122292,
      "loss": 4.9029,
      "mean_token_accuracy": 0.22554488629102706,
      "num_tokens": 123664587.0,
      "step": 59590
    },
    {
      "entropy": 5.766827964782715,
      "epoch": 5.405932510885341,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002493069295777964,
      "loss": 4.8633,
      "mean_token_accuracy": 0.23397847265005112,
      "num_tokens": 123674474.0,
      "step": 59595
    },
    {
      "entropy": 5.745809030532837,
      "epoch": 5.406386066763425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024927478507470285,
      "loss": 4.7797,
      "mean_token_accuracy": 0.22939232587814332,
      "num_tokens": 123685108.0,
      "step": 59600
    },
    {
      "entropy": 5.73759880065918,
      "epoch": 5.40683962264151,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024924264110361334,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23042570799589157,
      "num_tokens": 123695723.0,
      "step": 59605
    },
    {
      "entropy": 5.801217269897461,
      "epoch": 5.407293178519594,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024921049766519257,
      "loss": 4.9646,
      "mean_token_accuracy": 0.22677515298128129,
      "num_tokens": 123706448.0,
      "step": 59610
    },
    {
      "entropy": 5.8496284008026125,
      "epoch": 5.407746734397678,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002491783547601052,
      "loss": 4.9601,
      "mean_token_accuracy": 0.2264284983277321,
      "num_tokens": 123717373.0,
      "step": 59615
    },
    {
      "entropy": 5.811253786087036,
      "epoch": 5.4082002902757615,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002491462123890161,
      "loss": 4.872,
      "mean_token_accuracy": 0.22875997722148894,
      "num_tokens": 123727174.0,
      "step": 59620
    },
    {
      "entropy": 5.763002967834472,
      "epoch": 5.408653846153846,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002491140705525899,
      "loss": 4.8748,
      "mean_token_accuracy": 0.2332538515329361,
      "num_tokens": 123736962.0,
      "step": 59625
    },
    {
      "entropy": 5.772865676879883,
      "epoch": 5.40910740203193,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024908192925149125,
      "loss": 4.8504,
      "mean_token_accuracy": 0.2308148831129074,
      "num_tokens": 123748808.0,
      "step": 59630
    },
    {
      "entropy": 5.70525426864624,
      "epoch": 5.409560957910014,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000249049788486385,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23627904057502747,
      "num_tokens": 123759211.0,
      "step": 59635
    },
    {
      "entropy": 5.793389511108399,
      "epoch": 5.410014513788099,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002490176482579357,
      "loss": 4.9751,
      "mean_token_accuracy": 0.2212551549077034,
      "num_tokens": 123769568.0,
      "step": 59640
    },
    {
      "entropy": 5.85509238243103,
      "epoch": 5.410468069666183,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024898550856680815,
      "loss": 4.9009,
      "mean_token_accuracy": 0.22252778261899947,
      "num_tokens": 123779503.0,
      "step": 59645
    },
    {
      "entropy": 5.757020759582519,
      "epoch": 5.410921625544267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024895336941366676,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22844638228416442,
      "num_tokens": 123790194.0,
      "step": 59650
    },
    {
      "entropy": 5.89770245552063,
      "epoch": 5.411375181422351,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024892123079917647,
      "loss": 5.0872,
      "mean_token_accuracy": 0.21237576007843018,
      "num_tokens": 123800055.0,
      "step": 59655
    },
    {
      "entropy": 5.842959976196289,
      "epoch": 5.411828737300436,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002488890927240017,
      "loss": 4.824,
      "mean_token_accuracy": 0.23725035190582275,
      "num_tokens": 123809358.0,
      "step": 59660
    },
    {
      "entropy": 5.745514822006226,
      "epoch": 5.41228229317852,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024885695518880706,
      "loss": 4.8422,
      "mean_token_accuracy": 0.22791779339313506,
      "num_tokens": 123820294.0,
      "step": 59665
    },
    {
      "entropy": 5.741038036346436,
      "epoch": 5.412735849056604,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024882481819425733,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22362202703952788,
      "num_tokens": 123830580.0,
      "step": 59670
    },
    {
      "entropy": 5.755595684051514,
      "epoch": 5.413189404934688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024879268174101706,
      "loss": 4.8624,
      "mean_token_accuracy": 0.22443924695253373,
      "num_tokens": 123842089.0,
      "step": 59675
    },
    {
      "entropy": 5.71263747215271,
      "epoch": 5.413642960812772,
      "grad_norm": 1.5,
      "learning_rate": 0.00024876054582975075,
      "loss": 4.7743,
      "mean_token_accuracy": 0.24245260655879974,
      "num_tokens": 123851852.0,
      "step": 59680
    },
    {
      "entropy": 5.7802557945251465,
      "epoch": 5.414096516690856,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002487284104611231,
      "loss": 4.8384,
      "mean_token_accuracy": 0.23043641448020935,
      "num_tokens": 123863193.0,
      "step": 59685
    },
    {
      "entropy": 5.806909799575806,
      "epoch": 5.41455007256894,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002486962756357986,
      "loss": 4.9845,
      "mean_token_accuracy": 0.2172803223133087,
      "num_tokens": 123874771.0,
      "step": 59690
    },
    {
      "entropy": 5.798520088195801,
      "epoch": 5.415003628447025,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002486641413544418,
      "loss": 4.8566,
      "mean_token_accuracy": 0.23639950305223464,
      "num_tokens": 123885613.0,
      "step": 59695
    },
    {
      "entropy": 5.729588794708252,
      "epoch": 5.415457184325109,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002486320076177173,
      "loss": 4.7879,
      "mean_token_accuracy": 0.23562992215156556,
      "num_tokens": 123896181.0,
      "step": 59700
    },
    {
      "entropy": 5.73174409866333,
      "epoch": 5.415910740203193,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024859987442628963,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23331409692764282,
      "num_tokens": 123906245.0,
      "step": 59705
    },
    {
      "entropy": 5.7394078254699705,
      "epoch": 5.4163642960812775,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002485677417808232,
      "loss": 4.8305,
      "mean_token_accuracy": 0.2372388333082199,
      "num_tokens": 123916596.0,
      "step": 59710
    },
    {
      "entropy": 5.7146471500396725,
      "epoch": 5.416817851959362,
      "grad_norm": 1.375,
      "learning_rate": 0.0002485356096819826,
      "loss": 4.8073,
      "mean_token_accuracy": 0.24059313237667085,
      "num_tokens": 123927125.0,
      "step": 59715
    },
    {
      "entropy": 5.7896981716156,
      "epoch": 5.417271407837446,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002485034781304324,
      "loss": 4.886,
      "mean_token_accuracy": 0.22117627263069153,
      "num_tokens": 123937457.0,
      "step": 59720
    },
    {
      "entropy": 5.748226308822632,
      "epoch": 5.417724963715529,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000248471347126837,
      "loss": 4.9065,
      "mean_token_accuracy": 0.23366164416074753,
      "num_tokens": 123947178.0,
      "step": 59725
    },
    {
      "entropy": 5.7607550621032715,
      "epoch": 5.418178519593614,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024843921667186094,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2319886177778244,
      "num_tokens": 123957103.0,
      "step": 59730
    },
    {
      "entropy": 5.722415113449097,
      "epoch": 5.418632075471698,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002484070867661686,
      "loss": 4.8324,
      "mean_token_accuracy": 0.22724045515060426,
      "num_tokens": 123967311.0,
      "step": 59735
    },
    {
      "entropy": 5.809472751617432,
      "epoch": 5.419085631349782,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002483749574104245,
      "loss": 4.8802,
      "mean_token_accuracy": 0.2314649000763893,
      "num_tokens": 123977964.0,
      "step": 59740
    },
    {
      "entropy": 5.757829570770264,
      "epoch": 5.419539187227866,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002483428286052931,
      "loss": 4.8375,
      "mean_token_accuracy": 0.23085639178752898,
      "num_tokens": 123988345.0,
      "step": 59745
    },
    {
      "entropy": 5.741592645645142,
      "epoch": 5.419992743105951,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002483107003514387,
      "loss": 4.8643,
      "mean_token_accuracy": 0.23159322738647461,
      "num_tokens": 123998056.0,
      "step": 59750
    },
    {
      "entropy": 5.744367504119873,
      "epoch": 5.420446298984035,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024827857264952583,
      "loss": 4.8998,
      "mean_token_accuracy": 0.22296811938285827,
      "num_tokens": 124009104.0,
      "step": 59755
    },
    {
      "entropy": 5.734332656860351,
      "epoch": 5.420899854862119,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024824644550021894,
      "loss": 4.8357,
      "mean_token_accuracy": 0.22571689039468765,
      "num_tokens": 124018975.0,
      "step": 59760
    },
    {
      "entropy": 5.721537637710571,
      "epoch": 5.4213534107402035,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002482143189041824,
      "loss": 4.7943,
      "mean_token_accuracy": 0.23638955056667327,
      "num_tokens": 124028770.0,
      "step": 59765
    },
    {
      "entropy": 5.7076679229736325,
      "epoch": 5.421806966618288,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024818219286208053,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23723320066928863,
      "num_tokens": 124038625.0,
      "step": 59770
    },
    {
      "entropy": 5.757748651504516,
      "epoch": 5.422260522496371,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002481500673745778,
      "loss": 4.843,
      "mean_token_accuracy": 0.23751198798418044,
      "num_tokens": 124049063.0,
      "step": 59775
    },
    {
      "entropy": 5.804155683517456,
      "epoch": 5.422714078374455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002481179424423385,
      "loss": 4.9418,
      "mean_token_accuracy": 0.23321450054645537,
      "num_tokens": 124059564.0,
      "step": 59780
    },
    {
      "entropy": 5.763774156570435,
      "epoch": 5.42316763425254,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002480858180660269,
      "loss": 4.8459,
      "mean_token_accuracy": 0.23185279667377473,
      "num_tokens": 124069397.0,
      "step": 59785
    },
    {
      "entropy": 5.797355604171753,
      "epoch": 5.423621190130624,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002480536942463075,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22721763402223588,
      "num_tokens": 124080173.0,
      "step": 59790
    },
    {
      "entropy": 5.808654403686523,
      "epoch": 5.424074746008708,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024802157098384464,
      "loss": 4.9012,
      "mean_token_accuracy": 0.2275164544582367,
      "num_tokens": 124091445.0,
      "step": 59795
    },
    {
      "entropy": 5.766126441955566,
      "epoch": 5.4245283018867925,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024798944827930246,
      "loss": 4.8701,
      "mean_token_accuracy": 0.22648181021213531,
      "num_tokens": 124100849.0,
      "step": 59800
    },
    {
      "entropy": 5.76282377243042,
      "epoch": 5.424981857764877,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002479573261333454,
      "loss": 4.8182,
      "mean_token_accuracy": 0.2331443816423416,
      "num_tokens": 124110536.0,
      "step": 59805
    },
    {
      "entropy": 5.720163488388062,
      "epoch": 5.425435413642961,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002479252045466378,
      "loss": 4.8631,
      "mean_token_accuracy": 0.23894426226615906,
      "num_tokens": 124121329.0,
      "step": 59810
    },
    {
      "entropy": 5.683354711532592,
      "epoch": 5.425888969521045,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002478930835198439,
      "loss": 4.8646,
      "mean_token_accuracy": 0.22875872552394866,
      "num_tokens": 124131611.0,
      "step": 59815
    },
    {
      "entropy": 5.721369886398316,
      "epoch": 5.4263425253991295,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024786096305362776,
      "loss": 4.7856,
      "mean_token_accuracy": 0.24700887799263,
      "num_tokens": 124141757.0,
      "step": 59820
    },
    {
      "entropy": 5.766316652297974,
      "epoch": 5.426796081277214,
      "grad_norm": 1.375,
      "learning_rate": 0.00024782884314865394,
      "loss": 4.884,
      "mean_token_accuracy": 0.22598913311958313,
      "num_tokens": 124152410.0,
      "step": 59825
    },
    {
      "entropy": 5.756416273117066,
      "epoch": 5.427249637155297,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024779672380558654,
      "loss": 4.8295,
      "mean_token_accuracy": 0.24305659979581834,
      "num_tokens": 124163206.0,
      "step": 59830
    },
    {
      "entropy": 5.7685798645019535,
      "epoch": 5.4277031930333814,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002477646050250898,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2328029453754425,
      "num_tokens": 124173208.0,
      "step": 59835
    },
    {
      "entropy": 5.691294813156128,
      "epoch": 5.428156748911466,
      "grad_norm": 1.34375,
      "learning_rate": 0.000247732486807828,
      "loss": 4.7733,
      "mean_token_accuracy": 0.23365473598241807,
      "num_tokens": 124183759.0,
      "step": 59840
    },
    {
      "entropy": 5.85578408241272,
      "epoch": 5.42861030478955,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002477003691544654,
      "loss": 4.9996,
      "mean_token_accuracy": 0.22313280701637267,
      "num_tokens": 124194517.0,
      "step": 59845
    },
    {
      "entropy": 5.717498683929444,
      "epoch": 5.429063860667634,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002476682520656661,
      "loss": 4.7963,
      "mean_token_accuracy": 0.22666844725608826,
      "num_tokens": 124204752.0,
      "step": 59850
    },
    {
      "entropy": 5.7456443309783936,
      "epoch": 5.4295174165457185,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024763613554209423,
      "loss": 4.8915,
      "mean_token_accuracy": 0.2274913877248764,
      "num_tokens": 124214231.0,
      "step": 59855
    },
    {
      "entropy": 5.775132894515991,
      "epoch": 5.429970972423803,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002476040195844141,
      "loss": 4.8718,
      "mean_token_accuracy": 0.2353430911898613,
      "num_tokens": 124223984.0,
      "step": 59860
    },
    {
      "entropy": 5.7449146747589115,
      "epoch": 5.430424528301887,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024757190419328977,
      "loss": 4.8152,
      "mean_token_accuracy": 0.22966797798871993,
      "num_tokens": 124234170.0,
      "step": 59865
    },
    {
      "entropy": 5.731916856765747,
      "epoch": 5.430878084179971,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002475397893693855,
      "loss": 4.8446,
      "mean_token_accuracy": 0.23171538710594178,
      "num_tokens": 124244756.0,
      "step": 59870
    },
    {
      "entropy": 5.815624094009399,
      "epoch": 5.431331640058055,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024750767511336544,
      "loss": 4.9179,
      "mean_token_accuracy": 0.22891057431697845,
      "num_tokens": 124256148.0,
      "step": 59875
    },
    {
      "entropy": 5.722897958755493,
      "epoch": 5.431785195936139,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002474755614258935,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23117295652627945,
      "num_tokens": 124266774.0,
      "step": 59880
    },
    {
      "entropy": 5.763158988952637,
      "epoch": 5.432238751814223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002474434483076341,
      "loss": 4.9037,
      "mean_token_accuracy": 0.22752007693052292,
      "num_tokens": 124276747.0,
      "step": 59885
    },
    {
      "entropy": 5.76399598121643,
      "epoch": 5.4326923076923075,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002474113357592512,
      "loss": 4.8908,
      "mean_token_accuracy": 0.22528979331254959,
      "num_tokens": 124286610.0,
      "step": 59890
    },
    {
      "entropy": 5.853134727478027,
      "epoch": 5.433145863570392,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002473792237814088,
      "loss": 4.92,
      "mean_token_accuracy": 0.22414422035217285,
      "num_tokens": 124296438.0,
      "step": 59895
    },
    {
      "entropy": 5.83100323677063,
      "epoch": 5.433599419448476,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002473471123747712,
      "loss": 4.9471,
      "mean_token_accuracy": 0.22057733088731765,
      "num_tokens": 124305509.0,
      "step": 59900
    },
    {
      "entropy": 5.787449741363526,
      "epoch": 5.43405297532656,
      "grad_norm": 1.375,
      "learning_rate": 0.0002473150015400023,
      "loss": 4.8474,
      "mean_token_accuracy": 0.2314277097582817,
      "num_tokens": 124315911.0,
      "step": 59905
    },
    {
      "entropy": 5.742554426193237,
      "epoch": 5.4345065312046446,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002472828912777663,
      "loss": 4.8055,
      "mean_token_accuracy": 0.23973665982484818,
      "num_tokens": 124326670.0,
      "step": 59910
    },
    {
      "entropy": 5.718345260620117,
      "epoch": 5.434960087082729,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002472507815887271,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23159376978874208,
      "num_tokens": 124336851.0,
      "step": 59915
    },
    {
      "entropy": 5.764206838607788,
      "epoch": 5.435413642960813,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024721867247354886,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23153139799833297,
      "num_tokens": 124347056.0,
      "step": 59920
    },
    {
      "entropy": 5.768034029006958,
      "epoch": 5.435867198838897,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024718656393289554,
      "loss": 4.842,
      "mean_token_accuracy": 0.2299325555562973,
      "num_tokens": 124357684.0,
      "step": 59925
    },
    {
      "entropy": 5.687374305725098,
      "epoch": 5.436320754716981,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002471544559674311,
      "loss": 4.7218,
      "mean_token_accuracy": 0.24933360368013383,
      "num_tokens": 124368100.0,
      "step": 59930
    },
    {
      "entropy": 5.772447919845581,
      "epoch": 5.436774310595065,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002471223485778196,
      "loss": 4.9672,
      "mean_token_accuracy": 0.21772756427526474,
      "num_tokens": 124378434.0,
      "step": 59935
    },
    {
      "entropy": 5.780262088775634,
      "epoch": 5.437227866473149,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024709024176472515,
      "loss": 4.8844,
      "mean_token_accuracy": 0.22894364297389985,
      "num_tokens": 124389730.0,
      "step": 59940
    },
    {
      "entropy": 5.844293403625488,
      "epoch": 5.4376814223512335,
      "grad_norm": 1.25,
      "learning_rate": 0.00024705813552881157,
      "loss": 4.9853,
      "mean_token_accuracy": 0.21512582898139954,
      "num_tokens": 124401566.0,
      "step": 59945
    },
    {
      "entropy": 5.761141014099121,
      "epoch": 5.438134978229318,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024702602987074283,
      "loss": 4.8044,
      "mean_token_accuracy": 0.23960418105125428,
      "num_tokens": 124411365.0,
      "step": 59950
    },
    {
      "entropy": 5.714420223236084,
      "epoch": 5.438588534107402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002469939247911829,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23056236356496812,
      "num_tokens": 124422193.0,
      "step": 59955
    },
    {
      "entropy": 5.7533043384552,
      "epoch": 5.439042089985486,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002469618202907958,
      "loss": 4.9485,
      "mean_token_accuracy": 0.2233745276927948,
      "num_tokens": 124432564.0,
      "step": 59960
    },
    {
      "entropy": 5.80151104927063,
      "epoch": 5.439495645863571,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002469297163702454,
      "loss": 4.9211,
      "mean_token_accuracy": 0.22846785932779312,
      "num_tokens": 124442975.0,
      "step": 59965
    },
    {
      "entropy": 5.74159803390503,
      "epoch": 5.439949201741655,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002468976130301956,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23003299981355668,
      "num_tokens": 124453084.0,
      "step": 59970
    },
    {
      "entropy": 5.836034870147705,
      "epoch": 5.440402757619739,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024686551027131037,
      "loss": 4.9637,
      "mean_token_accuracy": 0.21668197214603424,
      "num_tokens": 124462921.0,
      "step": 59975
    },
    {
      "entropy": 5.793696546554566,
      "epoch": 5.440856313497823,
      "grad_norm": 1.375,
      "learning_rate": 0.0002468334080942535,
      "loss": 4.8254,
      "mean_token_accuracy": 0.22643526941537856,
      "num_tokens": 124473558.0,
      "step": 59980
    },
    {
      "entropy": 5.766199588775635,
      "epoch": 5.441309869375907,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002468013064996889,
      "loss": 4.8834,
      "mean_token_accuracy": 0.23409228026866913,
      "num_tokens": 124483543.0,
      "step": 59985
    },
    {
      "entropy": 5.858816480636596,
      "epoch": 5.441763425253991,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024676920548828056,
      "loss": 4.992,
      "mean_token_accuracy": 0.22431669980287552,
      "num_tokens": 124494278.0,
      "step": 59990
    },
    {
      "entropy": 5.8671026706695555,
      "epoch": 5.442216981132075,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002467371050606922,
      "loss": 4.9717,
      "mean_token_accuracy": 0.2174194261431694,
      "num_tokens": 124507007.0,
      "step": 59995
    },
    {
      "entropy": 5.7715260028839115,
      "epoch": 5.44267053701016,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024670500521758765,
      "loss": 4.799,
      "mean_token_accuracy": 0.2384910836815834,
      "num_tokens": 124517288.0,
      "step": 60000
    },
    {
      "epoch": 5.44267053701016,
      "eval_entropy": 5.462682273485003,
      "eval_loss": 4.949685573577881,
      "eval_mean_token_accuracy": 0.23308427074409988,
      "eval_num_tokens": 124517288.0,
      "eval_runtime": 20.3889,
      "eval_samples_per_second": 1734.28,
      "eval_steps_per_second": 216.785,
      "step": 60000
    },
    {
      "entropy": 5.666421842575073,
      "epoch": 5.443124092888244,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024667290595963085,
      "loss": 4.7565,
      "mean_token_accuracy": 0.23507855236530303,
      "num_tokens": 124527079.0,
      "step": 60005
    },
    {
      "entropy": 5.715499305725098,
      "epoch": 5.443577648766328,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002466408072874856,
      "loss": 4.82,
      "mean_token_accuracy": 0.23751953542232512,
      "num_tokens": 124536997.0,
      "step": 60010
    },
    {
      "entropy": 5.776473617553711,
      "epoch": 5.444031204644412,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024660870920181565,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23501832336187362,
      "num_tokens": 124547368.0,
      "step": 60015
    },
    {
      "entropy": 5.837204790115356,
      "epoch": 5.444484760522497,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002465766117032849,
      "loss": 4.898,
      "mean_token_accuracy": 0.22624534219503403,
      "num_tokens": 124557888.0,
      "step": 60020
    },
    {
      "entropy": 5.765932989120484,
      "epoch": 5.444938316400581,
      "grad_norm": 1.34375,
      "learning_rate": 0.000246544514792557,
      "loss": 4.806,
      "mean_token_accuracy": 0.2358730062842369,
      "num_tokens": 124567495.0,
      "step": 60025
    },
    {
      "entropy": 5.766256380081177,
      "epoch": 5.445391872278664,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002465124184702958,
      "loss": 4.9285,
      "mean_token_accuracy": 0.2213456928730011,
      "num_tokens": 124577893.0,
      "step": 60030
    },
    {
      "entropy": 5.756577253341675,
      "epoch": 5.4458454281567485,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024648032273716505,
      "loss": 4.8677,
      "mean_token_accuracy": 0.231549733877182,
      "num_tokens": 124588453.0,
      "step": 60035
    },
    {
      "entropy": 5.716476106643677,
      "epoch": 5.446298984034833,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024644822759382846,
      "loss": 4.7904,
      "mean_token_accuracy": 0.23781109899282454,
      "num_tokens": 124598726.0,
      "step": 60040
    },
    {
      "entropy": 5.823815584182739,
      "epoch": 5.446752539912917,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002464161330409498,
      "loss": 4.9105,
      "mean_token_accuracy": 0.23048828542232513,
      "num_tokens": 124610218.0,
      "step": 60045
    },
    {
      "entropy": 5.805804347991943,
      "epoch": 5.447206095791001,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024638403907919284,
      "loss": 4.9327,
      "mean_token_accuracy": 0.22087409049272538,
      "num_tokens": 124619920.0,
      "step": 60050
    },
    {
      "entropy": 5.855745792388916,
      "epoch": 5.447659651669086,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002463519457092212,
      "loss": 5.0009,
      "mean_token_accuracy": 0.22144560813903807,
      "num_tokens": 124630815.0,
      "step": 60055
    },
    {
      "entropy": 5.8152707576751705,
      "epoch": 5.44811320754717,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024631985293169863,
      "loss": 4.9006,
      "mean_token_accuracy": 0.2220796301960945,
      "num_tokens": 124641783.0,
      "step": 60060
    },
    {
      "entropy": 5.779336071014404,
      "epoch": 5.448566763425254,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002462877607472888,
      "loss": 4.8652,
      "mean_token_accuracy": 0.22484051436185837,
      "num_tokens": 124650943.0,
      "step": 60065
    },
    {
      "entropy": 5.7915795803070065,
      "epoch": 5.449020319303338,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024625566915665545,
      "loss": 4.8815,
      "mean_token_accuracy": 0.22837481945753096,
      "num_tokens": 124662065.0,
      "step": 60070
    },
    {
      "entropy": 5.788620090484619,
      "epoch": 5.449473875181423,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024622357816046216,
      "loss": 4.8354,
      "mean_token_accuracy": 0.22926126569509506,
      "num_tokens": 124672071.0,
      "step": 60075
    },
    {
      "entropy": 5.816858005523682,
      "epoch": 5.449927431059507,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024619148775937254,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22484310567378998,
      "num_tokens": 124683085.0,
      "step": 60080
    },
    {
      "entropy": 5.777716398239136,
      "epoch": 5.45038098693759,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024615939795405033,
      "loss": 4.9523,
      "mean_token_accuracy": 0.2213885635137558,
      "num_tokens": 124693026.0,
      "step": 60085
    },
    {
      "entropy": 5.789899635314941,
      "epoch": 5.450834542815675,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002461273087451592,
      "loss": 4.9211,
      "mean_token_accuracy": 0.2183824747800827,
      "num_tokens": 124704307.0,
      "step": 60090
    },
    {
      "entropy": 5.7095136642456055,
      "epoch": 5.451288098693759,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024609522013336265,
      "loss": 4.8132,
      "mean_token_accuracy": 0.22841770052909852,
      "num_tokens": 124715659.0,
      "step": 60095
    },
    {
      "entropy": 5.7296405792236325,
      "epoch": 5.451741654571843,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024606313211932425,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23446231633424758,
      "num_tokens": 124727064.0,
      "step": 60100
    },
    {
      "entropy": 5.736378955841064,
      "epoch": 5.452195210449927,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002460310447037077,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23586781322956085,
      "num_tokens": 124737896.0,
      "step": 60105
    },
    {
      "entropy": 5.694975328445435,
      "epoch": 5.452648766328012,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002459989578871765,
      "loss": 4.7244,
      "mean_token_accuracy": 0.2377829447388649,
      "num_tokens": 124748785.0,
      "step": 60110
    },
    {
      "entropy": 5.7960937976837155,
      "epoch": 5.453102322206096,
      "grad_norm": 1.375,
      "learning_rate": 0.00024596687167039425,
      "loss": 4.9018,
      "mean_token_accuracy": 0.22916053384542465,
      "num_tokens": 124760372.0,
      "step": 60115
    },
    {
      "entropy": 5.7661549091339115,
      "epoch": 5.45355587808418,
      "grad_norm": 1.28125,
      "learning_rate": 0.00024593478605402445,
      "loss": 4.87,
      "mean_token_accuracy": 0.2253403589129448,
      "num_tokens": 124771372.0,
      "step": 60120
    },
    {
      "entropy": 5.765401887893677,
      "epoch": 5.4540094339622645,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002459027010387308,
      "loss": 4.9092,
      "mean_token_accuracy": 0.23338212668895722,
      "num_tokens": 124781734.0,
      "step": 60125
    },
    {
      "entropy": 5.737032747268676,
      "epoch": 5.454462989840349,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002458706166251766,
      "loss": 4.827,
      "mean_token_accuracy": 0.23745880872011185,
      "num_tokens": 124792141.0,
      "step": 60130
    },
    {
      "entropy": 5.758210754394531,
      "epoch": 5.454916545718432,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002458385328140255,
      "loss": 4.899,
      "mean_token_accuracy": 0.2272140473127365,
      "num_tokens": 124803154.0,
      "step": 60135
    },
    {
      "entropy": 5.736509132385254,
      "epoch": 5.455370101596516,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000245806449605941,
      "loss": 4.8262,
      "mean_token_accuracy": 0.22903357446193695,
      "num_tokens": 124813160.0,
      "step": 60140
    },
    {
      "entropy": 5.82094373703003,
      "epoch": 5.455823657474601,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024577436700158655,
      "loss": 4.8832,
      "mean_token_accuracy": 0.22914521247148514,
      "num_tokens": 124823445.0,
      "step": 60145
    },
    {
      "entropy": 5.75492901802063,
      "epoch": 5.456277213352685,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002457422850016257,
      "loss": 4.8395,
      "mean_token_accuracy": 0.23069310635328294,
      "num_tokens": 124834598.0,
      "step": 60150
    },
    {
      "entropy": 5.71151032447815,
      "epoch": 5.456730769230769,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002457102036067217,
      "loss": 4.7987,
      "mean_token_accuracy": 0.2371416926383972,
      "num_tokens": 124844292.0,
      "step": 60155
    },
    {
      "entropy": 5.6822922706604,
      "epoch": 5.457184325108853,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024567812281753825,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22413105219602586,
      "num_tokens": 124854119.0,
      "step": 60160
    },
    {
      "entropy": 5.665872097015381,
      "epoch": 5.457637880986938,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002456460426347387,
      "loss": 4.7243,
      "mean_token_accuracy": 0.2449154570698738,
      "num_tokens": 124864441.0,
      "step": 60165
    },
    {
      "entropy": 5.746146106719971,
      "epoch": 5.458091436865022,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024561396305898645,
      "loss": 4.9312,
      "mean_token_accuracy": 0.22139687985181808,
      "num_tokens": 124874825.0,
      "step": 60170
    },
    {
      "entropy": 5.749810743331909,
      "epoch": 5.458544992743106,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024558188409094496,
      "loss": 4.828,
      "mean_token_accuracy": 0.23107898533344268,
      "num_tokens": 124883929.0,
      "step": 60175
    },
    {
      "entropy": 5.736826801300049,
      "epoch": 5.4589985486211905,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00024554980573127756,
      "loss": 4.8201,
      "mean_token_accuracy": 0.22945181280374527,
      "num_tokens": 124894829.0,
      "step": 60180
    },
    {
      "entropy": 5.744828128814698,
      "epoch": 5.459452104499274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024551772798064766,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23513370901346206,
      "num_tokens": 124905403.0,
      "step": 60185
    },
    {
      "entropy": 5.76446418762207,
      "epoch": 5.459905660377358,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024548565083971875,
      "loss": 4.9032,
      "mean_token_accuracy": 0.22286794781684877,
      "num_tokens": 124915309.0,
      "step": 60190
    },
    {
      "entropy": 5.681346225738525,
      "epoch": 5.460359216255442,
      "grad_norm": 1.46875,
      "learning_rate": 0.000245453574309154,
      "loss": 4.7564,
      "mean_token_accuracy": 0.23936910480260848,
      "num_tokens": 124924341.0,
      "step": 60195
    },
    {
      "entropy": 5.7840405941009525,
      "epoch": 5.460812772133527,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002454214983896169,
      "loss": 4.9017,
      "mean_token_accuracy": 0.22243474423885345,
      "num_tokens": 124934031.0,
      "step": 60200
    },
    {
      "entropy": 5.851454067230224,
      "epoch": 5.461266328011611,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002453894230817708,
      "loss": 4.9917,
      "mean_token_accuracy": 0.22447481602430344,
      "num_tokens": 124945059.0,
      "step": 60205
    },
    {
      "entropy": 5.8525162696838375,
      "epoch": 5.461719883889695,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024535734838627895,
      "loss": 4.9623,
      "mean_token_accuracy": 0.2187575414776802,
      "num_tokens": 124956182.0,
      "step": 60210
    },
    {
      "entropy": 5.806403636932373,
      "epoch": 5.4621734397677795,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002453252743038046,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22349131852388382,
      "num_tokens": 124967067.0,
      "step": 60215
    },
    {
      "entropy": 5.7407286167144775,
      "epoch": 5.462626995645864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002452932008350112,
      "loss": 4.8029,
      "mean_token_accuracy": 0.23110162168741227,
      "num_tokens": 124978270.0,
      "step": 60220
    },
    {
      "entropy": 5.810261964797974,
      "epoch": 5.463080551523948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00024526112798056206,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22014161348342895,
      "num_tokens": 124990406.0,
      "step": 60225
    },
    {
      "entropy": 5.762283658981323,
      "epoch": 5.463534107402032,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002452290557411203,
      "loss": 4.8486,
      "mean_token_accuracy": 0.2334250122308731,
      "num_tokens": 125000165.0,
      "step": 60230
    },
    {
      "entropy": 5.859541130065918,
      "epoch": 5.4639876632801165,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024519698411734923,
      "loss": 5.026,
      "mean_token_accuracy": 0.2176520362496376,
      "num_tokens": 125010815.0,
      "step": 60235
    },
    {
      "entropy": 5.868329763412476,
      "epoch": 5.4644412191582,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024516491310991214,
      "loss": 4.9401,
      "mean_token_accuracy": 0.22472713887691498,
      "num_tokens": 125020125.0,
      "step": 60240
    },
    {
      "entropy": 5.794366312026978,
      "epoch": 5.464894775036284,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002451328427194723,
      "loss": 4.8728,
      "mean_token_accuracy": 0.22266993522644044,
      "num_tokens": 125030253.0,
      "step": 60245
    },
    {
      "entropy": 5.735705327987671,
      "epoch": 5.4653483309143684,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024510077294669286,
      "loss": 4.824,
      "mean_token_accuracy": 0.23035874515771865,
      "num_tokens": 125040949.0,
      "step": 60250
    },
    {
      "entropy": 5.681847715377808,
      "epoch": 5.465801886792453,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000245068703792237,
      "loss": 4.7751,
      "mean_token_accuracy": 0.23731114268302916,
      "num_tokens": 125050549.0,
      "step": 60255
    },
    {
      "entropy": 5.719869232177734,
      "epoch": 5.466255442670537,
      "grad_norm": 1.328125,
      "learning_rate": 0.000245036635256768,
      "loss": 4.8672,
      "mean_token_accuracy": 0.2288963720202446,
      "num_tokens": 125061291.0,
      "step": 60260
    },
    {
      "entropy": 5.770724153518676,
      "epoch": 5.466708998548621,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024500456734094907,
      "loss": 4.85,
      "mean_token_accuracy": 0.22631922215223313,
      "num_tokens": 125071381.0,
      "step": 60265
    },
    {
      "entropy": 5.837673664093018,
      "epoch": 5.4671625544267055,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024497250004544335,
      "loss": 4.8456,
      "mean_token_accuracy": 0.23299288302659987,
      "num_tokens": 125081995.0,
      "step": 60270
    },
    {
      "entropy": 5.80779128074646,
      "epoch": 5.46761611030479,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024494043337091394,
      "loss": 4.8869,
      "mean_token_accuracy": 0.23123323917388916,
      "num_tokens": 125092436.0,
      "step": 60275
    },
    {
      "entropy": 5.745406913757324,
      "epoch": 5.468069666182874,
      "grad_norm": 1.375,
      "learning_rate": 0.000244908367318024,
      "loss": 4.8584,
      "mean_token_accuracy": 0.23096413761377335,
      "num_tokens": 125103228.0,
      "step": 60280
    },
    {
      "entropy": 5.7776207447052,
      "epoch": 5.468523222060958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002448763018874367,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23327501863241196,
      "num_tokens": 125113616.0,
      "step": 60285
    },
    {
      "entropy": 5.814168882369995,
      "epoch": 5.468976777939042,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002448442370798151,
      "loss": 4.92,
      "mean_token_accuracy": 0.2219358742237091,
      "num_tokens": 125124586.0,
      "step": 60290
    },
    {
      "entropy": 5.765963172912597,
      "epoch": 5.469430333817126,
      "grad_norm": 1.1875,
      "learning_rate": 0.00024481217289582246,
      "loss": 4.833,
      "mean_token_accuracy": 0.23420577198266984,
      "num_tokens": 125134457.0,
      "step": 60295
    },
    {
      "entropy": 5.717283296585083,
      "epoch": 5.46988388969521,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002447801093361217,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23614609241485596,
      "num_tokens": 125145116.0,
      "step": 60300
    },
    {
      "entropy": 5.808165884017944,
      "epoch": 5.4703374455732945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00024474804640137605,
      "loss": 4.9342,
      "mean_token_accuracy": 0.22332337498664856,
      "num_tokens": 125156164.0,
      "step": 60305
    },
    {
      "entropy": 5.711843061447143,
      "epoch": 5.470791001451379,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002447159840922485,
      "loss": 4.821,
      "mean_token_accuracy": 0.23414939641952515,
      "num_tokens": 125165955.0,
      "step": 60310
    },
    {
      "entropy": 5.678966951370239,
      "epoch": 5.471244557329463,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024468392240940205,
      "loss": 4.8077,
      "mean_token_accuracy": 0.23793211430311204,
      "num_tokens": 125176801.0,
      "step": 60315
    },
    {
      "entropy": 5.803739309310913,
      "epoch": 5.471698113207547,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002446518613534999,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22336412370204925,
      "num_tokens": 125186192.0,
      "step": 60320
    },
    {
      "entropy": 5.744718599319458,
      "epoch": 5.4721516690856316,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002446198009252049,
      "loss": 4.873,
      "mean_token_accuracy": 0.22363604307174684,
      "num_tokens": 125197542.0,
      "step": 60325
    },
    {
      "entropy": 5.7559957027435305,
      "epoch": 5.472605224963716,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002445877411251802,
      "loss": 4.8192,
      "mean_token_accuracy": 0.22820854932069778,
      "num_tokens": 125207066.0,
      "step": 60330
    },
    {
      "entropy": 5.72601547241211,
      "epoch": 5.4730587808418,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002445556819540887,
      "loss": 4.7499,
      "mean_token_accuracy": 0.2408626452088356,
      "num_tokens": 125215745.0,
      "step": 60335
    },
    {
      "entropy": 5.770017004013061,
      "epoch": 5.4735123367198835,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002445236234125936,
      "loss": 4.94,
      "mean_token_accuracy": 0.2264385089278221,
      "num_tokens": 125226976.0,
      "step": 60340
    },
    {
      "entropy": 5.834124612808227,
      "epoch": 5.473965892597968,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024449156550135766,
      "loss": 4.9334,
      "mean_token_accuracy": 0.22259876877069473,
      "num_tokens": 125239223.0,
      "step": 60345
    },
    {
      "entropy": 5.69725604057312,
      "epoch": 5.474419448476052,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024445950822104385,
      "loss": 4.7671,
      "mean_token_accuracy": 0.2350001722574234,
      "num_tokens": 125248962.0,
      "step": 60350
    },
    {
      "entropy": 5.763346385955811,
      "epoch": 5.474873004354136,
      "grad_norm": 1.375,
      "learning_rate": 0.0002444274515723153,
      "loss": 4.8225,
      "mean_token_accuracy": 0.23135105073451995,
      "num_tokens": 125258326.0,
      "step": 60355
    },
    {
      "entropy": 5.758658504486084,
      "epoch": 5.4753265602322205,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002443953955558348,
      "loss": 4.838,
      "mean_token_accuracy": 0.22859419733285904,
      "num_tokens": 125267711.0,
      "step": 60360
    },
    {
      "entropy": 5.76811671257019,
      "epoch": 5.475780116110305,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024436334017226535,
      "loss": 4.8983,
      "mean_token_accuracy": 0.23091135919094086,
      "num_tokens": 125279267.0,
      "step": 60365
    },
    {
      "entropy": 5.728475904464721,
      "epoch": 5.476233671988389,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002443312854222697,
      "loss": 4.8699,
      "mean_token_accuracy": 0.23332827240228654,
      "num_tokens": 125290671.0,
      "step": 60370
    },
    {
      "entropy": 5.752922773361206,
      "epoch": 5.476687227866473,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024429923130651107,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23155155181884765,
      "num_tokens": 125300387.0,
      "step": 60375
    },
    {
      "entropy": 5.832582283020019,
      "epoch": 5.477140783744558,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024426717782565204,
      "loss": 4.9632,
      "mean_token_accuracy": 0.21660177260637284,
      "num_tokens": 125310940.0,
      "step": 60380
    },
    {
      "entropy": 5.769948053359985,
      "epoch": 5.477594339622642,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002442351249803556,
      "loss": 4.8021,
      "mean_token_accuracy": 0.2324775919318199,
      "num_tokens": 125320716.0,
      "step": 60385
    },
    {
      "entropy": 5.776958322525024,
      "epoch": 5.478047895500726,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002442030727712847,
      "loss": 4.885,
      "mean_token_accuracy": 0.22363611310720444,
      "num_tokens": 125331175.0,
      "step": 60390
    },
    {
      "entropy": 5.816863775253296,
      "epoch": 5.4785014513788095,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024417102119910205,
      "loss": 4.8836,
      "mean_token_accuracy": 0.2236055001616478,
      "num_tokens": 125341396.0,
      "step": 60395
    },
    {
      "entropy": 5.727754783630371,
      "epoch": 5.478955007256894,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00024413897026447052,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23354504406452178,
      "num_tokens": 125352032.0,
      "step": 60400
    },
    {
      "entropy": 5.684459161758423,
      "epoch": 5.479408563134978,
      "grad_norm": 1.4375,
      "learning_rate": 0.000244106919968053,
      "loss": 4.7898,
      "mean_token_accuracy": 0.23344236016273498,
      "num_tokens": 125362909.0,
      "step": 60405
    },
    {
      "entropy": 5.851068639755249,
      "epoch": 5.479862119013062,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024407487031051224,
      "loss": 5.0171,
      "mean_token_accuracy": 0.21457452625036239,
      "num_tokens": 125373880.0,
      "step": 60410
    },
    {
      "entropy": 5.73221435546875,
      "epoch": 5.480315674891147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024404282129251097,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23684699684381486,
      "num_tokens": 125384364.0,
      "step": 60415
    },
    {
      "entropy": 5.741836929321289,
      "epoch": 5.480769230769231,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024401077291471212,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23593341261148454,
      "num_tokens": 125394985.0,
      "step": 60420
    },
    {
      "entropy": 5.825339508056641,
      "epoch": 5.481222786647315,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024397872517777837,
      "loss": 4.9297,
      "mean_token_accuracy": 0.23247765451669694,
      "num_tokens": 125405734.0,
      "step": 60425
    },
    {
      "entropy": 5.776834964752197,
      "epoch": 5.481676342525399,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024394667808237246,
      "loss": 4.936,
      "mean_token_accuracy": 0.22502511888742446,
      "num_tokens": 125417143.0,
      "step": 60430
    },
    {
      "entropy": 5.7202925205230715,
      "epoch": 5.482129898403484,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024391463162915718,
      "loss": 4.7948,
      "mean_token_accuracy": 0.23534522205591202,
      "num_tokens": 125426830.0,
      "step": 60435
    },
    {
      "entropy": 5.798351383209228,
      "epoch": 5.482583454281568,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024388258581879525,
      "loss": 4.9218,
      "mean_token_accuracy": 0.2170767143368721,
      "num_tokens": 125438332.0,
      "step": 60440
    },
    {
      "entropy": 5.786859846115112,
      "epoch": 5.483037010159651,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024385054065194934,
      "loss": 4.8516,
      "mean_token_accuracy": 0.2341096967458725,
      "num_tokens": 125447313.0,
      "step": 60445
    },
    {
      "entropy": 5.781339693069458,
      "epoch": 5.4834905660377355,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024381849612928224,
      "loss": 4.8355,
      "mean_token_accuracy": 0.233433273434639,
      "num_tokens": 125457758.0,
      "step": 60450
    },
    {
      "entropy": 5.733104515075683,
      "epoch": 5.48394412191582,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024378645225145657,
      "loss": 4.8356,
      "mean_token_accuracy": 0.23349330425262452,
      "num_tokens": 125467748.0,
      "step": 60455
    },
    {
      "entropy": 5.698206996917724,
      "epoch": 5.484397677793904,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000243754409019135,
      "loss": 4.7263,
      "mean_token_accuracy": 0.2378639444708824,
      "num_tokens": 125477968.0,
      "step": 60460
    },
    {
      "entropy": 5.819231700897217,
      "epoch": 5.484851233671988,
      "grad_norm": 1.53125,
      "learning_rate": 0.00024372236643298024,
      "loss": 4.968,
      "mean_token_accuracy": 0.22502818554639817,
      "num_tokens": 125487651.0,
      "step": 60465
    },
    {
      "entropy": 5.763676023483276,
      "epoch": 5.485304789550073,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002436903244936548,
      "loss": 4.9195,
      "mean_token_accuracy": 0.2268074184656143,
      "num_tokens": 125498093.0,
      "step": 60470
    },
    {
      "entropy": 5.726015186309814,
      "epoch": 5.485758345428157,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024365828320182154,
      "loss": 4.7883,
      "mean_token_accuracy": 0.2431759849190712,
      "num_tokens": 125508333.0,
      "step": 60475
    },
    {
      "entropy": 5.794683408737183,
      "epoch": 5.486211901306241,
      "grad_norm": 1.484375,
      "learning_rate": 0.00024362624255814302,
      "loss": 4.8896,
      "mean_token_accuracy": 0.2281597822904587,
      "num_tokens": 125518872.0,
      "step": 60480
    },
    {
      "entropy": 5.751872825622558,
      "epoch": 5.486665457184325,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00024359420256328169,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22258259654045104,
      "num_tokens": 125530944.0,
      "step": 60485
    },
    {
      "entropy": 5.734519195556641,
      "epoch": 5.48711901306241,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002435621632179003,
      "loss": 4.8451,
      "mean_token_accuracy": 0.23531339317560196,
      "num_tokens": 125541324.0,
      "step": 60490
    },
    {
      "entropy": 5.699677038192749,
      "epoch": 5.487572568940493,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024353012452266138,
      "loss": 4.8131,
      "mean_token_accuracy": 0.233228699862957,
      "num_tokens": 125550489.0,
      "step": 60495
    },
    {
      "entropy": 5.772171306610107,
      "epoch": 5.488026124818577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024349808647822752,
      "loss": 4.9125,
      "mean_token_accuracy": 0.22305946350097655,
      "num_tokens": 125560602.0,
      "step": 60500
    },
    {
      "entropy": 5.728610610961914,
      "epoch": 5.488479680696662,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002434660490852612,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23855020701885224,
      "num_tokens": 125570054.0,
      "step": 60505
    },
    {
      "entropy": 5.735528039932251,
      "epoch": 5.488933236574746,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024343401234442507,
      "loss": 4.7907,
      "mean_token_accuracy": 0.23450353145599365,
      "num_tokens": 125579626.0,
      "step": 60510
    },
    {
      "entropy": 5.735679769515992,
      "epoch": 5.48938679245283,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024340197625638167,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23205209970474244,
      "num_tokens": 125590352.0,
      "step": 60515
    },
    {
      "entropy": 5.751881408691406,
      "epoch": 5.489840348330914,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024336994082179336,
      "loss": 4.8616,
      "mean_token_accuracy": 0.2280761942267418,
      "num_tokens": 125600897.0,
      "step": 60520
    },
    {
      "entropy": 5.698219299316406,
      "epoch": 5.490293904208999,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024333790604132277,
      "loss": 4.7629,
      "mean_token_accuracy": 0.2318425640463829,
      "num_tokens": 125612223.0,
      "step": 60525
    },
    {
      "entropy": 5.8267193794250485,
      "epoch": 5.490747460087083,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024330587191563236,
      "loss": 4.9901,
      "mean_token_accuracy": 0.21909252256155015,
      "num_tokens": 125623289.0,
      "step": 60530
    },
    {
      "entropy": 5.834580278396606,
      "epoch": 5.491201015965167,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002432738384453846,
      "loss": 4.9136,
      "mean_token_accuracy": 0.22822960317134858,
      "num_tokens": 125633712.0,
      "step": 60535
    },
    {
      "entropy": 5.727538394927978,
      "epoch": 5.4916545718432515,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024324180563124186,
      "loss": 4.8602,
      "mean_token_accuracy": 0.22695240527391433,
      "num_tokens": 125644089.0,
      "step": 60540
    },
    {
      "entropy": 5.780232715606689,
      "epoch": 5.492108127721336,
      "grad_norm": 1.5,
      "learning_rate": 0.00024320977347386675,
      "loss": 4.8747,
      "mean_token_accuracy": 0.224610598385334,
      "num_tokens": 125653363.0,
      "step": 60545
    },
    {
      "entropy": 5.889667224884033,
      "epoch": 5.492561683599419,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002431777419739216,
      "loss": 4.9863,
      "mean_token_accuracy": 0.21761952340602875,
      "num_tokens": 125663774.0,
      "step": 60550
    },
    {
      "entropy": 5.635402822494507,
      "epoch": 5.493015239477503,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024314571113206884,
      "loss": 4.7182,
      "mean_token_accuracy": 0.23358422368764878,
      "num_tokens": 125674225.0,
      "step": 60555
    },
    {
      "entropy": 5.7379796504974365,
      "epoch": 5.493468795355588,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00024311368094897097,
      "loss": 4.8018,
      "mean_token_accuracy": 0.23171792179346085,
      "num_tokens": 125684128.0,
      "step": 60560
    },
    {
      "entropy": 5.774091482162476,
      "epoch": 5.493922351233672,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024308165142529026,
      "loss": 4.8899,
      "mean_token_accuracy": 0.2221697524189949,
      "num_tokens": 125694704.0,
      "step": 60565
    },
    {
      "entropy": 5.782389450073242,
      "epoch": 5.494375907111756,
      "grad_norm": 1.234375,
      "learning_rate": 0.00024304962256168911,
      "loss": 4.896,
      "mean_token_accuracy": 0.233230796456337,
      "num_tokens": 125705216.0,
      "step": 60570
    },
    {
      "entropy": 5.808228540420532,
      "epoch": 5.49482946298984,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024301759435882992,
      "loss": 4.9013,
      "mean_token_accuracy": 0.23067556023597718,
      "num_tokens": 125715690.0,
      "step": 60575
    },
    {
      "entropy": 5.769849967956543,
      "epoch": 5.495283018867925,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024298556681737505,
      "loss": 4.8942,
      "mean_token_accuracy": 0.22950443625450134,
      "num_tokens": 125726179.0,
      "step": 60580
    },
    {
      "entropy": 5.762535429000854,
      "epoch": 5.495736574746009,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002429535399379868,
      "loss": 4.9013,
      "mean_token_accuracy": 0.2300898790359497,
      "num_tokens": 125736498.0,
      "step": 60585
    },
    {
      "entropy": 5.700122690200805,
      "epoch": 5.496190130624093,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024292151372132747,
      "loss": 4.7844,
      "mean_token_accuracy": 0.23126632124185562,
      "num_tokens": 125746560.0,
      "step": 60590
    },
    {
      "entropy": 5.709147930145264,
      "epoch": 5.496643686502177,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024288948816805945,
      "loss": 4.7516,
      "mean_token_accuracy": 0.24329727590084077,
      "num_tokens": 125756973.0,
      "step": 60595
    },
    {
      "entropy": 5.746280527114868,
      "epoch": 5.497097242380261,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000242857463278845,
      "loss": 4.8563,
      "mean_token_accuracy": 0.23223625868558884,
      "num_tokens": 125767889.0,
      "step": 60600
    },
    {
      "entropy": 5.765365552902222,
      "epoch": 5.497550798258345,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002428254390543464,
      "loss": 4.8969,
      "mean_token_accuracy": 0.23216029554605483,
      "num_tokens": 125778879.0,
      "step": 60605
    },
    {
      "entropy": 5.801500272750855,
      "epoch": 5.498004354136429,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024279341549522594,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2255736246705055,
      "num_tokens": 125788654.0,
      "step": 60610
    },
    {
      "entropy": 5.72476806640625,
      "epoch": 5.498457910014514,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002427613926021458,
      "loss": 4.8737,
      "mean_token_accuracy": 0.22967825680971146,
      "num_tokens": 125799343.0,
      "step": 60615
    },
    {
      "entropy": 5.774102830886841,
      "epoch": 5.498911465892598,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024272937037576832,
      "loss": 4.891,
      "mean_token_accuracy": 0.22370796352624894,
      "num_tokens": 125810415.0,
      "step": 60620
    },
    {
      "entropy": 5.750460767745972,
      "epoch": 5.499365021770682,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024269734881675564,
      "loss": 4.8235,
      "mean_token_accuracy": 0.22759006023406983,
      "num_tokens": 125821016.0,
      "step": 60625
    },
    {
      "entropy": 5.752955436706543,
      "epoch": 5.4998185776487665,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024266532792577007,
      "loss": 4.8469,
      "mean_token_accuracy": 0.2315983444452286,
      "num_tokens": 125830881.0,
      "step": 60630
    },
    {
      "entropy": 5.734161281585694,
      "epoch": 5.500272133526851,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024263330770347375,
      "loss": 4.7904,
      "mean_token_accuracy": 0.2377526968717575,
      "num_tokens": 125839896.0,
      "step": 60635
    },
    {
      "entropy": 5.787393426895141,
      "epoch": 5.500725689404935,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024260128815052883,
      "loss": 4.9221,
      "mean_token_accuracy": 0.23002861738204955,
      "num_tokens": 125850434.0,
      "step": 60640
    },
    {
      "entropy": 5.705767393112183,
      "epoch": 5.501179245283019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002425692692675976,
      "loss": 4.7616,
      "mean_token_accuracy": 0.23198652714490892,
      "num_tokens": 125860806.0,
      "step": 60645
    },
    {
      "entropy": 5.808223056793213,
      "epoch": 5.501632801161103,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024253725105534207,
      "loss": 4.9085,
      "mean_token_accuracy": 0.23343318253755568,
      "num_tokens": 125871649.0,
      "step": 60650
    },
    {
      "entropy": 5.744110250473023,
      "epoch": 5.502086357039187,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024250523351442455,
      "loss": 4.8686,
      "mean_token_accuracy": 0.217412967979908,
      "num_tokens": 125881855.0,
      "step": 60655
    },
    {
      "entropy": 5.759748125076294,
      "epoch": 5.502539912917271,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024247321664550698,
      "loss": 4.8956,
      "mean_token_accuracy": 0.23414877504110337,
      "num_tokens": 125891147.0,
      "step": 60660
    },
    {
      "entropy": 5.7606837272644045,
      "epoch": 5.5029934687953554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00024244120044925166,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23162750601768495,
      "num_tokens": 125900165.0,
      "step": 60665
    },
    {
      "entropy": 5.774419975280762,
      "epoch": 5.50344702467344,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024240918492632058,
      "loss": 4.8053,
      "mean_token_accuracy": 0.22923588156700134,
      "num_tokens": 125910540.0,
      "step": 60670
    },
    {
      "entropy": 5.790562534332276,
      "epoch": 5.503900580551524,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024237717007737582,
      "loss": 4.8215,
      "mean_token_accuracy": 0.22813387364149093,
      "num_tokens": 125920704.0,
      "step": 60675
    },
    {
      "entropy": 5.830491590499878,
      "epoch": 5.504354136429608,
      "grad_norm": 1.4375,
      "learning_rate": 0.00024234515590307955,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22434040158987045,
      "num_tokens": 125931417.0,
      "step": 60680
    },
    {
      "entropy": 5.697303056716919,
      "epoch": 5.5048076923076925,
      "grad_norm": 1.375,
      "learning_rate": 0.00024231314240409375,
      "loss": 4.8291,
      "mean_token_accuracy": 0.24191376864910125,
      "num_tokens": 125942429.0,
      "step": 60685
    },
    {
      "entropy": 5.722144508361817,
      "epoch": 5.505261248185777,
      "grad_norm": 1.578125,
      "learning_rate": 0.00024228112958108052,
      "loss": 4.8124,
      "mean_token_accuracy": 0.24242394119501115,
      "num_tokens": 125952611.0,
      "step": 60690
    },
    {
      "entropy": 5.733221387863159,
      "epoch": 5.505714804063861,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024224911743470195,
      "loss": 4.8351,
      "mean_token_accuracy": 0.22894175350666046,
      "num_tokens": 125963156.0,
      "step": 60695
    },
    {
      "entropy": 5.7373944282531735,
      "epoch": 5.506168359941945,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024221710596561985,
      "loss": 4.8378,
      "mean_token_accuracy": 0.2299693524837494,
      "num_tokens": 125974207.0,
      "step": 60700
    },
    {
      "entropy": 5.743368530273438,
      "epoch": 5.506621915820029,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002421850951744964,
      "loss": 4.7777,
      "mean_token_accuracy": 0.23272951394319535,
      "num_tokens": 125984340.0,
      "step": 60705
    },
    {
      "entropy": 5.799983119964599,
      "epoch": 5.507075471698113,
      "grad_norm": 1.34375,
      "learning_rate": 0.00024215308506199352,
      "loss": 4.9035,
      "mean_token_accuracy": 0.2321609452366829,
      "num_tokens": 125994211.0,
      "step": 60710
    },
    {
      "entropy": 5.847073125839233,
      "epoch": 5.507529027576197,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024212107562877322,
      "loss": 4.9632,
      "mean_token_accuracy": 0.22216108441352844,
      "num_tokens": 126005352.0,
      "step": 60715
    },
    {
      "entropy": 5.82396936416626,
      "epoch": 5.5079825834542815,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024208906687549743,
      "loss": 4.9294,
      "mean_token_accuracy": 0.22552372962236406,
      "num_tokens": 126015833.0,
      "step": 60720
    },
    {
      "entropy": 5.804620981216431,
      "epoch": 5.508436139332366,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002420570588028282,
      "loss": 4.8708,
      "mean_token_accuracy": 0.2305353105068207,
      "num_tokens": 126026556.0,
      "step": 60725
    },
    {
      "entropy": 5.747214698791504,
      "epoch": 5.50888969521045,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024202505141142732,
      "loss": 4.8708,
      "mean_token_accuracy": 0.22889342606067659,
      "num_tokens": 126037561.0,
      "step": 60730
    },
    {
      "entropy": 5.811886739730835,
      "epoch": 5.509343251088534,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024199304470195676,
      "loss": 4.9851,
      "mean_token_accuracy": 0.2233654409646988,
      "num_tokens": 126048008.0,
      "step": 60735
    },
    {
      "entropy": 5.781643056869507,
      "epoch": 5.5097968069666186,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002419610386750785,
      "loss": 4.8683,
      "mean_token_accuracy": 0.22852229326963425,
      "num_tokens": 126058622.0,
      "step": 60740
    },
    {
      "entropy": 5.837650775909424,
      "epoch": 5.510250362844703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024192903333145437,
      "loss": 4.8724,
      "mean_token_accuracy": 0.2263342872262001,
      "num_tokens": 126069235.0,
      "step": 60745
    },
    {
      "entropy": 5.7148364067077635,
      "epoch": 5.510703918722786,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002418970286717463,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23118884414434432,
      "num_tokens": 126079247.0,
      "step": 60750
    },
    {
      "entropy": 5.782214069366455,
      "epoch": 5.5111574746008705,
      "grad_norm": 1.515625,
      "learning_rate": 0.00024186502469661602,
      "loss": 4.889,
      "mean_token_accuracy": 0.22286295592784883,
      "num_tokens": 126089060.0,
      "step": 60755
    },
    {
      "entropy": 5.761290645599365,
      "epoch": 5.511611030478955,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024183302140672554,
      "loss": 4.843,
      "mean_token_accuracy": 0.2312833249568939,
      "num_tokens": 126099988.0,
      "step": 60760
    },
    {
      "entropy": 5.7582591533660885,
      "epoch": 5.512064586357039,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024180101880273664,
      "loss": 4.8642,
      "mean_token_accuracy": 0.22761610746383668,
      "num_tokens": 126110102.0,
      "step": 60765
    },
    {
      "entropy": 5.817187118530273,
      "epoch": 5.512518142235123,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024176901688531104,
      "loss": 4.8709,
      "mean_token_accuracy": 0.23154381662607193,
      "num_tokens": 126120342.0,
      "step": 60770
    },
    {
      "entropy": 5.787964344024658,
      "epoch": 5.5129716981132075,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002417370156551107,
      "loss": 4.9112,
      "mean_token_accuracy": 0.22543468922376633,
      "num_tokens": 126131121.0,
      "step": 60775
    },
    {
      "entropy": 5.820700597763062,
      "epoch": 5.513425253991292,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024170501511279734,
      "loss": 4.8944,
      "mean_token_accuracy": 0.229422864317894,
      "num_tokens": 126141004.0,
      "step": 60780
    },
    {
      "entropy": 5.755273294448853,
      "epoch": 5.513878809869376,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024167301525903273,
      "loss": 4.8577,
      "mean_token_accuracy": 0.23510974496603013,
      "num_tokens": 126150692.0,
      "step": 60785
    },
    {
      "entropy": 5.800183439254761,
      "epoch": 5.51433236574746,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024164101609447868,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22585658878087997,
      "num_tokens": 126161659.0,
      "step": 60790
    },
    {
      "entropy": 5.713426542282105,
      "epoch": 5.514785921625545,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00024160901761979698,
      "loss": 4.8408,
      "mean_token_accuracy": 0.2308924451470375,
      "num_tokens": 126171958.0,
      "step": 60795
    },
    {
      "entropy": 5.765686464309693,
      "epoch": 5.515239477503629,
      "grad_norm": 1.46875,
      "learning_rate": 0.00024157701983564927,
      "loss": 4.8544,
      "mean_token_accuracy": 0.23210838586091995,
      "num_tokens": 126181491.0,
      "step": 60800
    },
    {
      "entropy": 5.786379289627075,
      "epoch": 5.515693033381712,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024154502274269723,
      "loss": 4.8793,
      "mean_token_accuracy": 0.22609312534332277,
      "num_tokens": 126192318.0,
      "step": 60805
    },
    {
      "entropy": 5.755172967910767,
      "epoch": 5.5161465892597965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00024151302634160276,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23742668330669403,
      "num_tokens": 126202375.0,
      "step": 60810
    },
    {
      "entropy": 5.751914215087891,
      "epoch": 5.516600145137881,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002414810306330274,
      "loss": 4.8245,
      "mean_token_accuracy": 0.23538895845413207,
      "num_tokens": 126211172.0,
      "step": 60815
    },
    {
      "entropy": 5.825482416152954,
      "epoch": 5.517053701015965,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002414490356176328,
      "loss": 4.9288,
      "mean_token_accuracy": 0.22697273790836334,
      "num_tokens": 126221040.0,
      "step": 60820
    },
    {
      "entropy": 5.719668388366699,
      "epoch": 5.517507256894049,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024141704129608077,
      "loss": 4.8086,
      "mean_token_accuracy": 0.23215322941541672,
      "num_tokens": 126230016.0,
      "step": 60825
    },
    {
      "entropy": 5.777716255187988,
      "epoch": 5.517960812772134,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024138504766903286,
      "loss": 4.9219,
      "mean_token_accuracy": 0.22413024604320525,
      "num_tokens": 126240343.0,
      "step": 60830
    },
    {
      "entropy": 5.77862286567688,
      "epoch": 5.518414368650218,
      "grad_norm": 1.25,
      "learning_rate": 0.00024135305473715074,
      "loss": 4.893,
      "mean_token_accuracy": 0.2283478632569313,
      "num_tokens": 126251063.0,
      "step": 60835
    },
    {
      "entropy": 5.731346273422242,
      "epoch": 5.518867924528302,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024132106250109598,
      "loss": 4.7744,
      "mean_token_accuracy": 0.23461494892835616,
      "num_tokens": 126260312.0,
      "step": 60840
    },
    {
      "entropy": 5.709058952331543,
      "epoch": 5.519321480406386,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024128907096153024,
      "loss": 4.7832,
      "mean_token_accuracy": 0.2378908082842827,
      "num_tokens": 126270607.0,
      "step": 60845
    },
    {
      "entropy": 5.657137250900268,
      "epoch": 5.51977503628447,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024125708011911518,
      "loss": 4.7306,
      "mean_token_accuracy": 0.24231739044189454,
      "num_tokens": 126280885.0,
      "step": 60850
    },
    {
      "entropy": 5.761921977996826,
      "epoch": 5.520228592162555,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002412250899745122,
      "loss": 4.9969,
      "mean_token_accuracy": 0.21804598867893218,
      "num_tokens": 126291206.0,
      "step": 60855
    },
    {
      "entropy": 5.688448238372803,
      "epoch": 5.520682148040638,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00024119310052838301,
      "loss": 4.7982,
      "mean_token_accuracy": 0.2322043940424919,
      "num_tokens": 126302434.0,
      "step": 60860
    },
    {
      "entropy": 5.7990288734436035,
      "epoch": 5.5211357039187225,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024116111178138918,
      "loss": 4.9141,
      "mean_token_accuracy": 0.2186686098575592,
      "num_tokens": 126313114.0,
      "step": 60865
    },
    {
      "entropy": 5.767472171783448,
      "epoch": 5.521589259796807,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002411291237341921,
      "loss": 4.7912,
      "mean_token_accuracy": 0.23635710924863815,
      "num_tokens": 126323172.0,
      "step": 60870
    },
    {
      "entropy": 5.725008535385132,
      "epoch": 5.522042815674891,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024109713638745335,
      "loss": 4.7923,
      "mean_token_accuracy": 0.23412148803472518,
      "num_tokens": 126334690.0,
      "step": 60875
    },
    {
      "entropy": 5.7411637783050535,
      "epoch": 5.522496371552975,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024106514974183452,
      "loss": 4.808,
      "mean_token_accuracy": 0.23991503417491913,
      "num_tokens": 126345097.0,
      "step": 60880
    },
    {
      "entropy": 5.777131843566894,
      "epoch": 5.52294992743106,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000241033163797997,
      "loss": 4.8566,
      "mean_token_accuracy": 0.22728825360536575,
      "num_tokens": 126356130.0,
      "step": 60885
    },
    {
      "entropy": 5.749288749694824,
      "epoch": 5.523403483309144,
      "grad_norm": 1.890625,
      "learning_rate": 0.0002410011785566023,
      "loss": 4.8586,
      "mean_token_accuracy": 0.23175693452358245,
      "num_tokens": 126366640.0,
      "step": 60890
    },
    {
      "entropy": 5.820151472091675,
      "epoch": 5.523857039187228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00024096919401831197,
      "loss": 4.9246,
      "mean_token_accuracy": 0.2194299355149269,
      "num_tokens": 126378465.0,
      "step": 60895
    },
    {
      "entropy": 5.728661680221558,
      "epoch": 5.524310595065312,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00024093721018378733,
      "loss": 4.8334,
      "mean_token_accuracy": 0.23015864342451095,
      "num_tokens": 126388506.0,
      "step": 60900
    },
    {
      "entropy": 5.803047323226929,
      "epoch": 5.524764150943396,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024090522705368984,
      "loss": 4.905,
      "mean_token_accuracy": 0.22111255228519439,
      "num_tokens": 126398674.0,
      "step": 60905
    },
    {
      "entropy": 5.679454374313354,
      "epoch": 5.52521770682148,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024087324462868098,
      "loss": 4.7183,
      "mean_token_accuracy": 0.2398838832974434,
      "num_tokens": 126409671.0,
      "step": 60910
    },
    {
      "entropy": 5.769996595382691,
      "epoch": 5.525671262699564,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00024084126290942215,
      "loss": 4.9398,
      "mean_token_accuracy": 0.22977309823036193,
      "num_tokens": 126419797.0,
      "step": 60915
    },
    {
      "entropy": 5.761795616149902,
      "epoch": 5.526124818577649,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00024080928189657464,
      "loss": 4.9511,
      "mean_token_accuracy": 0.22375517189502717,
      "num_tokens": 126430964.0,
      "step": 60920
    },
    {
      "entropy": 5.787273406982422,
      "epoch": 5.526578374455733,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002407773015907999,
      "loss": 4.8746,
      "mean_token_accuracy": 0.22505122125148774,
      "num_tokens": 126440909.0,
      "step": 60925
    },
    {
      "entropy": 5.814329195022583,
      "epoch": 5.527031930333817,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00024074532199275928,
      "loss": 4.8511,
      "mean_token_accuracy": 0.23268292397260665,
      "num_tokens": 126450492.0,
      "step": 60930
    },
    {
      "entropy": 5.730631923675537,
      "epoch": 5.527485486211901,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024071334310311415,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23327966332435607,
      "num_tokens": 126462209.0,
      "step": 60935
    },
    {
      "entropy": 5.795416069030762,
      "epoch": 5.527939042089986,
      "grad_norm": 1.375,
      "learning_rate": 0.00024068136492252584,
      "loss": 4.9043,
      "mean_token_accuracy": 0.22215009927749635,
      "num_tokens": 126473226.0,
      "step": 60940
    },
    {
      "entropy": 5.810486364364624,
      "epoch": 5.52839259796807,
      "grad_norm": 1.390625,
      "learning_rate": 0.00024064938745165564,
      "loss": 4.835,
      "mean_token_accuracy": 0.23301441371440887,
      "num_tokens": 126483419.0,
      "step": 60945
    },
    {
      "entropy": 5.750939750671387,
      "epoch": 5.528846153846154,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024061741069116488,
      "loss": 4.8107,
      "mean_token_accuracy": 0.23444368243217467,
      "num_tokens": 126494076.0,
      "step": 60950
    },
    {
      "entropy": 5.776532125473023,
      "epoch": 5.5292997097242385,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002405854346417148,
      "loss": 4.8972,
      "mean_token_accuracy": 0.2299799770116806,
      "num_tokens": 126504339.0,
      "step": 60955
    },
    {
      "entropy": 5.664638805389404,
      "epoch": 5.529753265602322,
      "grad_norm": 1.375,
      "learning_rate": 0.00024055345930396667,
      "loss": 4.732,
      "mean_token_accuracy": 0.23514699339866638,
      "num_tokens": 126514110.0,
      "step": 60960
    },
    {
      "entropy": 5.711789608001709,
      "epoch": 5.530206821480406,
      "grad_norm": 1.296875,
      "learning_rate": 0.00024052148467858183,
      "loss": 4.7867,
      "mean_token_accuracy": 0.2348064437508583,
      "num_tokens": 126524520.0,
      "step": 60965
    },
    {
      "entropy": 5.744262266159057,
      "epoch": 5.53066037735849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00024048951076622145,
      "loss": 4.8808,
      "mean_token_accuracy": 0.23074570298194885,
      "num_tokens": 126536069.0,
      "step": 60970
    },
    {
      "entropy": 5.689115381240844,
      "epoch": 5.531113933236575,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024045753756754673,
      "loss": 4.7545,
      "mean_token_accuracy": 0.24482348561286926,
      "num_tokens": 126545374.0,
      "step": 60975
    },
    {
      "entropy": 5.719000816345215,
      "epoch": 5.531567489114659,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024042556508321904,
      "loss": 4.8622,
      "mean_token_accuracy": 0.22903051525354384,
      "num_tokens": 126555995.0,
      "step": 60980
    },
    {
      "entropy": 5.73126859664917,
      "epoch": 5.532021044992743,
      "grad_norm": 1.3125,
      "learning_rate": 0.00024039359331389944,
      "loss": 4.8371,
      "mean_token_accuracy": 0.22933810949325562,
      "num_tokens": 126566281.0,
      "step": 60985
    },
    {
      "entropy": 5.764046859741211,
      "epoch": 5.532474600870827,
      "grad_norm": 1.703125,
      "learning_rate": 0.0002403616222602491,
      "loss": 4.8694,
      "mean_token_accuracy": 0.235222664475441,
      "num_tokens": 126575333.0,
      "step": 60990
    },
    {
      "entropy": 5.7557168960571286,
      "epoch": 5.532928156748912,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00024032965192292928,
      "loss": 4.8813,
      "mean_token_accuracy": 0.23420776575803756,
      "num_tokens": 126586071.0,
      "step": 60995
    },
    {
      "entropy": 5.686702394485474,
      "epoch": 5.533381712626996,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024029768230260113,
      "loss": 4.8344,
      "mean_token_accuracy": 0.22658687978982925,
      "num_tokens": 126596641.0,
      "step": 61000
    },
    {
      "entropy": 5.739862155914307,
      "epoch": 5.533835268505079,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002402657133999257,
      "loss": 4.8883,
      "mean_token_accuracy": 0.2306861087679863,
      "num_tokens": 126607521.0,
      "step": 61005
    },
    {
      "entropy": 5.758076333999634,
      "epoch": 5.5342888243831645,
      "grad_norm": 1.453125,
      "learning_rate": 0.00024023374521556413,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2393776535987854,
      "num_tokens": 126617513.0,
      "step": 61010
    },
    {
      "entropy": 5.751321840286255,
      "epoch": 5.534742380261248,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00024020177775017766,
      "loss": 4.8296,
      "mean_token_accuracy": 0.23456534147262573,
      "num_tokens": 126627552.0,
      "step": 61015
    },
    {
      "entropy": 5.731550455093384,
      "epoch": 5.535195936139332,
      "grad_norm": 1.40625,
      "learning_rate": 0.00024016981100442725,
      "loss": 4.8267,
      "mean_token_accuracy": 0.23553713858127595,
      "num_tokens": 126637587.0,
      "step": 61020
    },
    {
      "entropy": 5.674509716033936,
      "epoch": 5.535649492017416,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00024013784497897402,
      "loss": 4.7212,
      "mean_token_accuracy": 0.24265725314617156,
      "num_tokens": 126647123.0,
      "step": 61025
    },
    {
      "entropy": 5.778433465957642,
      "epoch": 5.536103047895501,
      "grad_norm": 1.421875,
      "learning_rate": 0.00024010587967447904,
      "loss": 4.8901,
      "mean_token_accuracy": 0.23130233138799666,
      "num_tokens": 126657489.0,
      "step": 61030
    },
    {
      "entropy": 5.797156143188476,
      "epoch": 5.536556603773585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00024007391509160337,
      "loss": 4.8632,
      "mean_token_accuracy": 0.23263555020093918,
      "num_tokens": 126667691.0,
      "step": 61035
    },
    {
      "entropy": 5.8180474758148195,
      "epoch": 5.537010159651669,
      "grad_norm": 1.328125,
      "learning_rate": 0.00024004195123100808,
      "loss": 4.916,
      "mean_token_accuracy": 0.2212613418698311,
      "num_tokens": 126678614.0,
      "step": 61040
    },
    {
      "entropy": 5.7676490306854244,
      "epoch": 5.5374637155297535,
      "grad_norm": 1.359375,
      "learning_rate": 0.00024000998809335406,
      "loss": 4.8575,
      "mean_token_accuracy": 0.2300331711769104,
      "num_tokens": 126688241.0,
      "step": 61045
    },
    {
      "entropy": 5.773014259338379,
      "epoch": 5.537917271407838,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023997802567930245,
      "loss": 4.8378,
      "mean_token_accuracy": 0.22980307340621947,
      "num_tokens": 126698056.0,
      "step": 61050
    },
    {
      "entropy": 5.717348051071167,
      "epoch": 5.538370827285922,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023994606398951418,
      "loss": 4.7928,
      "mean_token_accuracy": 0.2381496548652649,
      "num_tokens": 126709336.0,
      "step": 61055
    },
    {
      "entropy": 5.78118143081665,
      "epoch": 5.538824383164005,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023991410302465023,
      "loss": 4.8856,
      "mean_token_accuracy": 0.23248049467802048,
      "num_tokens": 126720917.0,
      "step": 61060
    },
    {
      "entropy": 5.750292253494263,
      "epoch": 5.53927793904209,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023988214278537157,
      "loss": 4.8396,
      "mean_token_accuracy": 0.22239184230566025,
      "num_tokens": 126732059.0,
      "step": 61065
    },
    {
      "entropy": 5.777815818786621,
      "epoch": 5.539731494920174,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002398501832723392,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22729071080684662,
      "num_tokens": 126743354.0,
      "step": 61070
    },
    {
      "entropy": 5.737503337860107,
      "epoch": 5.540185050798258,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023981822448621397,
      "loss": 4.8201,
      "mean_token_accuracy": 0.23843912929296493,
      "num_tokens": 126753468.0,
      "step": 61075
    },
    {
      "entropy": 5.762113809585571,
      "epoch": 5.5406386066763424,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023978626642765676,
      "loss": 4.8453,
      "mean_token_accuracy": 0.23023263961076737,
      "num_tokens": 126763229.0,
      "step": 61080
    },
    {
      "entropy": 5.759349393844604,
      "epoch": 5.541092162554427,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002397543090973286,
      "loss": 4.8356,
      "mean_token_accuracy": 0.2315279319882393,
      "num_tokens": 126773997.0,
      "step": 61085
    },
    {
      "entropy": 5.765229988098144,
      "epoch": 5.541545718432511,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023972235249589035,
      "loss": 4.8029,
      "mean_token_accuracy": 0.2293090358376503,
      "num_tokens": 126784124.0,
      "step": 61090
    },
    {
      "entropy": 5.844376420974731,
      "epoch": 5.541999274310595,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023969039662400271,
      "loss": 4.9286,
      "mean_token_accuracy": 0.2319178119301796,
      "num_tokens": 126795206.0,
      "step": 61095
    },
    {
      "entropy": 5.805587959289551,
      "epoch": 5.5424528301886795,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023965844148232673,
      "loss": 4.949,
      "mean_token_accuracy": 0.22258429378271102,
      "num_tokens": 126805798.0,
      "step": 61100
    },
    {
      "entropy": 5.766224718093872,
      "epoch": 5.542906386066764,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023962648707152322,
      "loss": 4.8181,
      "mean_token_accuracy": 0.24659052789211272,
      "num_tokens": 126816189.0,
      "step": 61105
    },
    {
      "entropy": 5.786282539367676,
      "epoch": 5.543359941944848,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023959453339225296,
      "loss": 4.9142,
      "mean_token_accuracy": 0.23044454902410508,
      "num_tokens": 126827219.0,
      "step": 61110
    },
    {
      "entropy": 5.8337846279144285,
      "epoch": 5.543813497822931,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023956258044517675,
      "loss": 4.9307,
      "mean_token_accuracy": 0.22594169974327089,
      "num_tokens": 126836930.0,
      "step": 61115
    },
    {
      "entropy": 5.772285461425781,
      "epoch": 5.544267053701016,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023953062823095546,
      "loss": 4.859,
      "mean_token_accuracy": 0.23026665151119233,
      "num_tokens": 126846163.0,
      "step": 61120
    },
    {
      "entropy": 5.765929412841797,
      "epoch": 5.5447206095791,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023949867675024982,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22792547047138215,
      "num_tokens": 126857141.0,
      "step": 61125
    },
    {
      "entropy": 5.826832914352417,
      "epoch": 5.545174165457184,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002394667260037206,
      "loss": 4.9481,
      "mean_token_accuracy": 0.22786979377269745,
      "num_tokens": 126866872.0,
      "step": 61130
    },
    {
      "entropy": 5.670581865310669,
      "epoch": 5.5456277213352685,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023943477599202846,
      "loss": 4.6989,
      "mean_token_accuracy": 0.24829467982053757,
      "num_tokens": 126877367.0,
      "step": 61135
    },
    {
      "entropy": 5.737587738037109,
      "epoch": 5.546081277213353,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002394028267158342,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22654692828655243,
      "num_tokens": 126888161.0,
      "step": 61140
    },
    {
      "entropy": 5.801478576660156,
      "epoch": 5.546534833091437,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002393708781757986,
      "loss": 4.924,
      "mean_token_accuracy": 0.2297637939453125,
      "num_tokens": 126897981.0,
      "step": 61145
    },
    {
      "entropy": 5.797157144546508,
      "epoch": 5.546988388969521,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023933893037258235,
      "loss": 4.887,
      "mean_token_accuracy": 0.22764071226119995,
      "num_tokens": 126908696.0,
      "step": 61150
    },
    {
      "entropy": 5.867232704162598,
      "epoch": 5.5474419448476056,
      "grad_norm": 1.375,
      "learning_rate": 0.00023930698330684602,
      "loss": 4.9175,
      "mean_token_accuracy": 0.2211208835244179,
      "num_tokens": 126919336.0,
      "step": 61155
    },
    {
      "entropy": 5.786465835571289,
      "epoch": 5.547895500725689,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023927503697925045,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2279188498854637,
      "num_tokens": 126929975.0,
      "step": 61160
    },
    {
      "entropy": 5.77167239189148,
      "epoch": 5.548349056603773,
      "grad_norm": 1.5,
      "learning_rate": 0.00023924309139045625,
      "loss": 4.8679,
      "mean_token_accuracy": 0.23297925740480424,
      "num_tokens": 126940489.0,
      "step": 61165
    },
    {
      "entropy": 5.713829374313354,
      "epoch": 5.5488026124818575,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023921114654112396,
      "loss": 4.8882,
      "mean_token_accuracy": 0.2315928280353546,
      "num_tokens": 126950681.0,
      "step": 61170
    },
    {
      "entropy": 5.773396778106689,
      "epoch": 5.549256168359942,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023917920243191426,
      "loss": 4.8605,
      "mean_token_accuracy": 0.22625950574874878,
      "num_tokens": 126961267.0,
      "step": 61175
    },
    {
      "entropy": 5.861931228637696,
      "epoch": 5.549709724238026,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023914725906348783,
      "loss": 4.9532,
      "mean_token_accuracy": 0.22469299137592316,
      "num_tokens": 126972100.0,
      "step": 61180
    },
    {
      "entropy": 5.7556768417358395,
      "epoch": 5.55016328011611,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023911531643650526,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2294782429933548,
      "num_tokens": 126982806.0,
      "step": 61185
    },
    {
      "entropy": 5.743021869659424,
      "epoch": 5.5506168359941945,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023908337455162698,
      "loss": 4.7664,
      "mean_token_accuracy": 0.24095361232757567,
      "num_tokens": 126992218.0,
      "step": 61190
    },
    {
      "entropy": 5.744309711456299,
      "epoch": 5.551070391872279,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023905143340951373,
      "loss": 4.7692,
      "mean_token_accuracy": 0.23936820775270462,
      "num_tokens": 127002783.0,
      "step": 61195
    },
    {
      "entropy": 5.77211561203003,
      "epoch": 5.551523947750363,
      "grad_norm": 1.25,
      "learning_rate": 0.000239019493010826,
      "loss": 4.9531,
      "mean_token_accuracy": 0.22267627269029616,
      "num_tokens": 127013720.0,
      "step": 61200
    },
    {
      "entropy": 5.812718677520752,
      "epoch": 5.551977503628447,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002389875533562243,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2279920220375061,
      "num_tokens": 127024443.0,
      "step": 61205
    },
    {
      "entropy": 5.7030140399932865,
      "epoch": 5.552431059506532,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023895561444636914,
      "loss": 4.7074,
      "mean_token_accuracy": 0.23774086833000183,
      "num_tokens": 127035565.0,
      "step": 61210
    },
    {
      "entropy": 5.726567173004151,
      "epoch": 5.552884615384615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023892367628192119,
      "loss": 4.773,
      "mean_token_accuracy": 0.24193715304136276,
      "num_tokens": 127044762.0,
      "step": 61215
    },
    {
      "entropy": 5.708119964599609,
      "epoch": 5.553338171262699,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023889173886354072,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23502082526683807,
      "num_tokens": 127054893.0,
      "step": 61220
    },
    {
      "entropy": 5.733253765106201,
      "epoch": 5.5537917271407835,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023885980219188829,
      "loss": 4.8803,
      "mean_token_accuracy": 0.23069918900728226,
      "num_tokens": 127065063.0,
      "step": 61225
    },
    {
      "entropy": 5.739117193222046,
      "epoch": 5.554245283018868,
      "grad_norm": 1.625,
      "learning_rate": 0.00023882786626762436,
      "loss": 4.8408,
      "mean_token_accuracy": 0.22934121787548065,
      "num_tokens": 127075894.0,
      "step": 61230
    },
    {
      "entropy": 5.831407117843628,
      "epoch": 5.554698838896952,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002387959310914094,
      "loss": 4.8898,
      "mean_token_accuracy": 0.234576116502285,
      "num_tokens": 127086299.0,
      "step": 61235
    },
    {
      "entropy": 5.788647651672363,
      "epoch": 5.555152394775036,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023876399666390374,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22396669536828995,
      "num_tokens": 127096913.0,
      "step": 61240
    },
    {
      "entropy": 5.832722234725952,
      "epoch": 5.555605950653121,
      "grad_norm": 1.5,
      "learning_rate": 0.00023873206298576793,
      "loss": 4.959,
      "mean_token_accuracy": 0.2194732591509819,
      "num_tokens": 127107467.0,
      "step": 61245
    },
    {
      "entropy": 5.814874696731567,
      "epoch": 5.556059506531205,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002387001300576623,
      "loss": 4.95,
      "mean_token_accuracy": 0.21685249954462052,
      "num_tokens": 127119121.0,
      "step": 61250
    },
    {
      "entropy": 5.752050542831421,
      "epoch": 5.556513062409289,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023866819788024725,
      "loss": 4.8555,
      "mean_token_accuracy": 0.2263732969760895,
      "num_tokens": 127129305.0,
      "step": 61255
    },
    {
      "entropy": 5.789967679977417,
      "epoch": 5.556966618287373,
      "grad_norm": 1.375,
      "learning_rate": 0.00023863626645418301,
      "loss": 4.8328,
      "mean_token_accuracy": 0.2297664389014244,
      "num_tokens": 127138867.0,
      "step": 61260
    },
    {
      "entropy": 5.81579008102417,
      "epoch": 5.557420174165458,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023860433578013014,
      "loss": 4.8999,
      "mean_token_accuracy": 0.2300606369972229,
      "num_tokens": 127148702.0,
      "step": 61265
    },
    {
      "entropy": 5.7727278709411625,
      "epoch": 5.557873730043541,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002385724058587489,
      "loss": 4.8232,
      "mean_token_accuracy": 0.23840027302503586,
      "num_tokens": 127158533.0,
      "step": 61270
    },
    {
      "entropy": 5.6972719669342045,
      "epoch": 5.558327285921625,
      "grad_norm": 1.375,
      "learning_rate": 0.0002385404766906995,
      "loss": 4.7387,
      "mean_token_accuracy": 0.24634252190589906,
      "num_tokens": 127169115.0,
      "step": 61275
    },
    {
      "entropy": 5.704015302658081,
      "epoch": 5.5587808417997095,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023850854827664242,
      "loss": 4.8414,
      "mean_token_accuracy": 0.23313488364219664,
      "num_tokens": 127180396.0,
      "step": 61280
    },
    {
      "entropy": 5.742135858535766,
      "epoch": 5.559234397677794,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023847662061723786,
      "loss": 4.8447,
      "mean_token_accuracy": 0.23099036365747452,
      "num_tokens": 127191341.0,
      "step": 61285
    },
    {
      "entropy": 5.685102272033691,
      "epoch": 5.559687953555878,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002384446937131461,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24167555272579194,
      "num_tokens": 127201321.0,
      "step": 61290
    },
    {
      "entropy": 5.855008697509765,
      "epoch": 5.560141509433962,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023841276756502735,
      "loss": 4.9387,
      "mean_token_accuracy": 0.22547147870063783,
      "num_tokens": 127211490.0,
      "step": 61295
    },
    {
      "entropy": 5.785648393630981,
      "epoch": 5.560595065312047,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002383808421735419,
      "loss": 4.9147,
      "mean_token_accuracy": 0.22335413545370103,
      "num_tokens": 127221719.0,
      "step": 61300
    },
    {
      "entropy": 5.75805082321167,
      "epoch": 5.561048621190131,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023834891753934996,
      "loss": 4.8858,
      "mean_token_accuracy": 0.2243228331208229,
      "num_tokens": 127232543.0,
      "step": 61305
    },
    {
      "entropy": 5.777977371215821,
      "epoch": 5.561502177068215,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023831699366311173,
      "loss": 4.9055,
      "mean_token_accuracy": 0.22864255905151368,
      "num_tokens": 127242549.0,
      "step": 61310
    },
    {
      "entropy": 5.794892644882202,
      "epoch": 5.5619557329462985,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023828507054548737,
      "loss": 4.8373,
      "mean_token_accuracy": 0.23443833291530608,
      "num_tokens": 127252539.0,
      "step": 61315
    },
    {
      "entropy": 5.808511066436767,
      "epoch": 5.562409288824383,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023825314818713721,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22909856140613555,
      "num_tokens": 127262639.0,
      "step": 61320
    },
    {
      "entropy": 5.756712865829468,
      "epoch": 5.562862844702467,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023822122658872125,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23209603130817413,
      "num_tokens": 127272609.0,
      "step": 61325
    },
    {
      "entropy": 5.749996566772461,
      "epoch": 5.563316400580551,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002381893057508997,
      "loss": 4.9417,
      "mean_token_accuracy": 0.22804958373308182,
      "num_tokens": 127283539.0,
      "step": 61330
    },
    {
      "entropy": 5.806922817230225,
      "epoch": 5.563769956458636,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023815738567433277,
      "loss": 4.9609,
      "mean_token_accuracy": 0.2239489197731018,
      "num_tokens": 127294566.0,
      "step": 61335
    },
    {
      "entropy": 5.725141000747681,
      "epoch": 5.56422351233672,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023812546635968042,
      "loss": 4.7758,
      "mean_token_accuracy": 0.23897470831871032,
      "num_tokens": 127304197.0,
      "step": 61340
    },
    {
      "entropy": 5.7173796653747555,
      "epoch": 5.564677068214804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023809354780760279,
      "loss": 4.7462,
      "mean_token_accuracy": 0.24442499727010727,
      "num_tokens": 127313834.0,
      "step": 61345
    },
    {
      "entropy": 5.704560708999634,
      "epoch": 5.565130624092888,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023806163001876009,
      "loss": 4.8196,
      "mean_token_accuracy": 0.22651606053113937,
      "num_tokens": 127324880.0,
      "step": 61350
    },
    {
      "entropy": 5.706297254562378,
      "epoch": 5.565584179970973,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023802971299381222,
      "loss": 4.8561,
      "mean_token_accuracy": 0.2231311783194542,
      "num_tokens": 127334732.0,
      "step": 61355
    },
    {
      "entropy": 5.70858678817749,
      "epoch": 5.566037735849057,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002379977967334193,
      "loss": 4.7997,
      "mean_token_accuracy": 0.24338916391134263,
      "num_tokens": 127345224.0,
      "step": 61360
    },
    {
      "entropy": 5.779675722122192,
      "epoch": 5.566491291727141,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023796588123824143,
      "loss": 4.8638,
      "mean_token_accuracy": 0.22491228431463242,
      "num_tokens": 127356041.0,
      "step": 61365
    },
    {
      "entropy": 5.73808479309082,
      "epoch": 5.566944847605225,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023793396650893852,
      "loss": 4.821,
      "mean_token_accuracy": 0.2394078940153122,
      "num_tokens": 127365576.0,
      "step": 61370
    },
    {
      "entropy": 5.782843828201294,
      "epoch": 5.567398403483309,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023790205254617063,
      "loss": 4.9605,
      "mean_token_accuracy": 0.21946114003658296,
      "num_tokens": 127375933.0,
      "step": 61375
    },
    {
      "entropy": 5.722122287750244,
      "epoch": 5.567851959361393,
      "grad_norm": 1.546875,
      "learning_rate": 0.00023787013935059776,
      "loss": 4.8897,
      "mean_token_accuracy": 0.2222125381231308,
      "num_tokens": 127386187.0,
      "step": 61380
    },
    {
      "entropy": 5.791303396224976,
      "epoch": 5.568305515239477,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023783822692287988,
      "loss": 4.8992,
      "mean_token_accuracy": 0.22946493923664094,
      "num_tokens": 127395537.0,
      "step": 61385
    },
    {
      "entropy": 5.797808313369751,
      "epoch": 5.568759071117562,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023780631526367697,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2208385184407234,
      "num_tokens": 127405745.0,
      "step": 61390
    },
    {
      "entropy": 5.749021339416504,
      "epoch": 5.569212626995646,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023777440437364888,
      "loss": 4.858,
      "mean_token_accuracy": 0.23107463121414185,
      "num_tokens": 127415430.0,
      "step": 61395
    },
    {
      "entropy": 5.775242328643799,
      "epoch": 5.56966618287373,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023774249425345562,
      "loss": 4.8891,
      "mean_token_accuracy": 0.22671545892953873,
      "num_tokens": 127426058.0,
      "step": 61400
    },
    {
      "entropy": 5.795022487640381,
      "epoch": 5.570119738751814,
      "grad_norm": 1.375,
      "learning_rate": 0.00023771058490375707,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23043540716171265,
      "num_tokens": 127437401.0,
      "step": 61405
    },
    {
      "entropy": 5.796159744262695,
      "epoch": 5.570573294629899,
      "grad_norm": 1.5,
      "learning_rate": 0.00023767867632521313,
      "loss": 4.9724,
      "mean_token_accuracy": 0.21879100650548935,
      "num_tokens": 127446927.0,
      "step": 61410
    },
    {
      "entropy": 5.7296514987945555,
      "epoch": 5.571026850507983,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023764676851848365,
      "loss": 4.8206,
      "mean_token_accuracy": 0.2312238484621048,
      "num_tokens": 127457468.0,
      "step": 61415
    },
    {
      "entropy": 5.778930711746216,
      "epoch": 5.571480406386067,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023761486148422855,
      "loss": 4.8469,
      "mean_token_accuracy": 0.22607595175504686,
      "num_tokens": 127467722.0,
      "step": 61420
    },
    {
      "entropy": 5.776580286026001,
      "epoch": 5.571933962264151,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002375829552231077,
      "loss": 4.8223,
      "mean_token_accuracy": 0.22696424275636673,
      "num_tokens": 127478735.0,
      "step": 61425
    },
    {
      "entropy": 5.76103253364563,
      "epoch": 5.572387518142235,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023755104973578072,
      "loss": 4.8204,
      "mean_token_accuracy": 0.232679483294487,
      "num_tokens": 127489500.0,
      "step": 61430
    },
    {
      "entropy": 5.746566581726074,
      "epoch": 5.572841074020319,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023751914502290772,
      "loss": 4.8499,
      "mean_token_accuracy": 0.22591961473226546,
      "num_tokens": 127499515.0,
      "step": 61435
    },
    {
      "entropy": 5.8134149551391605,
      "epoch": 5.573294629898403,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023748724108514825,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22803219705820083,
      "num_tokens": 127509310.0,
      "step": 61440
    },
    {
      "entropy": 5.8651103496551515,
      "epoch": 5.573748185776488,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002374553379231622,
      "loss": 4.8835,
      "mean_token_accuracy": 0.2281198650598526,
      "num_tokens": 127519413.0,
      "step": 61445
    },
    {
      "entropy": 5.802383852005005,
      "epoch": 5.574201741654572,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023742343553760938,
      "loss": 4.8635,
      "mean_token_accuracy": 0.2286461815237999,
      "num_tokens": 127529100.0,
      "step": 61450
    },
    {
      "entropy": 5.7834783554077145,
      "epoch": 5.574655297532656,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023739153392914947,
      "loss": 4.8634,
      "mean_token_accuracy": 0.23197731226682664,
      "num_tokens": 127539538.0,
      "step": 61455
    },
    {
      "entropy": 5.81930022239685,
      "epoch": 5.5751088534107405,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023735963309844222,
      "loss": 4.8532,
      "mean_token_accuracy": 0.22966714054346085,
      "num_tokens": 127549351.0,
      "step": 61460
    },
    {
      "entropy": 5.819626379013061,
      "epoch": 5.575562409288825,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023732773304614734,
      "loss": 4.9549,
      "mean_token_accuracy": 0.21805561035871507,
      "num_tokens": 127559676.0,
      "step": 61465
    },
    {
      "entropy": 5.784291124343872,
      "epoch": 5.576015965166908,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002372958337729246,
      "loss": 4.8877,
      "mean_token_accuracy": 0.22932808995246887,
      "num_tokens": 127569448.0,
      "step": 61470
    },
    {
      "entropy": 5.785489273071289,
      "epoch": 5.576469521044992,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023726393527943358,
      "loss": 4.8398,
      "mean_token_accuracy": 0.23185816258192063,
      "num_tokens": 127580466.0,
      "step": 61475
    },
    {
      "entropy": 5.723159503936768,
      "epoch": 5.576923076923077,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023723203756633395,
      "loss": 4.8493,
      "mean_token_accuracy": 0.23577948063611984,
      "num_tokens": 127590955.0,
      "step": 61480
    },
    {
      "entropy": 5.6732320308685305,
      "epoch": 5.577376632801161,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023720014063428546,
      "loss": 4.8338,
      "mean_token_accuracy": 0.2328068047761917,
      "num_tokens": 127601410.0,
      "step": 61485
    },
    {
      "entropy": 5.730936813354492,
      "epoch": 5.577830188679245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023716824448394768,
      "loss": 4.785,
      "mean_token_accuracy": 0.23862657994031905,
      "num_tokens": 127611275.0,
      "step": 61490
    },
    {
      "entropy": 5.699485492706299,
      "epoch": 5.5782837445573294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002371363491159803,
      "loss": 4.7679,
      "mean_token_accuracy": 0.23307028263807297,
      "num_tokens": 127621660.0,
      "step": 61495
    },
    {
      "entropy": 5.794744539260864,
      "epoch": 5.578737300435414,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023710445453104274,
      "loss": 4.9499,
      "mean_token_accuracy": 0.21777986735105515,
      "num_tokens": 127631534.0,
      "step": 61500
    },
    {
      "entropy": 5.688975095748901,
      "epoch": 5.579190856313498,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023707256072979483,
      "loss": 4.8288,
      "mean_token_accuracy": 0.2288837656378746,
      "num_tokens": 127641742.0,
      "step": 61505
    },
    {
      "entropy": 5.737190866470337,
      "epoch": 5.579644412191582,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000237040667712896,
      "loss": 4.8782,
      "mean_token_accuracy": 0.2291557163000107,
      "num_tokens": 127653832.0,
      "step": 61510
    },
    {
      "entropy": 5.745658540725708,
      "epoch": 5.5800979680696665,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002370087754810058,
      "loss": 4.8023,
      "mean_token_accuracy": 0.23108193129301072,
      "num_tokens": 127664072.0,
      "step": 61515
    },
    {
      "entropy": 5.723881006240845,
      "epoch": 5.580551523947751,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023697688403478384,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23804794996976852,
      "num_tokens": 127673725.0,
      "step": 61520
    },
    {
      "entropy": 5.7085456371307375,
      "epoch": 5.581005079825834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023694499337488962,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23745090067386626,
      "num_tokens": 127684616.0,
      "step": 61525
    },
    {
      "entropy": 5.749031352996826,
      "epoch": 5.581458635703918,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002369131035019826,
      "loss": 4.9038,
      "mean_token_accuracy": 0.23188718557357788,
      "num_tokens": 127695395.0,
      "step": 61530
    },
    {
      "entropy": 5.731037950515747,
      "epoch": 5.581912191582003,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023688121441672234,
      "loss": 4.8173,
      "mean_token_accuracy": 0.23768060952425002,
      "num_tokens": 127704712.0,
      "step": 61535
    },
    {
      "entropy": 5.719976329803467,
      "epoch": 5.582365747460087,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023684932611976823,
      "loss": 4.8534,
      "mean_token_accuracy": 0.22874655425548554,
      "num_tokens": 127715913.0,
      "step": 61540
    },
    {
      "entropy": 5.745455312728882,
      "epoch": 5.582819303338171,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023681743861177985,
      "loss": 4.8369,
      "mean_token_accuracy": 0.22743297666311263,
      "num_tokens": 127725639.0,
      "step": 61545
    },
    {
      "entropy": 5.698141431808471,
      "epoch": 5.5832728592162555,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023678555189341656,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23017718642950058,
      "num_tokens": 127737285.0,
      "step": 61550
    },
    {
      "entropy": 5.733644151687622,
      "epoch": 5.58372641509434,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002367536659653378,
      "loss": 4.7821,
      "mean_token_accuracy": 0.24150873571634293,
      "num_tokens": 127746492.0,
      "step": 61555
    },
    {
      "entropy": 5.7185605525970455,
      "epoch": 5.584179970972424,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002367217808282029,
      "loss": 4.8267,
      "mean_token_accuracy": 0.23510129600763321,
      "num_tokens": 127756491.0,
      "step": 61560
    },
    {
      "entropy": 5.77636456489563,
      "epoch": 5.584633526850508,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023668989648267135,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22720146775245667,
      "num_tokens": 127767934.0,
      "step": 61565
    },
    {
      "entropy": 5.811403131484985,
      "epoch": 5.585087082728592,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023665801292940253,
      "loss": 4.8753,
      "mean_token_accuracy": 0.23391941636800767,
      "num_tokens": 127778219.0,
      "step": 61570
    },
    {
      "entropy": 5.8351226329803465,
      "epoch": 5.585540638606677,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023662613016905577,
      "loss": 4.8769,
      "mean_token_accuracy": 0.2341575115919113,
      "num_tokens": 127787484.0,
      "step": 61575
    },
    {
      "entropy": 5.7166603088378904,
      "epoch": 5.58599419448476,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023659424820229042,
      "loss": 4.7429,
      "mean_token_accuracy": 0.23953025341033934,
      "num_tokens": 127797409.0,
      "step": 61580
    },
    {
      "entropy": 5.723410034179688,
      "epoch": 5.5864477503628445,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002365623670297658,
      "loss": 4.7766,
      "mean_token_accuracy": 0.23664687871932982,
      "num_tokens": 127807919.0,
      "step": 61585
    },
    {
      "entropy": 5.744383049011231,
      "epoch": 5.586901306240929,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0002365304866521412,
      "loss": 4.8627,
      "mean_token_accuracy": 0.22784863114356996,
      "num_tokens": 127817225.0,
      "step": 61590
    },
    {
      "entropy": 5.695826053619385,
      "epoch": 5.587354862119013,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023649860707007593,
      "loss": 4.7965,
      "mean_token_accuracy": 0.24038872569799424,
      "num_tokens": 127827287.0,
      "step": 61595
    },
    {
      "entropy": 5.7118239402771,
      "epoch": 5.587808417997097,
      "grad_norm": 1.375,
      "learning_rate": 0.00023646672828422932,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23238642066717147,
      "num_tokens": 127838264.0,
      "step": 61600
    },
    {
      "entropy": 5.779437446594239,
      "epoch": 5.5882619738751815,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023643485029526057,
      "loss": 4.8924,
      "mean_token_accuracy": 0.22994166761636733,
      "num_tokens": 127848214.0,
      "step": 61605
    },
    {
      "entropy": 5.79988956451416,
      "epoch": 5.588715529753266,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002364029731038289,
      "loss": 4.8413,
      "mean_token_accuracy": 0.23354928344488143,
      "num_tokens": 127857912.0,
      "step": 61610
    },
    {
      "entropy": 5.825675821304321,
      "epoch": 5.58916908563135,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002363710967105937,
      "loss": 4.88,
      "mean_token_accuracy": 0.22779999673366547,
      "num_tokens": 127867749.0,
      "step": 61615
    },
    {
      "entropy": 5.793728160858154,
      "epoch": 5.589622641509434,
      "grad_norm": 1.375,
      "learning_rate": 0.00023633922111621398,
      "loss": 4.9324,
      "mean_token_accuracy": 0.22559473067522048,
      "num_tokens": 127878410.0,
      "step": 61620
    },
    {
      "entropy": 5.8055832862854,
      "epoch": 5.590076197387518,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023630734632134898,
      "loss": 4.9329,
      "mean_token_accuracy": 0.22311871945858003,
      "num_tokens": 127888375.0,
      "step": 61625
    },
    {
      "entropy": 5.765535306930542,
      "epoch": 5.590529753265602,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023627547232665802,
      "loss": 4.869,
      "mean_token_accuracy": 0.22577641159296036,
      "num_tokens": 127898010.0,
      "step": 61630
    },
    {
      "entropy": 5.7207982540130615,
      "epoch": 5.590983309143686,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002362435991328001,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2368711531162262,
      "num_tokens": 127908780.0,
      "step": 61635
    },
    {
      "entropy": 5.765975999832153,
      "epoch": 5.5914368650217705,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023621172674043444,
      "loss": 4.8884,
      "mean_token_accuracy": 0.23134151399135588,
      "num_tokens": 127918897.0,
      "step": 61640
    },
    {
      "entropy": 5.733240222930908,
      "epoch": 5.591890420899855,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002361798551502201,
      "loss": 4.8223,
      "mean_token_accuracy": 0.23011681139469148,
      "num_tokens": 127929687.0,
      "step": 61645
    },
    {
      "entropy": 5.789556455612183,
      "epoch": 5.592343976777939,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002361479843628163,
      "loss": 4.8638,
      "mean_token_accuracy": 0.23096280992031099,
      "num_tokens": 127940676.0,
      "step": 61650
    },
    {
      "entropy": 5.759036874771118,
      "epoch": 5.592797532656023,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002361161143788821,
      "loss": 4.8271,
      "mean_token_accuracy": 0.23178229182958604,
      "num_tokens": 127950891.0,
      "step": 61655
    },
    {
      "entropy": 5.738532447814942,
      "epoch": 5.593251088534108,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002360842451990765,
      "loss": 4.8116,
      "mean_token_accuracy": 0.24147382080554963,
      "num_tokens": 127961871.0,
      "step": 61660
    },
    {
      "entropy": 5.746682453155517,
      "epoch": 5.593704644412192,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023605237682405871,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2329203486442566,
      "num_tokens": 127970602.0,
      "step": 61665
    },
    {
      "entropy": 5.81123538017273,
      "epoch": 5.594158200290276,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023602050925448765,
      "loss": 4.9524,
      "mean_token_accuracy": 0.21450824290513992,
      "num_tokens": 127981672.0,
      "step": 61670
    },
    {
      "entropy": 5.739222478866577,
      "epoch": 5.59461175616836,
      "grad_norm": 1.515625,
      "learning_rate": 0.00023598864249102243,
      "loss": 4.828,
      "mean_token_accuracy": 0.2341065898537636,
      "num_tokens": 127990995.0,
      "step": 61675
    },
    {
      "entropy": 5.762585067749024,
      "epoch": 5.595065312046444,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023595677653432195,
      "loss": 4.8183,
      "mean_token_accuracy": 0.23296509683132172,
      "num_tokens": 128000490.0,
      "step": 61680
    },
    {
      "entropy": 5.799787378311157,
      "epoch": 5.595518867924528,
      "grad_norm": 1.375,
      "learning_rate": 0.00023592491138504535,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22751777470111847,
      "num_tokens": 128011257.0,
      "step": 61685
    },
    {
      "entropy": 5.777444553375244,
      "epoch": 5.595972423802612,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002358930470438515,
      "loss": 4.9043,
      "mean_token_accuracy": 0.2234334498643875,
      "num_tokens": 128021297.0,
      "step": 61690
    },
    {
      "entropy": 5.790564727783203,
      "epoch": 5.5964259796806965,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002358611835113994,
      "loss": 4.9278,
      "mean_token_accuracy": 0.22663778215646743,
      "num_tokens": 128031170.0,
      "step": 61695
    },
    {
      "entropy": 5.724302577972412,
      "epoch": 5.596879535558781,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023582932078834801,
      "loss": 4.7555,
      "mean_token_accuracy": 0.23965077549219133,
      "num_tokens": 128040845.0,
      "step": 61700
    },
    {
      "entropy": 5.713896322250366,
      "epoch": 5.597333091436865,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023579745887535624,
      "loss": 4.8472,
      "mean_token_accuracy": 0.23068763315677643,
      "num_tokens": 128051082.0,
      "step": 61705
    },
    {
      "entropy": 5.800951910018921,
      "epoch": 5.597786647314949,
      "grad_norm": 1.234375,
      "learning_rate": 0.00023576559777308304,
      "loss": 4.8772,
      "mean_token_accuracy": 0.22767064422369004,
      "num_tokens": 128062611.0,
      "step": 61710
    },
    {
      "entropy": 5.793698358535766,
      "epoch": 5.598240203193034,
      "grad_norm": 1.640625,
      "learning_rate": 0.00023573373748218723,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2390742227435112,
      "num_tokens": 128071608.0,
      "step": 61715
    },
    {
      "entropy": 5.7374755859375,
      "epoch": 5.598693759071118,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002357018780033278,
      "loss": 4.8265,
      "mean_token_accuracy": 0.22839527875185012,
      "num_tokens": 128081912.0,
      "step": 61720
    },
    {
      "entropy": 5.801742458343506,
      "epoch": 5.599147314949201,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023567001933716354,
      "loss": 4.866,
      "mean_token_accuracy": 0.2309807151556015,
      "num_tokens": 128092015.0,
      "step": 61725
    },
    {
      "entropy": 5.778424644470215,
      "epoch": 5.599600870827286,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023563816148435324,
      "loss": 4.7849,
      "mean_token_accuracy": 0.24123229533433915,
      "num_tokens": 128102148.0,
      "step": 61730
    },
    {
      "entropy": 5.736553001403808,
      "epoch": 5.60005442670537,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002356063044455558,
      "loss": 4.9008,
      "mean_token_accuracy": 0.23138103932142257,
      "num_tokens": 128112163.0,
      "step": 61735
    },
    {
      "entropy": 5.770172452926635,
      "epoch": 5.600507982583454,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023557444822143008,
      "loss": 4.8633,
      "mean_token_accuracy": 0.22967558205127717,
      "num_tokens": 128123059.0,
      "step": 61740
    },
    {
      "entropy": 5.708918762207031,
      "epoch": 5.600961538461538,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023554259281263474,
      "loss": 4.8285,
      "mean_token_accuracy": 0.22718839347362518,
      "num_tokens": 128133171.0,
      "step": 61745
    },
    {
      "entropy": 5.736393165588379,
      "epoch": 5.601415094339623,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023551073821982872,
      "loss": 4.8511,
      "mean_token_accuracy": 0.228737673163414,
      "num_tokens": 128143240.0,
      "step": 61750
    },
    {
      "entropy": 5.797930908203125,
      "epoch": 5.601868650217707,
      "grad_norm": 1.296875,
      "learning_rate": 0.00023547888444367066,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22946744561195373,
      "num_tokens": 128154434.0,
      "step": 61755
    },
    {
      "entropy": 5.776775789260864,
      "epoch": 5.602322206095791,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023544703148481932,
      "loss": 4.8458,
      "mean_token_accuracy": 0.23529551774263383,
      "num_tokens": 128164684.0,
      "step": 61760
    },
    {
      "entropy": 5.785068321228027,
      "epoch": 5.602775761973875,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00023541517934393347,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23395693004131318,
      "num_tokens": 128174118.0,
      "step": 61765
    },
    {
      "entropy": 5.672942018508911,
      "epoch": 5.60322931785196,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002353833280216718,
      "loss": 4.8122,
      "mean_token_accuracy": 0.2369011178612709,
      "num_tokens": 128184780.0,
      "step": 61770
    },
    {
      "entropy": 5.690045022964478,
      "epoch": 5.603682873730044,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000235351477518693,
      "loss": 4.7223,
      "mean_token_accuracy": 0.23485401570796965,
      "num_tokens": 128194178.0,
      "step": 61775
    },
    {
      "entropy": 5.63191237449646,
      "epoch": 5.604136429608127,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002353196278356557,
      "loss": 4.7344,
      "mean_token_accuracy": 0.24055743515491484,
      "num_tokens": 128205817.0,
      "step": 61780
    },
    {
      "entropy": 5.729196214675904,
      "epoch": 5.604589985486212,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023528777897321868,
      "loss": 4.7841,
      "mean_token_accuracy": 0.23955894857645035,
      "num_tokens": 128215753.0,
      "step": 61785
    },
    {
      "entropy": 5.724500560760498,
      "epoch": 5.605043541364296,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023525593093204052,
      "loss": 4.8069,
      "mean_token_accuracy": 0.23820528239011765,
      "num_tokens": 128225763.0,
      "step": 61790
    },
    {
      "entropy": 5.870219135284424,
      "epoch": 5.60549709724238,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023522408371277976,
      "loss": 5.0131,
      "mean_token_accuracy": 0.2226038694381714,
      "num_tokens": 128237185.0,
      "step": 61795
    },
    {
      "entropy": 5.708606624603272,
      "epoch": 5.605950653120464,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023519223731609511,
      "loss": 4.8336,
      "mean_token_accuracy": 0.2413829118013382,
      "num_tokens": 128247578.0,
      "step": 61800
    },
    {
      "entropy": 5.6824699401855465,
      "epoch": 5.606404208998549,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023516039174264515,
      "loss": 4.7288,
      "mean_token_accuracy": 0.23450824171304702,
      "num_tokens": 128258391.0,
      "step": 61805
    },
    {
      "entropy": 5.727898645401001,
      "epoch": 5.606857764876633,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023512854699308844,
      "loss": 4.8422,
      "mean_token_accuracy": 0.2285187840461731,
      "num_tokens": 128269498.0,
      "step": 61810
    },
    {
      "entropy": 5.759931802749634,
      "epoch": 5.607311320754717,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023509670306808356,
      "loss": 4.8513,
      "mean_token_accuracy": 0.2327726364135742,
      "num_tokens": 128279544.0,
      "step": 61815
    },
    {
      "entropy": 5.68633542060852,
      "epoch": 5.607764876632801,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023506485996828901,
      "loss": 4.7437,
      "mean_token_accuracy": 0.24626922309398652,
      "num_tokens": 128290376.0,
      "step": 61820
    },
    {
      "entropy": 5.669070911407471,
      "epoch": 5.608218432510886,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023503301769436336,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23303680270910263,
      "num_tokens": 128300598.0,
      "step": 61825
    },
    {
      "entropy": 5.72937421798706,
      "epoch": 5.60867198838897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023500117624696497,
      "loss": 4.8358,
      "mean_token_accuracy": 0.23530269414186478,
      "num_tokens": 128310866.0,
      "step": 61830
    },
    {
      "entropy": 5.7252332210540775,
      "epoch": 5.609125544267053,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002349693356267526,
      "loss": 4.8813,
      "mean_token_accuracy": 0.226187863945961,
      "num_tokens": 128321666.0,
      "step": 61835
    },
    {
      "entropy": 5.744535493850708,
      "epoch": 5.609579100145138,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023493749583438451,
      "loss": 4.778,
      "mean_token_accuracy": 0.24024010747671126,
      "num_tokens": 128331457.0,
      "step": 61840
    },
    {
      "entropy": 5.760304641723633,
      "epoch": 5.610032656023222,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023490565687051923,
      "loss": 4.8277,
      "mean_token_accuracy": 0.22812644988298417,
      "num_tokens": 128342796.0,
      "step": 61845
    },
    {
      "entropy": 5.704354238510132,
      "epoch": 5.610486211901306,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002348738187358151,
      "loss": 4.8751,
      "mean_token_accuracy": 0.2332711100578308,
      "num_tokens": 128353758.0,
      "step": 61850
    },
    {
      "entropy": 5.765225219726562,
      "epoch": 5.61093976777939,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023484198143093073,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23289713859558106,
      "num_tokens": 128364205.0,
      "step": 61855
    },
    {
      "entropy": 5.792587566375732,
      "epoch": 5.611393323657475,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023481014495652441,
      "loss": 4.941,
      "mean_token_accuracy": 0.22003959119319916,
      "num_tokens": 128375632.0,
      "step": 61860
    },
    {
      "entropy": 5.839235782623291,
      "epoch": 5.611846879535559,
      "grad_norm": 1.59375,
      "learning_rate": 0.00023477830931325445,
      "loss": 4.8774,
      "mean_token_accuracy": 0.23566273748874664,
      "num_tokens": 128384605.0,
      "step": 61865
    },
    {
      "entropy": 5.727094745635986,
      "epoch": 5.612300435413643,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023474647450177937,
      "loss": 4.761,
      "mean_token_accuracy": 0.2422578066587448,
      "num_tokens": 128394026.0,
      "step": 61870
    },
    {
      "entropy": 5.660543489456177,
      "epoch": 5.6127539912917275,
      "grad_norm": 1.625,
      "learning_rate": 0.0002347146405227575,
      "loss": 4.7738,
      "mean_token_accuracy": 0.2390811800956726,
      "num_tokens": 128403791.0,
      "step": 61875
    },
    {
      "entropy": 5.710608291625976,
      "epoch": 5.613207547169811,
      "grad_norm": 1.578125,
      "learning_rate": 0.00023468280737684705,
      "loss": 4.8649,
      "mean_token_accuracy": 0.23094376921653748,
      "num_tokens": 128414313.0,
      "step": 61880
    },
    {
      "entropy": 5.762955570220948,
      "epoch": 5.613661103047895,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002346509750647064,
      "loss": 4.8655,
      "mean_token_accuracy": 0.2371579110622406,
      "num_tokens": 128424619.0,
      "step": 61885
    },
    {
      "entropy": 5.700330781936645,
      "epoch": 5.614114658925979,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023461914358699387,
      "loss": 4.8094,
      "mean_token_accuracy": 0.2385867491364479,
      "num_tokens": 128433691.0,
      "step": 61890
    },
    {
      "entropy": 5.808100938796997,
      "epoch": 5.614568214804064,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023458731294436787,
      "loss": 4.892,
      "mean_token_accuracy": 0.22867459207773208,
      "num_tokens": 128443730.0,
      "step": 61895
    },
    {
      "entropy": 5.738594484329224,
      "epoch": 5.615021770682148,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023455548313748638,
      "loss": 4.7725,
      "mean_token_accuracy": 0.23966053873300552,
      "num_tokens": 128454293.0,
      "step": 61900
    },
    {
      "entropy": 5.740762615203858,
      "epoch": 5.615475326560232,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023452365416700788,
      "loss": 4.8674,
      "mean_token_accuracy": 0.2324304088950157,
      "num_tokens": 128465025.0,
      "step": 61905
    },
    {
      "entropy": 5.749097919464111,
      "epoch": 5.6159288824383164,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023449182603359054,
      "loss": 4.8845,
      "mean_token_accuracy": 0.22577962577342986,
      "num_tokens": 128476763.0,
      "step": 61910
    },
    {
      "entropy": 5.699400806427002,
      "epoch": 5.616382438316401,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023445999873789248,
      "loss": 4.7843,
      "mean_token_accuracy": 0.23027238696813584,
      "num_tokens": 128486593.0,
      "step": 61915
    },
    {
      "entropy": 5.734932804107666,
      "epoch": 5.616835994194485,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023442817228057206,
      "loss": 4.8929,
      "mean_token_accuracy": 0.23248597532510756,
      "num_tokens": 128497733.0,
      "step": 61920
    },
    {
      "entropy": 5.7753386974334715,
      "epoch": 5.617289550072569,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002343963466622874,
      "loss": 4.8437,
      "mean_token_accuracy": 0.2356115460395813,
      "num_tokens": 128507657.0,
      "step": 61925
    },
    {
      "entropy": 5.768081378936768,
      "epoch": 5.6177431059506535,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023436452188369656,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23764755725860595,
      "num_tokens": 128517803.0,
      "step": 61930
    },
    {
      "entropy": 5.751744651794434,
      "epoch": 5.618196661828737,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023433269794545774,
      "loss": 4.8777,
      "mean_token_accuracy": 0.22998761683702468,
      "num_tokens": 128527612.0,
      "step": 61935
    },
    {
      "entropy": 5.767443227767944,
      "epoch": 5.618650217706821,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023430087484822915,
      "loss": 4.8748,
      "mean_token_accuracy": 0.23472413569688796,
      "num_tokens": 128538527.0,
      "step": 61940
    },
    {
      "entropy": 5.7711926937103275,
      "epoch": 5.619103773584905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023426905259266883,
      "loss": 4.9114,
      "mean_token_accuracy": 0.23102874755859376,
      "num_tokens": 128549659.0,
      "step": 61945
    },
    {
      "entropy": 5.811004495620727,
      "epoch": 5.61955732946299,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023423723117943485,
      "loss": 4.9343,
      "mean_token_accuracy": 0.22814526855945588,
      "num_tokens": 128560605.0,
      "step": 61950
    },
    {
      "entropy": 5.812346506118774,
      "epoch": 5.620010885341074,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023420541060918538,
      "loss": 4.8326,
      "mean_token_accuracy": 0.24311988204717636,
      "num_tokens": 128571092.0,
      "step": 61955
    },
    {
      "entropy": 5.71406135559082,
      "epoch": 5.620464441219158,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002341735908825784,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23582851886749268,
      "num_tokens": 128582183.0,
      "step": 61960
    },
    {
      "entropy": 5.754461145401001,
      "epoch": 5.6209179970972425,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023414177200027197,
      "loss": 4.9122,
      "mean_token_accuracy": 0.2260160341858864,
      "num_tokens": 128593704.0,
      "step": 61965
    },
    {
      "entropy": 5.740417003631592,
      "epoch": 5.621371552975327,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023410995396292407,
      "loss": 4.8338,
      "mean_token_accuracy": 0.23093851506710053,
      "num_tokens": 128603878.0,
      "step": 61970
    },
    {
      "entropy": 5.7184412479400635,
      "epoch": 5.621825108853411,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023407813677119282,
      "loss": 4.7903,
      "mean_token_accuracy": 0.2360624611377716,
      "num_tokens": 128614609.0,
      "step": 61975
    },
    {
      "entropy": 5.762820816040039,
      "epoch": 5.622278664731495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023404632042573604,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2332321122288704,
      "num_tokens": 128624806.0,
      "step": 61980
    },
    {
      "entropy": 5.739219999313354,
      "epoch": 5.62273222060958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023401450492721183,
      "loss": 4.8159,
      "mean_token_accuracy": 0.23640191704034805,
      "num_tokens": 128635939.0,
      "step": 61985
    },
    {
      "entropy": 5.83322958946228,
      "epoch": 5.623185776487663,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002339826902762781,
      "loss": 4.9136,
      "mean_token_accuracy": 0.21985674053430557,
      "num_tokens": 128646404.0,
      "step": 61990
    },
    {
      "entropy": 5.744362020492554,
      "epoch": 5.623639332365747,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023395087647359275,
      "loss": 4.8071,
      "mean_token_accuracy": 0.23409707993268966,
      "num_tokens": 128656426.0,
      "step": 61995
    },
    {
      "entropy": 5.7766337394714355,
      "epoch": 5.6240928882438315,
      "grad_norm": 1.375,
      "learning_rate": 0.00023391906351981378,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22400241941213608,
      "num_tokens": 128665854.0,
      "step": 62000
    },
    {
      "entropy": 5.648719358444214,
      "epoch": 5.624546444121916,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000233887251415599,
      "loss": 4.7349,
      "mean_token_accuracy": 0.23522628098726273,
      "num_tokens": 128676006.0,
      "step": 62005
    },
    {
      "entropy": 5.702200508117675,
      "epoch": 5.625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023385544016160637,
      "loss": 4.7551,
      "mean_token_accuracy": 0.2315069854259491,
      "num_tokens": 128686108.0,
      "step": 62010
    },
    {
      "entropy": 5.740422248840332,
      "epoch": 5.625453555878084,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002338236297584937,
      "loss": 4.8693,
      "mean_token_accuracy": 0.22856408804655076,
      "num_tokens": 128695656.0,
      "step": 62015
    },
    {
      "entropy": 5.737794685363769,
      "epoch": 5.6259071117561685,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023379182020691892,
      "loss": 4.8843,
      "mean_token_accuracy": 0.22793468683958054,
      "num_tokens": 128705414.0,
      "step": 62020
    },
    {
      "entropy": 5.7704675674438475,
      "epoch": 5.626360667634253,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023376001150753972,
      "loss": 4.8347,
      "mean_token_accuracy": 0.23125788569450378,
      "num_tokens": 128716532.0,
      "step": 62025
    },
    {
      "entropy": 5.714646244049073,
      "epoch": 5.626814223512337,
      "grad_norm": 1.34375,
      "learning_rate": 0.000233728203661014,
      "loss": 4.8581,
      "mean_token_accuracy": 0.22654867470264434,
      "num_tokens": 128727156.0,
      "step": 62030
    },
    {
      "entropy": 5.8132239818573,
      "epoch": 5.62726777939042,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002336963966679996,
      "loss": 4.8553,
      "mean_token_accuracy": 0.22314040809869767,
      "num_tokens": 128737160.0,
      "step": 62035
    },
    {
      "entropy": 5.726498222351074,
      "epoch": 5.627721335268505,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023366459052915424,
      "loss": 4.8275,
      "mean_token_accuracy": 0.22661434561014177,
      "num_tokens": 128748657.0,
      "step": 62040
    },
    {
      "entropy": 5.734697341918945,
      "epoch": 5.628174891146589,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002336327852451356,
      "loss": 4.7988,
      "mean_token_accuracy": 0.23656877279281616,
      "num_tokens": 128759200.0,
      "step": 62045
    },
    {
      "entropy": 5.730139446258545,
      "epoch": 5.628628447024673,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002336009808166016,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2315222978591919,
      "num_tokens": 128769351.0,
      "step": 62050
    },
    {
      "entropy": 5.8052661418914795,
      "epoch": 5.6290820029027575,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023356917724420983,
      "loss": 4.9211,
      "mean_token_accuracy": 0.22721150666475295,
      "num_tokens": 128779540.0,
      "step": 62055
    },
    {
      "entropy": 5.832570362091064,
      "epoch": 5.629535558780842,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000233537374528618,
      "loss": 4.9326,
      "mean_token_accuracy": 0.21626625061035157,
      "num_tokens": 128790159.0,
      "step": 62060
    },
    {
      "entropy": 5.819000577926635,
      "epoch": 5.629989114658926,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00023350557267048382,
      "loss": 4.9194,
      "mean_token_accuracy": 0.22146227955818176,
      "num_tokens": 128801679.0,
      "step": 62065
    },
    {
      "entropy": 5.788735485076904,
      "epoch": 5.63044267053701,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023347377167046501,
      "loss": 4.8818,
      "mean_token_accuracy": 0.22808743864297867,
      "num_tokens": 128812511.0,
      "step": 62070
    },
    {
      "entropy": 5.697423505783081,
      "epoch": 5.630896226415095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023344197152921925,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23785140067338945,
      "num_tokens": 128822703.0,
      "step": 62075
    },
    {
      "entropy": 5.743999099731445,
      "epoch": 5.631349782293179,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000233410172247404,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23071907013654708,
      "num_tokens": 128832429.0,
      "step": 62080
    },
    {
      "entropy": 5.770981168746948,
      "epoch": 5.631803338171263,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023337837382567705,
      "loss": 4.8653,
      "mean_token_accuracy": 0.22735790461301802,
      "num_tokens": 128843429.0,
      "step": 62085
    },
    {
      "entropy": 5.761844062805176,
      "epoch": 5.6322568940493465,
      "grad_norm": 1.5,
      "learning_rate": 0.0002333465762646959,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23691022843122483,
      "num_tokens": 128852967.0,
      "step": 62090
    },
    {
      "entropy": 5.822368717193603,
      "epoch": 5.632710449927431,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002333147795651182,
      "loss": 4.9399,
      "mean_token_accuracy": 0.22434593886137008,
      "num_tokens": 128863811.0,
      "step": 62095
    },
    {
      "entropy": 5.7554412364959715,
      "epoch": 5.633164005805515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023328298372760143,
      "loss": 4.8115,
      "mean_token_accuracy": 0.23267441540956496,
      "num_tokens": 128874377.0,
      "step": 62100
    },
    {
      "entropy": 5.770128202438355,
      "epoch": 5.633617561683599,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002332511887528032,
      "loss": 4.8071,
      "mean_token_accuracy": 0.22740649580955505,
      "num_tokens": 128884355.0,
      "step": 62105
    },
    {
      "entropy": 5.760368967056275,
      "epoch": 5.6340711175616836,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002332193946413811,
      "loss": 4.8606,
      "mean_token_accuracy": 0.2351283073425293,
      "num_tokens": 128894385.0,
      "step": 62110
    },
    {
      "entropy": 5.841277742385865,
      "epoch": 5.634524673439768,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023318760139399248,
      "loss": 4.9719,
      "mean_token_accuracy": 0.2220888093113899,
      "num_tokens": 128906414.0,
      "step": 62115
    },
    {
      "entropy": 5.792251873016357,
      "epoch": 5.634978229317852,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023315580901129497,
      "loss": 4.8767,
      "mean_token_accuracy": 0.23493178933858871,
      "num_tokens": 128916302.0,
      "step": 62120
    },
    {
      "entropy": 5.794486045837402,
      "epoch": 5.635431785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023312401749394596,
      "loss": 4.876,
      "mean_token_accuracy": 0.22721494883298873,
      "num_tokens": 128926366.0,
      "step": 62125
    },
    {
      "entropy": 5.704292821884155,
      "epoch": 5.635885341074021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002330922268426029,
      "loss": 4.8035,
      "mean_token_accuracy": 0.2343246281147003,
      "num_tokens": 128936716.0,
      "step": 62130
    },
    {
      "entropy": 5.768261432647705,
      "epoch": 5.636338896952105,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00023306043705792335,
      "loss": 4.8369,
      "mean_token_accuracy": 0.23090745955705644,
      "num_tokens": 128948802.0,
      "step": 62135
    },
    {
      "entropy": 5.778226661682129,
      "epoch": 5.636792452830189,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023302864814056463,
      "loss": 4.8876,
      "mean_token_accuracy": 0.22487218230962752,
      "num_tokens": 128959914.0,
      "step": 62140
    },
    {
      "entropy": 5.757935857772827,
      "epoch": 5.6372460087082725,
      "grad_norm": 1.34375,
      "learning_rate": 0.00023299686009118414,
      "loss": 4.8472,
      "mean_token_accuracy": 0.23101726770401002,
      "num_tokens": 128970666.0,
      "step": 62145
    },
    {
      "entropy": 5.730387401580811,
      "epoch": 5.637699564586357,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023296507291043929,
      "loss": 4.8317,
      "mean_token_accuracy": 0.22856396138668061,
      "num_tokens": 128981302.0,
      "step": 62150
    },
    {
      "entropy": 5.771584320068359,
      "epoch": 5.638153120464441,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002329332865989875,
      "loss": 4.8704,
      "mean_token_accuracy": 0.2290557086467743,
      "num_tokens": 128991242.0,
      "step": 62155
    },
    {
      "entropy": 5.790273666381836,
      "epoch": 5.638606676342525,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000232901501157486,
      "loss": 4.8517,
      "mean_token_accuracy": 0.2361661449074745,
      "num_tokens": 129000778.0,
      "step": 62160
    },
    {
      "entropy": 5.778160810470581,
      "epoch": 5.63906023222061,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023286971658659217,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2291703924536705,
      "num_tokens": 129011799.0,
      "step": 62165
    },
    {
      "entropy": 5.755846166610718,
      "epoch": 5.639513788098694,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023283793288696338,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23567822724580764,
      "num_tokens": 129022038.0,
      "step": 62170
    },
    {
      "entropy": 5.75643744468689,
      "epoch": 5.639967343976778,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023280615005925688,
      "loss": 4.8433,
      "mean_token_accuracy": 0.22712468802928926,
      "num_tokens": 129032337.0,
      "step": 62175
    },
    {
      "entropy": 5.790838623046875,
      "epoch": 5.640420899854862,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023277436810412993,
      "loss": 4.9065,
      "mean_token_accuracy": 0.22772819846868514,
      "num_tokens": 129042391.0,
      "step": 62180
    },
    {
      "entropy": 5.700743103027344,
      "epoch": 5.640874455732947,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023274258702223977,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2243838906288147,
      "num_tokens": 129052872.0,
      "step": 62185
    },
    {
      "entropy": 5.755791473388672,
      "epoch": 5.64132801161103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002327108068142438,
      "loss": 4.8587,
      "mean_token_accuracy": 0.22805603593587875,
      "num_tokens": 129063483.0,
      "step": 62190
    },
    {
      "entropy": 5.770305013656616,
      "epoch": 5.641781567489114,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000232679027480799,
      "loss": 4.8032,
      "mean_token_accuracy": 0.2332320049405098,
      "num_tokens": 129075357.0,
      "step": 62195
    },
    {
      "entropy": 5.7982463359832765,
      "epoch": 5.642235123367199,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023264724902256274,
      "loss": 4.8904,
      "mean_token_accuracy": 0.22883277982473374,
      "num_tokens": 129086424.0,
      "step": 62200
    },
    {
      "entropy": 5.7036731243133545,
      "epoch": 5.642688679245283,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023261547144019212,
      "loss": 4.7582,
      "mean_token_accuracy": 0.233970907330513,
      "num_tokens": 129096350.0,
      "step": 62205
    },
    {
      "entropy": 5.741555070877075,
      "epoch": 5.643142235123367,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00023258369473434443,
      "loss": 4.8193,
      "mean_token_accuracy": 0.2354196861386299,
      "num_tokens": 129106858.0,
      "step": 62210
    },
    {
      "entropy": 5.83533992767334,
      "epoch": 5.643595791001451,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023255191890567668,
      "loss": 4.9114,
      "mean_token_accuracy": 0.2269252508878708,
      "num_tokens": 129116545.0,
      "step": 62215
    },
    {
      "entropy": 5.77319221496582,
      "epoch": 5.644049346879536,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023252014395484612,
      "loss": 4.8494,
      "mean_token_accuracy": 0.23472396284341812,
      "num_tokens": 129126053.0,
      "step": 62220
    },
    {
      "entropy": 5.775715303421021,
      "epoch": 5.64450290275762,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023248836988250975,
      "loss": 4.8947,
      "mean_token_accuracy": 0.2218783304095268,
      "num_tokens": 129136297.0,
      "step": 62225
    },
    {
      "entropy": 5.726774406433106,
      "epoch": 5.644956458635704,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002324565966893248,
      "loss": 4.8138,
      "mean_token_accuracy": 0.23540071099996568,
      "num_tokens": 129146809.0,
      "step": 62230
    },
    {
      "entropy": 5.754907131195068,
      "epoch": 5.645410014513788,
      "grad_norm": 1.453125,
      "learning_rate": 0.00023242482437594818,
      "loss": 4.9216,
      "mean_token_accuracy": 0.226126067340374,
      "num_tokens": 129157682.0,
      "step": 62235
    },
    {
      "entropy": 5.736856889724732,
      "epoch": 5.645863570391873,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023239305294303714,
      "loss": 4.8787,
      "mean_token_accuracy": 0.2231192857027054,
      "num_tokens": 129167637.0,
      "step": 62240
    },
    {
      "entropy": 5.710333395004272,
      "epoch": 5.646317126269956,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023236128239124853,
      "loss": 4.7288,
      "mean_token_accuracy": 0.2334202140569687,
      "num_tokens": 129177449.0,
      "step": 62245
    },
    {
      "entropy": 5.713738822937012,
      "epoch": 5.64677068214804,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023232951272123952,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23146334886550904,
      "num_tokens": 129186740.0,
      "step": 62250
    },
    {
      "entropy": 5.784978628158569,
      "epoch": 5.647224238026125,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023229774393366704,
      "loss": 4.9163,
      "mean_token_accuracy": 0.2261568933725357,
      "num_tokens": 129197371.0,
      "step": 62255
    },
    {
      "entropy": 5.7687629699707035,
      "epoch": 5.647677793904209,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023226597602918813,
      "loss": 4.9226,
      "mean_token_accuracy": 0.22695029377937317,
      "num_tokens": 129208326.0,
      "step": 62260
    },
    {
      "entropy": 5.735763597488403,
      "epoch": 5.648131349782293,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002322342090084597,
      "loss": 4.8604,
      "mean_token_accuracy": 0.22571678459644318,
      "num_tokens": 129219434.0,
      "step": 62265
    },
    {
      "entropy": 5.692404413223267,
      "epoch": 5.648584905660377,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023220244287213866,
      "loss": 4.7892,
      "mean_token_accuracy": 0.2352328807115555,
      "num_tokens": 129230582.0,
      "step": 62270
    },
    {
      "entropy": 5.715869569778443,
      "epoch": 5.649038461538462,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023217067762088212,
      "loss": 4.7512,
      "mean_token_accuracy": 0.24195851236581803,
      "num_tokens": 129240476.0,
      "step": 62275
    },
    {
      "entropy": 5.716146183013916,
      "epoch": 5.649492017416546,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023213891325534687,
      "loss": 4.8015,
      "mean_token_accuracy": 0.23634894341230392,
      "num_tokens": 129250566.0,
      "step": 62280
    },
    {
      "entropy": 5.738188171386719,
      "epoch": 5.64994557329463,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002321071497761897,
      "loss": 4.9205,
      "mean_token_accuracy": 0.22927965372800826,
      "num_tokens": 129261173.0,
      "step": 62285
    },
    {
      "entropy": 5.821591997146607,
      "epoch": 5.650399129172714,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023207538718406773,
      "loss": 4.9518,
      "mean_token_accuracy": 0.230065056681633,
      "num_tokens": 129271775.0,
      "step": 62290
    },
    {
      "entropy": 5.84927134513855,
      "epoch": 5.650852685050799,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00023204362547963765,
      "loss": 4.9505,
      "mean_token_accuracy": 0.22251350581645965,
      "num_tokens": 129282203.0,
      "step": 62295
    },
    {
      "entropy": 5.85210075378418,
      "epoch": 5.651306240928882,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023201186466355634,
      "loss": 4.9098,
      "mean_token_accuracy": 0.22447742968797685,
      "num_tokens": 129292584.0,
      "step": 62300
    },
    {
      "entropy": 5.7928492546081545,
      "epoch": 5.651759796806966,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023198010473648057,
      "loss": 4.9065,
      "mean_token_accuracy": 0.23012103736400605,
      "num_tokens": 129303301.0,
      "step": 62305
    },
    {
      "entropy": 5.753168106079102,
      "epoch": 5.652213352685051,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00023194834569906724,
      "loss": 4.8271,
      "mean_token_accuracy": 0.2317702054977417,
      "num_tokens": 129314229.0,
      "step": 62310
    },
    {
      "entropy": 5.7226808071136475,
      "epoch": 5.652666908563135,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023191658755197315,
      "loss": 4.8084,
      "mean_token_accuracy": 0.233319254219532,
      "num_tokens": 129324998.0,
      "step": 62315
    },
    {
      "entropy": 5.642700624465943,
      "epoch": 5.653120464441219,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023188483029585487,
      "loss": 4.7116,
      "mean_token_accuracy": 0.23884087204933166,
      "num_tokens": 129336133.0,
      "step": 62320
    },
    {
      "entropy": 5.699928569793701,
      "epoch": 5.6535740203193035,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002318530739313694,
      "loss": 4.83,
      "mean_token_accuracy": 0.23366150707006456,
      "num_tokens": 129346416.0,
      "step": 62325
    },
    {
      "entropy": 5.774641370773315,
      "epoch": 5.654027576197388,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00023182131845917337,
      "loss": 4.8763,
      "mean_token_accuracy": 0.22835716903209685,
      "num_tokens": 129357214.0,
      "step": 62330
    },
    {
      "entropy": 5.7227489948272705,
      "epoch": 5.654481132075472,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002317895638799234,
      "loss": 4.7374,
      "mean_token_accuracy": 0.23570508062839507,
      "num_tokens": 129366796.0,
      "step": 62335
    },
    {
      "entropy": 5.754882669448852,
      "epoch": 5.654934687953556,
      "grad_norm": 1.390625,
      "learning_rate": 0.00023175781019427633,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23155032098293304,
      "num_tokens": 129376791.0,
      "step": 62340
    },
    {
      "entropy": 5.723834800720215,
      "epoch": 5.65538824383164,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002317260574028887,
      "loss": 4.8828,
      "mean_token_accuracy": 0.23002263456583022,
      "num_tokens": 129387589.0,
      "step": 62345
    },
    {
      "entropy": 5.819589710235595,
      "epoch": 5.655841799709724,
      "grad_norm": 1.5,
      "learning_rate": 0.0002316943055064173,
      "loss": 4.9184,
      "mean_token_accuracy": 0.22608767747879027,
      "num_tokens": 129397041.0,
      "step": 62350
    },
    {
      "entropy": 5.683071804046631,
      "epoch": 5.656295355587808,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002316625545055186,
      "loss": 4.7304,
      "mean_token_accuracy": 0.23937471359968185,
      "num_tokens": 129406984.0,
      "step": 62355
    },
    {
      "entropy": 5.742789793014526,
      "epoch": 5.656748911465892,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023163080440084943,
      "loss": 4.7981,
      "mean_token_accuracy": 0.22956838458776474,
      "num_tokens": 129417809.0,
      "step": 62360
    },
    {
      "entropy": 5.644474458694458,
      "epoch": 5.657202467343977,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023159905519306624,
      "loss": 4.794,
      "mean_token_accuracy": 0.23629727214574814,
      "num_tokens": 129427838.0,
      "step": 62365
    },
    {
      "entropy": 5.755648469924926,
      "epoch": 5.657656023222061,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00023156730688282556,
      "loss": 4.9066,
      "mean_token_accuracy": 0.22694755643606185,
      "num_tokens": 129437536.0,
      "step": 62370
    },
    {
      "entropy": 5.788776445388794,
      "epoch": 5.658109579100145,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00023153555947078415,
      "loss": 4.8379,
      "mean_token_accuracy": 0.23463085740804673,
      "num_tokens": 129447930.0,
      "step": 62375
    },
    {
      "entropy": 5.775259780883789,
      "epoch": 5.6585631349782295,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023150381295759844,
      "loss": 4.8265,
      "mean_token_accuracy": 0.23745596408843994,
      "num_tokens": 129457740.0,
      "step": 62380
    },
    {
      "entropy": 5.841782093048096,
      "epoch": 5.659016690856314,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002314720673439249,
      "loss": 4.999,
      "mean_token_accuracy": 0.22141723483800888,
      "num_tokens": 129468340.0,
      "step": 62385
    },
    {
      "entropy": 5.796716785430908,
      "epoch": 5.659470246734398,
      "grad_norm": 1.40625,
      "learning_rate": 0.00023144032263042015,
      "loss": 4.8975,
      "mean_token_accuracy": 0.22586537003517151,
      "num_tokens": 129478880.0,
      "step": 62390
    },
    {
      "entropy": 5.751505947113037,
      "epoch": 5.659923802612482,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023140857881774057,
      "loss": 4.8265,
      "mean_token_accuracy": 0.2330569803714752,
      "num_tokens": 129489362.0,
      "step": 62395
    },
    {
      "entropy": 5.804844713211059,
      "epoch": 5.660377358490566,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00023137683590654273,
      "loss": 4.912,
      "mean_token_accuracy": 0.2217870682477951,
      "num_tokens": 129500294.0,
      "step": 62400
    },
    {
      "entropy": 5.759037828445434,
      "epoch": 5.66083091436865,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023134509389748315,
      "loss": 4.8259,
      "mean_token_accuracy": 0.2310274600982666,
      "num_tokens": 129510258.0,
      "step": 62405
    },
    {
      "entropy": 5.755191373825073,
      "epoch": 5.661284470246734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000231313352791218,
      "loss": 4.8629,
      "mean_token_accuracy": 0.23688496947288512,
      "num_tokens": 129520390.0,
      "step": 62410
    },
    {
      "entropy": 5.729142475128174,
      "epoch": 5.6617380261248185,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00023128161258840387,
      "loss": 4.8089,
      "mean_token_accuracy": 0.2348996505141258,
      "num_tokens": 129531165.0,
      "step": 62415
    },
    {
      "entropy": 5.789650726318359,
      "epoch": 5.662191582002903,
      "grad_norm": 1.328125,
      "learning_rate": 0.00023124987328969714,
      "loss": 4.9227,
      "mean_token_accuracy": 0.22433397322893142,
      "num_tokens": 129541468.0,
      "step": 62420
    },
    {
      "entropy": 5.81687183380127,
      "epoch": 5.662645137880987,
      "grad_norm": 1.375,
      "learning_rate": 0.00023121813489575428,
      "loss": 4.8424,
      "mean_token_accuracy": 0.22647175043821335,
      "num_tokens": 129551536.0,
      "step": 62425
    },
    {
      "entropy": 5.745698928833008,
      "epoch": 5.663098693759071,
      "grad_norm": 1.625,
      "learning_rate": 0.00023118639740723134,
      "loss": 4.8617,
      "mean_token_accuracy": 0.2286305695772171,
      "num_tokens": 129561101.0,
      "step": 62430
    },
    {
      "entropy": 5.850694179534912,
      "epoch": 5.6635522496371555,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023115466082478505,
      "loss": 4.9775,
      "mean_token_accuracy": 0.21691211462020873,
      "num_tokens": 129572040.0,
      "step": 62435
    },
    {
      "entropy": 5.737083768844604,
      "epoch": 5.66400580551524,
      "grad_norm": 1.46875,
      "learning_rate": 0.00023112292514907147,
      "loss": 4.8198,
      "mean_token_accuracy": 0.22920384556055068,
      "num_tokens": 129582243.0,
      "step": 62440
    },
    {
      "entropy": 5.6978456497192385,
      "epoch": 5.664459361393323,
      "grad_norm": 1.421875,
      "learning_rate": 0.000231091190380747,
      "loss": 4.7861,
      "mean_token_accuracy": 0.2318025916814804,
      "num_tokens": 129592950.0,
      "step": 62445
    },
    {
      "entropy": 5.708763074874878,
      "epoch": 5.664912917271408,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002310594565204679,
      "loss": 4.843,
      "mean_token_accuracy": 0.2355784982442856,
      "num_tokens": 129603936.0,
      "step": 62450
    },
    {
      "entropy": 5.785477924346924,
      "epoch": 5.665366473149492,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002310277235688904,
      "loss": 4.8935,
      "mean_token_accuracy": 0.22909920364618303,
      "num_tokens": 129613132.0,
      "step": 62455
    },
    {
      "entropy": 5.794998693466186,
      "epoch": 5.665820029027576,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023099599152667083,
      "loss": 4.9403,
      "mean_token_accuracy": 0.22758332639932632,
      "num_tokens": 129624016.0,
      "step": 62460
    },
    {
      "entropy": 5.803298091888427,
      "epoch": 5.66627358490566,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00023096426039446533,
      "loss": 4.8827,
      "mean_token_accuracy": 0.22861838936805726,
      "num_tokens": 129634423.0,
      "step": 62465
    },
    {
      "entropy": 5.746685123443603,
      "epoch": 5.6667271407837445,
      "grad_norm": 1.375,
      "learning_rate": 0.00023093253017293015,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22785753905773162,
      "num_tokens": 129644133.0,
      "step": 62470
    },
    {
      "entropy": 5.696043539047241,
      "epoch": 5.667180696661829,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023090080086272152,
      "loss": 4.8014,
      "mean_token_accuracy": 0.23412881046533585,
      "num_tokens": 129653761.0,
      "step": 62475
    },
    {
      "entropy": 5.695357847213745,
      "epoch": 5.667634252539913,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00023086907246449556,
      "loss": 4.8543,
      "mean_token_accuracy": 0.22181057333946227,
      "num_tokens": 129664628.0,
      "step": 62480
    },
    {
      "entropy": 5.702380800247193,
      "epoch": 5.668087808417997,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00023083734497890835,
      "loss": 4.8233,
      "mean_token_accuracy": 0.2446413591504097,
      "num_tokens": 129675301.0,
      "step": 62485
    },
    {
      "entropy": 5.8008496284484865,
      "epoch": 5.668541364296082,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023080561840661617,
      "loss": 4.9029,
      "mean_token_accuracy": 0.21443067491054535,
      "num_tokens": 129685599.0,
      "step": 62490
    },
    {
      "entropy": 5.7972657680511475,
      "epoch": 5.668994920174166,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002307738927482751,
      "loss": 4.8987,
      "mean_token_accuracy": 0.22100016176700593,
      "num_tokens": 129695927.0,
      "step": 62495
    },
    {
      "entropy": 5.746980237960815,
      "epoch": 5.669448476052249,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023074216800454106,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2267930820584297,
      "num_tokens": 129706422.0,
      "step": 62500
    },
    {
      "entropy": 5.762116718292236,
      "epoch": 5.6699020319303335,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002307104441760704,
      "loss": 4.8767,
      "mean_token_accuracy": 0.2229721039533615,
      "num_tokens": 129716125.0,
      "step": 62505
    },
    {
      "entropy": 5.7657371997833256,
      "epoch": 5.670355587808418,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00023067872126351902,
      "loss": 4.8211,
      "mean_token_accuracy": 0.2350431889295578,
      "num_tokens": 129727017.0,
      "step": 62510
    },
    {
      "entropy": 5.7481584548950195,
      "epoch": 5.670809143686502,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023064699926754297,
      "loss": 4.8326,
      "mean_token_accuracy": 0.23253362029790878,
      "num_tokens": 129737420.0,
      "step": 62515
    },
    {
      "entropy": 5.740907001495361,
      "epoch": 5.671262699564586,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00023061527818879818,
      "loss": 4.8083,
      "mean_token_accuracy": 0.22867179811000823,
      "num_tokens": 129747090.0,
      "step": 62520
    },
    {
      "entropy": 5.774267911911011,
      "epoch": 5.6717162554426706,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023058355802794083,
      "loss": 4.8711,
      "mean_token_accuracy": 0.23567916303873063,
      "num_tokens": 129757600.0,
      "step": 62525
    },
    {
      "entropy": 5.739974594116211,
      "epoch": 5.672169811320755,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00023055183878562685,
      "loss": 4.8342,
      "mean_token_accuracy": 0.2269715577363968,
      "num_tokens": 129767815.0,
      "step": 62530
    },
    {
      "entropy": 5.7463315486907955,
      "epoch": 5.672623367198839,
      "grad_norm": 1.265625,
      "learning_rate": 0.00023052012046251207,
      "loss": 4.8107,
      "mean_token_accuracy": 0.2358715355396271,
      "num_tokens": 129779222.0,
      "step": 62535
    },
    {
      "entropy": 5.7933612823486325,
      "epoch": 5.673076923076923,
      "grad_norm": 1.28125,
      "learning_rate": 0.00023048840305925266,
      "loss": 4.8962,
      "mean_token_accuracy": 0.23050930351018906,
      "num_tokens": 129790985.0,
      "step": 62540
    },
    {
      "entropy": 5.811457967758178,
      "epoch": 5.673530478955008,
      "grad_norm": 1.5,
      "learning_rate": 0.00023045668657650437,
      "loss": 4.7718,
      "mean_token_accuracy": 0.24343864470720292,
      "num_tokens": 129802386.0,
      "step": 62545
    },
    {
      "entropy": 5.726950788497925,
      "epoch": 5.673984034833092,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00023042497101492305,
      "loss": 4.8357,
      "mean_token_accuracy": 0.23494350016117097,
      "num_tokens": 129812345.0,
      "step": 62550
    },
    {
      "entropy": 5.768697690963745,
      "epoch": 5.674437590711175,
      "grad_norm": 1.5,
      "learning_rate": 0.00023039325637516485,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22450954020023345,
      "num_tokens": 129823430.0,
      "step": 62555
    },
    {
      "entropy": 5.8249236106872555,
      "epoch": 5.6748911465892595,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002303615426578854,
      "loss": 4.946,
      "mean_token_accuracy": 0.22606947273015976,
      "num_tokens": 129833715.0,
      "step": 62560
    },
    {
      "entropy": 5.784312343597412,
      "epoch": 5.675344702467344,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023032982986374062,
      "loss": 4.9031,
      "mean_token_accuracy": 0.2249494433403015,
      "num_tokens": 129845561.0,
      "step": 62565
    },
    {
      "entropy": 5.733760643005371,
      "epoch": 5.675798258345428,
      "grad_norm": 1.421875,
      "learning_rate": 0.00023029811799338628,
      "loss": 4.8478,
      "mean_token_accuracy": 0.24187968522310258,
      "num_tokens": 129855228.0,
      "step": 62570
    },
    {
      "entropy": 5.755346965789795,
      "epoch": 5.676251814223512,
      "grad_norm": 1.484375,
      "learning_rate": 0.00023026640704747837,
      "loss": 4.8327,
      "mean_token_accuracy": 0.23495341539382936,
      "num_tokens": 129865788.0,
      "step": 62575
    },
    {
      "entropy": 5.8352467060089115,
      "epoch": 5.676705370101597,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002302346970266725,
      "loss": 4.9361,
      "mean_token_accuracy": 0.22718936800956727,
      "num_tokens": 129875788.0,
      "step": 62580
    },
    {
      "entropy": 5.757521343231201,
      "epoch": 5.677158925979681,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00023020298793162452,
      "loss": 4.8393,
      "mean_token_accuracy": 0.2309536635875702,
      "num_tokens": 129886892.0,
      "step": 62585
    },
    {
      "entropy": 5.739808511734009,
      "epoch": 5.677612481857765,
      "grad_norm": 1.359375,
      "learning_rate": 0.00023017127976299013,
      "loss": 4.9451,
      "mean_token_accuracy": 0.22130272388458253,
      "num_tokens": 129897834.0,
      "step": 62590
    },
    {
      "entropy": 5.757444667816162,
      "epoch": 5.678066037735849,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00023013957252142515,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23116319477558137,
      "num_tokens": 129908038.0,
      "step": 62595
    },
    {
      "entropy": 5.824409675598145,
      "epoch": 5.678519593613933,
      "grad_norm": 1.4375,
      "learning_rate": 0.00023010786620758522,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22250084429979325,
      "num_tokens": 129918386.0,
      "step": 62600
    },
    {
      "entropy": 5.803698444366455,
      "epoch": 5.678973149492017,
      "grad_norm": 1.4375,
      "learning_rate": 0.000230076160822126,
      "loss": 4.9026,
      "mean_token_accuracy": 0.2248530000448227,
      "num_tokens": 129928776.0,
      "step": 62605
    },
    {
      "entropy": 5.74831109046936,
      "epoch": 5.679426705370101,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002300444563657033,
      "loss": 4.7938,
      "mean_token_accuracy": 0.23707152158021927,
      "num_tokens": 129939234.0,
      "step": 62610
    },
    {
      "entropy": 5.690375328063965,
      "epoch": 5.679880261248186,
      "grad_norm": 1.3125,
      "learning_rate": 0.00023001275283897273,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23925906419754028,
      "num_tokens": 129949506.0,
      "step": 62615
    },
    {
      "entropy": 5.753827619552612,
      "epoch": 5.68033381712627,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002299810502425898,
      "loss": 4.8182,
      "mean_token_accuracy": 0.24014124721288682,
      "num_tokens": 129959815.0,
      "step": 62620
    },
    {
      "entropy": 5.747087717056274,
      "epoch": 5.680787373004354,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022994934857721027,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22677279263734818,
      "num_tokens": 129969598.0,
      "step": 62625
    },
    {
      "entropy": 5.831044292449951,
      "epoch": 5.681240928882438,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022991764784348974,
      "loss": 4.8812,
      "mean_token_accuracy": 0.235447433590889,
      "num_tokens": 129979497.0,
      "step": 62630
    },
    {
      "entropy": 5.747351789474488,
      "epoch": 5.681694484760523,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022988594804208362,
      "loss": 4.8315,
      "mean_token_accuracy": 0.23263418823480606,
      "num_tokens": 129989716.0,
      "step": 62635
    },
    {
      "entropy": 5.7490565299987795,
      "epoch": 5.682148040638607,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022985424917364766,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23253291398286818,
      "num_tokens": 129999676.0,
      "step": 62640
    },
    {
      "entropy": 5.780193424224853,
      "epoch": 5.682601596516691,
      "grad_norm": 1.5,
      "learning_rate": 0.00022982255123883734,
      "loss": 4.8445,
      "mean_token_accuracy": 0.2311609774827957,
      "num_tokens": 130009346.0,
      "step": 62645
    },
    {
      "entropy": 5.743049049377442,
      "epoch": 5.683055152394775,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002297908542383082,
      "loss": 4.8816,
      "mean_token_accuracy": 0.21938689053058624,
      "num_tokens": 130021072.0,
      "step": 62650
    },
    {
      "entropy": 5.7124796390533445,
      "epoch": 5.683508708272859,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022975915817271565,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23041363507509233,
      "num_tokens": 130030899.0,
      "step": 62655
    },
    {
      "entropy": 5.722499084472656,
      "epoch": 5.683962264150943,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022972746304271525,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23826833367347716,
      "num_tokens": 130040345.0,
      "step": 62660
    },
    {
      "entropy": 5.7821611881256105,
      "epoch": 5.684415820029027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022969576884896247,
      "loss": 4.895,
      "mean_token_accuracy": 0.22594158798456193,
      "num_tokens": 130050627.0,
      "step": 62665
    },
    {
      "entropy": 5.757815170288086,
      "epoch": 5.684869375907112,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022966407559211273,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23199768662452697,
      "num_tokens": 130060592.0,
      "step": 62670
    },
    {
      "entropy": 5.777463579177857,
      "epoch": 5.685322931785196,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022963238327282143,
      "loss": 4.9186,
      "mean_token_accuracy": 0.22722258418798447,
      "num_tokens": 130070285.0,
      "step": 62675
    },
    {
      "entropy": 5.81440486907959,
      "epoch": 5.68577648766328,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002296006918917441,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2276184782385826,
      "num_tokens": 130080356.0,
      "step": 62680
    },
    {
      "entropy": 5.783275127410889,
      "epoch": 5.686230043541364,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022956900144953596,
      "loss": 4.9082,
      "mean_token_accuracy": 0.22095756828784943,
      "num_tokens": 130091532.0,
      "step": 62685
    },
    {
      "entropy": 5.677399396896362,
      "epoch": 5.686683599419449,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002295373119468524,
      "loss": 4.7771,
      "mean_token_accuracy": 0.23671059012413026,
      "num_tokens": 130101853.0,
      "step": 62690
    },
    {
      "entropy": 5.77581033706665,
      "epoch": 5.687137155297533,
      "grad_norm": 1.28125,
      "learning_rate": 0.00022950562338434888,
      "loss": 4.8997,
      "mean_token_accuracy": 0.2304423227906227,
      "num_tokens": 130111801.0,
      "step": 62695
    },
    {
      "entropy": 5.782634210586548,
      "epoch": 5.687590711175617,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022947393576268067,
      "loss": 4.8596,
      "mean_token_accuracy": 0.23484150022268296,
      "num_tokens": 130123007.0,
      "step": 62700
    },
    {
      "entropy": 5.799100399017334,
      "epoch": 5.6880442670537015,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000229442249082503,
      "loss": 4.8821,
      "mean_token_accuracy": 0.2287486106157303,
      "num_tokens": 130133699.0,
      "step": 62705
    },
    {
      "entropy": 5.845475673675537,
      "epoch": 5.688497822931785,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002294105633444713,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22511152029037476,
      "num_tokens": 130144178.0,
      "step": 62710
    },
    {
      "entropy": 5.6862608909606935,
      "epoch": 5.688951378809869,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002293788785492408,
      "loss": 4.795,
      "mean_token_accuracy": 0.2306189641356468,
      "num_tokens": 130155218.0,
      "step": 62715
    },
    {
      "entropy": 5.706049299240112,
      "epoch": 5.689404934687953,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022934719469746662,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2376090407371521,
      "num_tokens": 130166346.0,
      "step": 62720
    },
    {
      "entropy": 5.7653721332550045,
      "epoch": 5.689858490566038,
      "grad_norm": 1.5,
      "learning_rate": 0.00022931551178980415,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23199670612812043,
      "num_tokens": 130176354.0,
      "step": 62725
    },
    {
      "entropy": 5.74111704826355,
      "epoch": 5.690312046444122,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002292838298269085,
      "loss": 4.7858,
      "mean_token_accuracy": 0.23969070613384247,
      "num_tokens": 130185920.0,
      "step": 62730
    },
    {
      "entropy": 5.779671716690063,
      "epoch": 5.690765602322206,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022925214880943496,
      "loss": 4.8998,
      "mean_token_accuracy": 0.2258180186152458,
      "num_tokens": 130196670.0,
      "step": 62735
    },
    {
      "entropy": 5.797641134262085,
      "epoch": 5.6912191582002905,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022922046873803854,
      "loss": 4.8595,
      "mean_token_accuracy": 0.2325635328888893,
      "num_tokens": 130206567.0,
      "step": 62740
    },
    {
      "entropy": 5.749904489517212,
      "epoch": 5.691672714078375,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002291887896133746,
      "loss": 4.8682,
      "mean_token_accuracy": 0.2300350621342659,
      "num_tokens": 130217835.0,
      "step": 62745
    },
    {
      "entropy": 5.691327381134033,
      "epoch": 5.692126269956459,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022915711143609808,
      "loss": 4.8009,
      "mean_token_accuracy": 0.24109305143356324,
      "num_tokens": 130227968.0,
      "step": 62750
    },
    {
      "entropy": 5.765854692459106,
      "epoch": 5.692579825834542,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002291254342068642,
      "loss": 4.8689,
      "mean_token_accuracy": 0.23436411172151567,
      "num_tokens": 130237863.0,
      "step": 62755
    },
    {
      "entropy": 5.766698884963989,
      "epoch": 5.693033381712627,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022909375792632802,
      "loss": 4.8868,
      "mean_token_accuracy": 0.23117901980876923,
      "num_tokens": 130249296.0,
      "step": 62760
    },
    {
      "entropy": 5.746799945831299,
      "epoch": 5.693486937590711,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002290620825951447,
      "loss": 4.7835,
      "mean_token_accuracy": 0.23745742291212082,
      "num_tokens": 130259117.0,
      "step": 62765
    },
    {
      "entropy": 5.7860959529876705,
      "epoch": 5.693940493468795,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022903040821396913,
      "loss": 4.9057,
      "mean_token_accuracy": 0.22312599420547485,
      "num_tokens": 130269678.0,
      "step": 62770
    },
    {
      "entropy": 5.782543802261353,
      "epoch": 5.694394049346879,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002289987347834565,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22671292424201966,
      "num_tokens": 130280238.0,
      "step": 62775
    },
    {
      "entropy": 5.777627468109131,
      "epoch": 5.694847605224964,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022896706230426166,
      "loss": 4.8586,
      "mean_token_accuracy": 0.22788255363702775,
      "num_tokens": 130290586.0,
      "step": 62780
    },
    {
      "entropy": 5.815192365646363,
      "epoch": 5.695301161103048,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022893539077703984,
      "loss": 4.8399,
      "mean_token_accuracy": 0.23634149730205536,
      "num_tokens": 130300921.0,
      "step": 62785
    },
    {
      "entropy": 5.7521758556365965,
      "epoch": 5.695754716981132,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022890372020244578,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23304354697465895,
      "num_tokens": 130311420.0,
      "step": 62790
    },
    {
      "entropy": 5.791241264343261,
      "epoch": 5.6962082728592165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022887205058113458,
      "loss": 4.9326,
      "mean_token_accuracy": 0.22267668694257736,
      "num_tokens": 130322411.0,
      "step": 62795
    },
    {
      "entropy": 5.793440246582032,
      "epoch": 5.696661828737301,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002288403819137611,
      "loss": 4.8684,
      "mean_token_accuracy": 0.23216068297624587,
      "num_tokens": 130332422.0,
      "step": 62800
    },
    {
      "entropy": 5.760962963104248,
      "epoch": 5.697115384615385,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002288087142009803,
      "loss": 4.8143,
      "mean_token_accuracy": 0.23470404893159866,
      "num_tokens": 130342221.0,
      "step": 62805
    },
    {
      "entropy": 5.788327312469482,
      "epoch": 5.697568940493468,
      "grad_norm": 1.296875,
      "learning_rate": 0.000228777047443447,
      "loss": 4.8477,
      "mean_token_accuracy": 0.23425359576940535,
      "num_tokens": 130352882.0,
      "step": 62810
    },
    {
      "entropy": 5.756702947616577,
      "epoch": 5.698022496371553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022874538164181625,
      "loss": 4.8642,
      "mean_token_accuracy": 0.220935395359993,
      "num_tokens": 130363521.0,
      "step": 62815
    },
    {
      "entropy": 5.734642839431762,
      "epoch": 5.698476052249637,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022871371679674275,
      "loss": 4.9042,
      "mean_token_accuracy": 0.2307838663458824,
      "num_tokens": 130374737.0,
      "step": 62820
    },
    {
      "entropy": 5.838747930526734,
      "epoch": 5.698929608127721,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022868205290888134,
      "loss": 4.9169,
      "mean_token_accuracy": 0.22837896794080734,
      "num_tokens": 130385677.0,
      "step": 62825
    },
    {
      "entropy": 5.810202264785767,
      "epoch": 5.6993831640058055,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000228650389978887,
      "loss": 4.8701,
      "mean_token_accuracy": 0.23119571954011917,
      "num_tokens": 130396708.0,
      "step": 62830
    },
    {
      "entropy": 5.737503337860107,
      "epoch": 5.69983671988389,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002286187280074143,
      "loss": 4.851,
      "mean_token_accuracy": 0.23531675040721894,
      "num_tokens": 130406789.0,
      "step": 62835
    },
    {
      "entropy": 5.734456777572632,
      "epoch": 5.700290275761974,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002285870669951181,
      "loss": 4.7864,
      "mean_token_accuracy": 0.24800721257925035,
      "num_tokens": 130416876.0,
      "step": 62840
    },
    {
      "entropy": 5.794735240936279,
      "epoch": 5.700743831640058,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022855540694265326,
      "loss": 4.8338,
      "mean_token_accuracy": 0.236070217192173,
      "num_tokens": 130427580.0,
      "step": 62845
    },
    {
      "entropy": 5.769813060760498,
      "epoch": 5.7011973875181425,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022852374785067438,
      "loss": 4.8614,
      "mean_token_accuracy": 0.23205424398183822,
      "num_tokens": 130437870.0,
      "step": 62850
    },
    {
      "entropy": 5.74259729385376,
      "epoch": 5.701650943396227,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022849208971983625,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22958768904209137,
      "num_tokens": 130447852.0,
      "step": 62855
    },
    {
      "entropy": 5.7650071144104,
      "epoch": 5.702104499274311,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022846043255079357,
      "loss": 4.8694,
      "mean_token_accuracy": 0.2308114841580391,
      "num_tokens": 130458322.0,
      "step": 62860
    },
    {
      "entropy": 5.764510726928711,
      "epoch": 5.7025580551523944,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000228428776344201,
      "loss": 4.7644,
      "mean_token_accuracy": 0.2408124253153801,
      "num_tokens": 130468344.0,
      "step": 62865
    },
    {
      "entropy": 5.72079439163208,
      "epoch": 5.703011611030479,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022839712110071315,
      "loss": 4.7689,
      "mean_token_accuracy": 0.2352310299873352,
      "num_tokens": 130478288.0,
      "step": 62870
    },
    {
      "entropy": 5.718899202346802,
      "epoch": 5.703465166908563,
      "grad_norm": 1.375,
      "learning_rate": 0.0002283654668209848,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2350244179368019,
      "num_tokens": 130488223.0,
      "step": 62875
    },
    {
      "entropy": 5.66985559463501,
      "epoch": 5.703918722786647,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022833381350567046,
      "loss": 4.7231,
      "mean_token_accuracy": 0.23981060534715654,
      "num_tokens": 130498602.0,
      "step": 62880
    },
    {
      "entropy": 5.727999496459961,
      "epoch": 5.7043722786647315,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022830216115542467,
      "loss": 4.7937,
      "mean_token_accuracy": 0.23874205499887466,
      "num_tokens": 130509479.0,
      "step": 62885
    },
    {
      "entropy": 5.7419686794281,
      "epoch": 5.704825834542816,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002282705097709022,
      "loss": 4.8585,
      "mean_token_accuracy": 0.2293197989463806,
      "num_tokens": 130520606.0,
      "step": 62890
    },
    {
      "entropy": 5.7708159446716305,
      "epoch": 5.7052793904209,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00022823885935275746,
      "loss": 4.9368,
      "mean_token_accuracy": 0.2248995706439018,
      "num_tokens": 130532447.0,
      "step": 62895
    },
    {
      "entropy": 5.742708778381347,
      "epoch": 5.705732946298984,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022820720990164502,
      "loss": 4.8114,
      "mean_token_accuracy": 0.228180430829525,
      "num_tokens": 130542047.0,
      "step": 62900
    },
    {
      "entropy": 5.724402713775635,
      "epoch": 5.706186502177069,
      "grad_norm": 1.59375,
      "learning_rate": 0.00022817556141821932,
      "loss": 4.8002,
      "mean_token_accuracy": 0.23183899968862534,
      "num_tokens": 130552631.0,
      "step": 62905
    },
    {
      "entropy": 5.734602260589599,
      "epoch": 5.706640058055152,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022814391390313504,
      "loss": 4.8289,
      "mean_token_accuracy": 0.2312895119190216,
      "num_tokens": 130562323.0,
      "step": 62910
    },
    {
      "entropy": 5.740685844421387,
      "epoch": 5.707093613933236,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022811226735704654,
      "loss": 4.8336,
      "mean_token_accuracy": 0.23025530725717544,
      "num_tokens": 130573310.0,
      "step": 62915
    },
    {
      "entropy": 5.80313196182251,
      "epoch": 5.7075471698113205,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022808062178060828,
      "loss": 4.8773,
      "mean_token_accuracy": 0.22679204791784285,
      "num_tokens": 130584059.0,
      "step": 62920
    },
    {
      "entropy": 5.737573862075806,
      "epoch": 5.708000725689405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022804897717447476,
      "loss": 4.7955,
      "mean_token_accuracy": 0.23219710886478423,
      "num_tokens": 130594101.0,
      "step": 62925
    },
    {
      "entropy": 5.69411301612854,
      "epoch": 5.708454281567489,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022801733353930038,
      "loss": 4.802,
      "mean_token_accuracy": 0.2335984706878662,
      "num_tokens": 130604531.0,
      "step": 62930
    },
    {
      "entropy": 5.708703327178955,
      "epoch": 5.708907837445573,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002279856908757394,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2395538255572319,
      "num_tokens": 130614213.0,
      "step": 62935
    },
    {
      "entropy": 5.698153972625732,
      "epoch": 5.7093613933236576,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002279540491844464,
      "loss": 4.7821,
      "mean_token_accuracy": 0.23677754402160645,
      "num_tokens": 130624809.0,
      "step": 62940
    },
    {
      "entropy": 5.741017436981201,
      "epoch": 5.709814949201742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022792240846607558,
      "loss": 4.8579,
      "mean_token_accuracy": 0.23170955032110213,
      "num_tokens": 130633874.0,
      "step": 62945
    },
    {
      "entropy": 5.68378176689148,
      "epoch": 5.710268505079826,
      "grad_norm": 1.375,
      "learning_rate": 0.00022789076872128144,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23585027009248732,
      "num_tokens": 130644256.0,
      "step": 62950
    },
    {
      "entropy": 5.708943462371826,
      "epoch": 5.71072206095791,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022785912995071817,
      "loss": 4.799,
      "mean_token_accuracy": 0.2344634562730789,
      "num_tokens": 130654318.0,
      "step": 62955
    },
    {
      "entropy": 5.741167688369751,
      "epoch": 5.711175616835995,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002278274921550401,
      "loss": 4.8403,
      "mean_token_accuracy": 0.2265452966094017,
      "num_tokens": 130664256.0,
      "step": 62960
    },
    {
      "entropy": 5.842983055114746,
      "epoch": 5.711629172714078,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022779585533490155,
      "loss": 4.9611,
      "mean_token_accuracy": 0.22581115812063218,
      "num_tokens": 130674238.0,
      "step": 62965
    },
    {
      "entropy": 5.758849000930786,
      "epoch": 5.712082728592162,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022776421949095667,
      "loss": 4.8492,
      "mean_token_accuracy": 0.22355429381132125,
      "num_tokens": 130685308.0,
      "step": 62970
    },
    {
      "entropy": 5.766688966751099,
      "epoch": 5.7125362844702465,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022773258462385981,
      "loss": 4.8221,
      "mean_token_accuracy": 0.23550186902284623,
      "num_tokens": 130695773.0,
      "step": 62975
    },
    {
      "entropy": 5.81067476272583,
      "epoch": 5.712989840348331,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002277009507342651,
      "loss": 4.8649,
      "mean_token_accuracy": 0.2336650550365448,
      "num_tokens": 130705730.0,
      "step": 62980
    },
    {
      "entropy": 5.790423059463501,
      "epoch": 5.713443396226415,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022766931782282685,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2291719138622284,
      "num_tokens": 130717154.0,
      "step": 62985
    },
    {
      "entropy": 5.7402599334716795,
      "epoch": 5.713896952104499,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022763768589019912,
      "loss": 4.8568,
      "mean_token_accuracy": 0.23454699367284776,
      "num_tokens": 130726963.0,
      "step": 62990
    },
    {
      "entropy": 5.749005460739136,
      "epoch": 5.714350507982584,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022760605493703616,
      "loss": 4.9113,
      "mean_token_accuracy": 0.22327701151371002,
      "num_tokens": 130736736.0,
      "step": 62995
    },
    {
      "entropy": 5.694872522354126,
      "epoch": 5.714804063860668,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022757442496399206,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23921666592359542,
      "num_tokens": 130746758.0,
      "step": 63000
    },
    {
      "epoch": 5.714804063860668,
      "eval_entropy": 5.43687910569739,
      "eval_loss": 4.943871021270752,
      "eval_mean_token_accuracy": 0.23370465590572195,
      "eval_num_tokens": 130746758.0,
      "eval_runtime": 20.5463,
      "eval_samples_per_second": 1720.99,
      "eval_steps_per_second": 215.124,
      "step": 63000
    },
    {
      "entropy": 5.723702239990234,
      "epoch": 5.715257619738752,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022754279597172084,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23986051380634307,
      "num_tokens": 130756962.0,
      "step": 63005
    },
    {
      "entropy": 5.749063205718994,
      "epoch": 5.7157111756168355,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002275111679608768,
      "loss": 4.7997,
      "mean_token_accuracy": 0.2324761927127838,
      "num_tokens": 130768553.0,
      "step": 63010
    },
    {
      "entropy": 5.805951404571533,
      "epoch": 5.716164731494921,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022747954093211388,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22605715692043304,
      "num_tokens": 130778683.0,
      "step": 63015
    },
    {
      "entropy": 5.708277845382691,
      "epoch": 5.716618287373004,
      "grad_norm": 1.5,
      "learning_rate": 0.0002274479148860862,
      "loss": 4.7926,
      "mean_token_accuracy": 0.23773394674062728,
      "num_tokens": 130789109.0,
      "step": 63020
    },
    {
      "entropy": 5.6789782524108885,
      "epoch": 5.717071843251088,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002274162898234477,
      "loss": 4.7975,
      "mean_token_accuracy": 0.23765368312597274,
      "num_tokens": 130800376.0,
      "step": 63025
    },
    {
      "entropy": 5.798116683959961,
      "epoch": 5.717525399129173,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022738466574485244,
      "loss": 4.9153,
      "mean_token_accuracy": 0.226673686504364,
      "num_tokens": 130811688.0,
      "step": 63030
    },
    {
      "entropy": 5.758401155471802,
      "epoch": 5.717978955007257,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022735304265095445,
      "loss": 4.8532,
      "mean_token_accuracy": 0.23014710396528243,
      "num_tokens": 130822137.0,
      "step": 63035
    },
    {
      "entropy": 5.729073762893677,
      "epoch": 5.718432510885341,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022732142054240767,
      "loss": 4.7764,
      "mean_token_accuracy": 0.23057634234428406,
      "num_tokens": 130831635.0,
      "step": 63040
    },
    {
      "entropy": 5.757547855377197,
      "epoch": 5.718886066763425,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022728979941986612,
      "loss": 4.8844,
      "mean_token_accuracy": 0.23117374032735824,
      "num_tokens": 130841277.0,
      "step": 63045
    },
    {
      "entropy": 5.658376407623291,
      "epoch": 5.71933962264151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022725817928398367,
      "loss": 4.7278,
      "mean_token_accuracy": 0.24414318650960923,
      "num_tokens": 130851252.0,
      "step": 63050
    },
    {
      "entropy": 5.7642663478851315,
      "epoch": 5.719793178519594,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022722656013541415,
      "loss": 4.8489,
      "mean_token_accuracy": 0.2318326562643051,
      "num_tokens": 130861756.0,
      "step": 63055
    },
    {
      "entropy": 5.734080362319946,
      "epoch": 5.720246734397678,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022719494197481155,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23559747487306595,
      "num_tokens": 130871284.0,
      "step": 63060
    },
    {
      "entropy": 5.7411370277404785,
      "epoch": 5.7207002902757615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002271633248028298,
      "loss": 4.8083,
      "mean_token_accuracy": 0.23725079149007797,
      "num_tokens": 130882320.0,
      "step": 63065
    },
    {
      "entropy": 5.748185825347901,
      "epoch": 5.721153846153846,
      "grad_norm": 1.265625,
      "learning_rate": 0.00022713170862012267,
      "loss": 4.8421,
      "mean_token_accuracy": 0.22672594338655472,
      "num_tokens": 130893711.0,
      "step": 63070
    },
    {
      "entropy": 5.705765914916992,
      "epoch": 5.72160740203193,
      "grad_norm": 1.421875,
      "learning_rate": 0.000227100093427344,
      "loss": 4.8334,
      "mean_token_accuracy": 0.22980556786060333,
      "num_tokens": 130903543.0,
      "step": 63075
    },
    {
      "entropy": 5.699061679840088,
      "epoch": 5.722060957910014,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002270684792251476,
      "loss": 4.7135,
      "mean_token_accuracy": 0.24379598051309587,
      "num_tokens": 130913819.0,
      "step": 63080
    },
    {
      "entropy": 5.718132257461548,
      "epoch": 5.722514513788099,
      "grad_norm": 1.375,
      "learning_rate": 0.00022703686601418727,
      "loss": 4.8249,
      "mean_token_accuracy": 0.2428395539522171,
      "num_tokens": 130924628.0,
      "step": 63085
    },
    {
      "entropy": 5.746065950393676,
      "epoch": 5.722968069666183,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022700525379511672,
      "loss": 4.8313,
      "mean_token_accuracy": 0.22841412723064422,
      "num_tokens": 130934163.0,
      "step": 63090
    },
    {
      "entropy": 5.774480533599854,
      "epoch": 5.723421625544267,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022697364256858983,
      "loss": 4.7921,
      "mean_token_accuracy": 0.2358289524912834,
      "num_tokens": 130944449.0,
      "step": 63095
    },
    {
      "entropy": 5.780919599533081,
      "epoch": 5.723875181422351,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022694203233526023,
      "loss": 4.8475,
      "mean_token_accuracy": 0.23510490655899047,
      "num_tokens": 130954595.0,
      "step": 63100
    },
    {
      "entropy": 5.758744812011718,
      "epoch": 5.724328737300436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00022691042309578164,
      "loss": 4.8833,
      "mean_token_accuracy": 0.2291286051273346,
      "num_tokens": 130965405.0,
      "step": 63105
    },
    {
      "entropy": 5.753524494171143,
      "epoch": 5.72478229317852,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022687881485080775,
      "loss": 4.8717,
      "mean_token_accuracy": 0.2235974684357643,
      "num_tokens": 130975984.0,
      "step": 63110
    },
    {
      "entropy": 5.756491184234619,
      "epoch": 5.725235849056604,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002268472076009922,
      "loss": 4.8335,
      "mean_token_accuracy": 0.23411109894514084,
      "num_tokens": 130987038.0,
      "step": 63115
    },
    {
      "entropy": 5.74836916923523,
      "epoch": 5.725689404934688,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002268156013469887,
      "loss": 4.7724,
      "mean_token_accuracy": 0.23559966534376145,
      "num_tokens": 130996872.0,
      "step": 63120
    },
    {
      "entropy": 5.7788066387176515,
      "epoch": 5.726142960812772,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022678399608945082,
      "loss": 4.8167,
      "mean_token_accuracy": 0.23198606967926025,
      "num_tokens": 131008128.0,
      "step": 63125
    },
    {
      "entropy": 5.793580675125122,
      "epoch": 5.726596516690856,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022675239182903223,
      "loss": 4.9636,
      "mean_token_accuracy": 0.22415110766887664,
      "num_tokens": 131019014.0,
      "step": 63130
    },
    {
      "entropy": 5.781585836410523,
      "epoch": 5.72705007256894,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022672078856638645,
      "loss": 4.8887,
      "mean_token_accuracy": 0.23506686240434646,
      "num_tokens": 131029148.0,
      "step": 63135
    },
    {
      "entropy": 5.712237310409546,
      "epoch": 5.727503628447025,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022668918630216705,
      "loss": 4.8241,
      "mean_token_accuracy": 0.23420713990926742,
      "num_tokens": 131039290.0,
      "step": 63140
    },
    {
      "entropy": 5.726251125335693,
      "epoch": 5.727957184325109,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002266575850370276,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23517630100250245,
      "num_tokens": 131050264.0,
      "step": 63145
    },
    {
      "entropy": 5.721894693374634,
      "epoch": 5.728410740203193,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022662598477162166,
      "loss": 4.7956,
      "mean_token_accuracy": 0.23163548707962037,
      "num_tokens": 131060089.0,
      "step": 63150
    },
    {
      "entropy": 5.80484676361084,
      "epoch": 5.7288642960812775,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002265943855066026,
      "loss": 4.9239,
      "mean_token_accuracy": 0.22248965501785278,
      "num_tokens": 131072336.0,
      "step": 63155
    },
    {
      "entropy": 5.769718933105469,
      "epoch": 5.729317851959362,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000226562787242624,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23079734295606613,
      "num_tokens": 131082028.0,
      "step": 63160
    },
    {
      "entropy": 5.7916546821594235,
      "epoch": 5.729771407837445,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022653118998033932,
      "loss": 4.8921,
      "mean_token_accuracy": 0.22995831221342086,
      "num_tokens": 131091899.0,
      "step": 63165
    },
    {
      "entropy": 5.713601922988891,
      "epoch": 5.73022496371553,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00022649959372040207,
      "loss": 4.7847,
      "mean_token_accuracy": 0.23135925978422164,
      "num_tokens": 131102385.0,
      "step": 63170
    },
    {
      "entropy": 5.742674160003662,
      "epoch": 5.730678519593614,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00022646799846346544,
      "loss": 4.9234,
      "mean_token_accuracy": 0.22318374514579772,
      "num_tokens": 131112407.0,
      "step": 63175
    },
    {
      "entropy": 5.7188973903656,
      "epoch": 5.731132075471698,
      "grad_norm": 1.453125,
      "learning_rate": 0.000226436404210183,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23385708332061766,
      "num_tokens": 131122450.0,
      "step": 63180
    },
    {
      "entropy": 5.717384958267212,
      "epoch": 5.731585631349782,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002264048109612082,
      "loss": 4.7873,
      "mean_token_accuracy": 0.236953704059124,
      "num_tokens": 131133409.0,
      "step": 63185
    },
    {
      "entropy": 5.759009504318238,
      "epoch": 5.732039187227866,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022637321871719419,
      "loss": 4.89,
      "mean_token_accuracy": 0.23442719131708145,
      "num_tokens": 131142878.0,
      "step": 63190
    },
    {
      "entropy": 5.758086156845093,
      "epoch": 5.732492743105951,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002263416274787944,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2349141165614128,
      "num_tokens": 131152442.0,
      "step": 63195
    },
    {
      "entropy": 5.862503147125244,
      "epoch": 5.732946298984035,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002263100372466622,
      "loss": 4.8643,
      "mean_token_accuracy": 0.23474084883928298,
      "num_tokens": 131163581.0,
      "step": 63200
    },
    {
      "entropy": 5.727890253067017,
      "epoch": 5.733399854862119,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022627844802145086,
      "loss": 4.7661,
      "mean_token_accuracy": 0.23494701236486434,
      "num_tokens": 131173198.0,
      "step": 63205
    },
    {
      "entropy": 5.775779771804809,
      "epoch": 5.7338534107402035,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022624685980381353,
      "loss": 4.8936,
      "mean_token_accuracy": 0.22756629586219787,
      "num_tokens": 131183073.0,
      "step": 63210
    },
    {
      "entropy": 5.7405167579650875,
      "epoch": 5.734306966618288,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022621527259440366,
      "loss": 4.8473,
      "mean_token_accuracy": 0.23301422894001006,
      "num_tokens": 131194222.0,
      "step": 63215
    },
    {
      "entropy": 5.8266222953796385,
      "epoch": 5.734760522496371,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022618368639387438,
      "loss": 4.9357,
      "mean_token_accuracy": 0.22812913060188295,
      "num_tokens": 131204724.0,
      "step": 63220
    },
    {
      "entropy": 5.843273735046386,
      "epoch": 5.735214078374455,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022615210120287887,
      "loss": 4.9768,
      "mean_token_accuracy": 0.22261869460344313,
      "num_tokens": 131214079.0,
      "step": 63225
    },
    {
      "entropy": 5.792194128036499,
      "epoch": 5.73566763425254,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022612051702207032,
      "loss": 4.873,
      "mean_token_accuracy": 0.22128077149391173,
      "num_tokens": 131224494.0,
      "step": 63230
    },
    {
      "entropy": 5.730325031280517,
      "epoch": 5.736121190130624,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000226088933852102,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23229456096887588,
      "num_tokens": 131235604.0,
      "step": 63235
    },
    {
      "entropy": 5.67391996383667,
      "epoch": 5.736574746008708,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022605735169362701,
      "loss": 4.7573,
      "mean_token_accuracy": 0.23577645123004914,
      "num_tokens": 131246857.0,
      "step": 63240
    },
    {
      "entropy": 5.770697975158692,
      "epoch": 5.7370283018867925,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002260257705472984,
      "loss": 4.8849,
      "mean_token_accuracy": 0.2238420695066452,
      "num_tokens": 131257508.0,
      "step": 63245
    },
    {
      "entropy": 5.745258808135986,
      "epoch": 5.737481857764877,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022599419041376934,
      "loss": 4.8901,
      "mean_token_accuracy": 0.2285746842622757,
      "num_tokens": 131267820.0,
      "step": 63250
    },
    {
      "entropy": 5.818406009674073,
      "epoch": 5.737935413642961,
      "grad_norm": 1.375,
      "learning_rate": 0.000225962611293693,
      "loss": 4.9643,
      "mean_token_accuracy": 0.22615649700164794,
      "num_tokens": 131278307.0,
      "step": 63255
    },
    {
      "entropy": 5.770679235458374,
      "epoch": 5.738388969521045,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022593103318772228,
      "loss": 4.8272,
      "mean_token_accuracy": 0.23159335404634476,
      "num_tokens": 131288576.0,
      "step": 63260
    },
    {
      "entropy": 5.764786195755005,
      "epoch": 5.7388425253991295,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002258994560965103,
      "loss": 4.8215,
      "mean_token_accuracy": 0.2295350581407547,
      "num_tokens": 131299419.0,
      "step": 63265
    },
    {
      "entropy": 5.74167766571045,
      "epoch": 5.739296081277214,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022586788002071007,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23072513490915297,
      "num_tokens": 131309722.0,
      "step": 63270
    },
    {
      "entropy": 5.806350946426392,
      "epoch": 5.739749637155297,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022583630496097456,
      "loss": 4.8092,
      "mean_token_accuracy": 0.24129804074764252,
      "num_tokens": 131318955.0,
      "step": 63275
    },
    {
      "entropy": 5.812300395965576,
      "epoch": 5.7402031930333814,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022580473091795688,
      "loss": 4.8932,
      "mean_token_accuracy": 0.223804335296154,
      "num_tokens": 131329113.0,
      "step": 63280
    },
    {
      "entropy": 5.785334157943725,
      "epoch": 5.740656748911466,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002257731578923099,
      "loss": 4.9103,
      "mean_token_accuracy": 0.22070534527301788,
      "num_tokens": 131339473.0,
      "step": 63285
    },
    {
      "entropy": 5.821410608291626,
      "epoch": 5.74111030478955,
      "grad_norm": 1.5,
      "learning_rate": 0.00022574158588468647,
      "loss": 5.006,
      "mean_token_accuracy": 0.21922341734170914,
      "num_tokens": 131349654.0,
      "step": 63290
    },
    {
      "entropy": 5.866785287857056,
      "epoch": 5.741563860667634,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00022571001489573968,
      "loss": 4.9317,
      "mean_token_accuracy": 0.22826533615589142,
      "num_tokens": 131359870.0,
      "step": 63295
    },
    {
      "entropy": 5.745858097076416,
      "epoch": 5.7420174165457185,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022567844492612233,
      "loss": 4.7641,
      "mean_token_accuracy": 0.23677221983671187,
      "num_tokens": 131369568.0,
      "step": 63300
    },
    {
      "entropy": 5.912884378433228,
      "epoch": 5.742470972423803,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022564687597648726,
      "loss": 5.0673,
      "mean_token_accuracy": 0.21802129596471786,
      "num_tokens": 131381254.0,
      "step": 63305
    },
    {
      "entropy": 5.699623107910156,
      "epoch": 5.742924528301887,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022561530804748737,
      "loss": 4.7539,
      "mean_token_accuracy": 0.24257195889949798,
      "num_tokens": 131391182.0,
      "step": 63310
    },
    {
      "entropy": 5.819729518890381,
      "epoch": 5.743378084179971,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002255837411397756,
      "loss": 4.9796,
      "mean_token_accuracy": 0.22403106987476348,
      "num_tokens": 131402125.0,
      "step": 63315
    },
    {
      "entropy": 5.775352334976196,
      "epoch": 5.743831640058055,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022555217525400457,
      "loss": 4.8439,
      "mean_token_accuracy": 0.2206995666027069,
      "num_tokens": 131411994.0,
      "step": 63320
    },
    {
      "entropy": 5.793735313415527,
      "epoch": 5.744285195936139,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002255206103908271,
      "loss": 4.8951,
      "mean_token_accuracy": 0.22176494747400283,
      "num_tokens": 131423864.0,
      "step": 63325
    },
    {
      "entropy": 5.705172729492188,
      "epoch": 5.744738751814223,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022548904655089608,
      "loss": 4.7795,
      "mean_token_accuracy": 0.2331220895051956,
      "num_tokens": 131434320.0,
      "step": 63330
    },
    {
      "entropy": 5.763970136642456,
      "epoch": 5.7451923076923075,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022545748373486424,
      "loss": 4.8471,
      "mean_token_accuracy": 0.2232184886932373,
      "num_tokens": 131444628.0,
      "step": 63335
    },
    {
      "entropy": 5.743957662582398,
      "epoch": 5.745645863570392,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022542592194338418,
      "loss": 4.8043,
      "mean_token_accuracy": 0.24134710431098938,
      "num_tokens": 131455644.0,
      "step": 63340
    },
    {
      "entropy": 5.858719062805176,
      "epoch": 5.746099419448476,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022539436117710877,
      "loss": 4.9271,
      "mean_token_accuracy": 0.23051225394010544,
      "num_tokens": 131466138.0,
      "step": 63345
    },
    {
      "entropy": 5.818234252929687,
      "epoch": 5.74655297532656,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022536280143669053,
      "loss": 4.9662,
      "mean_token_accuracy": 0.2256313219666481,
      "num_tokens": 131476153.0,
      "step": 63350
    },
    {
      "entropy": 5.750279664993286,
      "epoch": 5.7470065312046446,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022533124272278222,
      "loss": 4.8551,
      "mean_token_accuracy": 0.23294132500886916,
      "num_tokens": 131485973.0,
      "step": 63355
    },
    {
      "entropy": 5.755166149139404,
      "epoch": 5.747460087082729,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00022529968503603654,
      "loss": 4.8531,
      "mean_token_accuracy": 0.2261092409491539,
      "num_tokens": 131496428.0,
      "step": 63360
    },
    {
      "entropy": 5.788383626937867,
      "epoch": 5.747913642960813,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022526812837710597,
      "loss": 4.8789,
      "mean_token_accuracy": 0.2227666512131691,
      "num_tokens": 131507836.0,
      "step": 63365
    },
    {
      "entropy": 5.754771614074707,
      "epoch": 5.748367198838897,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002252365727466432,
      "loss": 4.8729,
      "mean_token_accuracy": 0.22338528484106063,
      "num_tokens": 131518548.0,
      "step": 63370
    },
    {
      "entropy": 5.720959234237671,
      "epoch": 5.748820754716981,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022520501814530074,
      "loss": 4.7527,
      "mean_token_accuracy": 0.23531978130340575,
      "num_tokens": 131528784.0,
      "step": 63375
    },
    {
      "entropy": 5.728237199783325,
      "epoch": 5.749274310595065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022517346457373127,
      "loss": 4.824,
      "mean_token_accuracy": 0.23728896230459212,
      "num_tokens": 131539617.0,
      "step": 63380
    },
    {
      "entropy": 5.7587440490722654,
      "epoch": 5.749727866473149,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022514191203258722,
      "loss": 4.897,
      "mean_token_accuracy": 0.22176202535629272,
      "num_tokens": 131550279.0,
      "step": 63385
    },
    {
      "entropy": 5.768814182281494,
      "epoch": 5.7501814223512335,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002251103605225211,
      "loss": 4.8525,
      "mean_token_accuracy": 0.23007745891809464,
      "num_tokens": 131561494.0,
      "step": 63390
    },
    {
      "entropy": 5.770186424255371,
      "epoch": 5.750634978229318,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022507881004418546,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23707843124866484,
      "num_tokens": 131572166.0,
      "step": 63395
    },
    {
      "entropy": 5.809416151046753,
      "epoch": 5.751088534107402,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022504726059823277,
      "loss": 4.9122,
      "mean_token_accuracy": 0.22450665533542633,
      "num_tokens": 131582224.0,
      "step": 63400
    },
    {
      "entropy": 5.721942472457886,
      "epoch": 5.751542089985486,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022501571218531554,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23227176815271378,
      "num_tokens": 131592612.0,
      "step": 63405
    },
    {
      "entropy": 5.6989439010620115,
      "epoch": 5.751995645863571,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022498416480608597,
      "loss": 4.8094,
      "mean_token_accuracy": 0.23851267248392105,
      "num_tokens": 131603002.0,
      "step": 63410
    },
    {
      "entropy": 5.740689754486084,
      "epoch": 5.752449201741655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002249526184611967,
      "loss": 4.821,
      "mean_token_accuracy": 0.22762681394815446,
      "num_tokens": 131614045.0,
      "step": 63415
    },
    {
      "entropy": 5.786373281478882,
      "epoch": 5.752902757619739,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002249210731513,
      "loss": 4.8115,
      "mean_token_accuracy": 0.23974277228116989,
      "num_tokens": 131623395.0,
      "step": 63420
    },
    {
      "entropy": 5.753016376495362,
      "epoch": 5.753356313497823,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022488952887704823,
      "loss": 4.8745,
      "mean_token_accuracy": 0.23159576207399368,
      "num_tokens": 131633985.0,
      "step": 63425
    },
    {
      "entropy": 5.741438865661621,
      "epoch": 5.753809869375907,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022485798563909383,
      "loss": 4.755,
      "mean_token_accuracy": 0.23914728164672852,
      "num_tokens": 131644136.0,
      "step": 63430
    },
    {
      "entropy": 5.702110004425049,
      "epoch": 5.754263425253991,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002248264434380891,
      "loss": 4.8079,
      "mean_token_accuracy": 0.2285836398601532,
      "num_tokens": 131654197.0,
      "step": 63435
    },
    {
      "entropy": 5.7885829448699955,
      "epoch": 5.754716981132075,
      "grad_norm": 1.46875,
      "learning_rate": 0.00022479490227468623,
      "loss": 4.9135,
      "mean_token_accuracy": 0.22893003672361373,
      "num_tokens": 131664697.0,
      "step": 63440
    },
    {
      "entropy": 5.728383493423462,
      "epoch": 5.75517053701016,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022476336214953757,
      "loss": 4.8164,
      "mean_token_accuracy": 0.23760607838630676,
      "num_tokens": 131674822.0,
      "step": 63445
    },
    {
      "entropy": 5.737560415267945,
      "epoch": 5.755624092888244,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022473182306329548,
      "loss": 4.8083,
      "mean_token_accuracy": 0.2286667764186859,
      "num_tokens": 131684716.0,
      "step": 63450
    },
    {
      "entropy": 5.759054040908813,
      "epoch": 5.756077648766328,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022470028501661205,
      "loss": 4.9356,
      "mean_token_accuracy": 0.2206907719373703,
      "num_tokens": 131696608.0,
      "step": 63455
    },
    {
      "entropy": 5.735082674026489,
      "epoch": 5.756531204644412,
      "grad_norm": 1.375,
      "learning_rate": 0.00022466874801013948,
      "loss": 4.8587,
      "mean_token_accuracy": 0.23004749566316604,
      "num_tokens": 131707247.0,
      "step": 63460
    },
    {
      "entropy": 5.739834260940552,
      "epoch": 5.756984760522497,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022463721204453007,
      "loss": 4.7908,
      "mean_token_accuracy": 0.24193430095911025,
      "num_tokens": 131716877.0,
      "step": 63465
    },
    {
      "entropy": 5.697592639923096,
      "epoch": 5.757438316400581,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000224605677120436,
      "loss": 4.7737,
      "mean_token_accuracy": 0.23327502757310867,
      "num_tokens": 131728061.0,
      "step": 63470
    },
    {
      "entropy": 5.677018404006958,
      "epoch": 5.757891872278664,
      "grad_norm": 1.546875,
      "learning_rate": 0.00022457414323850928,
      "loss": 4.7195,
      "mean_token_accuracy": 0.24035023301839828,
      "num_tokens": 131737679.0,
      "step": 63475
    },
    {
      "entropy": 5.818303012847901,
      "epoch": 5.7583454281567485,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002245426103994022,
      "loss": 4.947,
      "mean_token_accuracy": 0.21759410798549653,
      "num_tokens": 131747570.0,
      "step": 63480
    },
    {
      "entropy": 5.662282133102417,
      "epoch": 5.758798984034833,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002245110786037668,
      "loss": 4.7006,
      "mean_token_accuracy": 0.23677029460668564,
      "num_tokens": 131757216.0,
      "step": 63485
    },
    {
      "entropy": 5.669548559188843,
      "epoch": 5.759252539912917,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022447954785225516,
      "loss": 4.7707,
      "mean_token_accuracy": 0.2314198151230812,
      "num_tokens": 131767964.0,
      "step": 63490
    },
    {
      "entropy": 5.725184345245362,
      "epoch": 5.759706095791001,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022444801814551924,
      "loss": 4.9203,
      "mean_token_accuracy": 0.2213355854153633,
      "num_tokens": 131779155.0,
      "step": 63495
    },
    {
      "entropy": 5.771312475204468,
      "epoch": 5.760159651669086,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002244164894842113,
      "loss": 4.8499,
      "mean_token_accuracy": 0.23723388612270355,
      "num_tokens": 131789091.0,
      "step": 63500
    },
    {
      "entropy": 5.701703405380249,
      "epoch": 5.76061320754717,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022438496186898322,
      "loss": 4.8851,
      "mean_token_accuracy": 0.2255299910902977,
      "num_tokens": 131799205.0,
      "step": 63505
    },
    {
      "entropy": 5.708455991744995,
      "epoch": 5.761066763425254,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022435343530048695,
      "loss": 4.8392,
      "mean_token_accuracy": 0.22796089202165604,
      "num_tokens": 131808803.0,
      "step": 63510
    },
    {
      "entropy": 5.741676139831543,
      "epoch": 5.761520319303338,
      "grad_norm": 1.296875,
      "learning_rate": 0.00022432190977937462,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23007998317480088,
      "num_tokens": 131818682.0,
      "step": 63515
    },
    {
      "entropy": 5.762245988845825,
      "epoch": 5.761973875181423,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022429038530629813,
      "loss": 4.8618,
      "mean_token_accuracy": 0.2222379446029663,
      "num_tokens": 131829705.0,
      "step": 63520
    },
    {
      "entropy": 5.748649930953979,
      "epoch": 5.762427431059507,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022425886188190937,
      "loss": 4.8175,
      "mean_token_accuracy": 0.23511715978384018,
      "num_tokens": 131839206.0,
      "step": 63525
    },
    {
      "entropy": 5.6995893001556395,
      "epoch": 5.76288098693759,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022422733950686025,
      "loss": 4.8285,
      "mean_token_accuracy": 0.2321699008345604,
      "num_tokens": 131848695.0,
      "step": 63530
    },
    {
      "entropy": 5.760444402694702,
      "epoch": 5.763334542815675,
      "grad_norm": 1.375,
      "learning_rate": 0.00022419581818180275,
      "loss": 4.8839,
      "mean_token_accuracy": 0.22410583794116973,
      "num_tokens": 131858443.0,
      "step": 63535
    },
    {
      "entropy": 5.799008798599243,
      "epoch": 5.763788098693759,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022416429790738862,
      "loss": 4.8877,
      "mean_token_accuracy": 0.23391833007335663,
      "num_tokens": 131868280.0,
      "step": 63540
    },
    {
      "entropy": 5.791013431549072,
      "epoch": 5.764241654571843,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022413277868426972,
      "loss": 4.9247,
      "mean_token_accuracy": 0.22420626133680344,
      "num_tokens": 131880013.0,
      "step": 63545
    },
    {
      "entropy": 5.792615222930908,
      "epoch": 5.764695210449927,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022410126051309803,
      "loss": 4.9188,
      "mean_token_accuracy": 0.22403724193573,
      "num_tokens": 131890828.0,
      "step": 63550
    },
    {
      "entropy": 5.753800916671753,
      "epoch": 5.765148766328012,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002240697433945252,
      "loss": 4.8169,
      "mean_token_accuracy": 0.23380247354507447,
      "num_tokens": 131900817.0,
      "step": 63555
    },
    {
      "entropy": 5.7875776290893555,
      "epoch": 5.765602322206096,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000224038227329203,
      "loss": 4.8495,
      "mean_token_accuracy": 0.22641795575618745,
      "num_tokens": 131911237.0,
      "step": 63560
    },
    {
      "entropy": 5.764220714569092,
      "epoch": 5.76605587808418,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002240067123177833,
      "loss": 4.8816,
      "mean_token_accuracy": 0.2275848224759102,
      "num_tokens": 131921429.0,
      "step": 63565
    },
    {
      "entropy": 5.715973949432373,
      "epoch": 5.7665094339622645,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002239751983609178,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23034892678260804,
      "num_tokens": 131932812.0,
      "step": 63570
    },
    {
      "entropy": 5.852840375900269,
      "epoch": 5.766962989840349,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002239436854592582,
      "loss": 4.8734,
      "mean_token_accuracy": 0.22805277556180953,
      "num_tokens": 131943360.0,
      "step": 63575
    },
    {
      "entropy": 5.780640602111816,
      "epoch": 5.767416545718433,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022391217361345612,
      "loss": 4.8365,
      "mean_token_accuracy": 0.22830145210027694,
      "num_tokens": 131953713.0,
      "step": 63580
    },
    {
      "entropy": 5.7203577041625975,
      "epoch": 5.767870101596516,
      "grad_norm": 1.34375,
      "learning_rate": 0.00022388066282416337,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23624206483364105,
      "num_tokens": 131964959.0,
      "step": 63585
    },
    {
      "entropy": 5.773099136352539,
      "epoch": 5.768323657474601,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022384915309203151,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22656911760568618,
      "num_tokens": 131975127.0,
      "step": 63590
    },
    {
      "entropy": 5.73150897026062,
      "epoch": 5.768777213352685,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022381764441771224,
      "loss": 4.8321,
      "mean_token_accuracy": 0.23583097010850906,
      "num_tokens": 131985125.0,
      "step": 63595
    },
    {
      "entropy": 5.742377471923828,
      "epoch": 5.769230769230769,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002237861368018571,
      "loss": 4.8348,
      "mean_token_accuracy": 0.22521643191576005,
      "num_tokens": 131995745.0,
      "step": 63600
    },
    {
      "entropy": 5.750640344619751,
      "epoch": 5.769684325108853,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022375463024511773,
      "loss": 4.8326,
      "mean_token_accuracy": 0.2379637196660042,
      "num_tokens": 132006335.0,
      "step": 63605
    },
    {
      "entropy": 5.770081233978272,
      "epoch": 5.770137880986938,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022372312474814565,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22661036998033524,
      "num_tokens": 132016589.0,
      "step": 63610
    },
    {
      "entropy": 5.795874881744385,
      "epoch": 5.770591436865022,
      "grad_norm": 1.65625,
      "learning_rate": 0.00022369162031159236,
      "loss": 4.8365,
      "mean_token_accuracy": 0.23074077665805817,
      "num_tokens": 132026430.0,
      "step": 63615
    },
    {
      "entropy": 5.774061298370361,
      "epoch": 5.771044992743106,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002236601169361095,
      "loss": 4.8598,
      "mean_token_accuracy": 0.2266649603843689,
      "num_tokens": 132036790.0,
      "step": 63620
    },
    {
      "entropy": 5.7562864303588865,
      "epoch": 5.7714985486211905,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002236286146223485,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22936234921216964,
      "num_tokens": 132047379.0,
      "step": 63625
    },
    {
      "entropy": 5.748395347595215,
      "epoch": 5.771952104499274,
      "grad_norm": 1.5,
      "learning_rate": 0.00022359711337096083,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22583646178245545,
      "num_tokens": 132058628.0,
      "step": 63630
    },
    {
      "entropy": 5.742564487457275,
      "epoch": 5.772405660377358,
      "grad_norm": 1.53125,
      "learning_rate": 0.00022356561318259794,
      "loss": 4.9211,
      "mean_token_accuracy": 0.22687187194824218,
      "num_tokens": 132068133.0,
      "step": 63635
    },
    {
      "entropy": 5.731548213958741,
      "epoch": 5.772859216255442,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002235341140579113,
      "loss": 4.7843,
      "mean_token_accuracy": 0.23727545589208604,
      "num_tokens": 132077813.0,
      "step": 63640
    },
    {
      "entropy": 5.7571028709411625,
      "epoch": 5.773312772133527,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022350261599755227,
      "loss": 4.887,
      "mean_token_accuracy": 0.233127261698246,
      "num_tokens": 132087823.0,
      "step": 63645
    },
    {
      "entropy": 5.831214714050293,
      "epoch": 5.773766328011611,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002234711190021723,
      "loss": 4.9107,
      "mean_token_accuracy": 0.22808133959770202,
      "num_tokens": 132098393.0,
      "step": 63650
    },
    {
      "entropy": 5.7713240623474125,
      "epoch": 5.774219883889695,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002234396230724227,
      "loss": 4.889,
      "mean_token_accuracy": 0.22497794926166534,
      "num_tokens": 132109146.0,
      "step": 63655
    },
    {
      "entropy": 5.823916578292847,
      "epoch": 5.7746734397677795,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002234081282089549,
      "loss": 4.9646,
      "mean_token_accuracy": 0.22259293645620346,
      "num_tokens": 132119286.0,
      "step": 63660
    },
    {
      "entropy": 5.745236587524414,
      "epoch": 5.775126995645864,
      "grad_norm": 1.375,
      "learning_rate": 0.00022337663441242007,
      "loss": 4.7605,
      "mean_token_accuracy": 0.23843995332717896,
      "num_tokens": 132129123.0,
      "step": 63665
    },
    {
      "entropy": 5.7462958812713625,
      "epoch": 5.775580551523948,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002233451416834696,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22476766854524613,
      "num_tokens": 132139935.0,
      "step": 63670
    },
    {
      "entropy": 5.693197870254517,
      "epoch": 5.776034107402032,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022331365002275477,
      "loss": 4.8027,
      "mean_token_accuracy": 0.2361575797200203,
      "num_tokens": 132150407.0,
      "step": 63675
    },
    {
      "entropy": 5.7873743057250975,
      "epoch": 5.7764876632801165,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022328215943092684,
      "loss": 4.8745,
      "mean_token_accuracy": 0.23378397077322005,
      "num_tokens": 132160487.0,
      "step": 63680
    },
    {
      "entropy": 5.8879317283630375,
      "epoch": 5.7769412191582,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022325066990863714,
      "loss": 5.0367,
      "mean_token_accuracy": 0.2131169319152832,
      "num_tokens": 132170868.0,
      "step": 63685
    },
    {
      "entropy": 5.785963487625122,
      "epoch": 5.777394775036284,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022321918145653662,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23769945800304412,
      "num_tokens": 132181774.0,
      "step": 63690
    },
    {
      "entropy": 5.81717882156372,
      "epoch": 5.7778483309143684,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022318769407527668,
      "loss": 4.9085,
      "mean_token_accuracy": 0.23418955653905868,
      "num_tokens": 132191862.0,
      "step": 63695
    },
    {
      "entropy": 5.84136815071106,
      "epoch": 5.778301886792453,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022315620776550843,
      "loss": 4.9761,
      "mean_token_accuracy": 0.21919198483228683,
      "num_tokens": 132202385.0,
      "step": 63700
    },
    {
      "entropy": 5.704436159133911,
      "epoch": 5.778755442670537,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000223124722527883,
      "loss": 4.8199,
      "mean_token_accuracy": 0.2340528905391693,
      "num_tokens": 132213654.0,
      "step": 63705
    },
    {
      "entropy": 5.819426679611206,
      "epoch": 5.779208998548621,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00022309323836305152,
      "loss": 4.9332,
      "mean_token_accuracy": 0.223400342464447,
      "num_tokens": 132224816.0,
      "step": 63710
    },
    {
      "entropy": 5.755114507675171,
      "epoch": 5.7796625544267055,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002230617552716651,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23752570301294326,
      "num_tokens": 132234347.0,
      "step": 63715
    },
    {
      "entropy": 5.740732717514038,
      "epoch": 5.78011611030479,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022303027325437487,
      "loss": 4.7693,
      "mean_token_accuracy": 0.24013743698596954,
      "num_tokens": 132244413.0,
      "step": 63720
    },
    {
      "entropy": 5.790129804611206,
      "epoch": 5.780569666182874,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022299879231183173,
      "loss": 4.8643,
      "mean_token_accuracy": 0.2330336257815361,
      "num_tokens": 132254093.0,
      "step": 63725
    },
    {
      "entropy": 5.725061511993408,
      "epoch": 5.781023222060957,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022296731244468693,
      "loss": 4.7959,
      "mean_token_accuracy": 0.23112425059080124,
      "num_tokens": 132264449.0,
      "step": 63730
    },
    {
      "entropy": 5.748254680633545,
      "epoch": 5.781476777939043,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022293583365359131,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23458662629127502,
      "num_tokens": 132274784.0,
      "step": 63735
    },
    {
      "entropy": 5.746340227127075,
      "epoch": 5.781930333817126,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002229043559391959,
      "loss": 4.8029,
      "mean_token_accuracy": 0.23050864040851593,
      "num_tokens": 132284289.0,
      "step": 63740
    },
    {
      "entropy": 5.742184209823608,
      "epoch": 5.78238388969521,
      "grad_norm": 1.578125,
      "learning_rate": 0.00022287287930215167,
      "loss": 4.8319,
      "mean_token_accuracy": 0.23475530445575715,
      "num_tokens": 132293749.0,
      "step": 63745
    },
    {
      "entropy": 5.6839052677154545,
      "epoch": 5.7828374455732945,
      "grad_norm": 1.3125,
      "learning_rate": 0.00022284140374310963,
      "loss": 4.7665,
      "mean_token_accuracy": 0.23489245623350144,
      "num_tokens": 132304129.0,
      "step": 63750
    },
    {
      "entropy": 5.733480834960938,
      "epoch": 5.783291001451379,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022280992926272065,
      "loss": 4.8197,
      "mean_token_accuracy": 0.2372421219944954,
      "num_tokens": 132316102.0,
      "step": 63755
    },
    {
      "entropy": 5.759804773330688,
      "epoch": 5.783744557329463,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022277845586163558,
      "loss": 4.88,
      "mean_token_accuracy": 0.22767960727214814,
      "num_tokens": 132327421.0,
      "step": 63760
    },
    {
      "entropy": 5.826210308074951,
      "epoch": 5.784198113207547,
      "grad_norm": 1.375,
      "learning_rate": 0.00022274698354050543,
      "loss": 4.9111,
      "mean_token_accuracy": 0.22454860508441926,
      "num_tokens": 132338285.0,
      "step": 63765
    },
    {
      "entropy": 5.733132076263428,
      "epoch": 5.7846516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022271551229998093,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23194511532783507,
      "num_tokens": 132348288.0,
      "step": 63770
    },
    {
      "entropy": 5.737843132019043,
      "epoch": 5.785105224963716,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022268404214071297,
      "loss": 4.8191,
      "mean_token_accuracy": 0.22879954874515535,
      "num_tokens": 132358225.0,
      "step": 63775
    },
    {
      "entropy": 5.783758544921875,
      "epoch": 5.7855587808418,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022265257306335235,
      "loss": 4.9249,
      "mean_token_accuracy": 0.21918977797031403,
      "num_tokens": 132369000.0,
      "step": 63780
    },
    {
      "entropy": 5.637060117721558,
      "epoch": 5.7860123367198835,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022262110506854988,
      "loss": 4.7801,
      "mean_token_accuracy": 0.23730121552944183,
      "num_tokens": 132379461.0,
      "step": 63785
    },
    {
      "entropy": 5.737604093551636,
      "epoch": 5.786465892597968,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00022258963815695627,
      "loss": 4.839,
      "mean_token_accuracy": 0.22950417846441268,
      "num_tokens": 132389869.0,
      "step": 63790
    },
    {
      "entropy": 5.744702577590942,
      "epoch": 5.786919448476052,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022255817232922227,
      "loss": 4.8264,
      "mean_token_accuracy": 0.2333090201020241,
      "num_tokens": 132399463.0,
      "step": 63795
    },
    {
      "entropy": 5.767023038864136,
      "epoch": 5.787373004354136,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022252670758599868,
      "loss": 4.9069,
      "mean_token_accuracy": 0.22307186126708983,
      "num_tokens": 132410633.0,
      "step": 63800
    },
    {
      "entropy": 5.766717910766602,
      "epoch": 5.7878265602322205,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002224952439279362,
      "loss": 4.8647,
      "mean_token_accuracy": 0.23005347102880477,
      "num_tokens": 132421340.0,
      "step": 63805
    },
    {
      "entropy": 5.738311910629273,
      "epoch": 5.788280116110305,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0002224637813556853,
      "loss": 4.8523,
      "mean_token_accuracy": 0.2282591313123703,
      "num_tokens": 132430870.0,
      "step": 63810
    },
    {
      "entropy": 5.731258964538574,
      "epoch": 5.788733671988389,
      "grad_norm": 1.5,
      "learning_rate": 0.00022243231986989693,
      "loss": 4.7361,
      "mean_token_accuracy": 0.2389766052365303,
      "num_tokens": 132441120.0,
      "step": 63815
    },
    {
      "entropy": 5.747881937026977,
      "epoch": 5.789187227866473,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022240085947122153,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24033200293779372,
      "num_tokens": 132451149.0,
      "step": 63820
    },
    {
      "entropy": 5.7203270435333256,
      "epoch": 5.789640783744558,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022236940016030977,
      "loss": 4.8377,
      "mean_token_accuracy": 0.23665518313646317,
      "num_tokens": 132461566.0,
      "step": 63825
    },
    {
      "entropy": 5.724222898483276,
      "epoch": 5.790094339622642,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002223379419378122,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23649505972862245,
      "num_tokens": 132471486.0,
      "step": 63830
    },
    {
      "entropy": 5.761337089538574,
      "epoch": 5.790547895500726,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00022230648480437942,
      "loss": 4.8411,
      "mean_token_accuracy": 0.23242529332637787,
      "num_tokens": 132482391.0,
      "step": 63835
    },
    {
      "entropy": 5.749511003494263,
      "epoch": 5.7910014513788095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022227502876066203,
      "loss": 4.843,
      "mean_token_accuracy": 0.2368449464440346,
      "num_tokens": 132492235.0,
      "step": 63840
    },
    {
      "entropy": 5.785829210281372,
      "epoch": 5.791455007256894,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022224357380731035,
      "loss": 4.9332,
      "mean_token_accuracy": 0.22777434289455414,
      "num_tokens": 132501984.0,
      "step": 63845
    },
    {
      "entropy": 5.759360504150391,
      "epoch": 5.791908563134978,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002222121199449751,
      "loss": 4.8036,
      "mean_token_accuracy": 0.2344875916838646,
      "num_tokens": 132512608.0,
      "step": 63850
    },
    {
      "entropy": 5.731854248046875,
      "epoch": 5.792362119013062,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022218066717430662,
      "loss": 4.7392,
      "mean_token_accuracy": 0.24065372049808503,
      "num_tokens": 132522682.0,
      "step": 63855
    },
    {
      "entropy": 5.779045486450196,
      "epoch": 5.792815674891147,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0002221492154959554,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23344658464193344,
      "num_tokens": 132531680.0,
      "step": 63860
    },
    {
      "entropy": 5.747996425628662,
      "epoch": 5.793269230769231,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002221177649105719,
      "loss": 4.8082,
      "mean_token_accuracy": 0.23274241983890534,
      "num_tokens": 132542149.0,
      "step": 63865
    },
    {
      "entropy": 5.715740776062011,
      "epoch": 5.793722786647315,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00022208631541880653,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22788792997598648,
      "num_tokens": 132552249.0,
      "step": 63870
    },
    {
      "entropy": 5.776061964035034,
      "epoch": 5.794176342525399,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002220548670213096,
      "loss": 4.8622,
      "mean_token_accuracy": 0.23719941079616547,
      "num_tokens": 132562471.0,
      "step": 63875
    },
    {
      "entropy": 5.761414241790772,
      "epoch": 5.794629898403484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022202341971873152,
      "loss": 4.8783,
      "mean_token_accuracy": 0.23128251135349273,
      "num_tokens": 132572357.0,
      "step": 63880
    },
    {
      "entropy": 5.7178491115570065,
      "epoch": 5.795083454281567,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022199197351172262,
      "loss": 4.8929,
      "mean_token_accuracy": 0.23033079504966736,
      "num_tokens": 132582020.0,
      "step": 63885
    },
    {
      "entropy": 5.785088443756104,
      "epoch": 5.795537010159651,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022196052840093322,
      "loss": 4.9274,
      "mean_token_accuracy": 0.2223281055688858,
      "num_tokens": 132592018.0,
      "step": 63890
    },
    {
      "entropy": 5.749674558639526,
      "epoch": 5.7959905660377355,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002219290843870136,
      "loss": 4.8279,
      "mean_token_accuracy": 0.2332886502146721,
      "num_tokens": 132602433.0,
      "step": 63895
    },
    {
      "entropy": 5.672652769088745,
      "epoch": 5.79644412191582,
      "grad_norm": 5.28125,
      "learning_rate": 0.00022189764147061402,
      "loss": 4.8065,
      "mean_token_accuracy": 0.24251690357923508,
      "num_tokens": 132612633.0,
      "step": 63900
    },
    {
      "entropy": 5.7341893196105955,
      "epoch": 5.796897677793904,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002218661996523848,
      "loss": 4.8738,
      "mean_token_accuracy": 0.2283213496208191,
      "num_tokens": 132622432.0,
      "step": 63905
    },
    {
      "entropy": 5.737374782562256,
      "epoch": 5.797351233671988,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002218347589329761,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22728462219238282,
      "num_tokens": 132632770.0,
      "step": 63910
    },
    {
      "entropy": 5.674747133255005,
      "epoch": 5.797804789550073,
      "grad_norm": 1.4375,
      "learning_rate": 0.00022180331931303805,
      "loss": 4.7678,
      "mean_token_accuracy": 0.23692586421966552,
      "num_tokens": 132642745.0,
      "step": 63915
    },
    {
      "entropy": 5.647774791717529,
      "epoch": 5.798258345428157,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022177188079322102,
      "loss": 4.809,
      "mean_token_accuracy": 0.23140082508325577,
      "num_tokens": 132653023.0,
      "step": 63920
    },
    {
      "entropy": 5.780079936981201,
      "epoch": 5.798711901306241,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022174044337417505,
      "loss": 4.9527,
      "mean_token_accuracy": 0.22971645891666412,
      "num_tokens": 132662143.0,
      "step": 63925
    },
    {
      "entropy": 5.790542268753052,
      "epoch": 5.799165457184325,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022170900705655017,
      "loss": 4.8877,
      "mean_token_accuracy": 0.2223242774605751,
      "num_tokens": 132671541.0,
      "step": 63930
    },
    {
      "entropy": 5.725984191894531,
      "epoch": 5.79961901306241,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022167757184099669,
      "loss": 4.8292,
      "mean_token_accuracy": 0.2337482362985611,
      "num_tokens": 132681660.0,
      "step": 63935
    },
    {
      "entropy": 5.7451643466949465,
      "epoch": 5.800072568940493,
      "grad_norm": 1.375,
      "learning_rate": 0.0002216461377281646,
      "loss": 4.8439,
      "mean_token_accuracy": 0.23854232728481292,
      "num_tokens": 132692625.0,
      "step": 63940
    },
    {
      "entropy": 5.709821510314941,
      "epoch": 5.800526124818577,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022161470471870404,
      "loss": 4.7888,
      "mean_token_accuracy": 0.2368643432855606,
      "num_tokens": 132703082.0,
      "step": 63945
    },
    {
      "entropy": 5.7752281665802006,
      "epoch": 5.800979680696662,
      "grad_norm": 1.328125,
      "learning_rate": 0.00022158327281326484,
      "loss": 4.8526,
      "mean_token_accuracy": 0.23383124619722367,
      "num_tokens": 132713434.0,
      "step": 63950
    },
    {
      "entropy": 5.751251697540283,
      "epoch": 5.801433236574746,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00022155184201249728,
      "loss": 4.8466,
      "mean_token_accuracy": 0.22673978209495543,
      "num_tokens": 132724266.0,
      "step": 63955
    },
    {
      "entropy": 5.797741937637329,
      "epoch": 5.80188679245283,
      "grad_norm": 1.375,
      "learning_rate": 0.00022152041231705128,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23520492166280746,
      "num_tokens": 132734417.0,
      "step": 63960
    },
    {
      "entropy": 5.736817932128906,
      "epoch": 5.802340348330914,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002214889837275767,
      "loss": 4.8402,
      "mean_token_accuracy": 0.23349838554859162,
      "num_tokens": 132745596.0,
      "step": 63965
    },
    {
      "entropy": 5.742407035827637,
      "epoch": 5.802793904208999,
      "grad_norm": 1.484375,
      "learning_rate": 0.00022145755624472364,
      "loss": 4.8182,
      "mean_token_accuracy": 0.23429275155067444,
      "num_tokens": 132754852.0,
      "step": 63970
    },
    {
      "entropy": 5.763727474212646,
      "epoch": 5.803247460087083,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002214261298691419,
      "loss": 4.8718,
      "mean_token_accuracy": 0.22624333947896957,
      "num_tokens": 132767341.0,
      "step": 63975
    },
    {
      "entropy": 5.679670190811157,
      "epoch": 5.803701015965167,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002213947046014815,
      "loss": 4.713,
      "mean_token_accuracy": 0.24672281444072724,
      "num_tokens": 132777372.0,
      "step": 63980
    },
    {
      "entropy": 5.74154314994812,
      "epoch": 5.8041545718432515,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00022136328044239218,
      "loss": 4.772,
      "mean_token_accuracy": 0.23050163984298705,
      "num_tokens": 132786845.0,
      "step": 63985
    },
    {
      "entropy": 5.791355752944947,
      "epoch": 5.804608127721336,
      "grad_norm": 1.4375,
      "learning_rate": 0.000221331857392524,
      "loss": 4.9034,
      "mean_token_accuracy": 0.2285320281982422,
      "num_tokens": 132796699.0,
      "step": 63990
    },
    {
      "entropy": 5.746911573410034,
      "epoch": 5.805061683599419,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022130043545252664,
      "loss": 4.8371,
      "mean_token_accuracy": 0.23832879960536957,
      "num_tokens": 132806448.0,
      "step": 63995
    },
    {
      "entropy": 5.758840322494507,
      "epoch": 5.805515239477503,
      "grad_norm": 1.375,
      "learning_rate": 0.00022126901462304993,
      "loss": 4.8058,
      "mean_token_accuracy": 0.23287999927997588,
      "num_tokens": 132815570.0,
      "step": 64000
    },
    {
      "entropy": 5.655071830749511,
      "epoch": 5.805968795355588,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002212375949047437,
      "loss": 4.7543,
      "mean_token_accuracy": 0.23524679243564606,
      "num_tokens": 132825524.0,
      "step": 64005
    },
    {
      "entropy": 5.758178949356079,
      "epoch": 5.806422351233672,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00022120617629825773,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23274971544742584,
      "num_tokens": 132836312.0,
      "step": 64010
    },
    {
      "entropy": 5.656079816818237,
      "epoch": 5.806875907111756,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002211747588042417,
      "loss": 4.728,
      "mean_token_accuracy": 0.238116654753685,
      "num_tokens": 132845897.0,
      "step": 64015
    },
    {
      "entropy": 5.726676988601684,
      "epoch": 5.80732946298984,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022114334242334544,
      "loss": 4.7965,
      "mean_token_accuracy": 0.23969653844833375,
      "num_tokens": 132855882.0,
      "step": 64020
    },
    {
      "entropy": 5.726904439926147,
      "epoch": 5.807783018867925,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00022111192715621857,
      "loss": 4.8206,
      "mean_token_accuracy": 0.23517441302537917,
      "num_tokens": 132865989.0,
      "step": 64025
    },
    {
      "entropy": 5.767388963699341,
      "epoch": 5.808236574746009,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022108051300351078,
      "loss": 4.8712,
      "mean_token_accuracy": 0.23542069345712663,
      "num_tokens": 132876314.0,
      "step": 64030
    },
    {
      "entropy": 5.76947922706604,
      "epoch": 5.808690130624093,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002210490999658717,
      "loss": 4.7875,
      "mean_token_accuracy": 0.2382005885243416,
      "num_tokens": 132885498.0,
      "step": 64035
    },
    {
      "entropy": 5.709338140487671,
      "epoch": 5.809143686502177,
      "grad_norm": 1.515625,
      "learning_rate": 0.00022101768804395094,
      "loss": 4.7716,
      "mean_token_accuracy": 0.23217172622680665,
      "num_tokens": 132895389.0,
      "step": 64040
    },
    {
      "entropy": 5.762001705169678,
      "epoch": 5.809597242380261,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00022098627723839826,
      "loss": 4.821,
      "mean_token_accuracy": 0.23608357906341554,
      "num_tokens": 132905332.0,
      "step": 64045
    },
    {
      "entropy": 5.7451671123504635,
      "epoch": 5.810050798258345,
      "grad_norm": 1.375,
      "learning_rate": 0.000220954867549863,
      "loss": 4.8599,
      "mean_token_accuracy": 0.21790256053209306,
      "num_tokens": 132915402.0,
      "step": 64050
    },
    {
      "entropy": 5.745922994613648,
      "epoch": 5.810504354136429,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002209234589789949,
      "loss": 4.874,
      "mean_token_accuracy": 0.230851948261261,
      "num_tokens": 132925117.0,
      "step": 64055
    },
    {
      "entropy": 5.763046360015869,
      "epoch": 5.810957910014514,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022089205152644347,
      "loss": 4.8559,
      "mean_token_accuracy": 0.22360205203294753,
      "num_tokens": 132935163.0,
      "step": 64060
    },
    {
      "entropy": 5.8097920417785645,
      "epoch": 5.811411465892598,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002208606451928582,
      "loss": 4.9149,
      "mean_token_accuracy": 0.22924504727125167,
      "num_tokens": 132944971.0,
      "step": 64065
    },
    {
      "entropy": 5.818258428573609,
      "epoch": 5.811865021770682,
      "grad_norm": 1.5,
      "learning_rate": 0.0002208292399788886,
      "loss": 4.8973,
      "mean_token_accuracy": 0.22384159564971923,
      "num_tokens": 132955199.0,
      "step": 64070
    },
    {
      "entropy": 5.803219223022461,
      "epoch": 5.8123185776487665,
      "grad_norm": 1.375,
      "learning_rate": 0.0002207978358851841,
      "loss": 4.9193,
      "mean_token_accuracy": 0.22526053488254547,
      "num_tokens": 132966144.0,
      "step": 64075
    },
    {
      "entropy": 5.85648627281189,
      "epoch": 5.812772133526851,
      "grad_norm": 1.421875,
      "learning_rate": 0.00022076643291239417,
      "loss": 4.9074,
      "mean_token_accuracy": 0.22817719876766204,
      "num_tokens": 132976866.0,
      "step": 64080
    },
    {
      "entropy": 5.792990350723267,
      "epoch": 5.813225689404935,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002207350310611682,
      "loss": 4.9072,
      "mean_token_accuracy": 0.2269338071346283,
      "num_tokens": 132988178.0,
      "step": 64085
    },
    {
      "entropy": 5.781523656845093,
      "epoch": 5.813679245283019,
      "grad_norm": 1.359375,
      "learning_rate": 0.00022070363033215563,
      "loss": 4.9119,
      "mean_token_accuracy": 0.2301834613084793,
      "num_tokens": 132998631.0,
      "step": 64090
    },
    {
      "entropy": 5.763556051254272,
      "epoch": 5.814132801161103,
      "grad_norm": 1.390625,
      "learning_rate": 0.00022067223072600584,
      "loss": 4.8573,
      "mean_token_accuracy": 0.23033640384674073,
      "num_tokens": 133008864.0,
      "step": 64095
    },
    {
      "entropy": 5.705760765075683,
      "epoch": 5.814586357039187,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002206408322433681,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2245801642537117,
      "num_tokens": 133019771.0,
      "step": 64100
    },
    {
      "entropy": 5.800080966949463,
      "epoch": 5.815039912917271,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022060943488489188,
      "loss": 4.8963,
      "mean_token_accuracy": 0.22501684576272965,
      "num_tokens": 133030123.0,
      "step": 64105
    },
    {
      "entropy": 5.766630411148071,
      "epoch": 5.8154934687953554,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022057803865122645,
      "loss": 4.8135,
      "mean_token_accuracy": 0.2242105096578598,
      "num_tokens": 133040104.0,
      "step": 64110
    },
    {
      "entropy": 5.754782581329346,
      "epoch": 5.81594702467344,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000220546643543021,
      "loss": 4.7688,
      "mean_token_accuracy": 0.23807352036237717,
      "num_tokens": 133049594.0,
      "step": 64115
    },
    {
      "entropy": 5.700023889541626,
      "epoch": 5.816400580551524,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00022051524956092483,
      "loss": 4.7502,
      "mean_token_accuracy": 0.24338564723730088,
      "num_tokens": 133060233.0,
      "step": 64120
    },
    {
      "entropy": 5.7867943286895756,
      "epoch": 5.816854136429608,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002204838567055872,
      "loss": 4.8978,
      "mean_token_accuracy": 0.22507301568984986,
      "num_tokens": 133070486.0,
      "step": 64125
    },
    {
      "entropy": 5.784095001220703,
      "epoch": 5.8173076923076925,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022045246497765726,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2289177283644676,
      "num_tokens": 133080432.0,
      "step": 64130
    },
    {
      "entropy": 5.7441963195800785,
      "epoch": 5.817761248185777,
      "grad_norm": 1.6875,
      "learning_rate": 0.00022042107437778426,
      "loss": 4.8464,
      "mean_token_accuracy": 0.23522463738918303,
      "num_tokens": 133090349.0,
      "step": 64135
    },
    {
      "entropy": 5.730401468276978,
      "epoch": 5.818214804063861,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00022038968490661738,
      "loss": 4.7709,
      "mean_token_accuracy": 0.2399880349636078,
      "num_tokens": 133101055.0,
      "step": 64140
    },
    {
      "entropy": 5.755735778808594,
      "epoch": 5.818668359941945,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022035829656480565,
      "loss": 4.8314,
      "mean_token_accuracy": 0.22649094015359877,
      "num_tokens": 133111185.0,
      "step": 64145
    },
    {
      "entropy": 5.751525735855102,
      "epoch": 5.819121915820029,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00022032690935299838,
      "loss": 4.8492,
      "mean_token_accuracy": 0.2227715566754341,
      "num_tokens": 133121567.0,
      "step": 64150
    },
    {
      "entropy": 5.696672344207764,
      "epoch": 5.819575471698113,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002202955232718445,
      "loss": 4.756,
      "mean_token_accuracy": 0.24890464842319487,
      "num_tokens": 133132306.0,
      "step": 64155
    },
    {
      "entropy": 5.733574676513672,
      "epoch": 5.820029027576197,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002202641383219931,
      "loss": 4.8305,
      "mean_token_accuracy": 0.2366074487566948,
      "num_tokens": 133143418.0,
      "step": 64160
    },
    {
      "entropy": 5.767370128631592,
      "epoch": 5.8204825834542815,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00022023275450409335,
      "loss": 4.9,
      "mean_token_accuracy": 0.22040922641754152,
      "num_tokens": 133154420.0,
      "step": 64165
    },
    {
      "entropy": 5.742401456832885,
      "epoch": 5.820936139332366,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00022020137181879413,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23113688975572586,
      "num_tokens": 133163557.0,
      "step": 64170
    },
    {
      "entropy": 5.755724048614502,
      "epoch": 5.82138969521045,
      "grad_norm": 1.453125,
      "learning_rate": 0.00022016999026674457,
      "loss": 4.7632,
      "mean_token_accuracy": 0.23679130524396896,
      "num_tokens": 133173902.0,
      "step": 64175
    },
    {
      "entropy": 5.750830364227295,
      "epoch": 5.821843251088534,
      "grad_norm": 1.375,
      "learning_rate": 0.00022013860984859346,
      "loss": 4.7658,
      "mean_token_accuracy": 0.2441149353981018,
      "num_tokens": 133183488.0,
      "step": 64180
    },
    {
      "entropy": 5.761875867843628,
      "epoch": 5.8222968069666186,
      "grad_norm": 1.40625,
      "learning_rate": 0.00022010723056499,
      "loss": 4.8595,
      "mean_token_accuracy": 0.22835520505905152,
      "num_tokens": 133194014.0,
      "step": 64185
    },
    {
      "entropy": 5.8144759178161625,
      "epoch": 5.822750362844703,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000220075852416583,
      "loss": 4.9091,
      "mean_token_accuracy": 0.22934829592704772,
      "num_tokens": 133204019.0,
      "step": 64190
    },
    {
      "entropy": 5.726443958282471,
      "epoch": 5.823203918722786,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002200444754040213,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2386443793773651,
      "num_tokens": 133214290.0,
      "step": 64195
    },
    {
      "entropy": 5.71792950630188,
      "epoch": 5.8236574746008705,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002200130995279539,
      "loss": 4.8117,
      "mean_token_accuracy": 0.23205435872077942,
      "num_tokens": 133225102.0,
      "step": 64200
    },
    {
      "entropy": 5.728876447677612,
      "epoch": 5.824111030478955,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002199817247890296,
      "loss": 4.8585,
      "mean_token_accuracy": 0.22975656390190125,
      "num_tokens": 133235683.0,
      "step": 64205
    },
    {
      "entropy": 5.790160655975342,
      "epoch": 5.824564586357039,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021995035118789724,
      "loss": 4.8849,
      "mean_token_accuracy": 0.22715161591768265,
      "num_tokens": 133247664.0,
      "step": 64210
    },
    {
      "entropy": 5.73052978515625,
      "epoch": 5.825018142235123,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021991897872520566,
      "loss": 4.7771,
      "mean_token_accuracy": 0.24622739106416702,
      "num_tokens": 133257686.0,
      "step": 64215
    },
    {
      "entropy": 5.799735736846924,
      "epoch": 5.8254716981132075,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021988760740160364,
      "loss": 4.8224,
      "mean_token_accuracy": 0.23428070843219756,
      "num_tokens": 133267295.0,
      "step": 64220
    },
    {
      "entropy": 5.706882286071777,
      "epoch": 5.825925253991292,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021985623721773996,
      "loss": 4.712,
      "mean_token_accuracy": 0.2407909333705902,
      "num_tokens": 133277540.0,
      "step": 64225
    },
    {
      "entropy": 5.724161815643311,
      "epoch": 5.826378809869376,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021982486817426328,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23688258826732636,
      "num_tokens": 133288160.0,
      "step": 64230
    },
    {
      "entropy": 5.738337898254395,
      "epoch": 5.82683236574746,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021979350027182242,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23768595457077027,
      "num_tokens": 133298648.0,
      "step": 64235
    },
    {
      "entropy": 5.770078563690186,
      "epoch": 5.827285921625545,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021976213351106605,
      "loss": 4.8504,
      "mean_token_accuracy": 0.23433684408664704,
      "num_tokens": 133309320.0,
      "step": 64240
    },
    {
      "entropy": 5.778469800949097,
      "epoch": 5.827739477503629,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021973076789264285,
      "loss": 4.9232,
      "mean_token_accuracy": 0.22484929710626603,
      "num_tokens": 133318948.0,
      "step": 64245
    },
    {
      "entropy": 5.747474336624146,
      "epoch": 5.828193033381712,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002196994034172014,
      "loss": 4.8151,
      "mean_token_accuracy": 0.2280384749174118,
      "num_tokens": 133328573.0,
      "step": 64250
    },
    {
      "entropy": 5.708255100250244,
      "epoch": 5.8286465892597965,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021966804008539042,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23602599203586577,
      "num_tokens": 133338503.0,
      "step": 64255
    },
    {
      "entropy": 5.732471561431884,
      "epoch": 5.829100145137881,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021963667789785853,
      "loss": 4.8509,
      "mean_token_accuracy": 0.2312191143631935,
      "num_tokens": 133349733.0,
      "step": 64260
    },
    {
      "entropy": 5.743278312683105,
      "epoch": 5.829553701015965,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021960531685525413,
      "loss": 4.8541,
      "mean_token_accuracy": 0.23038170635700225,
      "num_tokens": 133360639.0,
      "step": 64265
    },
    {
      "entropy": 5.770643138885498,
      "epoch": 5.830007256894049,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021957395695822597,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23085969239473342,
      "num_tokens": 133370678.0,
      "step": 64270
    },
    {
      "entropy": 5.808866024017334,
      "epoch": 5.830460812772134,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021954259820742251,
      "loss": 4.8861,
      "mean_token_accuracy": 0.2348012775182724,
      "num_tokens": 133382140.0,
      "step": 64275
    },
    {
      "entropy": 5.74003529548645,
      "epoch": 5.830914368650218,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021951124060349221,
      "loss": 4.8214,
      "mean_token_accuracy": 0.22957319766283035,
      "num_tokens": 133392152.0,
      "step": 64280
    },
    {
      "entropy": 5.828607082366943,
      "epoch": 5.831367924528302,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021947988414708358,
      "loss": 4.9503,
      "mean_token_accuracy": 0.21988611966371535,
      "num_tokens": 133402543.0,
      "step": 64285
    },
    {
      "entropy": 5.829240322113037,
      "epoch": 5.831821480406386,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002194485288388452,
      "loss": 4.9295,
      "mean_token_accuracy": 0.22513594776391982,
      "num_tokens": 133413418.0,
      "step": 64290
    },
    {
      "entropy": 5.724407052993774,
      "epoch": 5.83227503628447,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002194171746794253,
      "loss": 4.8733,
      "mean_token_accuracy": 0.22959961742162704,
      "num_tokens": 133424342.0,
      "step": 64295
    },
    {
      "entropy": 5.850616216659546,
      "epoch": 5.832728592162555,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002193858216694724,
      "loss": 5.0066,
      "mean_token_accuracy": 0.21624718457460404,
      "num_tokens": 133435403.0,
      "step": 64300
    },
    {
      "entropy": 5.786234331130982,
      "epoch": 5.833182148040638,
      "grad_norm": 1.375,
      "learning_rate": 0.0002193544698096349,
      "loss": 4.9628,
      "mean_token_accuracy": 0.22568301856517792,
      "num_tokens": 133446267.0,
      "step": 64305
    },
    {
      "entropy": 5.770062065124511,
      "epoch": 5.8336357039187225,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021932311910056117,
      "loss": 4.8586,
      "mean_token_accuracy": 0.23133768290281295,
      "num_tokens": 133457156.0,
      "step": 64310
    },
    {
      "entropy": 5.750090456008911,
      "epoch": 5.834089259796807,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021929176954289947,
      "loss": 4.8348,
      "mean_token_accuracy": 0.23089662343263626,
      "num_tokens": 133467583.0,
      "step": 64315
    },
    {
      "entropy": 5.802462768554688,
      "epoch": 5.834542815674891,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021926042113729822,
      "loss": 4.9161,
      "mean_token_accuracy": 0.2201314315199852,
      "num_tokens": 133478019.0,
      "step": 64320
    },
    {
      "entropy": 5.692511320114136,
      "epoch": 5.834996371552975,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021922907388440566,
      "loss": 4.7772,
      "mean_token_accuracy": 0.24008396565914153,
      "num_tokens": 133487521.0,
      "step": 64325
    },
    {
      "entropy": 5.747450876235962,
      "epoch": 5.83544992743106,
      "grad_norm": 1.5,
      "learning_rate": 0.00021919772778487007,
      "loss": 4.8894,
      "mean_token_accuracy": 0.22834982275962828,
      "num_tokens": 133498453.0,
      "step": 64330
    },
    {
      "entropy": 5.788299703598023,
      "epoch": 5.835903483309144,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002191663828393396,
      "loss": 4.8944,
      "mean_token_accuracy": 0.2261227160692215,
      "num_tokens": 133509234.0,
      "step": 64335
    },
    {
      "entropy": 5.760857725143433,
      "epoch": 5.836357039187228,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021913503904846265,
      "loss": 4.8671,
      "mean_token_accuracy": 0.22749703228473664,
      "num_tokens": 133519983.0,
      "step": 64340
    },
    {
      "entropy": 5.76949987411499,
      "epoch": 5.836810595065312,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021910369641288735,
      "loss": 4.8051,
      "mean_token_accuracy": 0.2292675644159317,
      "num_tokens": 133530294.0,
      "step": 64345
    },
    {
      "entropy": 5.754980802536011,
      "epoch": 5.837264150943396,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021907235493326177,
      "loss": 4.8539,
      "mean_token_accuracy": 0.23609709590673447,
      "num_tokens": 133539979.0,
      "step": 64350
    },
    {
      "entropy": 5.745118761062622,
      "epoch": 5.83771770682148,
      "grad_norm": 1.375,
      "learning_rate": 0.00021904101461023417,
      "loss": 4.8582,
      "mean_token_accuracy": 0.2332402214407921,
      "num_tokens": 133550729.0,
      "step": 64355
    },
    {
      "entropy": 5.7421270370483395,
      "epoch": 5.838171262699564,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021900967544445272,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23435864299535752,
      "num_tokens": 133561143.0,
      "step": 64360
    },
    {
      "entropy": 5.766309976577759,
      "epoch": 5.838624818577649,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021897833743656536,
      "loss": 4.8538,
      "mean_token_accuracy": 0.24184090346097947,
      "num_tokens": 133572039.0,
      "step": 64365
    },
    {
      "entropy": 5.72725739479065,
      "epoch": 5.839078374455733,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021894700058722028,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23162374645471573,
      "num_tokens": 133581206.0,
      "step": 64370
    },
    {
      "entropy": 5.712641048431396,
      "epoch": 5.839531930333817,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021891566489706555,
      "loss": 4.8397,
      "mean_token_accuracy": 0.22844965755939484,
      "num_tokens": 133591724.0,
      "step": 64375
    },
    {
      "entropy": 5.778211212158203,
      "epoch": 5.839985486211901,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021888433036674915,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22242044955492019,
      "num_tokens": 133602282.0,
      "step": 64380
    },
    {
      "entropy": 5.7288477420806885,
      "epoch": 5.840439042089986,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000218852996996919,
      "loss": 4.8132,
      "mean_token_accuracy": 0.23765401542186737,
      "num_tokens": 133612389.0,
      "step": 64385
    },
    {
      "entropy": 5.77398157119751,
      "epoch": 5.84089259796807,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002188216647882233,
      "loss": 4.8647,
      "mean_token_accuracy": 0.2240688055753708,
      "num_tokens": 133622059.0,
      "step": 64390
    },
    {
      "entropy": 5.793232440948486,
      "epoch": 5.841346153846154,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021879033374130985,
      "loss": 4.8524,
      "mean_token_accuracy": 0.23049405515193938,
      "num_tokens": 133632423.0,
      "step": 64395
    },
    {
      "entropy": 5.764190292358398,
      "epoch": 5.8417997097242385,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021875900385682663,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23581282794475555,
      "num_tokens": 133643197.0,
      "step": 64400
    },
    {
      "entropy": 5.762211465835572,
      "epoch": 5.842253265602322,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021872767513542152,
      "loss": 4.87,
      "mean_token_accuracy": 0.22855498790740966,
      "num_tokens": 133653313.0,
      "step": 64405
    },
    {
      "entropy": 5.805374097824097,
      "epoch": 5.842706821480406,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021869634757774243,
      "loss": 4.8782,
      "mean_token_accuracy": 0.22676799744367598,
      "num_tokens": 133663082.0,
      "step": 64410
    },
    {
      "entropy": 5.714051580429077,
      "epoch": 5.84316037735849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021866502118443726,
      "loss": 4.7762,
      "mean_token_accuracy": 0.23267871737480164,
      "num_tokens": 133673213.0,
      "step": 64415
    },
    {
      "entropy": 5.738676834106445,
      "epoch": 5.843613933236575,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021863369595615374,
      "loss": 4.7513,
      "mean_token_accuracy": 0.23466638326644898,
      "num_tokens": 133682779.0,
      "step": 64420
    },
    {
      "entropy": 5.722029542922973,
      "epoch": 5.844067489114659,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002186023718935398,
      "loss": 4.8157,
      "mean_token_accuracy": 0.22786663025617598,
      "num_tokens": 133693404.0,
      "step": 64425
    },
    {
      "entropy": 5.704804563522339,
      "epoch": 5.844521044992743,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021857104899724323,
      "loss": 4.7897,
      "mean_token_accuracy": 0.2355131044983864,
      "num_tokens": 133703621.0,
      "step": 64430
    },
    {
      "entropy": 5.801505708694458,
      "epoch": 5.844974600870827,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002185397272679116,
      "loss": 4.9265,
      "mean_token_accuracy": 0.22538470923900605,
      "num_tokens": 133714944.0,
      "step": 64435
    },
    {
      "entropy": 5.7718470096588135,
      "epoch": 5.845428156748912,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021850840670619294,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2232121631503105,
      "num_tokens": 133725434.0,
      "step": 64440
    },
    {
      "entropy": 5.629910707473755,
      "epoch": 5.845881712626996,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021847708731273475,
      "loss": 4.6852,
      "mean_token_accuracy": 0.24573176801204683,
      "num_tokens": 133736629.0,
      "step": 64445
    },
    {
      "entropy": 5.770943641662598,
      "epoch": 5.846335268505079,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00021844576908818487,
      "loss": 4.8412,
      "mean_token_accuracy": 0.2352729082107544,
      "num_tokens": 133746967.0,
      "step": 64450
    },
    {
      "entropy": 5.787076616287232,
      "epoch": 5.8467888243831645,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021841445203319083,
      "loss": 4.8227,
      "mean_token_accuracy": 0.22918106168508529,
      "num_tokens": 133757723.0,
      "step": 64455
    },
    {
      "entropy": 5.7553393840789795,
      "epoch": 5.847242380261248,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002183831361484004,
      "loss": 4.8372,
      "mean_token_accuracy": 0.2250072628259659,
      "num_tokens": 133768718.0,
      "step": 64460
    },
    {
      "entropy": 5.75314736366272,
      "epoch": 5.847695936139332,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021835182143446102,
      "loss": 4.8459,
      "mean_token_accuracy": 0.22287366837263106,
      "num_tokens": 133779206.0,
      "step": 64465
    },
    {
      "entropy": 5.646875095367432,
      "epoch": 5.848149492017416,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021832050789202047,
      "loss": 4.7446,
      "mean_token_accuracy": 0.24202451556921006,
      "num_tokens": 133789668.0,
      "step": 64470
    },
    {
      "entropy": 5.73463659286499,
      "epoch": 5.848603047895501,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021828919552172632,
      "loss": 4.8226,
      "mean_token_accuracy": 0.23174909502267838,
      "num_tokens": 133800525.0,
      "step": 64475
    },
    {
      "entropy": 5.745654964447022,
      "epoch": 5.849056603773585,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021825788432422606,
      "loss": 4.7649,
      "mean_token_accuracy": 0.24025309830904007,
      "num_tokens": 133811887.0,
      "step": 64480
    },
    {
      "entropy": 5.805996084213257,
      "epoch": 5.849510159651669,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021822657430016712,
      "loss": 4.9694,
      "mean_token_accuracy": 0.22527358829975128,
      "num_tokens": 133822278.0,
      "step": 64485
    },
    {
      "entropy": 5.720743942260742,
      "epoch": 5.8499637155297535,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00021819526545019714,
      "loss": 4.8244,
      "mean_token_accuracy": 0.23110715597867965,
      "num_tokens": 133831705.0,
      "step": 64490
    },
    {
      "entropy": 5.76697268486023,
      "epoch": 5.850417271407838,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021816395777496357,
      "loss": 4.8231,
      "mean_token_accuracy": 0.2363029107451439,
      "num_tokens": 133842673.0,
      "step": 64495
    },
    {
      "entropy": 5.745232772827149,
      "epoch": 5.850870827285922,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021813265127511384,
      "loss": 4.8347,
      "mean_token_accuracy": 0.2293349951505661,
      "num_tokens": 133852792.0,
      "step": 64500
    },
    {
      "entropy": 5.791401004791259,
      "epoch": 5.851324383164005,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021810134595129528,
      "loss": 4.8228,
      "mean_token_accuracy": 0.23045375049114228,
      "num_tokens": 133862733.0,
      "step": 64505
    },
    {
      "entropy": 5.790577077865601,
      "epoch": 5.85177793904209,
      "grad_norm": 1.375,
      "learning_rate": 0.00021807004180415552,
      "loss": 4.9517,
      "mean_token_accuracy": 0.22202614545822144,
      "num_tokens": 133873757.0,
      "step": 64510
    },
    {
      "entropy": 5.797392749786377,
      "epoch": 5.852231494920174,
      "grad_norm": 1.296875,
      "learning_rate": 0.00021803873883434173,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23205045759677886,
      "num_tokens": 133883829.0,
      "step": 64515
    },
    {
      "entropy": 5.767114353179932,
      "epoch": 5.852685050798258,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021800743704250124,
      "loss": 4.9526,
      "mean_token_accuracy": 0.22894093990325928,
      "num_tokens": 133894323.0,
      "step": 64520
    },
    {
      "entropy": 5.686065292358398,
      "epoch": 5.8531386066763424,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021797613642928165,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23810014873743057,
      "num_tokens": 133904990.0,
      "step": 64525
    },
    {
      "entropy": 5.802216053009033,
      "epoch": 5.853592162554427,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021794483699533003,
      "loss": 4.8995,
      "mean_token_accuracy": 0.2319188043475151,
      "num_tokens": 133914668.0,
      "step": 64530
    },
    {
      "entropy": 5.734614896774292,
      "epoch": 5.854045718432511,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021791353874129376,
      "loss": 4.8013,
      "mean_token_accuracy": 0.23232101649045944,
      "num_tokens": 133924611.0,
      "step": 64535
    },
    {
      "entropy": 5.654428863525391,
      "epoch": 5.854499274310595,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021788224166782,
      "loss": 4.7381,
      "mean_token_accuracy": 0.2364375874400139,
      "num_tokens": 133935707.0,
      "step": 64540
    },
    {
      "entropy": 5.73713698387146,
      "epoch": 5.8549528301886795,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021785094577555603,
      "loss": 4.8549,
      "mean_token_accuracy": 0.23362389653921128,
      "num_tokens": 133945744.0,
      "step": 64545
    },
    {
      "entropy": 5.881212425231934,
      "epoch": 5.855406386066764,
      "grad_norm": 1.765625,
      "learning_rate": 0.00021781965106514907,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22900741845369338,
      "num_tokens": 133956073.0,
      "step": 64550
    },
    {
      "entropy": 5.73518934249878,
      "epoch": 5.855859941944848,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002177883575372463,
      "loss": 4.8213,
      "mean_token_accuracy": 0.23260203748941422,
      "num_tokens": 133966009.0,
      "step": 64555
    },
    {
      "entropy": 5.805143213272094,
      "epoch": 5.856313497822931,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021775706519249488,
      "loss": 4.9493,
      "mean_token_accuracy": 0.22673114389181137,
      "num_tokens": 133975759.0,
      "step": 64560
    },
    {
      "entropy": 5.7746528625488285,
      "epoch": 5.856767053701016,
      "grad_norm": 1.609375,
      "learning_rate": 0.00021772577403154193,
      "loss": 4.9321,
      "mean_token_accuracy": 0.2205115884542465,
      "num_tokens": 133986537.0,
      "step": 64565
    },
    {
      "entropy": 5.711092185974121,
      "epoch": 5.8572206095791,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002176944840550346,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23313798606395722,
      "num_tokens": 133997069.0,
      "step": 64570
    },
    {
      "entropy": 5.8683287620544435,
      "epoch": 5.857674165457184,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021766319526361988,
      "loss": 4.9406,
      "mean_token_accuracy": 0.21535890847444533,
      "num_tokens": 134008266.0,
      "step": 64575
    },
    {
      "entropy": 5.753750133514404,
      "epoch": 5.8581277213352685,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021763190765794493,
      "loss": 4.7452,
      "mean_token_accuracy": 0.24473093301057816,
      "num_tokens": 134017848.0,
      "step": 64580
    },
    {
      "entropy": 5.725297832489014,
      "epoch": 5.858581277213353,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021760062123865682,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23270809799432754,
      "num_tokens": 134027720.0,
      "step": 64585
    },
    {
      "entropy": 5.76985650062561,
      "epoch": 5.859034833091437,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002175693360064024,
      "loss": 4.941,
      "mean_token_accuracy": 0.22679513394832612,
      "num_tokens": 134037473.0,
      "step": 64590
    },
    {
      "entropy": 5.714684534072876,
      "epoch": 5.859488388969521,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021753805196182875,
      "loss": 4.8928,
      "mean_token_accuracy": 0.22671494632959366,
      "num_tokens": 134048373.0,
      "step": 64595
    },
    {
      "entropy": 5.776253747940063,
      "epoch": 5.8599419448476056,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002175067691055828,
      "loss": 4.8702,
      "mean_token_accuracy": 0.22687634825706482,
      "num_tokens": 134058631.0,
      "step": 64600
    },
    {
      "entropy": 5.707552862167359,
      "epoch": 5.860395500725689,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021747548743831162,
      "loss": 4.7659,
      "mean_token_accuracy": 0.23916254490613936,
      "num_tokens": 134069826.0,
      "step": 64605
    },
    {
      "entropy": 5.79184398651123,
      "epoch": 5.860849056603773,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021744420696066197,
      "loss": 4.9076,
      "mean_token_accuracy": 0.22898597568273543,
      "num_tokens": 134080248.0,
      "step": 64610
    },
    {
      "entropy": 5.799047803878784,
      "epoch": 5.8613026124818575,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021741292767328075,
      "loss": 4.9217,
      "mean_token_accuracy": 0.22415104657411575,
      "num_tokens": 134091549.0,
      "step": 64615
    },
    {
      "entropy": 5.814761638641357,
      "epoch": 5.861756168359942,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002173816495768149,
      "loss": 4.8947,
      "mean_token_accuracy": 0.22474958300590514,
      "num_tokens": 134101892.0,
      "step": 64620
    },
    {
      "entropy": 5.790245628356933,
      "epoch": 5.862209724238026,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002173503726719112,
      "loss": 4.8994,
      "mean_token_accuracy": 0.23402367830276488,
      "num_tokens": 134112073.0,
      "step": 64625
    },
    {
      "entropy": 5.721356821060181,
      "epoch": 5.86266328011611,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021731909695921653,
      "loss": 4.777,
      "mean_token_accuracy": 0.23347484618425368,
      "num_tokens": 134123015.0,
      "step": 64630
    },
    {
      "entropy": 5.72304368019104,
      "epoch": 5.8631168359941945,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021728782243937751,
      "loss": 4.8389,
      "mean_token_accuracy": 0.23473853915929793,
      "num_tokens": 134134742.0,
      "step": 64635
    },
    {
      "entropy": 5.766290092468262,
      "epoch": 5.863570391872279,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021725654911304115,
      "loss": 4.7527,
      "mean_token_accuracy": 0.23448158651590348,
      "num_tokens": 134145477.0,
      "step": 64640
    },
    {
      "entropy": 5.719954538345337,
      "epoch": 5.864023947750363,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002172252769808541,
      "loss": 4.8304,
      "mean_token_accuracy": 0.23558175265789033,
      "num_tokens": 134155987.0,
      "step": 64645
    },
    {
      "entropy": 5.809993839263916,
      "epoch": 5.864477503628447,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002171940060434629,
      "loss": 4.9108,
      "mean_token_accuracy": 0.22787827849388123,
      "num_tokens": 134165044.0,
      "step": 64650
    },
    {
      "entropy": 5.778594827651977,
      "epoch": 5.864931059506532,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021716273630151446,
      "loss": 4.9087,
      "mean_token_accuracy": 0.22225892543792725,
      "num_tokens": 134174786.0,
      "step": 64655
    },
    {
      "entropy": 5.752286100387574,
      "epoch": 5.865384615384615,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002171314677556554,
      "loss": 4.8216,
      "mean_token_accuracy": 0.23421185910701753,
      "num_tokens": 134184769.0,
      "step": 64660
    },
    {
      "entropy": 5.73875937461853,
      "epoch": 5.865838171262699,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021710020040653234,
      "loss": 4.8567,
      "mean_token_accuracy": 0.2327114760875702,
      "num_tokens": 134194288.0,
      "step": 64665
    },
    {
      "entropy": 5.832720232009888,
      "epoch": 5.8662917271407835,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002170689342547918,
      "loss": 4.9927,
      "mean_token_accuracy": 0.21951895207166672,
      "num_tokens": 134205150.0,
      "step": 64670
    },
    {
      "entropy": 5.710211610794067,
      "epoch": 5.866745283018868,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00021703766930108054,
      "loss": 4.8287,
      "mean_token_accuracy": 0.23149030059576034,
      "num_tokens": 134216221.0,
      "step": 64675
    },
    {
      "entropy": 5.780521392822266,
      "epoch": 5.867198838896952,
      "grad_norm": 1.5,
      "learning_rate": 0.00021700640554604507,
      "loss": 4.8625,
      "mean_token_accuracy": 0.2275157243013382,
      "num_tokens": 134226396.0,
      "step": 64680
    },
    {
      "entropy": 5.76653904914856,
      "epoch": 5.867652394775036,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021697514299033176,
      "loss": 4.8593,
      "mean_token_accuracy": 0.23120434433221818,
      "num_tokens": 134237564.0,
      "step": 64685
    },
    {
      "entropy": 5.761710691452026,
      "epoch": 5.868105950653121,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021694388163458745,
      "loss": 4.8026,
      "mean_token_accuracy": 0.23733714818954468,
      "num_tokens": 134247840.0,
      "step": 64690
    },
    {
      "entropy": 5.804352760314941,
      "epoch": 5.868559506531205,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002169126214794584,
      "loss": 4.8691,
      "mean_token_accuracy": 0.23402588963508605,
      "num_tokens": 134258366.0,
      "step": 64695
    },
    {
      "entropy": 5.7360944747924805,
      "epoch": 5.869013062409289,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021688136252559116,
      "loss": 4.7832,
      "mean_token_accuracy": 0.23758047670125962,
      "num_tokens": 134268551.0,
      "step": 64700
    },
    {
      "entropy": 5.828659009933472,
      "epoch": 5.869466618287373,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021685010477363214,
      "loss": 4.9233,
      "mean_token_accuracy": 0.22351398169994355,
      "num_tokens": 134279954.0,
      "step": 64705
    },
    {
      "entropy": 5.746769905090332,
      "epoch": 5.869920174165458,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021681884822422775,
      "loss": 4.8497,
      "mean_token_accuracy": 0.23013572096824647,
      "num_tokens": 134290719.0,
      "step": 64710
    },
    {
      "entropy": 5.73981785774231,
      "epoch": 5.870373730043541,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021678759287802448,
      "loss": 4.8955,
      "mean_token_accuracy": 0.22615503817796706,
      "num_tokens": 134300556.0,
      "step": 64715
    },
    {
      "entropy": 5.763929557800293,
      "epoch": 5.870827285921625,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021675633873566853,
      "loss": 4.9331,
      "mean_token_accuracy": 0.22300958782434463,
      "num_tokens": 134312265.0,
      "step": 64720
    },
    {
      "entropy": 5.760871267318725,
      "epoch": 5.8712808417997095,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021672508579780636,
      "loss": 4.8961,
      "mean_token_accuracy": 0.22413380891084672,
      "num_tokens": 134323124.0,
      "step": 64725
    },
    {
      "entropy": 5.757802486419678,
      "epoch": 5.871734397677794,
      "grad_norm": 1.6875,
      "learning_rate": 0.0002166938340650843,
      "loss": 4.828,
      "mean_token_accuracy": 0.23348827362060548,
      "num_tokens": 134331775.0,
      "step": 64730
    },
    {
      "entropy": 5.769988489151001,
      "epoch": 5.872187953555878,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021666258353814855,
      "loss": 4.8676,
      "mean_token_accuracy": 0.23287480026483537,
      "num_tokens": 134342471.0,
      "step": 64735
    },
    {
      "entropy": 5.743381500244141,
      "epoch": 5.872641509433962,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021663133421764554,
      "loss": 4.8382,
      "mean_token_accuracy": 0.23018198758363723,
      "num_tokens": 134353380.0,
      "step": 64740
    },
    {
      "entropy": 5.768298053741455,
      "epoch": 5.873095065312047,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021660008610422133,
      "loss": 4.9024,
      "mean_token_accuracy": 0.22597552835941315,
      "num_tokens": 134364060.0,
      "step": 64745
    },
    {
      "entropy": 5.7506317615509035,
      "epoch": 5.873548621190131,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021656883919852227,
      "loss": 4.8775,
      "mean_token_accuracy": 0.22315213829278946,
      "num_tokens": 134374337.0,
      "step": 64750
    },
    {
      "entropy": 5.773840761184692,
      "epoch": 5.874002177068215,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021653759350119444,
      "loss": 4.8405,
      "mean_token_accuracy": 0.23280860334634781,
      "num_tokens": 134385868.0,
      "step": 64755
    },
    {
      "entropy": 5.822838258743286,
      "epoch": 5.8744557329462985,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021650634901288412,
      "loss": 4.927,
      "mean_token_accuracy": 0.2224036157131195,
      "num_tokens": 134397324.0,
      "step": 64760
    },
    {
      "entropy": 5.828287982940674,
      "epoch": 5.874909288824383,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002164751057342374,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22365180850028993,
      "num_tokens": 134407408.0,
      "step": 64765
    },
    {
      "entropy": 5.75219759941101,
      "epoch": 5.875362844702467,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002164438636659004,
      "loss": 4.8435,
      "mean_token_accuracy": 0.22672894448041916,
      "num_tokens": 134416948.0,
      "step": 64770
    },
    {
      "entropy": 5.719744300842285,
      "epoch": 5.875816400580551,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021641262280851925,
      "loss": 4.8106,
      "mean_token_accuracy": 0.23787047117948532,
      "num_tokens": 134428274.0,
      "step": 64775
    },
    {
      "entropy": 5.690195751190186,
      "epoch": 5.876269956458636,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021638138316273998,
      "loss": 4.7572,
      "mean_token_accuracy": 0.2392630770802498,
      "num_tokens": 134437968.0,
      "step": 64780
    },
    {
      "entropy": 5.7134322166442875,
      "epoch": 5.87672351233672,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002163501447292086,
      "loss": 4.794,
      "mean_token_accuracy": 0.2313428446650505,
      "num_tokens": 134448349.0,
      "step": 64785
    },
    {
      "entropy": 5.752151107788086,
      "epoch": 5.877177068214804,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002163189075085712,
      "loss": 4.8394,
      "mean_token_accuracy": 0.23855798840522766,
      "num_tokens": 134458867.0,
      "step": 64790
    },
    {
      "entropy": 5.796602582931518,
      "epoch": 5.877630624092888,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021628767150147377,
      "loss": 4.921,
      "mean_token_accuracy": 0.2281259298324585,
      "num_tokens": 134469588.0,
      "step": 64795
    },
    {
      "entropy": 5.769102096557617,
      "epoch": 5.878084179970973,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021625643670856225,
      "loss": 4.8472,
      "mean_token_accuracy": 0.2331827089190483,
      "num_tokens": 134479698.0,
      "step": 64800
    },
    {
      "entropy": 5.748953723907471,
      "epoch": 5.878537735849057,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021622520313048255,
      "loss": 4.7417,
      "mean_token_accuracy": 0.23580833226442338,
      "num_tokens": 134490015.0,
      "step": 64805
    },
    {
      "entropy": 5.783118820190429,
      "epoch": 5.878991291727141,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021619397076788065,
      "loss": 4.8591,
      "mean_token_accuracy": 0.22649540454149247,
      "num_tokens": 134499536.0,
      "step": 64810
    },
    {
      "entropy": 5.732941627502441,
      "epoch": 5.879444847605225,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021616273962140242,
      "loss": 4.7778,
      "mean_token_accuracy": 0.23822992891073227,
      "num_tokens": 134510191.0,
      "step": 64815
    },
    {
      "entropy": 5.731414651870727,
      "epoch": 5.879898403483309,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021613150969169377,
      "loss": 4.7802,
      "mean_token_accuracy": 0.23745876103639602,
      "num_tokens": 134520387.0,
      "step": 64820
    },
    {
      "entropy": 5.726361703872681,
      "epoch": 5.880351959361393,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021610028097940044,
      "loss": 4.8515,
      "mean_token_accuracy": 0.2312743604183197,
      "num_tokens": 134530368.0,
      "step": 64825
    },
    {
      "entropy": 5.772447729110718,
      "epoch": 5.880805515239477,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002160690534851683,
      "loss": 4.9045,
      "mean_token_accuracy": 0.22292020171880722,
      "num_tokens": 134540275.0,
      "step": 64830
    },
    {
      "entropy": 5.793837404251098,
      "epoch": 5.881259071117562,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002160378272096432,
      "loss": 4.9067,
      "mean_token_accuracy": 0.22860902398824692,
      "num_tokens": 134551564.0,
      "step": 64835
    },
    {
      "entropy": 5.717478275299072,
      "epoch": 5.881712626995646,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021600660215347084,
      "loss": 4.8102,
      "mean_token_accuracy": 0.23181680887937545,
      "num_tokens": 134562929.0,
      "step": 64840
    },
    {
      "entropy": 5.744530200958252,
      "epoch": 5.88216618287373,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000215975378317297,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23270838856697082,
      "num_tokens": 134573171.0,
      "step": 64845
    },
    {
      "entropy": 5.739420318603516,
      "epoch": 5.882619738751814,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021594415570176735,
      "loss": 4.7902,
      "mean_token_accuracy": 0.23579539209604264,
      "num_tokens": 134583132.0,
      "step": 64850
    },
    {
      "entropy": 5.709088611602783,
      "epoch": 5.883073294629899,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021591293430752763,
      "loss": 4.8152,
      "mean_token_accuracy": 0.23727016299962997,
      "num_tokens": 134594070.0,
      "step": 64855
    },
    {
      "entropy": 5.720738363265991,
      "epoch": 5.883526850507983,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021588171413522344,
      "loss": 4.8378,
      "mean_token_accuracy": 0.22626406848430633,
      "num_tokens": 134603964.0,
      "step": 64860
    },
    {
      "entropy": 5.725980663299561,
      "epoch": 5.883980406386067,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00021585049518550053,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23382411897182465,
      "num_tokens": 134615951.0,
      "step": 64865
    },
    {
      "entropy": 5.737739086151123,
      "epoch": 5.884433962264151,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002158192774590045,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2269138589501381,
      "num_tokens": 134626397.0,
      "step": 64870
    },
    {
      "entropy": 5.716456317901612,
      "epoch": 5.884887518142235,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00021578806095638077,
      "loss": 4.8036,
      "mean_token_accuracy": 0.23386435359716415,
      "num_tokens": 134637264.0,
      "step": 64875
    },
    {
      "entropy": 5.767826271057129,
      "epoch": 5.885341074020319,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021575684567827502,
      "loss": 4.862,
      "mean_token_accuracy": 0.22609594315290452,
      "num_tokens": 134648170.0,
      "step": 64880
    },
    {
      "entropy": 5.778121900558472,
      "epoch": 5.885794629898403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021572563162533293,
      "loss": 4.8378,
      "mean_token_accuracy": 0.23072355091571808,
      "num_tokens": 134659174.0,
      "step": 64885
    },
    {
      "entropy": 5.75230827331543,
      "epoch": 5.886248185776488,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021569441879819973,
      "loss": 4.8692,
      "mean_token_accuracy": 0.23289230912923814,
      "num_tokens": 134669810.0,
      "step": 64890
    },
    {
      "entropy": 5.8015519142150875,
      "epoch": 5.886701741654572,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021566320719752113,
      "loss": 4.9566,
      "mean_token_accuracy": 0.22361188381910324,
      "num_tokens": 134679777.0,
      "step": 64895
    },
    {
      "entropy": 5.673054456710815,
      "epoch": 5.887155297532656,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021563199682394257,
      "loss": 4.7552,
      "mean_token_accuracy": 0.24024105370044707,
      "num_tokens": 134690052.0,
      "step": 64900
    },
    {
      "entropy": 5.768893194198609,
      "epoch": 5.8876088534107405,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021560078767810943,
      "loss": 4.951,
      "mean_token_accuracy": 0.22603296637535095,
      "num_tokens": 134701191.0,
      "step": 64905
    },
    {
      "entropy": 5.758406066894532,
      "epoch": 5.888062409288825,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002155695797606671,
      "loss": 4.9173,
      "mean_token_accuracy": 0.23136545568704606,
      "num_tokens": 134711852.0,
      "step": 64910
    },
    {
      "entropy": 5.694084167480469,
      "epoch": 5.888515965166908,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00021553837307226103,
      "loss": 4.7113,
      "mean_token_accuracy": 0.2472963809967041,
      "num_tokens": 134722391.0,
      "step": 64915
    },
    {
      "entropy": 5.782937383651733,
      "epoch": 5.888969521044992,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021550716761353656,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22340334951877594,
      "num_tokens": 134732274.0,
      "step": 64920
    },
    {
      "entropy": 5.752597379684448,
      "epoch": 5.889423076923077,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021547596338513896,
      "loss": 4.8226,
      "mean_token_accuracy": 0.2369839444756508,
      "num_tokens": 134742877.0,
      "step": 64925
    },
    {
      "entropy": 5.789031982421875,
      "epoch": 5.889876632801161,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021544476038771367,
      "loss": 4.8985,
      "mean_token_accuracy": 0.22467744052410127,
      "num_tokens": 134752821.0,
      "step": 64930
    },
    {
      "entropy": 5.729934501647949,
      "epoch": 5.890330188679245,
      "grad_norm": 1.578125,
      "learning_rate": 0.00021541355862190591,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23070691972970964,
      "num_tokens": 134763175.0,
      "step": 64935
    },
    {
      "entropy": 5.774036979675293,
      "epoch": 5.8907837445573294,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021538235808836093,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23826690912246704,
      "num_tokens": 134773037.0,
      "step": 64940
    },
    {
      "entropy": 5.7042014598846436,
      "epoch": 5.891237300435414,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021535115878772398,
      "loss": 4.7659,
      "mean_token_accuracy": 0.24118027836084366,
      "num_tokens": 134783306.0,
      "step": 64945
    },
    {
      "entropy": 5.73166675567627,
      "epoch": 5.891690856313498,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002153199607206403,
      "loss": 4.8329,
      "mean_token_accuracy": 0.2358483925461769,
      "num_tokens": 134793216.0,
      "step": 64950
    },
    {
      "entropy": 5.815846967697143,
      "epoch": 5.892144412191582,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021528876388775503,
      "loss": 4.9303,
      "mean_token_accuracy": 0.228115114569664,
      "num_tokens": 134802705.0,
      "step": 64955
    },
    {
      "entropy": 5.6894838333129885,
      "epoch": 5.8925979680696665,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021525756828971328,
      "loss": 4.8209,
      "mean_token_accuracy": 0.22993836551904678,
      "num_tokens": 134813151.0,
      "step": 64960
    },
    {
      "entropy": 5.75517954826355,
      "epoch": 5.893051523947751,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021522637392716037,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24018407613039017,
      "num_tokens": 134823281.0,
      "step": 64965
    },
    {
      "entropy": 5.772253561019897,
      "epoch": 5.893505079825834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002151951808007412,
      "loss": 4.8553,
      "mean_token_accuracy": 0.23012496531009674,
      "num_tokens": 134833965.0,
      "step": 64970
    },
    {
      "entropy": 5.769388628005982,
      "epoch": 5.893958635703918,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002151639889111009,
      "loss": 4.8643,
      "mean_token_accuracy": 0.22171814739704132,
      "num_tokens": 134844043.0,
      "step": 64975
    },
    {
      "entropy": 5.806267309188843,
      "epoch": 5.894412191582003,
      "grad_norm": 1.5,
      "learning_rate": 0.0002151327982588846,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23428940325975417,
      "num_tokens": 134853011.0,
      "step": 64980
    },
    {
      "entropy": 5.790991401672363,
      "epoch": 5.894865747460087,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021510160884473723,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2287479192018509,
      "num_tokens": 134863660.0,
      "step": 64985
    },
    {
      "entropy": 5.747065448760987,
      "epoch": 5.895319303338171,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021507042066930387,
      "loss": 4.8474,
      "mean_token_accuracy": 0.23423068970441818,
      "num_tokens": 134874026.0,
      "step": 64990
    },
    {
      "entropy": 5.814252281188965,
      "epoch": 5.8957728592162555,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021503923373322953,
      "loss": 4.9071,
      "mean_token_accuracy": 0.22323997020721437,
      "num_tokens": 134883381.0,
      "step": 64995
    },
    {
      "entropy": 5.724522066116333,
      "epoch": 5.89622641509434,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021500804803715906,
      "loss": 4.7733,
      "mean_token_accuracy": 0.2381947010755539,
      "num_tokens": 134893782.0,
      "step": 65000
    },
    {
      "entropy": 5.74653434753418,
      "epoch": 5.896679970972424,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002149768635817375,
      "loss": 4.8418,
      "mean_token_accuracy": 0.23397231549024583,
      "num_tokens": 134904698.0,
      "step": 65005
    },
    {
      "entropy": 5.70440993309021,
      "epoch": 5.897133526850508,
      "grad_norm": 1.328125,
      "learning_rate": 0.00021494568036760963,
      "loss": 4.7305,
      "mean_token_accuracy": 0.24428198039531707,
      "num_tokens": 134914764.0,
      "step": 65010
    },
    {
      "entropy": 5.759606266021729,
      "epoch": 5.897587082728592,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002149144983954205,
      "loss": 4.8909,
      "mean_token_accuracy": 0.22965790927410126,
      "num_tokens": 134925679.0,
      "step": 65015
    },
    {
      "entropy": 5.765037727355957,
      "epoch": 5.898040638606677,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021488331766581476,
      "loss": 4.8898,
      "mean_token_accuracy": 0.2277517706155777,
      "num_tokens": 134935233.0,
      "step": 65020
    },
    {
      "entropy": 5.725495529174805,
      "epoch": 5.89849419448476,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021485213817943732,
      "loss": 4.8162,
      "mean_token_accuracy": 0.2340220704674721,
      "num_tokens": 134945014.0,
      "step": 65025
    },
    {
      "entropy": 5.750791788101196,
      "epoch": 5.8989477503628445,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021482095993693306,
      "loss": 4.7755,
      "mean_token_accuracy": 0.23823024481534957,
      "num_tokens": 134955225.0,
      "step": 65030
    },
    {
      "entropy": 5.658940601348877,
      "epoch": 5.899401306240929,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00021478978293894662,
      "loss": 4.7948,
      "mean_token_accuracy": 0.24623603969812394,
      "num_tokens": 134965457.0,
      "step": 65035
    },
    {
      "entropy": 5.7617512226104735,
      "epoch": 5.899854862119013,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002147586071861229,
      "loss": 4.8774,
      "mean_token_accuracy": 0.2332090139389038,
      "num_tokens": 134976819.0,
      "step": 65040
    },
    {
      "entropy": 5.732456970214844,
      "epoch": 5.900308417997097,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021472743267910644,
      "loss": 4.835,
      "mean_token_accuracy": 0.23388670533895492,
      "num_tokens": 134987179.0,
      "step": 65045
    },
    {
      "entropy": 5.781571531295777,
      "epoch": 5.9007619738751815,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002146962594185421,
      "loss": 4.8274,
      "mean_token_accuracy": 0.23234590739011765,
      "num_tokens": 134997673.0,
      "step": 65050
    },
    {
      "entropy": 5.77727575302124,
      "epoch": 5.901215529753266,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002146650874050744,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22352958023548125,
      "num_tokens": 135008512.0,
      "step": 65055
    },
    {
      "entropy": 5.714623928070068,
      "epoch": 5.90166908563135,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021463391663934818,
      "loss": 4.7636,
      "mean_token_accuracy": 0.24217380136251448,
      "num_tokens": 135018600.0,
      "step": 65060
    },
    {
      "entropy": 5.688259267807007,
      "epoch": 5.902122641509434,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021460274712200789,
      "loss": 4.757,
      "mean_token_accuracy": 0.23625660240650176,
      "num_tokens": 135029248.0,
      "step": 65065
    },
    {
      "entropy": 5.683674335479736,
      "epoch": 5.902576197387518,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002145715788536981,
      "loss": 4.7946,
      "mean_token_accuracy": 0.23155660033226014,
      "num_tokens": 135040223.0,
      "step": 65070
    },
    {
      "entropy": 5.697850131988526,
      "epoch": 5.903029753265602,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021454041183506357,
      "loss": 4.7675,
      "mean_token_accuracy": 0.2350974664092064,
      "num_tokens": 135049759.0,
      "step": 65075
    },
    {
      "entropy": 5.756539344787598,
      "epoch": 5.903483309143686,
      "grad_norm": 1.5,
      "learning_rate": 0.0002145092460667487,
      "loss": 4.9426,
      "mean_token_accuracy": 0.22310224920511246,
      "num_tokens": 135059485.0,
      "step": 65080
    },
    {
      "entropy": 5.636148738861084,
      "epoch": 5.9039368650217705,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021447808154939797,
      "loss": 4.6902,
      "mean_token_accuracy": 0.24549688249826432,
      "num_tokens": 135069536.0,
      "step": 65085
    },
    {
      "entropy": 5.776242780685425,
      "epoch": 5.904390420899855,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021444691828365604,
      "loss": 4.8687,
      "mean_token_accuracy": 0.2333301156759262,
      "num_tokens": 135080020.0,
      "step": 65090
    },
    {
      "entropy": 5.801608896255493,
      "epoch": 5.904843976777939,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021441575627016723,
      "loss": 4.8778,
      "mean_token_accuracy": 0.22869940996170043,
      "num_tokens": 135090374.0,
      "step": 65095
    },
    {
      "entropy": 5.815768051147461,
      "epoch": 5.905297532656023,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021438459550957606,
      "loss": 4.9068,
      "mean_token_accuracy": 0.2287065029144287,
      "num_tokens": 135101437.0,
      "step": 65100
    },
    {
      "entropy": 5.771724987030029,
      "epoch": 5.905751088534108,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021435343600252683,
      "loss": 4.928,
      "mean_token_accuracy": 0.2256729394197464,
      "num_tokens": 135112113.0,
      "step": 65105
    },
    {
      "entropy": 5.649779319763184,
      "epoch": 5.906204644412192,
      "grad_norm": 1.265625,
      "learning_rate": 0.000214322277749664,
      "loss": 4.7504,
      "mean_token_accuracy": 0.2318393349647522,
      "num_tokens": 135123718.0,
      "step": 65110
    },
    {
      "entropy": 5.798305416107178,
      "epoch": 5.906658200290276,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021429112075163197,
      "loss": 4.8529,
      "mean_token_accuracy": 0.233294253051281,
      "num_tokens": 135134693.0,
      "step": 65115
    },
    {
      "entropy": 5.799834966659546,
      "epoch": 5.90711175616836,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021425996500907502,
      "loss": 4.8538,
      "mean_token_accuracy": 0.23455030471086502,
      "num_tokens": 135145057.0,
      "step": 65120
    },
    {
      "entropy": 5.775940465927124,
      "epoch": 5.907565312046444,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002142288105226375,
      "loss": 4.8314,
      "mean_token_accuracy": 0.2281556397676468,
      "num_tokens": 135155667.0,
      "step": 65125
    },
    {
      "entropy": 5.836379528045654,
      "epoch": 5.908018867924528,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021419765729296364,
      "loss": 4.9543,
      "mean_token_accuracy": 0.21759556978940964,
      "num_tokens": 135166677.0,
      "step": 65130
    },
    {
      "entropy": 5.782616901397705,
      "epoch": 5.908472423802612,
      "grad_norm": 1.5,
      "learning_rate": 0.0002141665053206977,
      "loss": 4.8675,
      "mean_token_accuracy": 0.22635830491781234,
      "num_tokens": 135177029.0,
      "step": 65135
    },
    {
      "entropy": 5.733906555175781,
      "epoch": 5.9089259796806965,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021413535460648392,
      "loss": 4.8723,
      "mean_token_accuracy": 0.22452800273895263,
      "num_tokens": 135187798.0,
      "step": 65140
    },
    {
      "entropy": 5.673496389389038,
      "epoch": 5.909379535558781,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021410420515096656,
      "loss": 4.818,
      "mean_token_accuracy": 0.2317204624414444,
      "num_tokens": 135198805.0,
      "step": 65145
    },
    {
      "entropy": 5.817495203018188,
      "epoch": 5.909833091436865,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021407305695478974,
      "loss": 4.8906,
      "mean_token_accuracy": 0.23422538489103317,
      "num_tokens": 135209781.0,
      "step": 65150
    },
    {
      "entropy": 5.730235385894775,
      "epoch": 5.910286647314949,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021404191001859763,
      "loss": 4.7769,
      "mean_token_accuracy": 0.23330365866422653,
      "num_tokens": 135219683.0,
      "step": 65155
    },
    {
      "entropy": 5.679609918594361,
      "epoch": 5.910740203193034,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021401076434303436,
      "loss": 4.7664,
      "mean_token_accuracy": 0.23673659861087798,
      "num_tokens": 135230085.0,
      "step": 65160
    },
    {
      "entropy": 5.790854215621948,
      "epoch": 5.911193759071118,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021397961992874405,
      "loss": 4.8875,
      "mean_token_accuracy": 0.228828501701355,
      "num_tokens": 135239601.0,
      "step": 65165
    },
    {
      "entropy": 5.78153715133667,
      "epoch": 5.911647314949201,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021394847677637068,
      "loss": 4.8323,
      "mean_token_accuracy": 0.23079716116189958,
      "num_tokens": 135249649.0,
      "step": 65170
    },
    {
      "entropy": 5.754029846191406,
      "epoch": 5.912100870827286,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002139173348865584,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23637532591819763,
      "num_tokens": 135259948.0,
      "step": 65175
    },
    {
      "entropy": 5.729983949661255,
      "epoch": 5.91255442670537,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021388619425995127,
      "loss": 4.8204,
      "mean_token_accuracy": 0.2292088508605957,
      "num_tokens": 135270650.0,
      "step": 65180
    },
    {
      "entropy": 5.755103492736817,
      "epoch": 5.913007982583454,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002138550548971931,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22839388102293015,
      "num_tokens": 135281547.0,
      "step": 65185
    },
    {
      "entropy": 5.728953456878662,
      "epoch": 5.913461538461538,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021382391679892804,
      "loss": 4.7864,
      "mean_token_accuracy": 0.23906180262565613,
      "num_tokens": 135290256.0,
      "step": 65190
    },
    {
      "entropy": 5.741478729248047,
      "epoch": 5.913915094339623,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021379277996579993,
      "loss": 4.8555,
      "mean_token_accuracy": 0.23251683861017228,
      "num_tokens": 135300542.0,
      "step": 65195
    },
    {
      "entropy": 5.721791934967041,
      "epoch": 5.914368650217707,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002137616443984528,
      "loss": 4.7768,
      "mean_token_accuracy": 0.2342704027891159,
      "num_tokens": 135311354.0,
      "step": 65200
    },
    {
      "entropy": 5.734398698806762,
      "epoch": 5.914822206095791,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021373051009753035,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2275175243616104,
      "num_tokens": 135322620.0,
      "step": 65205
    },
    {
      "entropy": 5.746233081817627,
      "epoch": 5.915275761973875,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002136993770636767,
      "loss": 4.8828,
      "mean_token_accuracy": 0.23279408365488052,
      "num_tokens": 135333199.0,
      "step": 65210
    },
    {
      "entropy": 5.753628778457641,
      "epoch": 5.91572931785196,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021366824529753554,
      "loss": 4.8716,
      "mean_token_accuracy": 0.23107558935880662,
      "num_tokens": 135343964.0,
      "step": 65215
    },
    {
      "entropy": 5.771047973632813,
      "epoch": 5.916182873730044,
      "grad_norm": 1.375,
      "learning_rate": 0.00021363711479975063,
      "loss": 4.824,
      "mean_token_accuracy": 0.2315003454685211,
      "num_tokens": 135354598.0,
      "step": 65220
    },
    {
      "entropy": 5.73129506111145,
      "epoch": 5.916636429608127,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00021360598557096583,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22425426989793779,
      "num_tokens": 135365228.0,
      "step": 65225
    },
    {
      "entropy": 5.738550090789795,
      "epoch": 5.917089985486212,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021357485761182488,
      "loss": 4.83,
      "mean_token_accuracy": 0.2288604974746704,
      "num_tokens": 135375612.0,
      "step": 65230
    },
    {
      "entropy": 5.747196531295776,
      "epoch": 5.917543541364296,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021354373092297157,
      "loss": 4.8716,
      "mean_token_accuracy": 0.23047059625387192,
      "num_tokens": 135386333.0,
      "step": 65235
    },
    {
      "entropy": 5.734304332733155,
      "epoch": 5.91799709724238,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021351260550504954,
      "loss": 4.851,
      "mean_token_accuracy": 0.23485122621059418,
      "num_tokens": 135396563.0,
      "step": 65240
    },
    {
      "entropy": 5.726317119598389,
      "epoch": 5.918450653120464,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002134814813587025,
      "loss": 4.8078,
      "mean_token_accuracy": 0.23334235548973084,
      "num_tokens": 135406895.0,
      "step": 65245
    },
    {
      "entropy": 5.786429786682129,
      "epoch": 5.918904208998549,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00021345035848457412,
      "loss": 4.8608,
      "mean_token_accuracy": 0.2351389929652214,
      "num_tokens": 135416946.0,
      "step": 65250
    },
    {
      "entropy": 5.795189762115479,
      "epoch": 5.919357764876633,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00021341923688330806,
      "loss": 4.8993,
      "mean_token_accuracy": 0.23128781765699385,
      "num_tokens": 135427662.0,
      "step": 65255
    },
    {
      "entropy": 5.742815685272217,
      "epoch": 5.919811320754717,
      "grad_norm": 1.375,
      "learning_rate": 0.0002133881165555478,
      "loss": 4.818,
      "mean_token_accuracy": 0.23094322234392167,
      "num_tokens": 135437416.0,
      "step": 65260
    },
    {
      "entropy": 5.745889329910279,
      "epoch": 5.920264876632801,
      "grad_norm": 1.5,
      "learning_rate": 0.000213356997501937,
      "loss": 4.8314,
      "mean_token_accuracy": 0.23190396577119826,
      "num_tokens": 135447643.0,
      "step": 65265
    },
    {
      "entropy": 5.731228542327881,
      "epoch": 5.920718432510886,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021332587972311924,
      "loss": 4.8658,
      "mean_token_accuracy": 0.2364930048584938,
      "num_tokens": 135458604.0,
      "step": 65270
    },
    {
      "entropy": 5.737176752090454,
      "epoch": 5.92117198838897,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002132947632197379,
      "loss": 4.8006,
      "mean_token_accuracy": 0.23351636677980422,
      "num_tokens": 135469591.0,
      "step": 65275
    },
    {
      "entropy": 5.6659791469573975,
      "epoch": 5.921625544267053,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002132636479924367,
      "loss": 4.7901,
      "mean_token_accuracy": 0.23196932822465896,
      "num_tokens": 135479656.0,
      "step": 65280
    },
    {
      "entropy": 5.7734418392181395,
      "epoch": 5.922079100145138,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021323253404185893,
      "loss": 4.8949,
      "mean_token_accuracy": 0.22175907641649245,
      "num_tokens": 135490838.0,
      "step": 65285
    },
    {
      "entropy": 5.684722232818603,
      "epoch": 5.922532656023222,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021320142136864806,
      "loss": 4.7499,
      "mean_token_accuracy": 0.23861986249685288,
      "num_tokens": 135500862.0,
      "step": 65290
    },
    {
      "entropy": 5.7962158679962155,
      "epoch": 5.922986211901306,
      "grad_norm": 1.96875,
      "learning_rate": 0.00021317030997344767,
      "loss": 4.8969,
      "mean_token_accuracy": 0.21915308237075806,
      "num_tokens": 135510521.0,
      "step": 65295
    },
    {
      "entropy": 5.721571779251098,
      "epoch": 5.92343976777939,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021313919985690088,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23691948354244233,
      "num_tokens": 135520348.0,
      "step": 65300
    },
    {
      "entropy": 5.749278450012207,
      "epoch": 5.923893323657475,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021310809101965134,
      "loss": 4.845,
      "mean_token_accuracy": 0.22682155072689056,
      "num_tokens": 135531482.0,
      "step": 65305
    },
    {
      "entropy": 5.696616744995117,
      "epoch": 5.924346879535559,
      "grad_norm": 1.375,
      "learning_rate": 0.0002130769834623421,
      "loss": 4.8245,
      "mean_token_accuracy": 0.22986329793930055,
      "num_tokens": 135541555.0,
      "step": 65310
    },
    {
      "entropy": 5.7518078804016115,
      "epoch": 5.924800435413643,
      "grad_norm": 1.234375,
      "learning_rate": 0.00021304587718561674,
      "loss": 4.8377,
      "mean_token_accuracy": 0.2314232885837555,
      "num_tokens": 135553459.0,
      "step": 65315
    },
    {
      "entropy": 5.851614475250244,
      "epoch": 5.9252539912917275,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021301477219011838,
      "loss": 4.937,
      "mean_token_accuracy": 0.2250560224056244,
      "num_tokens": 135563580.0,
      "step": 65320
    },
    {
      "entropy": 5.805643892288208,
      "epoch": 5.925707547169811,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002129836684764903,
      "loss": 4.8661,
      "mean_token_accuracy": 0.22816423326730728,
      "num_tokens": 135575131.0,
      "step": 65325
    },
    {
      "entropy": 5.858318710327149,
      "epoch": 5.926161103047895,
      "grad_norm": 1.25,
      "learning_rate": 0.00021295256604537583,
      "loss": 4.9081,
      "mean_token_accuracy": 0.22240505665540694,
      "num_tokens": 135585427.0,
      "step": 65330
    },
    {
      "entropy": 5.7450559616088865,
      "epoch": 5.926614658925979,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002129214648974181,
      "loss": 4.783,
      "mean_token_accuracy": 0.23306306153535844,
      "num_tokens": 135595118.0,
      "step": 65335
    },
    {
      "entropy": 5.824925899505615,
      "epoch": 5.927068214804064,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021289036503326027,
      "loss": 4.9293,
      "mean_token_accuracy": 0.22209297716617585,
      "num_tokens": 135606449.0,
      "step": 65340
    },
    {
      "entropy": 5.68394923210144,
      "epoch": 5.927521770682148,
      "grad_norm": 1.375,
      "learning_rate": 0.00021285926645354548,
      "loss": 4.8111,
      "mean_token_accuracy": 0.23638379126787185,
      "num_tokens": 135616250.0,
      "step": 65345
    },
    {
      "entropy": 5.741377973556519,
      "epoch": 5.927975326560232,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021282816915891694,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23059849292039872,
      "num_tokens": 135626882.0,
      "step": 65350
    },
    {
      "entropy": 5.707521867752075,
      "epoch": 5.9284288824383164,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021279707315001773,
      "loss": 4.7806,
      "mean_token_accuracy": 0.24469833374023436,
      "num_tokens": 135636824.0,
      "step": 65355
    },
    {
      "entropy": 5.819981384277344,
      "epoch": 5.928882438316401,
      "grad_norm": 1.28125,
      "learning_rate": 0.00021276597842749084,
      "loss": 4.9248,
      "mean_token_accuracy": 0.22377928495407104,
      "num_tokens": 135647310.0,
      "step": 65360
    },
    {
      "entropy": 5.799882221221924,
      "epoch": 5.929335994194485,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002127348849919794,
      "loss": 4.8699,
      "mean_token_accuracy": 0.23044497966766359,
      "num_tokens": 135657192.0,
      "step": 65365
    },
    {
      "entropy": 5.685503149032593,
      "epoch": 5.929789550072569,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021270379284412644,
      "loss": 4.7346,
      "mean_token_accuracy": 0.2445087894797325,
      "num_tokens": 135667780.0,
      "step": 65370
    },
    {
      "entropy": 5.703531074523926,
      "epoch": 5.9302431059506535,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002126727019845749,
      "loss": 4.736,
      "mean_token_accuracy": 0.24362110197544098,
      "num_tokens": 135677891.0,
      "step": 65375
    },
    {
      "entropy": 5.706798791885376,
      "epoch": 5.930696661828737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021264161241396767,
      "loss": 4.7986,
      "mean_token_accuracy": 0.23713549673557283,
      "num_tokens": 135687575.0,
      "step": 65380
    },
    {
      "entropy": 5.767254495620728,
      "epoch": 5.931150217706821,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002126105241329479,
      "loss": 4.9174,
      "mean_token_accuracy": 0.22450088560581208,
      "num_tokens": 135698592.0,
      "step": 65385
    },
    {
      "entropy": 5.629965734481812,
      "epoch": 5.931603773584905,
      "grad_norm": 1.484375,
      "learning_rate": 0.00021257943714215845,
      "loss": 4.7038,
      "mean_token_accuracy": 0.2436121165752411,
      "num_tokens": 135709159.0,
      "step": 65390
    },
    {
      "entropy": 5.806208610534668,
      "epoch": 5.93205732946299,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021254835144224205,
      "loss": 4.9475,
      "mean_token_accuracy": 0.2219046711921692,
      "num_tokens": 135718639.0,
      "step": 65395
    },
    {
      "entropy": 5.760971975326538,
      "epoch": 5.932510885341074,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021251726703384167,
      "loss": 4.8882,
      "mean_token_accuracy": 0.22536223381757736,
      "num_tokens": 135729494.0,
      "step": 65400
    },
    {
      "entropy": 5.774355411529541,
      "epoch": 5.932964441219158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00021248618391760005,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23217539191246034,
      "num_tokens": 135740456.0,
      "step": 65405
    },
    {
      "entropy": 5.701773929595947,
      "epoch": 5.9334179970972425,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021245510209416018,
      "loss": 4.7746,
      "mean_token_accuracy": 0.24640202820301055,
      "num_tokens": 135750363.0,
      "step": 65410
    },
    {
      "entropy": 5.707142639160156,
      "epoch": 5.933871552975327,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00021242402156416474,
      "loss": 4.7979,
      "mean_token_accuracy": 0.23788623958826066,
      "num_tokens": 135760505.0,
      "step": 65415
    },
    {
      "entropy": 5.749364948272705,
      "epoch": 5.934325108853411,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002123929423282564,
      "loss": 4.7875,
      "mean_token_accuracy": 0.23468256890773773,
      "num_tokens": 135770708.0,
      "step": 65420
    },
    {
      "entropy": 5.768815422058106,
      "epoch": 5.934778664731495,
      "grad_norm": 1.359375,
      "learning_rate": 0.000212361864387078,
      "loss": 4.9051,
      "mean_token_accuracy": 0.22813324630260468,
      "num_tokens": 135780761.0,
      "step": 65425
    },
    {
      "entropy": 5.781333494186401,
      "epoch": 5.93523222060958,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021233078774127225,
      "loss": 4.9199,
      "mean_token_accuracy": 0.23022902607917786,
      "num_tokens": 135790427.0,
      "step": 65430
    },
    {
      "entropy": 5.797046136856079,
      "epoch": 5.935685776487663,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021229971239148176,
      "loss": 4.9279,
      "mean_token_accuracy": 0.23203345090150834,
      "num_tokens": 135801962.0,
      "step": 65435
    },
    {
      "entropy": 5.74829421043396,
      "epoch": 5.936139332365747,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021226863833834913,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2270150750875473,
      "num_tokens": 135811942.0,
      "step": 65440
    },
    {
      "entropy": 5.7918213367462155,
      "epoch": 5.9365928882438315,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021223756558251712,
      "loss": 4.896,
      "mean_token_accuracy": 0.22402759790420532,
      "num_tokens": 135822102.0,
      "step": 65445
    },
    {
      "entropy": 5.770741176605225,
      "epoch": 5.937046444121916,
      "grad_norm": 1.453125,
      "learning_rate": 0.00021220649412462823,
      "loss": 4.8786,
      "mean_token_accuracy": 0.23453413248062133,
      "num_tokens": 135832804.0,
      "step": 65450
    },
    {
      "entropy": 5.7478940963745115,
      "epoch": 5.9375,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021217542396532503,
      "loss": 4.83,
      "mean_token_accuracy": 0.2308742582798004,
      "num_tokens": 135842415.0,
      "step": 65455
    },
    {
      "entropy": 5.766341304779052,
      "epoch": 5.937953555878084,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002121443551052501,
      "loss": 4.8359,
      "mean_token_accuracy": 0.2349383994936943,
      "num_tokens": 135853084.0,
      "step": 65460
    },
    {
      "entropy": 5.792606258392334,
      "epoch": 5.9384071117561685,
      "grad_norm": 1.640625,
      "learning_rate": 0.00021211328754504587,
      "loss": 4.8729,
      "mean_token_accuracy": 0.23196962624788284,
      "num_tokens": 135863215.0,
      "step": 65465
    },
    {
      "entropy": 5.711615324020386,
      "epoch": 5.938860667634253,
      "grad_norm": 1.515625,
      "learning_rate": 0.00021208222128535491,
      "loss": 4.8447,
      "mean_token_accuracy": 0.2283069983124733,
      "num_tokens": 135874167.0,
      "step": 65470
    },
    {
      "entropy": 5.70485553741455,
      "epoch": 5.939314223512337,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021205115632681965,
      "loss": 4.813,
      "mean_token_accuracy": 0.23800752311944962,
      "num_tokens": 135883309.0,
      "step": 65475
    },
    {
      "entropy": 5.808508205413818,
      "epoch": 5.93976777939042,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021202009267008254,
      "loss": 4.8985,
      "mean_token_accuracy": 0.2297166809439659,
      "num_tokens": 135893750.0,
      "step": 65480
    },
    {
      "entropy": 5.792537784576416,
      "epoch": 5.940221335268505,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00021198903031578598,
      "loss": 4.8195,
      "mean_token_accuracy": 0.23447550237178802,
      "num_tokens": 135904244.0,
      "step": 65485
    },
    {
      "entropy": 5.710324382781982,
      "epoch": 5.940674891146589,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021195796926457223,
      "loss": 4.8104,
      "mean_token_accuracy": 0.2309799760580063,
      "num_tokens": 135913538.0,
      "step": 65490
    },
    {
      "entropy": 5.749393892288208,
      "epoch": 5.941128447024673,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021192690951708383,
      "loss": 4.793,
      "mean_token_accuracy": 0.24033558815717698,
      "num_tokens": 135924711.0,
      "step": 65495
    },
    {
      "entropy": 5.737775135040283,
      "epoch": 5.9415820029027575,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021189585107396296,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22969312071800232,
      "num_tokens": 135935389.0,
      "step": 65500
    },
    {
      "entropy": 5.683243465423584,
      "epoch": 5.942035558780842,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00021186479393585196,
      "loss": 4.774,
      "mean_token_accuracy": 0.24654968082904816,
      "num_tokens": 135945311.0,
      "step": 65505
    },
    {
      "entropy": 5.737878942489624,
      "epoch": 5.942489114658926,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002118337381033932,
      "loss": 4.8128,
      "mean_token_accuracy": 0.2375430718064308,
      "num_tokens": 135954782.0,
      "step": 65510
    },
    {
      "entropy": 5.7548939228057865,
      "epoch": 5.94294267053701,
      "grad_norm": 1.3125,
      "learning_rate": 0.00021180268357722876,
      "loss": 4.805,
      "mean_token_accuracy": 0.23044235706329347,
      "num_tokens": 135965060.0,
      "step": 65515
    },
    {
      "entropy": 5.761497259140015,
      "epoch": 5.943396226415095,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021177163035800102,
      "loss": 4.9109,
      "mean_token_accuracy": 0.22844040393829346,
      "num_tokens": 135977373.0,
      "step": 65520
    },
    {
      "entropy": 5.7259986877441404,
      "epoch": 5.943849782293179,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002117405784463519,
      "loss": 4.8063,
      "mean_token_accuracy": 0.22716874480247498,
      "num_tokens": 135988426.0,
      "step": 65525
    },
    {
      "entropy": 5.739288091659546,
      "epoch": 5.944303338171263,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002117095278429239,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23392781615257263,
      "num_tokens": 135998636.0,
      "step": 65530
    },
    {
      "entropy": 5.796162509918213,
      "epoch": 5.9447568940493465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00021167847854835893,
      "loss": 4.8739,
      "mean_token_accuracy": 0.2342669427394867,
      "num_tokens": 136009665.0,
      "step": 65535
    },
    {
      "entropy": 5.743467950820923,
      "epoch": 5.945210449927431,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021164743056329915,
      "loss": 4.8495,
      "mean_token_accuracy": 0.2309972658753395,
      "num_tokens": 136020024.0,
      "step": 65540
    },
    {
      "entropy": 5.678156614303589,
      "epoch": 5.945664005805515,
      "grad_norm": 1.421875,
      "learning_rate": 0.00021161638388838666,
      "loss": 4.7485,
      "mean_token_accuracy": 0.2418784409761429,
      "num_tokens": 136030563.0,
      "step": 65545
    },
    {
      "entropy": 5.706792688369751,
      "epoch": 5.946117561683599,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021158533852426353,
      "loss": 4.7544,
      "mean_token_accuracy": 0.23999989330768584,
      "num_tokens": 136042000.0,
      "step": 65550
    },
    {
      "entropy": 5.718061399459839,
      "epoch": 5.9465711175616836,
      "grad_norm": 1.5,
      "learning_rate": 0.00021155429447157178,
      "loss": 4.8216,
      "mean_token_accuracy": 0.22959477156400682,
      "num_tokens": 136051570.0,
      "step": 65555
    },
    {
      "entropy": 5.764768648147583,
      "epoch": 5.947024673439768,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021152325173095328,
      "loss": 4.8635,
      "mean_token_accuracy": 0.23145010620355605,
      "num_tokens": 136062170.0,
      "step": 65560
    },
    {
      "entropy": 5.684288024902344,
      "epoch": 5.947478229317852,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002114922103030502,
      "loss": 4.8084,
      "mean_token_accuracy": 0.22815060317516328,
      "num_tokens": 136071216.0,
      "step": 65565
    },
    {
      "entropy": 5.750248289108276,
      "epoch": 5.947931785195936,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00021146117018850435,
      "loss": 4.8597,
      "mean_token_accuracy": 0.22290049195289613,
      "num_tokens": 136081359.0,
      "step": 65570
    },
    {
      "entropy": 5.765708351135254,
      "epoch": 5.948385341074021,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002114301313879577,
      "loss": 4.8503,
      "mean_token_accuracy": 0.23282091170549393,
      "num_tokens": 136092157.0,
      "step": 65575
    },
    {
      "entropy": 5.701197814941406,
      "epoch": 5.948838896952105,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002113990939020522,
      "loss": 4.7088,
      "mean_token_accuracy": 0.23804711401462555,
      "num_tokens": 136102077.0,
      "step": 65580
    },
    {
      "entropy": 5.7621691703796385,
      "epoch": 5.949292452830189,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021136805773142958,
      "loss": 4.8509,
      "mean_token_accuracy": 0.23054803013801575,
      "num_tokens": 136113736.0,
      "step": 65585
    },
    {
      "entropy": 5.828131914138794,
      "epoch": 5.9497460087082725,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021133702287673173,
      "loss": 4.9606,
      "mean_token_accuracy": 0.22309691905975343,
      "num_tokens": 136123818.0,
      "step": 65590
    },
    {
      "entropy": 5.755432176589966,
      "epoch": 5.950199564586357,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002113059893386005,
      "loss": 4.8231,
      "mean_token_accuracy": 0.2249840795993805,
      "num_tokens": 136133340.0,
      "step": 65595
    },
    {
      "entropy": 5.80881028175354,
      "epoch": 5.950653120464441,
      "grad_norm": 1.6875,
      "learning_rate": 0.00021127495711767763,
      "loss": 5.0015,
      "mean_token_accuracy": 0.2247139558196068,
      "num_tokens": 136144342.0,
      "step": 65600
    },
    {
      "entropy": 5.806127595901489,
      "epoch": 5.951106676342525,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021124392621460491,
      "loss": 4.9754,
      "mean_token_accuracy": 0.2203589126467705,
      "num_tokens": 136155246.0,
      "step": 65605
    },
    {
      "entropy": 5.724166631698608,
      "epoch": 5.95156023222061,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00021121289663002395,
      "loss": 4.8523,
      "mean_token_accuracy": 0.23911856710910798,
      "num_tokens": 136165654.0,
      "step": 65610
    },
    {
      "entropy": 5.755375719070434,
      "epoch": 5.952013788098694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002111818683645767,
      "loss": 4.8097,
      "mean_token_accuracy": 0.2388660117983818,
      "num_tokens": 136176301.0,
      "step": 65615
    },
    {
      "entropy": 5.753272485733032,
      "epoch": 5.952467343976778,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021115084141890456,
      "loss": 4.8207,
      "mean_token_accuracy": 0.23454942256212236,
      "num_tokens": 136187241.0,
      "step": 65620
    },
    {
      "entropy": 5.749810886383057,
      "epoch": 5.952920899854862,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002111198157936493,
      "loss": 4.8558,
      "mean_token_accuracy": 0.22775809913873674,
      "num_tokens": 136197436.0,
      "step": 65625
    },
    {
      "entropy": 5.630444097518921,
      "epoch": 5.953374455732947,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00021108879148945255,
      "loss": 4.7374,
      "mean_token_accuracy": 0.23987319320440292,
      "num_tokens": 136207739.0,
      "step": 65630
    },
    {
      "entropy": 5.696892499923706,
      "epoch": 5.95382801161103,
      "grad_norm": 1.546875,
      "learning_rate": 0.00021105776850695592,
      "loss": 4.8225,
      "mean_token_accuracy": 0.2319575995206833,
      "num_tokens": 136217230.0,
      "step": 65635
    },
    {
      "entropy": 5.744562864303589,
      "epoch": 5.954281567489114,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002110267468468009,
      "loss": 4.8316,
      "mean_token_accuracy": 0.23018891662359237,
      "num_tokens": 136226626.0,
      "step": 65640
    },
    {
      "entropy": 5.755486917495728,
      "epoch": 5.954735123367199,
      "grad_norm": 1.40625,
      "learning_rate": 0.000210995726509629,
      "loss": 4.805,
      "mean_token_accuracy": 0.22976332753896714,
      "num_tokens": 136236404.0,
      "step": 65645
    },
    {
      "entropy": 5.795773458480835,
      "epoch": 5.955188679245283,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00021096470749608183,
      "loss": 4.8913,
      "mean_token_accuracy": 0.23463335186243056,
      "num_tokens": 136246735.0,
      "step": 65650
    },
    {
      "entropy": 5.769292497634888,
      "epoch": 5.955642235123367,
      "grad_norm": 1.40625,
      "learning_rate": 0.00021093368980680083,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22596925795078276,
      "num_tokens": 136257286.0,
      "step": 65655
    },
    {
      "entropy": 5.6773566722869875,
      "epoch": 5.956095791001451,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021090267344242743,
      "loss": 4.8176,
      "mean_token_accuracy": 0.2303532600402832,
      "num_tokens": 136268245.0,
      "step": 65660
    },
    {
      "entropy": 5.7822717189788815,
      "epoch": 5.956549346879536,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021087165840360313,
      "loss": 4.93,
      "mean_token_accuracy": 0.22974329739809035,
      "num_tokens": 136279592.0,
      "step": 65665
    },
    {
      "entropy": 5.74695782661438,
      "epoch": 5.95700290275762,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00021084064469096926,
      "loss": 4.8137,
      "mean_token_accuracy": 0.23461296111345292,
      "num_tokens": 136290128.0,
      "step": 65670
    },
    {
      "entropy": 5.806404733657837,
      "epoch": 5.957456458635704,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002108096323051672,
      "loss": 4.8454,
      "mean_token_accuracy": 0.235225085914135,
      "num_tokens": 136300506.0,
      "step": 65675
    },
    {
      "entropy": 5.76794548034668,
      "epoch": 5.957910014513788,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021077862124683827,
      "loss": 4.8635,
      "mean_token_accuracy": 0.22850895822048187,
      "num_tokens": 136310205.0,
      "step": 65680
    },
    {
      "entropy": 5.7397332191467285,
      "epoch": 5.958363570391873,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002107476115166238,
      "loss": 4.8636,
      "mean_token_accuracy": 0.23107069730758667,
      "num_tokens": 136321870.0,
      "step": 65685
    },
    {
      "entropy": 5.747711706161499,
      "epoch": 5.958817126269956,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002107166031151652,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23347002118825913,
      "num_tokens": 136332534.0,
      "step": 65690
    },
    {
      "entropy": 5.770298814773559,
      "epoch": 5.95927068214804,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002106855960431035,
      "loss": 4.8819,
      "mean_token_accuracy": 0.22260531187057495,
      "num_tokens": 136341805.0,
      "step": 65695
    },
    {
      "entropy": 5.730735492706299,
      "epoch": 5.959724238026125,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021065459030108013,
      "loss": 4.8692,
      "mean_token_accuracy": 0.22691664546728135,
      "num_tokens": 136352664.0,
      "step": 65700
    },
    {
      "entropy": 5.7847284317016605,
      "epoch": 5.960177793904209,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00021062358588973628,
      "loss": 4.8484,
      "mean_token_accuracy": 0.22457588613033294,
      "num_tokens": 136362520.0,
      "step": 65705
    },
    {
      "entropy": 5.755375003814697,
      "epoch": 5.960631349782293,
      "grad_norm": 1.359375,
      "learning_rate": 0.00021059258280971298,
      "loss": 4.8076,
      "mean_token_accuracy": 0.23391106724739075,
      "num_tokens": 136371930.0,
      "step": 65710
    },
    {
      "entropy": 5.763120174407959,
      "epoch": 5.961084905660377,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00021056158106165157,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2358822777867317,
      "num_tokens": 136381698.0,
      "step": 65715
    },
    {
      "entropy": 5.781151199340821,
      "epoch": 5.961538461538462,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021053058064619308,
      "loss": 4.7782,
      "mean_token_accuracy": 0.24026547223329545,
      "num_tokens": 136390946.0,
      "step": 65720
    },
    {
      "entropy": 5.761228752136231,
      "epoch": 5.961992017416546,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00021049958156397854,
      "loss": 4.8094,
      "mean_token_accuracy": 0.23748520761728287,
      "num_tokens": 136400028.0,
      "step": 65725
    },
    {
      "entropy": 5.736727142333985,
      "epoch": 5.96244557329463,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002104685838156491,
      "loss": 4.8655,
      "mean_token_accuracy": 0.23357271850109101,
      "num_tokens": 136409626.0,
      "step": 65730
    },
    {
      "entropy": 5.738030910491943,
      "epoch": 5.962899129172714,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002104375874018458,
      "loss": 5.0096,
      "mean_token_accuracy": 0.21205532997846605,
      "num_tokens": 136420753.0,
      "step": 65735
    },
    {
      "entropy": 5.782754564285279,
      "epoch": 5.963352685050799,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002104065923232097,
      "loss": 4.8661,
      "mean_token_accuracy": 0.23177798390388488,
      "num_tokens": 136431342.0,
      "step": 65740
    },
    {
      "entropy": 5.785614824295044,
      "epoch": 5.963806240928882,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002103755985803817,
      "loss": 4.869,
      "mean_token_accuracy": 0.23164304792881013,
      "num_tokens": 136441419.0,
      "step": 65745
    },
    {
      "entropy": 5.87719578742981,
      "epoch": 5.964259796806966,
      "grad_norm": 1.53125,
      "learning_rate": 0.00021034460617400276,
      "loss": 4.9982,
      "mean_token_accuracy": 0.2226765990257263,
      "num_tokens": 136451685.0,
      "step": 65750
    },
    {
      "entropy": 5.762779760360718,
      "epoch": 5.964713352685051,
      "grad_norm": 1.4375,
      "learning_rate": 0.00021031361510471386,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2301624298095703,
      "num_tokens": 136461212.0,
      "step": 65755
    },
    {
      "entropy": 5.826099300384522,
      "epoch": 5.965166908563135,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002102826253731559,
      "loss": 4.8796,
      "mean_token_accuracy": 0.2280738100409508,
      "num_tokens": 136471531.0,
      "step": 65760
    },
    {
      "entropy": 5.7300676822662355,
      "epoch": 5.965620464441219,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002102516369799697,
      "loss": 4.8542,
      "mean_token_accuracy": 0.23734636157751082,
      "num_tokens": 136482109.0,
      "step": 65765
    },
    {
      "entropy": 5.752425765991211,
      "epoch": 5.9660740203193035,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002102206499257962,
      "loss": 4.8238,
      "mean_token_accuracy": 0.23556266576051713,
      "num_tokens": 136492981.0,
      "step": 65770
    },
    {
      "entropy": 5.785766744613648,
      "epoch": 5.966527576197388,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002101896642112761,
      "loss": 4.8399,
      "mean_token_accuracy": 0.2329142466187477,
      "num_tokens": 136502333.0,
      "step": 65775
    },
    {
      "entropy": 5.764764881134033,
      "epoch": 5.966981132075472,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002101586798370503,
      "loss": 4.8125,
      "mean_token_accuracy": 0.23453125059604646,
      "num_tokens": 136512940.0,
      "step": 65780
    },
    {
      "entropy": 5.7017333984375,
      "epoch": 5.967434687953556,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00021012769680375955,
      "loss": 4.7978,
      "mean_token_accuracy": 0.2332129642367363,
      "num_tokens": 136523557.0,
      "step": 65785
    },
    {
      "entropy": 5.7743347644805905,
      "epoch": 5.96788824383164,
      "grad_norm": 1.46875,
      "learning_rate": 0.00021009671511204453,
      "loss": 4.8878,
      "mean_token_accuracy": 0.22714188694953918,
      "num_tokens": 136533352.0,
      "step": 65790
    },
    {
      "entropy": 5.816037511825561,
      "epoch": 5.968341799709724,
      "grad_norm": 1.390625,
      "learning_rate": 0.00021006573476254597,
      "loss": 4.9118,
      "mean_token_accuracy": 0.22875984162092208,
      "num_tokens": 136543377.0,
      "step": 65795
    },
    {
      "entropy": 5.709797525405884,
      "epoch": 5.968795355587808,
      "grad_norm": 1.34375,
      "learning_rate": 0.00021003475575590458,
      "loss": 4.7699,
      "mean_token_accuracy": 0.24023972004652022,
      "num_tokens": 136553088.0,
      "step": 65800
    },
    {
      "entropy": 5.733905601501465,
      "epoch": 5.969248911465892,
      "grad_norm": 1.4375,
      "learning_rate": 0.000210003778092761,
      "loss": 4.874,
      "mean_token_accuracy": 0.2241554394364357,
      "num_tokens": 136563319.0,
      "step": 65805
    },
    {
      "entropy": 5.714648246765137,
      "epoch": 5.969702467343977,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002099728017737559,
      "loss": 4.8319,
      "mean_token_accuracy": 0.23295521587133408,
      "num_tokens": 136574278.0,
      "step": 65810
    },
    {
      "entropy": 5.751852083206177,
      "epoch": 5.970156023222061,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020994182679952978,
      "loss": 4.8736,
      "mean_token_accuracy": 0.23173246681690216,
      "num_tokens": 136583318.0,
      "step": 65815
    },
    {
      "entropy": 5.74559965133667,
      "epoch": 5.970609579100145,
      "grad_norm": 1.375,
      "learning_rate": 0.00020991085317072333,
      "loss": 4.8761,
      "mean_token_accuracy": 0.22656692266464235,
      "num_tokens": 136594226.0,
      "step": 65820
    },
    {
      "entropy": 5.800525903701782,
      "epoch": 5.9710631349782295,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020987988088797687,
      "loss": 4.8966,
      "mean_token_accuracy": 0.22769237756729127,
      "num_tokens": 136605540.0,
      "step": 65825
    },
    {
      "entropy": 5.7604522705078125,
      "epoch": 5.971516690856314,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002098489099519312,
      "loss": 4.8312,
      "mean_token_accuracy": 0.23338923454284669,
      "num_tokens": 136616154.0,
      "step": 65830
    },
    {
      "entropy": 5.738781547546386,
      "epoch": 5.971970246734398,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020981794036322666,
      "loss": 4.8572,
      "mean_token_accuracy": 0.22578487992286683,
      "num_tokens": 136626064.0,
      "step": 65835
    },
    {
      "entropy": 5.738336706161499,
      "epoch": 5.972423802612482,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002097869721225036,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23678650259971618,
      "num_tokens": 136636326.0,
      "step": 65840
    },
    {
      "entropy": 5.740685319900512,
      "epoch": 5.972877358490566,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020975600523040272,
      "loss": 4.8576,
      "mean_token_accuracy": 0.23421892672777175,
      "num_tokens": 136646675.0,
      "step": 65845
    },
    {
      "entropy": 5.692663526535034,
      "epoch": 5.97333091436865,
      "grad_norm": 1.375,
      "learning_rate": 0.00020972503968756418,
      "loss": 4.7849,
      "mean_token_accuracy": 0.2338703453540802,
      "num_tokens": 136657041.0,
      "step": 65850
    },
    {
      "entropy": 5.738545513153076,
      "epoch": 5.973784470246734,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020969407549462855,
      "loss": 4.8236,
      "mean_token_accuracy": 0.22741595804691314,
      "num_tokens": 136666156.0,
      "step": 65855
    },
    {
      "entropy": 5.71422929763794,
      "epoch": 5.9742380261248185,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020966311265223597,
      "loss": 4.7761,
      "mean_token_accuracy": 0.237815922498703,
      "num_tokens": 136675870.0,
      "step": 65860
    },
    {
      "entropy": 5.686468982696534,
      "epoch": 5.974691582002903,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002096321511610269,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23935675770044326,
      "num_tokens": 136686635.0,
      "step": 65865
    },
    {
      "entropy": 5.844439172744751,
      "epoch": 5.975145137880987,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020960119102164166,
      "loss": 5.0329,
      "mean_token_accuracy": 0.21272212862968445,
      "num_tokens": 136698365.0,
      "step": 65870
    },
    {
      "entropy": 5.746956634521484,
      "epoch": 5.975598693759071,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020957023223472037,
      "loss": 4.8547,
      "mean_token_accuracy": 0.2269283875823021,
      "num_tokens": 136709629.0,
      "step": 65875
    },
    {
      "entropy": 5.84520263671875,
      "epoch": 5.9760522496371555,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020953927480090346,
      "loss": 4.9184,
      "mean_token_accuracy": 0.2228563591837883,
      "num_tokens": 136718996.0,
      "step": 65880
    },
    {
      "entropy": 5.74919695854187,
      "epoch": 5.97650580551524,
      "grad_norm": 1.375,
      "learning_rate": 0.000209508318720831,
      "loss": 4.8552,
      "mean_token_accuracy": 0.23449087589979173,
      "num_tokens": 136729999.0,
      "step": 65885
    },
    {
      "entropy": 5.768736982345581,
      "epoch": 5.976959361393323,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020947736399514318,
      "loss": 4.8366,
      "mean_token_accuracy": 0.23621589541435242,
      "num_tokens": 136739809.0,
      "step": 65890
    },
    {
      "entropy": 5.75056266784668,
      "epoch": 5.977412917271408,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020944641062448004,
      "loss": 4.8236,
      "mean_token_accuracy": 0.23506750613451005,
      "num_tokens": 136750423.0,
      "step": 65895
    },
    {
      "entropy": 5.735143709182739,
      "epoch": 5.977866473149492,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020941545860948203,
      "loss": 4.8344,
      "mean_token_accuracy": 0.23334124982357024,
      "num_tokens": 136761907.0,
      "step": 65900
    },
    {
      "entropy": 5.845616722106934,
      "epoch": 5.978320029027576,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000209384507950789,
      "loss": 4.979,
      "mean_token_accuracy": 0.2175324559211731,
      "num_tokens": 136772437.0,
      "step": 65905
    },
    {
      "entropy": 5.836529207229614,
      "epoch": 5.97877358490566,
      "grad_norm": 1.453125,
      "learning_rate": 0.000209353558649041,
      "loss": 4.8803,
      "mean_token_accuracy": 0.23791563808917998,
      "num_tokens": 136782768.0,
      "step": 65910
    },
    {
      "entropy": 5.779751396179199,
      "epoch": 5.9792271407837445,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020932261070487817,
      "loss": 4.8419,
      "mean_token_accuracy": 0.23805829882621765,
      "num_tokens": 136792339.0,
      "step": 65915
    },
    {
      "entropy": 5.689732837677002,
      "epoch": 5.979680696661829,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020929166411894046,
      "loss": 4.7667,
      "mean_token_accuracy": 0.23990911692380906,
      "num_tokens": 136803242.0,
      "step": 65920
    },
    {
      "entropy": 5.808573150634766,
      "epoch": 5.980134252539913,
      "grad_norm": 1.25,
      "learning_rate": 0.00020926071889186787,
      "loss": 4.8745,
      "mean_token_accuracy": 0.22632445394992828,
      "num_tokens": 136812934.0,
      "step": 65925
    },
    {
      "entropy": 5.7465427875518795,
      "epoch": 5.980587808417997,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020922977502430047,
      "loss": 4.8021,
      "mean_token_accuracy": 0.24365047216415406,
      "num_tokens": 136822749.0,
      "step": 65930
    },
    {
      "entropy": 5.631460857391358,
      "epoch": 5.981041364296082,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020919883251687804,
      "loss": 4.6724,
      "mean_token_accuracy": 0.23925328701734544,
      "num_tokens": 136832573.0,
      "step": 65935
    },
    {
      "entropy": 5.67613787651062,
      "epoch": 5.981494920174166,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020916789137024052,
      "loss": 4.801,
      "mean_token_accuracy": 0.23820070028305054,
      "num_tokens": 136842372.0,
      "step": 65940
    },
    {
      "entropy": 5.729016399383545,
      "epoch": 5.981948476052249,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020913695158502772,
      "loss": 4.8004,
      "mean_token_accuracy": 0.23460334688425064,
      "num_tokens": 136853626.0,
      "step": 65945
    },
    {
      "entropy": 5.797708415985108,
      "epoch": 5.9824020319303335,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020910601316187955,
      "loss": 4.8817,
      "mean_token_accuracy": 0.22298693060874938,
      "num_tokens": 136864281.0,
      "step": 65950
    },
    {
      "entropy": 5.756408262252807,
      "epoch": 5.982855587808418,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002090750761014359,
      "loss": 4.8467,
      "mean_token_accuracy": 0.22648535221815108,
      "num_tokens": 136874451.0,
      "step": 65955
    },
    {
      "entropy": 5.750505161285401,
      "epoch": 5.983309143686502,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002090441404043364,
      "loss": 4.8884,
      "mean_token_accuracy": 0.22009871304035186,
      "num_tokens": 136884125.0,
      "step": 65960
    },
    {
      "entropy": 5.72385802268982,
      "epoch": 5.983762699564586,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020901320607122093,
      "loss": 4.8346,
      "mean_token_accuracy": 0.23265381455421447,
      "num_tokens": 136894278.0,
      "step": 65965
    },
    {
      "entropy": 5.7294622421264645,
      "epoch": 5.9842162554426706,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020898227310272915,
      "loss": 4.8499,
      "mean_token_accuracy": 0.22500678300857543,
      "num_tokens": 136904212.0,
      "step": 65970
    },
    {
      "entropy": 5.743808794021606,
      "epoch": 5.984669811320755,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020895134149950085,
      "loss": 4.7645,
      "mean_token_accuracy": 0.23891642987728118,
      "num_tokens": 136914591.0,
      "step": 65975
    },
    {
      "entropy": 5.715180730819702,
      "epoch": 5.985123367198839,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020892041126217555,
      "loss": 4.8135,
      "mean_token_accuracy": 0.23792488873004913,
      "num_tokens": 136925271.0,
      "step": 65980
    },
    {
      "entropy": 5.686546897888183,
      "epoch": 5.985576923076923,
      "grad_norm": 1.3125,
      "learning_rate": 0.000208889482391393,
      "loss": 4.8222,
      "mean_token_accuracy": 0.23465819358825685,
      "num_tokens": 136936575.0,
      "step": 65985
    },
    {
      "entropy": 5.753918123245239,
      "epoch": 5.986030478955008,
      "grad_norm": 1.28125,
      "learning_rate": 0.00020885855488779288,
      "loss": 4.8673,
      "mean_token_accuracy": 0.23075160682201384,
      "num_tokens": 136946488.0,
      "step": 65990
    },
    {
      "entropy": 5.738576889038086,
      "epoch": 5.986484034833092,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020882762875201456,
      "loss": 4.8798,
      "mean_token_accuracy": 0.23034765869379042,
      "num_tokens": 136956592.0,
      "step": 65995
    },
    {
      "entropy": 5.817387294769287,
      "epoch": 5.986937590711175,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020879670398469785,
      "loss": 4.8988,
      "mean_token_accuracy": 0.22049227207899094,
      "num_tokens": 136966878.0,
      "step": 66000
    },
    {
      "epoch": 5.986937590711175,
      "eval_entropy": 5.464585979923403,
      "eval_loss": 4.9393534660339355,
      "eval_mean_token_accuracy": 0.23471917223640426,
      "eval_num_tokens": 136966878.0,
      "eval_runtime": 20.3587,
      "eval_samples_per_second": 1736.853,
      "eval_steps_per_second": 217.107,
      "step": 66000
    },
    {
      "entropy": 5.75385012626648,
      "epoch": 5.9873911465892595,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020876578058648217,
      "loss": 4.8879,
      "mean_token_accuracy": 0.22600155472755432,
      "num_tokens": 136977626.0,
      "step": 66005
    },
    {
      "entropy": 5.689811372756958,
      "epoch": 5.987844702467344,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020873485855800694,
      "loss": 4.8329,
      "mean_token_accuracy": 0.23349374234676362,
      "num_tokens": 136987584.0,
      "step": 66010
    },
    {
      "entropy": 5.73044114112854,
      "epoch": 5.988298258345428,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002087039378999118,
      "loss": 4.8307,
      "mean_token_accuracy": 0.22448968887329102,
      "num_tokens": 136996556.0,
      "step": 66015
    },
    {
      "entropy": 5.778934574127197,
      "epoch": 5.988751814223512,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002086730186128361,
      "loss": 4.882,
      "mean_token_accuracy": 0.23097476959228516,
      "num_tokens": 137006381.0,
      "step": 66020
    },
    {
      "entropy": 5.756524848937988,
      "epoch": 5.989205370101597,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002086421006974193,
      "loss": 4.9038,
      "mean_token_accuracy": 0.2289631873369217,
      "num_tokens": 137016575.0,
      "step": 66025
    },
    {
      "entropy": 5.810823154449463,
      "epoch": 5.989658925979681,
      "grad_norm": 1.5,
      "learning_rate": 0.00020861118415430068,
      "loss": 4.9263,
      "mean_token_accuracy": 0.22235729694366455,
      "num_tokens": 137026825.0,
      "step": 66030
    },
    {
      "entropy": 5.784350442886352,
      "epoch": 5.990112481857765,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002085802689841198,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22895906269550323,
      "num_tokens": 137037064.0,
      "step": 66035
    },
    {
      "entropy": 5.700135135650635,
      "epoch": 5.990566037735849,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020854935518751583,
      "loss": 4.8286,
      "mean_token_accuracy": 0.23818369954824448,
      "num_tokens": 137048114.0,
      "step": 66040
    },
    {
      "entropy": 5.840141630172729,
      "epoch": 5.991019593613933,
      "grad_norm": 1.609375,
      "learning_rate": 0.00020851844276512815,
      "loss": 4.9903,
      "mean_token_accuracy": 0.2187919571995735,
      "num_tokens": 137057942.0,
      "step": 66045
    },
    {
      "entropy": 5.760671901702881,
      "epoch": 5.991473149492017,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000208487531717596,
      "loss": 4.7789,
      "mean_token_accuracy": 0.2422114834189415,
      "num_tokens": 137067422.0,
      "step": 66050
    },
    {
      "entropy": 5.745446825027466,
      "epoch": 5.991926705370101,
      "grad_norm": 1.625,
      "learning_rate": 0.00020845662204555866,
      "loss": 4.85,
      "mean_token_accuracy": 0.22586358189582825,
      "num_tokens": 137077613.0,
      "step": 66055
    },
    {
      "entropy": 5.732156896591187,
      "epoch": 5.992380261248186,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002084257137496553,
      "loss": 4.763,
      "mean_token_accuracy": 0.2349142238497734,
      "num_tokens": 137086934.0,
      "step": 66060
    },
    {
      "entropy": 5.743528509140015,
      "epoch": 5.99283381712627,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020839480683052514,
      "loss": 4.8105,
      "mean_token_accuracy": 0.24207396507263185,
      "num_tokens": 137097548.0,
      "step": 66065
    },
    {
      "entropy": 5.7381260871887205,
      "epoch": 5.993287373004354,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020836390128880738,
      "loss": 4.8536,
      "mean_token_accuracy": 0.2262520208954811,
      "num_tokens": 137107458.0,
      "step": 66070
    },
    {
      "entropy": 5.713402509689331,
      "epoch": 5.993740928882438,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020833299712514115,
      "loss": 4.782,
      "mean_token_accuracy": 0.23140530437231063,
      "num_tokens": 137117666.0,
      "step": 66075
    },
    {
      "entropy": 5.754132461547852,
      "epoch": 5.994194484760523,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020830209434016545,
      "loss": 4.8629,
      "mean_token_accuracy": 0.22437476962804795,
      "num_tokens": 137127599.0,
      "step": 66080
    },
    {
      "entropy": 5.73768892288208,
      "epoch": 5.994648040638607,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020827119293451946,
      "loss": 4.8644,
      "mean_token_accuracy": 0.23572910726070403,
      "num_tokens": 137137925.0,
      "step": 66085
    },
    {
      "entropy": 5.662273836135864,
      "epoch": 5.995101596516691,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020824029290884223,
      "loss": 4.7072,
      "mean_token_accuracy": 0.24964932054281236,
      "num_tokens": 137149054.0,
      "step": 66090
    },
    {
      "entropy": 5.755826187133789,
      "epoch": 5.995555152394775,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020820939426377278,
      "loss": 4.9158,
      "mean_token_accuracy": 0.22566907107830048,
      "num_tokens": 137159503.0,
      "step": 66095
    },
    {
      "entropy": 5.755894660949707,
      "epoch": 5.996008708272859,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020817849699995,
      "loss": 4.9074,
      "mean_token_accuracy": 0.23114770501852036,
      "num_tokens": 137170140.0,
      "step": 66100
    },
    {
      "entropy": 5.700460147857666,
      "epoch": 5.996462264150943,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000208147601118013,
      "loss": 4.8049,
      "mean_token_accuracy": 0.23600828647613525,
      "num_tokens": 137180591.0,
      "step": 66105
    },
    {
      "entropy": 5.739225292205811,
      "epoch": 5.996915820029027,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002081167066186006,
      "loss": 4.8263,
      "mean_token_accuracy": 0.23449210822582245,
      "num_tokens": 137191928.0,
      "step": 66110
    },
    {
      "entropy": 5.774206590652466,
      "epoch": 5.997369375907112,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002080858135023518,
      "loss": 4.9182,
      "mean_token_accuracy": 0.22588116377592088,
      "num_tokens": 137200909.0,
      "step": 66115
    },
    {
      "entropy": 5.734349298477173,
      "epoch": 5.997822931785196,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020805492176990538,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22606894075870515,
      "num_tokens": 137212277.0,
      "step": 66120
    },
    {
      "entropy": 5.7652946472167965,
      "epoch": 5.99827648766328,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002080240314219003,
      "loss": 4.8499,
      "mean_token_accuracy": 0.22776824682950975,
      "num_tokens": 137222761.0,
      "step": 66125
    },
    {
      "entropy": 5.76491346359253,
      "epoch": 5.998730043541364,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002079931424589752,
      "loss": 4.8365,
      "mean_token_accuracy": 0.2343516618013382,
      "num_tokens": 137233686.0,
      "step": 66130
    },
    {
      "entropy": 5.72426209449768,
      "epoch": 5.999183599419449,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020796225488176908,
      "loss": 4.8065,
      "mean_token_accuracy": 0.23374013006687164,
      "num_tokens": 137244122.0,
      "step": 66135
    },
    {
      "entropy": 5.674641513824463,
      "epoch": 5.999637155297533,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002079313686909206,
      "loss": 4.7268,
      "mean_token_accuracy": 0.23613055050373077,
      "num_tokens": 137255089.0,
      "step": 66140
    },
    {
      "entropy": 5.779843187332153,
      "epoch": 6.000090711175617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002079004838870685,
      "loss": 4.8614,
      "mean_token_accuracy": 0.2319876804947853,
      "num_tokens": 137266052.0,
      "step": 66145
    },
    {
      "entropy": 5.7377325057983395,
      "epoch": 6.000544267053701,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020786960047085147,
      "loss": 4.8093,
      "mean_token_accuracy": 0.23080038130283356,
      "num_tokens": 137276761.0,
      "step": 66150
    },
    {
      "entropy": 5.717046213150025,
      "epoch": 6.000997822931785,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020783871844290826,
      "loss": 4.796,
      "mean_token_accuracy": 0.24382565170526505,
      "num_tokens": 137286780.0,
      "step": 66155
    },
    {
      "entropy": 5.777522039413452,
      "epoch": 6.001451378809869,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020780783780387747,
      "loss": 4.8783,
      "mean_token_accuracy": 0.2293965309858322,
      "num_tokens": 137297211.0,
      "step": 66160
    },
    {
      "entropy": 5.736704635620117,
      "epoch": 6.001904934687953,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002077769585543976,
      "loss": 4.8665,
      "mean_token_accuracy": 0.23068630546331406,
      "num_tokens": 137307295.0,
      "step": 66165
    },
    {
      "entropy": 5.735664892196655,
      "epoch": 6.002358490566038,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020774608069510747,
      "loss": 4.7848,
      "mean_token_accuracy": 0.23730894178152084,
      "num_tokens": 137316677.0,
      "step": 66170
    },
    {
      "entropy": 5.718030500411987,
      "epoch": 6.002812046444122,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020771520422664552,
      "loss": 4.7609,
      "mean_token_accuracy": 0.24539123475551605,
      "num_tokens": 137326254.0,
      "step": 66175
    },
    {
      "entropy": 5.752772665023803,
      "epoch": 6.003265602322206,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002076843291496503,
      "loss": 4.8785,
      "mean_token_accuracy": 0.2279679298400879,
      "num_tokens": 137336974.0,
      "step": 66180
    },
    {
      "entropy": 5.674230098724365,
      "epoch": 6.0037191582002905,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020765345546476023,
      "loss": 4.7238,
      "mean_token_accuracy": 0.2425843060016632,
      "num_tokens": 137347170.0,
      "step": 66185
    },
    {
      "entropy": 5.7344084739685055,
      "epoch": 6.004172714078375,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020762258317261396,
      "loss": 4.8262,
      "mean_token_accuracy": 0.23801735192537307,
      "num_tokens": 137356908.0,
      "step": 66190
    },
    {
      "entropy": 5.68155345916748,
      "epoch": 6.004626269956459,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020759171227384982,
      "loss": 4.7279,
      "mean_token_accuracy": 0.23348651975393295,
      "num_tokens": 137367279.0,
      "step": 66195
    },
    {
      "entropy": 5.644135427474976,
      "epoch": 6.005079825834543,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020756084276910613,
      "loss": 4.7526,
      "mean_token_accuracy": 0.24292644262313842,
      "num_tokens": 137378296.0,
      "step": 66200
    },
    {
      "entropy": 5.67154893875122,
      "epoch": 6.005533381712627,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002075299746590215,
      "loss": 4.8087,
      "mean_token_accuracy": 0.23969862312078477,
      "num_tokens": 137389290.0,
      "step": 66205
    },
    {
      "entropy": 5.690626525878907,
      "epoch": 6.005986937590711,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002074991079442342,
      "loss": 4.747,
      "mean_token_accuracy": 0.2414974883198738,
      "num_tokens": 137399803.0,
      "step": 66210
    },
    {
      "entropy": 5.7718945980072025,
      "epoch": 6.006440493468795,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020746824262538255,
      "loss": 4.8401,
      "mean_token_accuracy": 0.2338146835565567,
      "num_tokens": 137410467.0,
      "step": 66215
    },
    {
      "entropy": 5.701465797424317,
      "epoch": 6.006894049346879,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002074373787031048,
      "loss": 4.7426,
      "mean_token_accuracy": 0.23594671189785005,
      "num_tokens": 137420830.0,
      "step": 66220
    },
    {
      "entropy": 5.703558921813965,
      "epoch": 6.007347605224964,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002074065161780393,
      "loss": 4.7879,
      "mean_token_accuracy": 0.2399976834654808,
      "num_tokens": 137431052.0,
      "step": 66225
    },
    {
      "entropy": 5.728919601440429,
      "epoch": 6.007801161103048,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020737565505082433,
      "loss": 4.857,
      "mean_token_accuracy": 0.22893405109643936,
      "num_tokens": 137440270.0,
      "step": 66230
    },
    {
      "entropy": 5.73771300315857,
      "epoch": 6.008254716981132,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020734479532209793,
      "loss": 4.7802,
      "mean_token_accuracy": 0.23643808811903,
      "num_tokens": 137450276.0,
      "step": 66235
    },
    {
      "entropy": 5.7695090770721436,
      "epoch": 6.0087082728592165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020731393699249844,
      "loss": 4.7947,
      "mean_token_accuracy": 0.24179090261459352,
      "num_tokens": 137461202.0,
      "step": 66240
    },
    {
      "entropy": 5.755919551849365,
      "epoch": 6.009161828737301,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002072830800626641,
      "loss": 4.7966,
      "mean_token_accuracy": 0.24412294179201127,
      "num_tokens": 137471202.0,
      "step": 66245
    },
    {
      "entropy": 5.793158054351807,
      "epoch": 6.009615384615385,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020725222453323283,
      "loss": 4.9073,
      "mean_token_accuracy": 0.2238181486725807,
      "num_tokens": 137481373.0,
      "step": 66250
    },
    {
      "entropy": 5.731910324096679,
      "epoch": 6.010068940493468,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020722137040484284,
      "loss": 4.8714,
      "mean_token_accuracy": 0.23067139089107513,
      "num_tokens": 137492045.0,
      "step": 66255
    },
    {
      "entropy": 5.737636375427246,
      "epoch": 6.010522496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020719051767813213,
      "loss": 4.8068,
      "mean_token_accuracy": 0.2338674023747444,
      "num_tokens": 137501910.0,
      "step": 66260
    },
    {
      "entropy": 5.776727104187012,
      "epoch": 6.010976052249637,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020715966635373884,
      "loss": 4.895,
      "mean_token_accuracy": 0.23036423772573472,
      "num_tokens": 137512722.0,
      "step": 66265
    },
    {
      "entropy": 5.816059303283692,
      "epoch": 6.011429608127721,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020712881643230092,
      "loss": 4.8514,
      "mean_token_accuracy": 0.24085594862699508,
      "num_tokens": 137524245.0,
      "step": 66270
    },
    {
      "entropy": 5.795741224288941,
      "epoch": 6.0118831640058055,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002070979679144564,
      "loss": 4.8694,
      "mean_token_accuracy": 0.22552659511566162,
      "num_tokens": 137533977.0,
      "step": 66275
    },
    {
      "entropy": 5.828777265548706,
      "epoch": 6.01233671988389,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020706712080084317,
      "loss": 4.8911,
      "mean_token_accuracy": 0.2272505059838295,
      "num_tokens": 137545801.0,
      "step": 66280
    },
    {
      "entropy": 5.790999174118042,
      "epoch": 6.012790275761974,
      "grad_norm": 1.625,
      "learning_rate": 0.00020703627509209915,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22576782256364822,
      "num_tokens": 137556109.0,
      "step": 66285
    },
    {
      "entropy": 5.7035070896148685,
      "epoch": 6.013243831640058,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020700543078886241,
      "loss": 4.7224,
      "mean_token_accuracy": 0.24791547954082488,
      "num_tokens": 137568060.0,
      "step": 66290
    },
    {
      "entropy": 5.744124126434326,
      "epoch": 6.0136973875181425,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002069745878917706,
      "loss": 4.8104,
      "mean_token_accuracy": 0.2391796216368675,
      "num_tokens": 137579167.0,
      "step": 66295
    },
    {
      "entropy": 5.7536098003387455,
      "epoch": 6.014150943396227,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020694374640146162,
      "loss": 4.8062,
      "mean_token_accuracy": 0.2346737027168274,
      "num_tokens": 137589050.0,
      "step": 66300
    },
    {
      "entropy": 5.738834619522095,
      "epoch": 6.01460449927431,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020691290631857335,
      "loss": 4.8249,
      "mean_token_accuracy": 0.24158884733915328,
      "num_tokens": 137599496.0,
      "step": 66305
    },
    {
      "entropy": 5.753095483779907,
      "epoch": 6.0150580551523944,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002068820676437435,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23840940445661546,
      "num_tokens": 137609272.0,
      "step": 66310
    },
    {
      "entropy": 5.69379563331604,
      "epoch": 6.015511611030479,
      "grad_norm": 1.453125,
      "learning_rate": 0.0002068512303776099,
      "loss": 4.7502,
      "mean_token_accuracy": 0.23459135740995407,
      "num_tokens": 137619541.0,
      "step": 66315
    },
    {
      "entropy": 5.741668176651001,
      "epoch": 6.015965166908563,
      "grad_norm": 1.375,
      "learning_rate": 0.0002068203945208102,
      "loss": 4.8415,
      "mean_token_accuracy": 0.23038293570280075,
      "num_tokens": 137629894.0,
      "step": 66320
    },
    {
      "entropy": 5.8608259677886965,
      "epoch": 6.016418722786647,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020678956007398215,
      "loss": 4.9505,
      "mean_token_accuracy": 0.22577807754278184,
      "num_tokens": 137639892.0,
      "step": 66325
    },
    {
      "entropy": 5.755406236648559,
      "epoch": 6.0168722786647315,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020675872703776327,
      "loss": 4.8311,
      "mean_token_accuracy": 0.22812165319919586,
      "num_tokens": 137650166.0,
      "step": 66330
    },
    {
      "entropy": 5.635398435592651,
      "epoch": 6.017325834542816,
      "grad_norm": 1.5,
      "learning_rate": 0.0002067278954127914,
      "loss": 4.7039,
      "mean_token_accuracy": 0.23843623101711273,
      "num_tokens": 137659676.0,
      "step": 66335
    },
    {
      "entropy": 5.732817316055298,
      "epoch": 6.0177793904209,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020669706519970405,
      "loss": 4.8642,
      "mean_token_accuracy": 0.2198810264468193,
      "num_tokens": 137670257.0,
      "step": 66340
    },
    {
      "entropy": 5.724808692932129,
      "epoch": 6.018232946298984,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020666623639913873,
      "loss": 4.7859,
      "mean_token_accuracy": 0.23115989267826081,
      "num_tokens": 137680797.0,
      "step": 66345
    },
    {
      "entropy": 5.782122659683227,
      "epoch": 6.018686502177069,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002066354090117331,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22711153328418732,
      "num_tokens": 137692199.0,
      "step": 66350
    },
    {
      "entropy": 5.739189147949219,
      "epoch": 6.019140058055152,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020660458303812469,
      "loss": 4.8769,
      "mean_token_accuracy": 0.21798599511384964,
      "num_tokens": 137703485.0,
      "step": 66355
    },
    {
      "entropy": 5.781496953964234,
      "epoch": 6.019593613933236,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00020657375847895089,
      "loss": 4.8295,
      "mean_token_accuracy": 0.23526073843240738,
      "num_tokens": 137714940.0,
      "step": 66360
    },
    {
      "entropy": 5.761775016784668,
      "epoch": 6.0200471698113205,
      "grad_norm": 1.5,
      "learning_rate": 0.00020654293533484914,
      "loss": 4.8417,
      "mean_token_accuracy": 0.22803981453180314,
      "num_tokens": 137724389.0,
      "step": 66365
    },
    {
      "entropy": 5.7065465450286865,
      "epoch": 6.020500725689405,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020651211360645693,
      "loss": 4.7041,
      "mean_token_accuracy": 0.24908433556556703,
      "num_tokens": 137734360.0,
      "step": 66370
    },
    {
      "entropy": 5.810867977142334,
      "epoch": 6.020954281567489,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002064812932944118,
      "loss": 4.9,
      "mean_token_accuracy": 0.22641763389110564,
      "num_tokens": 137745479.0,
      "step": 66375
    },
    {
      "entropy": 5.71432409286499,
      "epoch": 6.021407837445573,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020645047439935083,
      "loss": 4.747,
      "mean_token_accuracy": 0.23879550248384476,
      "num_tokens": 137755162.0,
      "step": 66380
    },
    {
      "entropy": 5.7199805736541744,
      "epoch": 6.0218613933236576,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020641965692191157,
      "loss": 4.8007,
      "mean_token_accuracy": 0.24253113567829132,
      "num_tokens": 137766090.0,
      "step": 66385
    },
    {
      "entropy": 5.750055599212646,
      "epoch": 6.022314949201742,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020638884086273136,
      "loss": 4.8166,
      "mean_token_accuracy": 0.23423076272010804,
      "num_tokens": 137775526.0,
      "step": 66390
    },
    {
      "entropy": 5.790779066085816,
      "epoch": 6.022768505079826,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020635802622244733,
      "loss": 4.9433,
      "mean_token_accuracy": 0.22892213463783265,
      "num_tokens": 137786005.0,
      "step": 66395
    },
    {
      "entropy": 5.747471809387207,
      "epoch": 6.02322206095791,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020632721300169676,
      "loss": 4.8003,
      "mean_token_accuracy": 0.23646320104599,
      "num_tokens": 137796609.0,
      "step": 66400
    },
    {
      "entropy": 5.788185405731201,
      "epoch": 6.023675616835995,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020629640120111702,
      "loss": 4.8725,
      "mean_token_accuracy": 0.22325920015573503,
      "num_tokens": 137806698.0,
      "step": 66405
    },
    {
      "entropy": 5.798457479476928,
      "epoch": 6.024129172714078,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002062655908213452,
      "loss": 4.9144,
      "mean_token_accuracy": 0.22789032608270646,
      "num_tokens": 137816457.0,
      "step": 66410
    },
    {
      "entropy": 5.75800724029541,
      "epoch": 6.024582728592162,
      "grad_norm": 1.515625,
      "learning_rate": 0.0002062347818630184,
      "loss": 4.851,
      "mean_token_accuracy": 0.23210099786520005,
      "num_tokens": 137826191.0,
      "step": 66415
    },
    {
      "entropy": 5.777729320526123,
      "epoch": 6.0250362844702465,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002062039743267739,
      "loss": 4.8673,
      "mean_token_accuracy": 0.22945381999015807,
      "num_tokens": 137837713.0,
      "step": 66420
    },
    {
      "entropy": 5.759807682037353,
      "epoch": 6.025489840348331,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020617316821324878,
      "loss": 4.7632,
      "mean_token_accuracy": 0.24335407018661498,
      "num_tokens": 137847960.0,
      "step": 66425
    },
    {
      "entropy": 5.790507078170776,
      "epoch": 6.025943396226415,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020614236352307997,
      "loss": 4.8835,
      "mean_token_accuracy": 0.22405871748924255,
      "num_tokens": 137858931.0,
      "step": 66430
    },
    {
      "entropy": 5.665984344482422,
      "epoch": 6.026396952104499,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002061115602569047,
      "loss": 4.7268,
      "mean_token_accuracy": 0.243442103266716,
      "num_tokens": 137869402.0,
      "step": 66435
    },
    {
      "entropy": 5.729118680953979,
      "epoch": 6.026850507982584,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020608075841535995,
      "loss": 4.7916,
      "mean_token_accuracy": 0.23749417066574097,
      "num_tokens": 137879908.0,
      "step": 66440
    },
    {
      "entropy": 5.724485588073731,
      "epoch": 6.027304063860668,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020604995799908266,
      "loss": 4.7784,
      "mean_token_accuracy": 0.23462527841329575,
      "num_tokens": 137888993.0,
      "step": 66445
    },
    {
      "entropy": 5.742689371109009,
      "epoch": 6.027757619738752,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020601915900870976,
      "loss": 4.8321,
      "mean_token_accuracy": 0.2362409085035324,
      "num_tokens": 137899560.0,
      "step": 66450
    },
    {
      "entropy": 5.707337331771851,
      "epoch": 6.028211175616836,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020598836144487825,
      "loss": 4.7797,
      "mean_token_accuracy": 0.2404465228319168,
      "num_tokens": 137909988.0,
      "step": 66455
    },
    {
      "entropy": 5.7322837829589846,
      "epoch": 6.02866473149492,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000205957565308225,
      "loss": 4.8066,
      "mean_token_accuracy": 0.23529352843761445,
      "num_tokens": 137919458.0,
      "step": 66460
    },
    {
      "entropy": 5.6868517875671385,
      "epoch": 6.029118287373004,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020592677059938687,
      "loss": 4.7783,
      "mean_token_accuracy": 0.23523883223533631,
      "num_tokens": 137929489.0,
      "step": 66465
    },
    {
      "entropy": 5.732542848587036,
      "epoch": 6.029571843251088,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020589597731900079,
      "loss": 4.7984,
      "mean_token_accuracy": 0.2392079398036003,
      "num_tokens": 137939775.0,
      "step": 66470
    },
    {
      "entropy": 5.734451580047607,
      "epoch": 6.030025399129173,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020586518546770354,
      "loss": 4.8031,
      "mean_token_accuracy": 0.23552582859992982,
      "num_tokens": 137950580.0,
      "step": 66475
    },
    {
      "entropy": 5.722465562820434,
      "epoch": 6.030478955007257,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020583439504613183,
      "loss": 4.7993,
      "mean_token_accuracy": 0.23777174949645996,
      "num_tokens": 137960836.0,
      "step": 66480
    },
    {
      "entropy": 5.740680980682373,
      "epoch": 6.030932510885341,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020580360605492237,
      "loss": 4.8531,
      "mean_token_accuracy": 0.23737360537052155,
      "num_tokens": 137972195.0,
      "step": 66485
    },
    {
      "entropy": 5.704543924331665,
      "epoch": 6.031386066763425,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020577281849471206,
      "loss": 4.7992,
      "mean_token_accuracy": 0.23055762350559234,
      "num_tokens": 137982314.0,
      "step": 66490
    },
    {
      "entropy": 5.693190240859986,
      "epoch": 6.03183962264151,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020574203236613754,
      "loss": 4.7607,
      "mean_token_accuracy": 0.2294461101293564,
      "num_tokens": 137992865.0,
      "step": 66495
    },
    {
      "entropy": 5.845944881439209,
      "epoch": 6.032293178519594,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020571124766983535,
      "loss": 4.8977,
      "mean_token_accuracy": 0.22600008249282838,
      "num_tokens": 138003048.0,
      "step": 66500
    },
    {
      "entropy": 5.761306381225586,
      "epoch": 6.032746734397678,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020568046440644228,
      "loss": 4.8652,
      "mean_token_accuracy": 0.2313055321574211,
      "num_tokens": 138013341.0,
      "step": 66505
    },
    {
      "entropy": 5.733290958404541,
      "epoch": 6.0332002902757615,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020564968257659482,
      "loss": 4.8098,
      "mean_token_accuracy": 0.2271922931075096,
      "num_tokens": 138023623.0,
      "step": 66510
    },
    {
      "entropy": 5.76942138671875,
      "epoch": 6.033653846153846,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002056189021809296,
      "loss": 4.9187,
      "mean_token_accuracy": 0.22061147689819335,
      "num_tokens": 138035556.0,
      "step": 66515
    },
    {
      "entropy": 5.721204328536987,
      "epoch": 6.03410740203193,
      "grad_norm": 1.59375,
      "learning_rate": 0.00020558812322008318,
      "loss": 4.7915,
      "mean_token_accuracy": 0.23757714480161668,
      "num_tokens": 138045849.0,
      "step": 66520
    },
    {
      "entropy": 5.731696128845215,
      "epoch": 6.034560957910014,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002055573456946921,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23449736982584,
      "num_tokens": 138057033.0,
      "step": 66525
    },
    {
      "entropy": 5.733651638031006,
      "epoch": 6.035014513788099,
      "grad_norm": 1.25,
      "learning_rate": 0.00020552656960539278,
      "loss": 4.8191,
      "mean_token_accuracy": 0.23583363592624665,
      "num_tokens": 138068638.0,
      "step": 66530
    },
    {
      "entropy": 5.768744707107544,
      "epoch": 6.035468069666183,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020549579495282168,
      "loss": 4.9065,
      "mean_token_accuracy": 0.23226333558559417,
      "num_tokens": 138079638.0,
      "step": 66535
    },
    {
      "entropy": 5.8073607921600345,
      "epoch": 6.035921625544267,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002054650217376153,
      "loss": 4.916,
      "mean_token_accuracy": 0.22984962314367294,
      "num_tokens": 138089753.0,
      "step": 66540
    },
    {
      "entropy": 5.813331031799317,
      "epoch": 6.036375181422351,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020543424996040993,
      "loss": 4.8998,
      "mean_token_accuracy": 0.23250449299812317,
      "num_tokens": 138099829.0,
      "step": 66545
    },
    {
      "entropy": 5.754285573959351,
      "epoch": 6.036828737300436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020540347962184202,
      "loss": 4.8043,
      "mean_token_accuracy": 0.23951468318700792,
      "num_tokens": 138110722.0,
      "step": 66550
    },
    {
      "entropy": 5.723865795135498,
      "epoch": 6.03728229317852,
      "grad_norm": 1.5,
      "learning_rate": 0.00020537271072254792,
      "loss": 4.8074,
      "mean_token_accuracy": 0.2351305827498436,
      "num_tokens": 138120753.0,
      "step": 66555
    },
    {
      "entropy": 5.724913549423218,
      "epoch": 6.037735849056604,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020534194326316386,
      "loss": 4.8123,
      "mean_token_accuracy": 0.22997965663671494,
      "num_tokens": 138131372.0,
      "step": 66560
    },
    {
      "entropy": 5.771505498886109,
      "epoch": 6.038189404934688,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020531117724432624,
      "loss": 4.885,
      "mean_token_accuracy": 0.22923787087202072,
      "num_tokens": 138141057.0,
      "step": 66565
    },
    {
      "entropy": 5.677001333236694,
      "epoch": 6.038642960812772,
      "grad_norm": 1.375,
      "learning_rate": 0.00020528041266667114,
      "loss": 4.6851,
      "mean_token_accuracy": 0.2446715384721756,
      "num_tokens": 138150638.0,
      "step": 66570
    },
    {
      "entropy": 5.737273216247559,
      "epoch": 6.039096516690856,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020524964953083492,
      "loss": 4.8403,
      "mean_token_accuracy": 0.23692838549613954,
      "num_tokens": 138160924.0,
      "step": 66575
    },
    {
      "entropy": 5.724655246734619,
      "epoch": 6.03955007256894,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002052188878374537,
      "loss": 4.8199,
      "mean_token_accuracy": 0.23315859586000443,
      "num_tokens": 138171865.0,
      "step": 66580
    },
    {
      "entropy": 5.671969270706176,
      "epoch": 6.040003628447025,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020518812758716365,
      "loss": 4.7634,
      "mean_token_accuracy": 0.23790756165981292,
      "num_tokens": 138181521.0,
      "step": 66585
    },
    {
      "entropy": 5.76436219215393,
      "epoch": 6.040457184325109,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020515736878060092,
      "loss": 4.9458,
      "mean_token_accuracy": 0.2265777289867401,
      "num_tokens": 138192265.0,
      "step": 66590
    },
    {
      "entropy": 5.715881633758545,
      "epoch": 6.040910740203193,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020512661141840167,
      "loss": 4.797,
      "mean_token_accuracy": 0.23840294778347015,
      "num_tokens": 138202912.0,
      "step": 66595
    },
    {
      "entropy": 5.760799360275269,
      "epoch": 6.0413642960812775,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020509585550120186,
      "loss": 4.8724,
      "mean_token_accuracy": 0.23188168108463286,
      "num_tokens": 138213376.0,
      "step": 66600
    },
    {
      "entropy": 5.738639974594117,
      "epoch": 6.041817851959362,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002050651010296375,
      "loss": 4.823,
      "mean_token_accuracy": 0.23864850848913194,
      "num_tokens": 138224190.0,
      "step": 66605
    },
    {
      "entropy": 5.752700424194336,
      "epoch": 6.042271407837446,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020503434800434479,
      "loss": 4.8465,
      "mean_token_accuracy": 0.23971777856349946,
      "num_tokens": 138235112.0,
      "step": 66610
    },
    {
      "entropy": 5.733872938156128,
      "epoch": 6.042724963715529,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002050035964259595,
      "loss": 4.8439,
      "mean_token_accuracy": 0.23032308071851731,
      "num_tokens": 138246534.0,
      "step": 66615
    },
    {
      "entropy": 5.687152957916259,
      "epoch": 6.043178519593614,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020497284629511769,
      "loss": 4.7724,
      "mean_token_accuracy": 0.2311318427324295,
      "num_tokens": 138257478.0,
      "step": 66620
    },
    {
      "entropy": 5.719878816604615,
      "epoch": 6.043632075471698,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002049420976124553,
      "loss": 4.8345,
      "mean_token_accuracy": 0.2353602334856987,
      "num_tokens": 138268099.0,
      "step": 66625
    },
    {
      "entropy": 5.780182504653931,
      "epoch": 6.044085631349782,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020491135037860818,
      "loss": 4.876,
      "mean_token_accuracy": 0.2344715565443039,
      "num_tokens": 138278982.0,
      "step": 66630
    },
    {
      "entropy": 5.784557771682739,
      "epoch": 6.044539187227866,
      "grad_norm": 1.5,
      "learning_rate": 0.00020488060459421215,
      "loss": 4.8638,
      "mean_token_accuracy": 0.2360062286257744,
      "num_tokens": 138289078.0,
      "step": 66635
    },
    {
      "entropy": 5.707110643386841,
      "epoch": 6.044992743105951,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020484986025990317,
      "loss": 4.7557,
      "mean_token_accuracy": 0.23790333569049835,
      "num_tokens": 138299177.0,
      "step": 66640
    },
    {
      "entropy": 5.7959897994995115,
      "epoch": 6.045446298984035,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020481911737631686,
      "loss": 4.9305,
      "mean_token_accuracy": 0.2224516302347183,
      "num_tokens": 138309576.0,
      "step": 66645
    },
    {
      "entropy": 5.776204919815063,
      "epoch": 6.045899854862119,
      "grad_norm": 1.625,
      "learning_rate": 0.00020478837594408916,
      "loss": 4.8651,
      "mean_token_accuracy": 0.23300817906856536,
      "num_tokens": 138318576.0,
      "step": 66650
    },
    {
      "entropy": 5.801296281814575,
      "epoch": 6.0463534107402035,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002047576359638557,
      "loss": 4.8993,
      "mean_token_accuracy": 0.22233835756778716,
      "num_tokens": 138328091.0,
      "step": 66655
    },
    {
      "entropy": 5.680299949645996,
      "epoch": 6.046806966618288,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020472689743625223,
      "loss": 4.8227,
      "mean_token_accuracy": 0.23387429416179656,
      "num_tokens": 138338925.0,
      "step": 66660
    },
    {
      "entropy": 5.764724063873291,
      "epoch": 6.047260522496371,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020469616036191446,
      "loss": 4.8339,
      "mean_token_accuracy": 0.22994198203086852,
      "num_tokens": 138349749.0,
      "step": 66665
    },
    {
      "entropy": 5.743840456008911,
      "epoch": 6.047714078374455,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020466542474147803,
      "loss": 4.7571,
      "mean_token_accuracy": 0.23860576897859573,
      "num_tokens": 138360224.0,
      "step": 66670
    },
    {
      "entropy": 5.74766435623169,
      "epoch": 6.04816763425254,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020463469057557855,
      "loss": 4.8331,
      "mean_token_accuracy": 0.23645564466714858,
      "num_tokens": 138370445.0,
      "step": 66675
    },
    {
      "entropy": 5.748660755157471,
      "epoch": 6.048621190130624,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002046039578648515,
      "loss": 4.8064,
      "mean_token_accuracy": 0.23274726569652557,
      "num_tokens": 138381456.0,
      "step": 66680
    },
    {
      "entropy": 5.746720933914185,
      "epoch": 6.049074746008708,
      "grad_norm": 1.5625,
      "learning_rate": 0.0002045732266099326,
      "loss": 4.8261,
      "mean_token_accuracy": 0.2353407621383667,
      "num_tokens": 138391403.0,
      "step": 66685
    },
    {
      "entropy": 5.758558511734009,
      "epoch": 6.0495283018867925,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002045424968114573,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23276012092828752,
      "num_tokens": 138402274.0,
      "step": 66690
    },
    {
      "entropy": 5.744263505935669,
      "epoch": 6.049981857764877,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020451176847006115,
      "loss": 4.8518,
      "mean_token_accuracy": 0.2367138907313347,
      "num_tokens": 138412417.0,
      "step": 66695
    },
    {
      "entropy": 5.673923110961914,
      "epoch": 6.050435413642961,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020448104158637953,
      "loss": 4.733,
      "mean_token_accuracy": 0.24298802018165588,
      "num_tokens": 138422996.0,
      "step": 66700
    },
    {
      "entropy": 5.75216007232666,
      "epoch": 6.050888969521045,
      "grad_norm": 1.5,
      "learning_rate": 0.00020445031616104798,
      "loss": 4.8467,
      "mean_token_accuracy": 0.23854378759860992,
      "num_tokens": 138432775.0,
      "step": 66705
    },
    {
      "entropy": 5.768328142166138,
      "epoch": 6.0513425253991295,
      "grad_norm": 1.375,
      "learning_rate": 0.00020441959219470184,
      "loss": 4.8334,
      "mean_token_accuracy": 0.2305882453918457,
      "num_tokens": 138443006.0,
      "step": 66710
    },
    {
      "entropy": 5.784128284454345,
      "epoch": 6.051796081277213,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020438886968797654,
      "loss": 4.9236,
      "mean_token_accuracy": 0.22191910147666932,
      "num_tokens": 138453892.0,
      "step": 66715
    },
    {
      "entropy": 5.707892036437988,
      "epoch": 6.052249637155297,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002043581486415075,
      "loss": 4.744,
      "mean_token_accuracy": 0.2464045524597168,
      "num_tokens": 138463355.0,
      "step": 66720
    },
    {
      "entropy": 5.740875196456909,
      "epoch": 6.0527031930333814,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020432742905592983,
      "loss": 4.809,
      "mean_token_accuracy": 0.22754669040441514,
      "num_tokens": 138473428.0,
      "step": 66725
    },
    {
      "entropy": 5.6622014999389645,
      "epoch": 6.053156748911466,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002042967109318789,
      "loss": 4.73,
      "mean_token_accuracy": 0.2350200355052948,
      "num_tokens": 138485083.0,
      "step": 66730
    },
    {
      "entropy": 5.633229970932007,
      "epoch": 6.05361030478955,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020426599426999015,
      "loss": 4.6999,
      "mean_token_accuracy": 0.2529819652438164,
      "num_tokens": 138495645.0,
      "step": 66735
    },
    {
      "entropy": 5.708830261230469,
      "epoch": 6.054063860667634,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020423527907089856,
      "loss": 4.8376,
      "mean_token_accuracy": 0.23331381976604462,
      "num_tokens": 138505638.0,
      "step": 66740
    },
    {
      "entropy": 5.766967630386352,
      "epoch": 6.0545174165457185,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020420456533523952,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2394075334072113,
      "num_tokens": 138516072.0,
      "step": 66745
    },
    {
      "entropy": 5.743108081817627,
      "epoch": 6.054970972423803,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000204173853063648,
      "loss": 4.7681,
      "mean_token_accuracy": 0.2404007688164711,
      "num_tokens": 138525529.0,
      "step": 66750
    },
    {
      "entropy": 5.749437665939331,
      "epoch": 6.055424528301887,
      "grad_norm": 1.609375,
      "learning_rate": 0.0002041431422567593,
      "loss": 4.7825,
      "mean_token_accuracy": 0.23815346658229827,
      "num_tokens": 138535459.0,
      "step": 66755
    },
    {
      "entropy": 5.727043867111206,
      "epoch": 6.055878084179971,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020411243291520848,
      "loss": 4.8691,
      "mean_token_accuracy": 0.22715040892362595,
      "num_tokens": 138546402.0,
      "step": 66760
    },
    {
      "entropy": 5.61678376197815,
      "epoch": 6.056331640058056,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020408172503963057,
      "loss": 4.7409,
      "mean_token_accuracy": 0.2361455500125885,
      "num_tokens": 138556510.0,
      "step": 66765
    },
    {
      "entropy": 5.698020982742309,
      "epoch": 6.056785195936139,
      "grad_norm": 1.5,
      "learning_rate": 0.00020405101863066072,
      "loss": 4.7425,
      "mean_token_accuracy": 0.23717524111270905,
      "num_tokens": 138567029.0,
      "step": 66770
    },
    {
      "entropy": 5.7266374111175535,
      "epoch": 6.057238751814223,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020402031368893387,
      "loss": 4.7949,
      "mean_token_accuracy": 0.23669066578149794,
      "num_tokens": 138577453.0,
      "step": 66775
    },
    {
      "entropy": 5.708195209503174,
      "epoch": 6.0576923076923075,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000203989610215085,
      "loss": 4.8064,
      "mean_token_accuracy": 0.24021437913179397,
      "num_tokens": 138587607.0,
      "step": 66780
    },
    {
      "entropy": 5.7560975551605225,
      "epoch": 6.058145863570392,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020395890820974904,
      "loss": 4.8228,
      "mean_token_accuracy": 0.22788803577423095,
      "num_tokens": 138597846.0,
      "step": 66785
    },
    {
      "entropy": 5.693230628967285,
      "epoch": 6.058599419448476,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020392820767356108,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23411106169223786,
      "num_tokens": 138608374.0,
      "step": 66790
    },
    {
      "entropy": 5.6886882305145265,
      "epoch": 6.05905297532656,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020389750860715582,
      "loss": 4.7866,
      "mean_token_accuracy": 0.23295072615146636,
      "num_tokens": 138617857.0,
      "step": 66795
    },
    {
      "entropy": 5.721652364730835,
      "epoch": 6.0595065312046446,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020386681101116816,
      "loss": 4.7905,
      "mean_token_accuracy": 0.23355917036533355,
      "num_tokens": 138627900.0,
      "step": 66800
    },
    {
      "entropy": 5.734973001480102,
      "epoch": 6.059960087082729,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020383611488623306,
      "loss": 4.7764,
      "mean_token_accuracy": 0.24295949041843415,
      "num_tokens": 138639520.0,
      "step": 66805
    },
    {
      "entropy": 5.733226156234741,
      "epoch": 6.060413642960813,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020380542023298526,
      "loss": 4.7284,
      "mean_token_accuracy": 0.24457197189331054,
      "num_tokens": 138648914.0,
      "step": 66810
    },
    {
      "entropy": 5.726370859146118,
      "epoch": 6.060867198838897,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020377472705205942,
      "loss": 4.7783,
      "mean_token_accuracy": 0.24528200775384904,
      "num_tokens": 138659793.0,
      "step": 66815
    },
    {
      "entropy": 5.691235733032227,
      "epoch": 6.061320754716981,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020374403534409037,
      "loss": 4.7675,
      "mean_token_accuracy": 0.2337031528353691,
      "num_tokens": 138670108.0,
      "step": 66820
    },
    {
      "entropy": 5.728797149658203,
      "epoch": 6.061774310595065,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002037133451097129,
      "loss": 4.8923,
      "mean_token_accuracy": 0.22651124447584153,
      "num_tokens": 138681112.0,
      "step": 66825
    },
    {
      "entropy": 5.702084732055664,
      "epoch": 6.062227866473149,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00020368265634956162,
      "loss": 4.8005,
      "mean_token_accuracy": 0.2348816603422165,
      "num_tokens": 138691075.0,
      "step": 66830
    },
    {
      "entropy": 5.772994804382324,
      "epoch": 6.0626814223512335,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020365196906427108,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2290772408246994,
      "num_tokens": 138700746.0,
      "step": 66835
    },
    {
      "entropy": 5.707300472259521,
      "epoch": 6.063134978229318,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0002036212832544761,
      "loss": 4.8029,
      "mean_token_accuracy": 0.24085324108600617,
      "num_tokens": 138711319.0,
      "step": 66840
    },
    {
      "entropy": 5.716060781478882,
      "epoch": 6.063588534107402,
      "grad_norm": 1.5,
      "learning_rate": 0.00020359059892081116,
      "loss": 4.7829,
      "mean_token_accuracy": 0.2299639731645584,
      "num_tokens": 138721298.0,
      "step": 66845
    },
    {
      "entropy": 5.786588144302368,
      "epoch": 6.064042089985486,
      "grad_norm": 1.640625,
      "learning_rate": 0.00020355991606391078,
      "loss": 4.846,
      "mean_token_accuracy": 0.23378260880708696,
      "num_tokens": 138731173.0,
      "step": 66850
    },
    {
      "entropy": 5.705037212371826,
      "epoch": 6.064495645863571,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020352923468440958,
      "loss": 4.7641,
      "mean_token_accuracy": 0.2394682914018631,
      "num_tokens": 138742490.0,
      "step": 66855
    },
    {
      "entropy": 5.7737609386444095,
      "epoch": 6.064949201741655,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020349855478294203,
      "loss": 4.9424,
      "mean_token_accuracy": 0.22394029796123505,
      "num_tokens": 138753885.0,
      "step": 66860
    },
    {
      "entropy": 5.716052532196045,
      "epoch": 6.065402757619739,
      "grad_norm": 1.578125,
      "learning_rate": 0.00020346787636014264,
      "loss": 4.7976,
      "mean_token_accuracy": 0.23837359845638276,
      "num_tokens": 138764330.0,
      "step": 66865
    },
    {
      "entropy": 5.734822416305542,
      "epoch": 6.0658563134978225,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020343719941664568,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23379025161266326,
      "num_tokens": 138774808.0,
      "step": 66870
    },
    {
      "entropy": 5.74218339920044,
      "epoch": 6.066309869375907,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020340652395308572,
      "loss": 4.8393,
      "mean_token_accuracy": 0.23576500415802001,
      "num_tokens": 138784977.0,
      "step": 66875
    },
    {
      "entropy": 5.809468507766724,
      "epoch": 6.066763425253991,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00020337584997009712,
      "loss": 4.8809,
      "mean_token_accuracy": 0.23207914531230928,
      "num_tokens": 138795730.0,
      "step": 66880
    },
    {
      "entropy": 5.716472530364991,
      "epoch": 6.067216981132075,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0002033451774683141,
      "loss": 4.7517,
      "mean_token_accuracy": 0.2379352405667305,
      "num_tokens": 138806248.0,
      "step": 66885
    },
    {
      "entropy": 5.790710878372193,
      "epoch": 6.06767053701016,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020331450644837113,
      "loss": 4.9128,
      "mean_token_accuracy": 0.22782996743917466,
      "num_tokens": 138816305.0,
      "step": 66890
    },
    {
      "entropy": 5.766047430038452,
      "epoch": 6.068124092888244,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020328383691090246,
      "loss": 4.915,
      "mean_token_accuracy": 0.2269142374396324,
      "num_tokens": 138826679.0,
      "step": 66895
    },
    {
      "entropy": 5.697878026962281,
      "epoch": 6.068577648766328,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020325316885654238,
      "loss": 4.753,
      "mean_token_accuracy": 0.23801006525754928,
      "num_tokens": 138836848.0,
      "step": 66900
    },
    {
      "entropy": 5.800212287902832,
      "epoch": 6.069031204644412,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020322250228592493,
      "loss": 4.8868,
      "mean_token_accuracy": 0.2292877182364464,
      "num_tokens": 138847898.0,
      "step": 66905
    },
    {
      "entropy": 5.752777433395385,
      "epoch": 6.069484760522497,
      "grad_norm": 1.34375,
      "learning_rate": 0.00020319183719968442,
      "loss": 4.8556,
      "mean_token_accuracy": 0.22281811088323594,
      "num_tokens": 138857893.0,
      "step": 66910
    },
    {
      "entropy": 5.771274757385254,
      "epoch": 6.069938316400581,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002031611735984551,
      "loss": 4.8106,
      "mean_token_accuracy": 0.2394083023071289,
      "num_tokens": 138868289.0,
      "step": 66915
    },
    {
      "entropy": 5.800396823883057,
      "epoch": 6.070391872278665,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020313051148287094,
      "loss": 4.8483,
      "mean_token_accuracy": 0.22489893138408662,
      "num_tokens": 138878018.0,
      "step": 66920
    },
    {
      "entropy": 5.704454565048218,
      "epoch": 6.0708454281567485,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020309985085356614,
      "loss": 4.7647,
      "mean_token_accuracy": 0.23185775130987168,
      "num_tokens": 138888390.0,
      "step": 66925
    },
    {
      "entropy": 5.706643676757812,
      "epoch": 6.071298984034833,
      "grad_norm": 1.359375,
      "learning_rate": 0.00020306919171117477,
      "loss": 4.812,
      "mean_token_accuracy": 0.23415087759494782,
      "num_tokens": 138898512.0,
      "step": 66930
    },
    {
      "entropy": 5.755786800384522,
      "epoch": 6.071752539912917,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020303853405633082,
      "loss": 4.8192,
      "mean_token_accuracy": 0.23638136833906173,
      "num_tokens": 138908711.0,
      "step": 66935
    },
    {
      "entropy": 5.786645889282227,
      "epoch": 6.072206095791001,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020300787788966835,
      "loss": 4.8692,
      "mean_token_accuracy": 0.23021041303873063,
      "num_tokens": 138919094.0,
      "step": 66940
    },
    {
      "entropy": 5.794741296768189,
      "epoch": 6.072659651669086,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020297722321182133,
      "loss": 4.8203,
      "mean_token_accuracy": 0.23356414586305618,
      "num_tokens": 138929064.0,
      "step": 66945
    },
    {
      "entropy": 5.795402479171753,
      "epoch": 6.07311320754717,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020294657002342364,
      "loss": 4.8481,
      "mean_token_accuracy": 0.23233685046434402,
      "num_tokens": 138939705.0,
      "step": 66950
    },
    {
      "entropy": 5.677492475509643,
      "epoch": 6.073566763425254,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00020291591832510925,
      "loss": 4.7302,
      "mean_token_accuracy": 0.2433949425816536,
      "num_tokens": 138949642.0,
      "step": 66955
    },
    {
      "entropy": 5.77375020980835,
      "epoch": 6.074020319303338,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000202885268117512,
      "loss": 4.8524,
      "mean_token_accuracy": 0.2304366335272789,
      "num_tokens": 138958670.0,
      "step": 66960
    },
    {
      "entropy": 5.749757099151611,
      "epoch": 6.074473875181423,
      "grad_norm": 1.5,
      "learning_rate": 0.00020285461940126588,
      "loss": 4.8674,
      "mean_token_accuracy": 0.227997887134552,
      "num_tokens": 138968731.0,
      "step": 66965
    },
    {
      "entropy": 5.799785709381103,
      "epoch": 6.074927431059507,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020282397217700454,
      "loss": 4.8895,
      "mean_token_accuracy": 0.22289057523012162,
      "num_tokens": 138979118.0,
      "step": 66970
    },
    {
      "entropy": 5.777488470077515,
      "epoch": 6.07538098693759,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020279332644536187,
      "loss": 4.8526,
      "mean_token_accuracy": 0.2400301143527031,
      "num_tokens": 138990369.0,
      "step": 66975
    },
    {
      "entropy": 5.816806268692017,
      "epoch": 6.075834542815675,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020276268220697163,
      "loss": 4.8663,
      "mean_token_accuracy": 0.23078053593635559,
      "num_tokens": 139001041.0,
      "step": 66980
    },
    {
      "entropy": 5.771488046646118,
      "epoch": 6.076288098693759,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020273203946246754,
      "loss": 4.8469,
      "mean_token_accuracy": 0.23564101457595826,
      "num_tokens": 139010921.0,
      "step": 66985
    },
    {
      "entropy": 5.761793899536133,
      "epoch": 6.076741654571843,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020270139821248323,
      "loss": 4.8684,
      "mean_token_accuracy": 0.22808921784162522,
      "num_tokens": 139021423.0,
      "step": 66990
    },
    {
      "entropy": 5.8042237758636475,
      "epoch": 6.077195210449927,
      "grad_norm": 1.5,
      "learning_rate": 0.0002026707584576525,
      "loss": 4.9154,
      "mean_token_accuracy": 0.22930804044008254,
      "num_tokens": 139032218.0,
      "step": 66995
    },
    {
      "entropy": 5.758068990707398,
      "epoch": 6.077648766328012,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002026401201986089,
      "loss": 4.824,
      "mean_token_accuracy": 0.23194829523563384,
      "num_tokens": 139041910.0,
      "step": 67000
    },
    {
      "entropy": 5.813280200958252,
      "epoch": 6.078102322206096,
      "grad_norm": 1.53125,
      "learning_rate": 0.00020260948343598596,
      "loss": 4.8293,
      "mean_token_accuracy": 0.2333728238940239,
      "num_tokens": 139051939.0,
      "step": 67005
    },
    {
      "entropy": 5.730562829971314,
      "epoch": 6.07855587808418,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020257884817041744,
      "loss": 4.8523,
      "mean_token_accuracy": 0.22822621017694472,
      "num_tokens": 139063219.0,
      "step": 67010
    },
    {
      "entropy": 5.763213777542115,
      "epoch": 6.0790094339622645,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020254821440253678,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23525612950325012,
      "num_tokens": 139074396.0,
      "step": 67015
    },
    {
      "entropy": 5.788669586181641,
      "epoch": 6.079462989840349,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002025175821329775,
      "loss": 4.894,
      "mean_token_accuracy": 0.23345830738544465,
      "num_tokens": 139084697.0,
      "step": 67020
    },
    {
      "entropy": 5.7052772521972654,
      "epoch": 6.079916545718432,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020248695136237312,
      "loss": 4.8044,
      "mean_token_accuracy": 0.24116694033145905,
      "num_tokens": 139094957.0,
      "step": 67025
    },
    {
      "entropy": 5.783423900604248,
      "epoch": 6.080370101596516,
      "grad_norm": 1.328125,
      "learning_rate": 0.00020245632209135706,
      "loss": 4.9383,
      "mean_token_accuracy": 0.2234399676322937,
      "num_tokens": 139105214.0,
      "step": 67030
    },
    {
      "entropy": 5.760763835906983,
      "epoch": 6.080823657474601,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020242569432056274,
      "loss": 4.8135,
      "mean_token_accuracy": 0.24258376508951188,
      "num_tokens": 139114574.0,
      "step": 67035
    },
    {
      "entropy": 5.755633449554443,
      "epoch": 6.081277213352685,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00020239506805062353,
      "loss": 4.756,
      "mean_token_accuracy": 0.23830052465200424,
      "num_tokens": 139124823.0,
      "step": 67040
    },
    {
      "entropy": 5.7073956489562985,
      "epoch": 6.081730769230769,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00020236444328217278,
      "loss": 4.7249,
      "mean_token_accuracy": 0.24276807755231858,
      "num_tokens": 139134870.0,
      "step": 67045
    },
    {
      "entropy": 5.689837169647217,
      "epoch": 6.082184325108853,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020233382001584393,
      "loss": 4.8283,
      "mean_token_accuracy": 0.23913905173540115,
      "num_tokens": 139145195.0,
      "step": 67050
    },
    {
      "entropy": 5.724210453033447,
      "epoch": 6.082637880986938,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020230319825227018,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23106130361557006,
      "num_tokens": 139156259.0,
      "step": 67055
    },
    {
      "entropy": 5.776947832107544,
      "epoch": 6.083091436865022,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002022725779920848,
      "loss": 4.875,
      "mean_token_accuracy": 0.22883941531181334,
      "num_tokens": 139166515.0,
      "step": 67060
    },
    {
      "entropy": 5.736971712112426,
      "epoch": 6.083544992743106,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020224195923592109,
      "loss": 4.8455,
      "mean_token_accuracy": 0.23025370836257936,
      "num_tokens": 139176927.0,
      "step": 67065
    },
    {
      "entropy": 5.783417463302612,
      "epoch": 6.0839985486211905,
      "grad_norm": 1.5625,
      "learning_rate": 0.00020221134198441216,
      "loss": 4.8648,
      "mean_token_accuracy": 0.22916150391101836,
      "num_tokens": 139187196.0,
      "step": 67070
    },
    {
      "entropy": 5.7601227283477785,
      "epoch": 6.084452104499274,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020218072623819116,
      "loss": 4.9427,
      "mean_token_accuracy": 0.22767623364925385,
      "num_tokens": 139198503.0,
      "step": 67075
    },
    {
      "entropy": 5.745536136627197,
      "epoch": 6.084905660377358,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0002021501119978914,
      "loss": 4.825,
      "mean_token_accuracy": 0.2345777213573456,
      "num_tokens": 139208449.0,
      "step": 67080
    },
    {
      "entropy": 5.864429616928101,
      "epoch": 6.085359216255442,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00020211949926414585,
      "loss": 4.9137,
      "mean_token_accuracy": 0.22565016001462937,
      "num_tokens": 139217725.0,
      "step": 67085
    },
    {
      "entropy": 5.794820737838745,
      "epoch": 6.085812772133527,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020208888803758762,
      "loss": 4.9009,
      "mean_token_accuracy": 0.23218864500522612,
      "num_tokens": 139227864.0,
      "step": 67090
    },
    {
      "entropy": 5.7951765060424805,
      "epoch": 6.086266328011611,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00020205827831884987,
      "loss": 4.8621,
      "mean_token_accuracy": 0.23516320586204528,
      "num_tokens": 139238679.0,
      "step": 67095
    },
    {
      "entropy": 5.71426739692688,
      "epoch": 6.086719883889695,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020202767010856542,
      "loss": 4.7368,
      "mean_token_accuracy": 0.2451995000243187,
      "num_tokens": 139248945.0,
      "step": 67100
    },
    {
      "entropy": 5.735973072052002,
      "epoch": 6.0871734397677795,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00020199706340736734,
      "loss": 4.8276,
      "mean_token_accuracy": 0.23586121797561646,
      "num_tokens": 139258180.0,
      "step": 67105
    },
    {
      "entropy": 5.758252096176148,
      "epoch": 6.087626995645864,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002019664582158886,
      "loss": 4.8487,
      "mean_token_accuracy": 0.23689637780189515,
      "num_tokens": 139267955.0,
      "step": 67110
    },
    {
      "entropy": 5.765208101272583,
      "epoch": 6.088080551523948,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00020193585453476216,
      "loss": 4.8257,
      "mean_token_accuracy": 0.22440292984247207,
      "num_tokens": 139277787.0,
      "step": 67115
    },
    {
      "entropy": 5.726996278762817,
      "epoch": 6.088534107402032,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020190525236462076,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23837389498949052,
      "num_tokens": 139288205.0,
      "step": 67120
    },
    {
      "entropy": 5.740935182571411,
      "epoch": 6.0889876632801165,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00020187465170609748,
      "loss": 4.8088,
      "mean_token_accuracy": 0.24053953737020492,
      "num_tokens": 139298124.0,
      "step": 67125
    },
    {
      "entropy": 5.866272068023681,
      "epoch": 6.0894412191582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00020184405255982497,
      "loss": 5.021,
      "mean_token_accuracy": 0.21459888964891433,
      "num_tokens": 139308692.0,
      "step": 67130
    },
    {
      "entropy": 5.771627712249756,
      "epoch": 6.089894775036284,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002018134549264361,
      "loss": 4.797,
      "mean_token_accuracy": 0.23579745292663573,
      "num_tokens": 139317932.0,
      "step": 67135
    },
    {
      "entropy": 5.780027389526367,
      "epoch": 6.0903483309143684,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002017828588065636,
      "loss": 4.8305,
      "mean_token_accuracy": 0.23166632950305938,
      "num_tokens": 139327419.0,
      "step": 67140
    },
    {
      "entropy": 5.720736885070801,
      "epoch": 6.090801886792453,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002017522642008403,
      "loss": 4.8161,
      "mean_token_accuracy": 0.23994386941194534,
      "num_tokens": 139338258.0,
      "step": 67145
    },
    {
      "entropy": 5.801737451553345,
      "epoch": 6.091255442670537,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002017216711098988,
      "loss": 4.8337,
      "mean_token_accuracy": 0.2353874623775482,
      "num_tokens": 139347591.0,
      "step": 67150
    },
    {
      "entropy": 5.697898721694946,
      "epoch": 6.091708998548621,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002016910795343719,
      "loss": 4.8063,
      "mean_token_accuracy": 0.23287779986858367,
      "num_tokens": 139357639.0,
      "step": 67155
    },
    {
      "entropy": 5.724838161468506,
      "epoch": 6.0921625544267055,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020166048947489206,
      "loss": 4.845,
      "mean_token_accuracy": 0.2324516847729683,
      "num_tokens": 139367988.0,
      "step": 67160
    },
    {
      "entropy": 5.72404522895813,
      "epoch": 6.09261611030479,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020162990093209205,
      "loss": 4.7816,
      "mean_token_accuracy": 0.23364427536725998,
      "num_tokens": 139377726.0,
      "step": 67165
    },
    {
      "entropy": 5.76822714805603,
      "epoch": 6.093069666182874,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002015993139066043,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23658154606819154,
      "num_tokens": 139387361.0,
      "step": 67170
    },
    {
      "entropy": 5.755673837661743,
      "epoch": 6.093523222060958,
      "grad_norm": 1.46875,
      "learning_rate": 0.00020156872839906153,
      "loss": 4.8745,
      "mean_token_accuracy": 0.2330319330096245,
      "num_tokens": 139397279.0,
      "step": 67175
    },
    {
      "entropy": 5.719816970825195,
      "epoch": 6.093976777939042,
      "grad_norm": 1.5,
      "learning_rate": 0.00020153814441009615,
      "loss": 4.7677,
      "mean_token_accuracy": 0.2328743815422058,
      "num_tokens": 139406715.0,
      "step": 67180
    },
    {
      "entropy": 5.805272626876831,
      "epoch": 6.094430333817126,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020150756194034063,
      "loss": 4.9442,
      "mean_token_accuracy": 0.22801343351602554,
      "num_tokens": 139417485.0,
      "step": 67185
    },
    {
      "entropy": 5.765566205978393,
      "epoch": 6.09488388969521,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0002014769809904275,
      "loss": 4.8299,
      "mean_token_accuracy": 0.23295444250106812,
      "num_tokens": 139428174.0,
      "step": 67190
    },
    {
      "entropy": 5.752988386154175,
      "epoch": 6.0953374455732945,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020144640156098914,
      "loss": 4.8146,
      "mean_token_accuracy": 0.23830806016921996,
      "num_tokens": 139438349.0,
      "step": 67195
    },
    {
      "entropy": 5.7974015235900875,
      "epoch": 6.095791001451379,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002014158236526579,
      "loss": 4.8678,
      "mean_token_accuracy": 0.22548991441726685,
      "num_tokens": 139448193.0,
      "step": 67200
    },
    {
      "entropy": 5.794012403488159,
      "epoch": 6.096244557329463,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0002013852472660662,
      "loss": 4.9006,
      "mean_token_accuracy": 0.228709614276886,
      "num_tokens": 139458418.0,
      "step": 67205
    },
    {
      "entropy": 5.730435562133789,
      "epoch": 6.096698113207547,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020135467240184636,
      "loss": 4.8761,
      "mean_token_accuracy": 0.23039024472236633,
      "num_tokens": 139468085.0,
      "step": 67210
    },
    {
      "entropy": 5.770635938644409,
      "epoch": 6.0971516690856316,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00020132409906063064,
      "loss": 4.9026,
      "mean_token_accuracy": 0.22442807108163834,
      "num_tokens": 139479250.0,
      "step": 67215
    },
    {
      "entropy": 5.761206007003784,
      "epoch": 6.097605224963716,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020129352724305134,
      "loss": 4.8574,
      "mean_token_accuracy": 0.23678766340017318,
      "num_tokens": 139488602.0,
      "step": 67220
    },
    {
      "entropy": 5.752795219421387,
      "epoch": 6.0980587808418,
      "grad_norm": 1.390625,
      "learning_rate": 0.00020126295694974068,
      "loss": 4.8709,
      "mean_token_accuracy": 0.22947044670581818,
      "num_tokens": 139499629.0,
      "step": 67225
    },
    {
      "entropy": 5.7343518257141115,
      "epoch": 6.0985123367198835,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020123238818133083,
      "loss": 4.8162,
      "mean_token_accuracy": 0.23312161415815352,
      "num_tokens": 139510567.0,
      "step": 67230
    },
    {
      "entropy": 5.746621608734131,
      "epoch": 6.098965892597968,
      "grad_norm": 1.546875,
      "learning_rate": 0.00020120182093845402,
      "loss": 4.7757,
      "mean_token_accuracy": 0.2384440705180168,
      "num_tokens": 139519893.0,
      "step": 67235
    },
    {
      "entropy": 5.819274187088013,
      "epoch": 6.099419448476052,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020117125522174236,
      "loss": 4.8749,
      "mean_token_accuracy": 0.22976595610380174,
      "num_tokens": 139530653.0,
      "step": 67240
    },
    {
      "entropy": 5.748059940338135,
      "epoch": 6.099873004354136,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002011406910318279,
      "loss": 4.7885,
      "mean_token_accuracy": 0.2369524821639061,
      "num_tokens": 139541032.0,
      "step": 67245
    },
    {
      "entropy": 5.7329816818237305,
      "epoch": 6.1003265602322205,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00020111012836934283,
      "loss": 4.8417,
      "mean_token_accuracy": 0.23845624774694443,
      "num_tokens": 139551838.0,
      "step": 67250
    },
    {
      "entropy": 5.736738586425782,
      "epoch": 6.100780116110305,
      "grad_norm": 1.484375,
      "learning_rate": 0.00020107956723491917,
      "loss": 4.7604,
      "mean_token_accuracy": 0.23458636850118636,
      "num_tokens": 139561672.0,
      "step": 67255
    },
    {
      "entropy": 5.698092842102051,
      "epoch": 6.101233671988389,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00020104900762918883,
      "loss": 4.7699,
      "mean_token_accuracy": 0.23945848196744918,
      "num_tokens": 139572710.0,
      "step": 67260
    },
    {
      "entropy": 5.641730690002442,
      "epoch": 6.101687227866473,
      "grad_norm": 1.421875,
      "learning_rate": 0.00020101844955278393,
      "loss": 4.6919,
      "mean_token_accuracy": 0.24010991752147676,
      "num_tokens": 139583020.0,
      "step": 67265
    },
    {
      "entropy": 5.8032842636108395,
      "epoch": 6.102140783744558,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00020098789300633624,
      "loss": 4.8733,
      "mean_token_accuracy": 0.23322055339813233,
      "num_tokens": 139592944.0,
      "step": 67270
    },
    {
      "entropy": 5.755306005477905,
      "epoch": 6.102594339622642,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020095733799047782,
      "loss": 4.8424,
      "mean_token_accuracy": 0.23514638096094131,
      "num_tokens": 139604415.0,
      "step": 67275
    },
    {
      "entropy": 5.799600982666016,
      "epoch": 6.103047895500726,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00020092678450584056,
      "loss": 4.8548,
      "mean_token_accuracy": 0.2349381133913994,
      "num_tokens": 139615456.0,
      "step": 67280
    },
    {
      "entropy": 5.756643581390381,
      "epoch": 6.1035014513788095,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002008962325530563,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2282392591238022,
      "num_tokens": 139625513.0,
      "step": 67285
    },
    {
      "entropy": 5.723760795593262,
      "epoch": 6.103955007256894,
      "grad_norm": 1.4375,
      "learning_rate": 0.0002008656821327567,
      "loss": 4.8508,
      "mean_token_accuracy": 0.22653703838586808,
      "num_tokens": 139635524.0,
      "step": 67290
    },
    {
      "entropy": 5.699713706970215,
      "epoch": 6.104408563134978,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0002008351332455738,
      "loss": 4.6919,
      "mean_token_accuracy": 0.2435808226466179,
      "num_tokens": 139645958.0,
      "step": 67295
    },
    {
      "entropy": 5.785110664367676,
      "epoch": 6.104862119013062,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020080458589213918,
      "loss": 4.8768,
      "mean_token_accuracy": 0.22777555286884307,
      "num_tokens": 139655022.0,
      "step": 67300
    },
    {
      "entropy": 5.826350355148316,
      "epoch": 6.105315674891147,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00020077404007308463,
      "loss": 4.9465,
      "mean_token_accuracy": 0.22507910281419755,
      "num_tokens": 139665913.0,
      "step": 67305
    },
    {
      "entropy": 5.79396915435791,
      "epoch": 6.105769230769231,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020074349578904195,
      "loss": 4.8557,
      "mean_token_accuracy": 0.23574003279209138,
      "num_tokens": 139675260.0,
      "step": 67310
    },
    {
      "entropy": 5.787536334991455,
      "epoch": 6.106222786647315,
      "grad_norm": 1.3125,
      "learning_rate": 0.00020071295304064258,
      "loss": 4.8174,
      "mean_token_accuracy": 0.24131570458412172,
      "num_tokens": 139685337.0,
      "step": 67315
    },
    {
      "entropy": 5.6750976085662845,
      "epoch": 6.106676342525399,
      "grad_norm": 1.40625,
      "learning_rate": 0.00020068241182851838,
      "loss": 4.7317,
      "mean_token_accuracy": 0.24385272413492204,
      "num_tokens": 139694977.0,
      "step": 67320
    },
    {
      "entropy": 5.778867292404175,
      "epoch": 6.107129898403484,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002006518721533007,
      "loss": 4.8892,
      "mean_token_accuracy": 0.22592318654060364,
      "num_tokens": 139705708.0,
      "step": 67325
    },
    {
      "entropy": 5.809373474121093,
      "epoch": 6.107583454281568,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020062133401562137,
      "loss": 4.8875,
      "mean_token_accuracy": 0.22530759274959564,
      "num_tokens": 139717268.0,
      "step": 67330
    },
    {
      "entropy": 5.719121026992798,
      "epoch": 6.108037010159651,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002005907974161118,
      "loss": 4.7673,
      "mean_token_accuracy": 0.2399125352501869,
      "num_tokens": 139728576.0,
      "step": 67335
    },
    {
      "entropy": 5.763119173049927,
      "epoch": 6.1084905660377355,
      "grad_norm": 1.296875,
      "learning_rate": 0.00020056026235540342,
      "loss": 4.9018,
      "mean_token_accuracy": 0.23073332458734513,
      "num_tokens": 139741136.0,
      "step": 67340
    },
    {
      "entropy": 5.680326128005982,
      "epoch": 6.10894412191582,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00020052972883412783,
      "loss": 4.7464,
      "mean_token_accuracy": 0.240944504737854,
      "num_tokens": 139751244.0,
      "step": 67345
    },
    {
      "entropy": 5.775876855850219,
      "epoch": 6.109397677793904,
      "grad_norm": 1.4375,
      "learning_rate": 0.00020049919685291646,
      "loss": 4.8963,
      "mean_token_accuracy": 0.2296706810593605,
      "num_tokens": 139760584.0,
      "step": 67350
    },
    {
      "entropy": 5.774984407424927,
      "epoch": 6.109851233671988,
      "grad_norm": 1.515625,
      "learning_rate": 0.00020046866641240064,
      "loss": 4.7572,
      "mean_token_accuracy": 0.23989014625549315,
      "num_tokens": 139770101.0,
      "step": 67355
    },
    {
      "entropy": 5.736812686920166,
      "epoch": 6.110304789550073,
      "grad_norm": 1.453125,
      "learning_rate": 0.00020043813751321183,
      "loss": 4.7792,
      "mean_token_accuracy": 0.23584046810865403,
      "num_tokens": 139780423.0,
      "step": 67360
    },
    {
      "entropy": 5.7569855690002445,
      "epoch": 6.110758345428157,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002004076101559813,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22386245876550676,
      "num_tokens": 139790999.0,
      "step": 67365
    },
    {
      "entropy": 5.708025550842285,
      "epoch": 6.111211901306241,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0002003770843413404,
      "loss": 4.7567,
      "mean_token_accuracy": 0.23442192524671554,
      "num_tokens": 139800842.0,
      "step": 67370
    },
    {
      "entropy": 5.679375267028808,
      "epoch": 6.111665457184325,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002003465600699204,
      "loss": 4.7338,
      "mean_token_accuracy": 0.24415872544050216,
      "num_tokens": 139810831.0,
      "step": 67375
    },
    {
      "entropy": 5.711828088760376,
      "epoch": 6.11211901306241,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0002003160373423526,
      "loss": 4.8395,
      "mean_token_accuracy": 0.24271852672100067,
      "num_tokens": 139822127.0,
      "step": 67380
    },
    {
      "entropy": 5.776581907272339,
      "epoch": 6.112572568940493,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002002855161592682,
      "loss": 4.8072,
      "mean_token_accuracy": 0.233468134701252,
      "num_tokens": 139832312.0,
      "step": 67385
    },
    {
      "entropy": 5.7553199291229244,
      "epoch": 6.113026124818577,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00020025499652129826,
      "loss": 4.8625,
      "mean_token_accuracy": 0.22666265219449996,
      "num_tokens": 139842755.0,
      "step": 67390
    },
    {
      "entropy": 5.734534120559692,
      "epoch": 6.113479680696662,
      "grad_norm": 1.703125,
      "learning_rate": 0.00020022447842907415,
      "loss": 4.7918,
      "mean_token_accuracy": 0.24023303091526033,
      "num_tokens": 139854540.0,
      "step": 67395
    },
    {
      "entropy": 5.762231063842774,
      "epoch": 6.113933236574746,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00020019396188322686,
      "loss": 4.835,
      "mean_token_accuracy": 0.2330259010195732,
      "num_tokens": 139864867.0,
      "step": 67400
    },
    {
      "entropy": 5.7422065258026125,
      "epoch": 6.11438679245283,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002001634468843875,
      "loss": 4.8215,
      "mean_token_accuracy": 0.22649752497673034,
      "num_tokens": 139876230.0,
      "step": 67405
    },
    {
      "entropy": 5.7263284683227536,
      "epoch": 6.114840348330914,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002001329334331871,
      "loss": 4.7776,
      "mean_token_accuracy": 0.24243558198213577,
      "num_tokens": 139887931.0,
      "step": 67410
    },
    {
      "entropy": 5.74350380897522,
      "epoch": 6.115293904208999,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002001024215302567,
      "loss": 4.8333,
      "mean_token_accuracy": 0.2308824434876442,
      "num_tokens": 139899003.0,
      "step": 67415
    },
    {
      "entropy": 5.8341292381286625,
      "epoch": 6.115747460087083,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00020007191117622737,
      "loss": 4.9076,
      "mean_token_accuracy": 0.2298718884587288,
      "num_tokens": 139909343.0,
      "step": 67420
    },
    {
      "entropy": 5.783523988723755,
      "epoch": 6.116201015965167,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002000414023717299,
      "loss": 4.8832,
      "mean_token_accuracy": 0.23980641961097718,
      "num_tokens": 139920208.0,
      "step": 67425
    },
    {
      "entropy": 5.690113353729248,
      "epoch": 6.1166545718432515,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00020001089511739547,
      "loss": 4.7429,
      "mean_token_accuracy": 0.2381434202194214,
      "num_tokens": 139931064.0,
      "step": 67430
    },
    {
      "entropy": 5.708019971847534,
      "epoch": 6.117108127721335,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019998038941385478,
      "loss": 4.7382,
      "mean_token_accuracy": 0.23925687968730927,
      "num_tokens": 139940630.0,
      "step": 67435
    },
    {
      "entropy": 5.7557495594024655,
      "epoch": 6.117561683599419,
      "grad_norm": 1.703125,
      "learning_rate": 0.0001999498852617387,
      "loss": 4.8822,
      "mean_token_accuracy": 0.2269471913576126,
      "num_tokens": 139950333.0,
      "step": 67440
    },
    {
      "entropy": 5.786212301254272,
      "epoch": 6.118015239477503,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001999193826616782,
      "loss": 4.9185,
      "mean_token_accuracy": 0.22670327723026276,
      "num_tokens": 139960298.0,
      "step": 67445
    },
    {
      "entropy": 5.816713476181031,
      "epoch": 6.118468795355588,
      "grad_norm": 1.640625,
      "learning_rate": 0.00019988888161430397,
      "loss": 4.8205,
      "mean_token_accuracy": 0.23447911590337753,
      "num_tokens": 139970422.0,
      "step": 67450
    },
    {
      "entropy": 5.68590989112854,
      "epoch": 6.118922351233672,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019985838212024682,
      "loss": 4.7402,
      "mean_token_accuracy": 0.23270778208971024,
      "num_tokens": 139980197.0,
      "step": 67455
    },
    {
      "entropy": 5.664391088485718,
      "epoch": 6.119375907111756,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019982788418013741,
      "loss": 4.7335,
      "mean_token_accuracy": 0.24273266792297363,
      "num_tokens": 139989474.0,
      "step": 67460
    },
    {
      "entropy": 5.8117955207824705,
      "epoch": 6.11982946298984,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001997973877946066,
      "loss": 4.9344,
      "mean_token_accuracy": 0.2334431752562523,
      "num_tokens": 139999256.0,
      "step": 67465
    },
    {
      "entropy": 5.691676998138428,
      "epoch": 6.120283018867925,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001997668929642849,
      "loss": 4.7729,
      "mean_token_accuracy": 0.23327729254961013,
      "num_tokens": 140010256.0,
      "step": 67470
    },
    {
      "entropy": 5.707353019714356,
      "epoch": 6.120736574746009,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019973639968980306,
      "loss": 4.7682,
      "mean_token_accuracy": 0.23994487375020981,
      "num_tokens": 140021175.0,
      "step": 67475
    },
    {
      "entropy": 5.845127725601197,
      "epoch": 6.121190130624093,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019970590797179167,
      "loss": 4.9254,
      "mean_token_accuracy": 0.22706100344657898,
      "num_tokens": 140030943.0,
      "step": 67480
    },
    {
      "entropy": 5.760222339630127,
      "epoch": 6.1216436865021775,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00019967541781088129,
      "loss": 4.8039,
      "mean_token_accuracy": 0.2282392367720604,
      "num_tokens": 140040919.0,
      "step": 67485
    },
    {
      "entropy": 5.798914909362793,
      "epoch": 6.122097242380261,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001996449292077025,
      "loss": 4.8827,
      "mean_token_accuracy": 0.2273378163576126,
      "num_tokens": 140051656.0,
      "step": 67490
    },
    {
      "entropy": 5.756993675231934,
      "epoch": 6.122550798258345,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019961444216288565,
      "loss": 4.8072,
      "mean_token_accuracy": 0.23390882164239885,
      "num_tokens": 140061655.0,
      "step": 67495
    },
    {
      "entropy": 5.694076442718506,
      "epoch": 6.123004354136429,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019958395667706147,
      "loss": 4.791,
      "mean_token_accuracy": 0.2346506580710411,
      "num_tokens": 140072985.0,
      "step": 67500
    },
    {
      "entropy": 5.693667316436768,
      "epoch": 6.123457910014514,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019955347275086023,
      "loss": 4.8113,
      "mean_token_accuracy": 0.2345726191997528,
      "num_tokens": 140083814.0,
      "step": 67505
    },
    {
      "entropy": 5.730710744857788,
      "epoch": 6.123911465892598,
      "grad_norm": 1.5,
      "learning_rate": 0.0001995229903849124,
      "loss": 4.8602,
      "mean_token_accuracy": 0.23268010169267656,
      "num_tokens": 140093664.0,
      "step": 67510
    },
    {
      "entropy": 5.782833480834961,
      "epoch": 6.124365021770682,
      "grad_norm": 1.5,
      "learning_rate": 0.00019949250957984838,
      "loss": 4.8642,
      "mean_token_accuracy": 0.23731240332126619,
      "num_tokens": 140103585.0,
      "step": 67515
    },
    {
      "entropy": 5.706340217590332,
      "epoch": 6.1248185776487665,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019946203033629844,
      "loss": 4.8214,
      "mean_token_accuracy": 0.2287755936384201,
      "num_tokens": 140112802.0,
      "step": 67520
    },
    {
      "entropy": 5.753205680847168,
      "epoch": 6.125272133526851,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001994315526548931,
      "loss": 4.8887,
      "mean_token_accuracy": 0.22667991667985915,
      "num_tokens": 140122980.0,
      "step": 67525
    },
    {
      "entropy": 5.771316719055176,
      "epoch": 6.125725689404935,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019940107653626245,
      "loss": 4.8563,
      "mean_token_accuracy": 0.22868633270263672,
      "num_tokens": 140134336.0,
      "step": 67530
    },
    {
      "entropy": 5.710371637344361,
      "epoch": 6.126179245283019,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019937060198103678,
      "loss": 4.817,
      "mean_token_accuracy": 0.23532509952783584,
      "num_tokens": 140144683.0,
      "step": 67535
    },
    {
      "entropy": 5.775339412689209,
      "epoch": 6.126632801161103,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019934012898984634,
      "loss": 4.8611,
      "mean_token_accuracy": 0.23025646507740022,
      "num_tokens": 140155536.0,
      "step": 67540
    },
    {
      "entropy": 5.819459772109985,
      "epoch": 6.127086357039187,
      "grad_norm": 1.34375,
      "learning_rate": 0.0001993096575633213,
      "loss": 4.8739,
      "mean_token_accuracy": 0.23017666786909102,
      "num_tokens": 140166391.0,
      "step": 67545
    },
    {
      "entropy": 5.794839763641358,
      "epoch": 6.127539912917271,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001992791877020918,
      "loss": 4.8996,
      "mean_token_accuracy": 0.2251337468624115,
      "num_tokens": 140176136.0,
      "step": 67550
    },
    {
      "entropy": 5.773949193954468,
      "epoch": 6.1279934687953554,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000199248719406788,
      "loss": 4.8191,
      "mean_token_accuracy": 0.2336352735757828,
      "num_tokens": 140187174.0,
      "step": 67555
    },
    {
      "entropy": 5.731536245346069,
      "epoch": 6.12844702467344,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019921825267804006,
      "loss": 4.8093,
      "mean_token_accuracy": 0.2376059740781784,
      "num_tokens": 140197512.0,
      "step": 67560
    },
    {
      "entropy": 5.761729907989502,
      "epoch": 6.128900580551524,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019918778751647792,
      "loss": 4.7511,
      "mean_token_accuracy": 0.24053036719560622,
      "num_tokens": 140208008.0,
      "step": 67565
    },
    {
      "entropy": 5.870375728607177,
      "epoch": 6.129354136429608,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019915732392273162,
      "loss": 4.9353,
      "mean_token_accuracy": 0.22769075334072114,
      "num_tokens": 140218278.0,
      "step": 67570
    },
    {
      "entropy": 5.73660454750061,
      "epoch": 6.1298076923076925,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019912686189743124,
      "loss": 4.818,
      "mean_token_accuracy": 0.23958082795143126,
      "num_tokens": 140228055.0,
      "step": 67575
    },
    {
      "entropy": 5.708625459671021,
      "epoch": 6.130261248185777,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019909640144120667,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24533315300941466,
      "num_tokens": 140237339.0,
      "step": 67580
    },
    {
      "entropy": 5.77033429145813,
      "epoch": 6.130714804063861,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001990659425546879,
      "loss": 4.8651,
      "mean_token_accuracy": 0.22688862532377244,
      "num_tokens": 140249220.0,
      "step": 67585
    },
    {
      "entropy": 5.813794183731079,
      "epoch": 6.131168359941944,
      "grad_norm": 1.5,
      "learning_rate": 0.00019903548523850473,
      "loss": 4.9023,
      "mean_token_accuracy": 0.2395605355501175,
      "num_tokens": 140259597.0,
      "step": 67590
    },
    {
      "entropy": 5.683323335647583,
      "epoch": 6.131621915820029,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019900502949328714,
      "loss": 4.75,
      "mean_token_accuracy": 0.2363306075334549,
      "num_tokens": 140269574.0,
      "step": 67595
    },
    {
      "entropy": 5.723536014556885,
      "epoch": 6.132075471698113,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019897457531966483,
      "loss": 4.7399,
      "mean_token_accuracy": 0.24100870937108992,
      "num_tokens": 140279001.0,
      "step": 67600
    },
    {
      "entropy": 5.7283392429351805,
      "epoch": 6.132529027576197,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019894412271826773,
      "loss": 4.7261,
      "mean_token_accuracy": 0.2338469311594963,
      "num_tokens": 140289107.0,
      "step": 67605
    },
    {
      "entropy": 5.768362283706665,
      "epoch": 6.1329825834542815,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001989136716897255,
      "loss": 4.8719,
      "mean_token_accuracy": 0.2235328286886215,
      "num_tokens": 140298796.0,
      "step": 67610
    },
    {
      "entropy": 5.727742481231689,
      "epoch": 6.133436139332366,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000198883222234668,
      "loss": 4.8023,
      "mean_token_accuracy": 0.24208150953054428,
      "num_tokens": 140308533.0,
      "step": 67615
    },
    {
      "entropy": 5.727155542373657,
      "epoch": 6.13388969521045,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019885277435372485,
      "loss": 4.8243,
      "mean_token_accuracy": 0.23523404598236083,
      "num_tokens": 140318454.0,
      "step": 67620
    },
    {
      "entropy": 5.785078525543213,
      "epoch": 6.134343251088534,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019882232804752564,
      "loss": 4.9409,
      "mean_token_accuracy": 0.22731386721134186,
      "num_tokens": 140329636.0,
      "step": 67625
    },
    {
      "entropy": 5.708003187179566,
      "epoch": 6.1347968069666186,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019879188331670022,
      "loss": 4.7615,
      "mean_token_accuracy": 0.23539189994335175,
      "num_tokens": 140339562.0,
      "step": 67630
    },
    {
      "entropy": 5.826677131652832,
      "epoch": 6.135250362844703,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019876144016187802,
      "loss": 4.9063,
      "mean_token_accuracy": 0.22654280811548233,
      "num_tokens": 140351276.0,
      "step": 67635
    },
    {
      "entropy": 5.6828038692474365,
      "epoch": 6.135703918722786,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019873099858368866,
      "loss": 4.7463,
      "mean_token_accuracy": 0.236710424721241,
      "num_tokens": 140361786.0,
      "step": 67640
    },
    {
      "entropy": 5.768073558807373,
      "epoch": 6.1361574746008705,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001987005585827617,
      "loss": 4.8583,
      "mean_token_accuracy": 0.22636977285146714,
      "num_tokens": 140371982.0,
      "step": 67645
    },
    {
      "entropy": 5.8299964427947994,
      "epoch": 6.136611030478955,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019867012015972662,
      "loss": 4.8955,
      "mean_token_accuracy": 0.2267414838075638,
      "num_tokens": 140382845.0,
      "step": 67650
    },
    {
      "entropy": 5.77206449508667,
      "epoch": 6.137064586357039,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019863968331521286,
      "loss": 4.9168,
      "mean_token_accuracy": 0.22169288247823715,
      "num_tokens": 140394161.0,
      "step": 67655
    },
    {
      "entropy": 5.8052863597869875,
      "epoch": 6.137518142235123,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019860924804984998,
      "loss": 4.8239,
      "mean_token_accuracy": 0.23352083414793015,
      "num_tokens": 140404938.0,
      "step": 67660
    },
    {
      "entropy": 5.7569568157196045,
      "epoch": 6.1379716981132075,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019857881436426733,
      "loss": 4.7919,
      "mean_token_accuracy": 0.23997202515602112,
      "num_tokens": 140414569.0,
      "step": 67665
    },
    {
      "entropy": 5.689466190338135,
      "epoch": 6.138425253991292,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019854838225909422,
      "loss": 4.7459,
      "mean_token_accuracy": 0.24115200638771056,
      "num_tokens": 140424501.0,
      "step": 67670
    },
    {
      "entropy": 5.713499927520752,
      "epoch": 6.138878809869376,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019851795173496007,
      "loss": 4.821,
      "mean_token_accuracy": 0.2324115142226219,
      "num_tokens": 140435357.0,
      "step": 67675
    },
    {
      "entropy": 5.712735271453857,
      "epoch": 6.13933236574746,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019848752279249415,
      "loss": 4.7682,
      "mean_token_accuracy": 0.2427240177989006,
      "num_tokens": 140445125.0,
      "step": 67680
    },
    {
      "entropy": 5.841021347045898,
      "epoch": 6.139785921625545,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001984570954323258,
      "loss": 4.9084,
      "mean_token_accuracy": 0.2257360965013504,
      "num_tokens": 140456024.0,
      "step": 67685
    },
    {
      "entropy": 5.742575693130493,
      "epoch": 6.140239477503629,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019842666965508415,
      "loss": 4.8675,
      "mean_token_accuracy": 0.23118859082460402,
      "num_tokens": 140466356.0,
      "step": 67690
    },
    {
      "entropy": 5.721516132354736,
      "epoch": 6.140693033381712,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019839624546139856,
      "loss": 4.8415,
      "mean_token_accuracy": 0.2335650786757469,
      "num_tokens": 140475560.0,
      "step": 67695
    },
    {
      "entropy": 5.688507175445556,
      "epoch": 6.1411465892597965,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019836582285189807,
      "loss": 4.7318,
      "mean_token_accuracy": 0.24285636097192764,
      "num_tokens": 140485521.0,
      "step": 67700
    },
    {
      "entropy": 5.801384401321411,
      "epoch": 6.141600145137881,
      "grad_norm": 1.5625,
      "learning_rate": 0.000198335401827212,
      "loss": 4.9207,
      "mean_token_accuracy": 0.22429685443639755,
      "num_tokens": 140495617.0,
      "step": 67705
    },
    {
      "entropy": 5.761858367919922,
      "epoch": 6.142053701015965,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019830498238796924,
      "loss": 4.8046,
      "mean_token_accuracy": 0.22795564234256743,
      "num_tokens": 140505760.0,
      "step": 67710
    },
    {
      "entropy": 5.816099786758423,
      "epoch": 6.142507256894049,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019827456453479903,
      "loss": 4.878,
      "mean_token_accuracy": 0.2348526746034622,
      "num_tokens": 140517135.0,
      "step": 67715
    },
    {
      "entropy": 5.829051399230957,
      "epoch": 6.142960812772134,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019824414826833037,
      "loss": 4.8722,
      "mean_token_accuracy": 0.22925180345773696,
      "num_tokens": 140527171.0,
      "step": 67720
    },
    {
      "entropy": 5.723122835159302,
      "epoch": 6.143414368650218,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019821373358919225,
      "loss": 4.7894,
      "mean_token_accuracy": 0.22991180866956712,
      "num_tokens": 140537871.0,
      "step": 67725
    },
    {
      "entropy": 5.7286923885345455,
      "epoch": 6.143867924528302,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019818332049801373,
      "loss": 4.856,
      "mean_token_accuracy": 0.2340627670288086,
      "num_tokens": 140547649.0,
      "step": 67730
    },
    {
      "entropy": 5.735935068130493,
      "epoch": 6.144321480406386,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019815290899542377,
      "loss": 4.8242,
      "mean_token_accuracy": 0.2401654064655304,
      "num_tokens": 140557506.0,
      "step": 67735
    },
    {
      "entropy": 5.695458889007568,
      "epoch": 6.144775036284471,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001981224990820511,
      "loss": 4.7755,
      "mean_token_accuracy": 0.23891450762748717,
      "num_tokens": 140567378.0,
      "step": 67740
    },
    {
      "entropy": 5.775875806808472,
      "epoch": 6.145228592162554,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019809209075852485,
      "loss": 4.821,
      "mean_token_accuracy": 0.2361915737390518,
      "num_tokens": 140578342.0,
      "step": 67745
    },
    {
      "entropy": 5.715877819061279,
      "epoch": 6.145682148040638,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019806168402547374,
      "loss": 4.8316,
      "mean_token_accuracy": 0.22721211910247802,
      "num_tokens": 140588280.0,
      "step": 67750
    },
    {
      "entropy": 5.716633558273315,
      "epoch": 6.1461357039187225,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001980312788835266,
      "loss": 4.7716,
      "mean_token_accuracy": 0.23560343831777572,
      "num_tokens": 140597789.0,
      "step": 67755
    },
    {
      "entropy": 5.696446418762207,
      "epoch": 6.146589259796807,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019800087533331222,
      "loss": 4.7581,
      "mean_token_accuracy": 0.23707371801137925,
      "num_tokens": 140607994.0,
      "step": 67760
    },
    {
      "entropy": 5.722074270248413,
      "epoch": 6.147042815674891,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019797047337545936,
      "loss": 4.8099,
      "mean_token_accuracy": 0.23522182255983354,
      "num_tokens": 140617989.0,
      "step": 67765
    },
    {
      "entropy": 5.82210693359375,
      "epoch": 6.147496371552975,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001979400730105968,
      "loss": 4.9914,
      "mean_token_accuracy": 0.21935938000679017,
      "num_tokens": 140629558.0,
      "step": 67770
    },
    {
      "entropy": 5.734259700775146,
      "epoch": 6.14794992743106,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001979096742393531,
      "loss": 4.7706,
      "mean_token_accuracy": 0.24161206781864167,
      "num_tokens": 140639356.0,
      "step": 67775
    },
    {
      "entropy": 5.6585822105407715,
      "epoch": 6.148403483309144,
      "grad_norm": 1.375,
      "learning_rate": 0.00019787927706235698,
      "loss": 4.6819,
      "mean_token_accuracy": 0.247771255671978,
      "num_tokens": 140649821.0,
      "step": 67780
    },
    {
      "entropy": 5.711633920669556,
      "epoch": 6.148857039187228,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019784888148023706,
      "loss": 4.7852,
      "mean_token_accuracy": 0.2368106573820114,
      "num_tokens": 140661341.0,
      "step": 67785
    },
    {
      "entropy": 5.763634395599365,
      "epoch": 6.149310595065312,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019781848749362196,
      "loss": 4.8788,
      "mean_token_accuracy": 0.23218161463737488,
      "num_tokens": 140672011.0,
      "step": 67790
    },
    {
      "entropy": 5.814073133468628,
      "epoch": 6.149764150943396,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019778809510314016,
      "loss": 4.881,
      "mean_token_accuracy": 0.22524976283311843,
      "num_tokens": 140682155.0,
      "step": 67795
    },
    {
      "entropy": 5.722126197814942,
      "epoch": 6.15021770682148,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001977577043094203,
      "loss": 4.8362,
      "mean_token_accuracy": 0.22955355942249298,
      "num_tokens": 140692758.0,
      "step": 67800
    },
    {
      "entropy": 5.8225998878479,
      "epoch": 6.150671262699564,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019772731511309079,
      "loss": 4.8922,
      "mean_token_accuracy": 0.22941372245550157,
      "num_tokens": 140702385.0,
      "step": 67805
    },
    {
      "entropy": 5.823269176483154,
      "epoch": 6.151124818577649,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019769692751478007,
      "loss": 4.9273,
      "mean_token_accuracy": 0.22427944242954254,
      "num_tokens": 140713857.0,
      "step": 67810
    },
    {
      "entropy": 5.776080131530762,
      "epoch": 6.151578374455733,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0001976665415151166,
      "loss": 4.8795,
      "mean_token_accuracy": 0.22644657492637635,
      "num_tokens": 140725406.0,
      "step": 67815
    },
    {
      "entropy": 5.760741901397705,
      "epoch": 6.152031930333817,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001976361571147287,
      "loss": 4.849,
      "mean_token_accuracy": 0.22870175093412398,
      "num_tokens": 140735365.0,
      "step": 67820
    },
    {
      "entropy": 5.741740036010742,
      "epoch": 6.152485486211901,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019760577431424486,
      "loss": 4.8574,
      "mean_token_accuracy": 0.23317452371120453,
      "num_tokens": 140745581.0,
      "step": 67825
    },
    {
      "entropy": 5.777467250823975,
      "epoch": 6.152939042089986,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019757539311429328,
      "loss": 4.88,
      "mean_token_accuracy": 0.23443654775619507,
      "num_tokens": 140755528.0,
      "step": 67830
    },
    {
      "entropy": 5.81812596321106,
      "epoch": 6.15339259796807,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019754501351550236,
      "loss": 4.9061,
      "mean_token_accuracy": 0.22824037373065947,
      "num_tokens": 140766892.0,
      "step": 67835
    },
    {
      "entropy": 5.844588994979858,
      "epoch": 6.153846153846154,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019751463551850024,
      "loss": 4.9145,
      "mean_token_accuracy": 0.2343122497200966,
      "num_tokens": 140777327.0,
      "step": 67840
    },
    {
      "entropy": 5.822874069213867,
      "epoch": 6.1542997097242385,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019748425912391522,
      "loss": 4.9047,
      "mean_token_accuracy": 0.2279055818915367,
      "num_tokens": 140787555.0,
      "step": 67845
    },
    {
      "entropy": 5.83050127029419,
      "epoch": 6.154753265602322,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0001974538843323755,
      "loss": 4.9011,
      "mean_token_accuracy": 0.22898381650447847,
      "num_tokens": 140797058.0,
      "step": 67850
    },
    {
      "entropy": 5.696348857879639,
      "epoch": 6.155206821480406,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001974235111445092,
      "loss": 4.7577,
      "mean_token_accuracy": 0.23945367634296416,
      "num_tokens": 140808420.0,
      "step": 67855
    },
    {
      "entropy": 5.687145566940307,
      "epoch": 6.15566037735849,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001973931395609444,
      "loss": 4.7494,
      "mean_token_accuracy": 0.24064653366804123,
      "num_tokens": 140818224.0,
      "step": 67860
    },
    {
      "entropy": 5.693771648406982,
      "epoch": 6.156113933236575,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001973627695823093,
      "loss": 4.783,
      "mean_token_accuracy": 0.23732432276010512,
      "num_tokens": 140827457.0,
      "step": 67865
    },
    {
      "entropy": 5.804610681533814,
      "epoch": 6.156567489114659,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001973324012092319,
      "loss": 4.8536,
      "mean_token_accuracy": 0.23772393614053727,
      "num_tokens": 140837675.0,
      "step": 67870
    },
    {
      "entropy": 5.774226713180542,
      "epoch": 6.157021044992743,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019730203444234029,
      "loss": 4.7987,
      "mean_token_accuracy": 0.24398029297590257,
      "num_tokens": 140847621.0,
      "step": 67875
    },
    {
      "entropy": 5.783900928497315,
      "epoch": 6.157474600870827,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019727166928226237,
      "loss": 4.8362,
      "mean_token_accuracy": 0.23893762230873108,
      "num_tokens": 140858193.0,
      "step": 67880
    },
    {
      "entropy": 5.7093900680542,
      "epoch": 6.157928156748912,
      "grad_norm": 1.375,
      "learning_rate": 0.00019724130572962612,
      "loss": 4.8293,
      "mean_token_accuracy": 0.22947366684675216,
      "num_tokens": 140868472.0,
      "step": 67885
    },
    {
      "entropy": 5.738605594635009,
      "epoch": 6.158381712626996,
      "grad_norm": 1.375,
      "learning_rate": 0.00019721094378505948,
      "loss": 4.8133,
      "mean_token_accuracy": 0.23526136875152587,
      "num_tokens": 140878409.0,
      "step": 67890
    },
    {
      "entropy": 5.722649574279785,
      "epoch": 6.15883526850508,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001971805834491904,
      "loss": 4.7515,
      "mean_token_accuracy": 0.2395060107111931,
      "num_tokens": 140889724.0,
      "step": 67895
    },
    {
      "entropy": 5.726387071609497,
      "epoch": 6.159288824383164,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001971502247226466,
      "loss": 4.8208,
      "mean_token_accuracy": 0.23535865247249604,
      "num_tokens": 140899603.0,
      "step": 67900
    },
    {
      "entropy": 5.749531364440918,
      "epoch": 6.159742380261248,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019711986760605606,
      "loss": 4.7646,
      "mean_token_accuracy": 0.23774496614933013,
      "num_tokens": 140909628.0,
      "step": 67905
    },
    {
      "entropy": 5.724930000305176,
      "epoch": 6.160195936139332,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019708951210004643,
      "loss": 4.8014,
      "mean_token_accuracy": 0.2352417975664139,
      "num_tokens": 140920023.0,
      "step": 67910
    },
    {
      "entropy": 5.7569445133209225,
      "epoch": 6.160649492017416,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019705915820524561,
      "loss": 4.9424,
      "mean_token_accuracy": 0.2287011668086052,
      "num_tokens": 140930522.0,
      "step": 67915
    },
    {
      "entropy": 5.760808420181275,
      "epoch": 6.161103047895501,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019702880592228122,
      "loss": 4.8355,
      "mean_token_accuracy": 0.23464999347925186,
      "num_tokens": 140939660.0,
      "step": 67920
    },
    {
      "entropy": 5.794876909255981,
      "epoch": 6.161556603773585,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019699845525178106,
      "loss": 4.9047,
      "mean_token_accuracy": 0.22395676523447036,
      "num_tokens": 140949807.0,
      "step": 67925
    },
    {
      "entropy": 5.7808233261108395,
      "epoch": 6.162010159651669,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019696810619437266,
      "loss": 4.886,
      "mean_token_accuracy": 0.2306601285934448,
      "num_tokens": 140960155.0,
      "step": 67930
    },
    {
      "entropy": 5.827651929855347,
      "epoch": 6.1624637155297535,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001969377587506837,
      "loss": 4.93,
      "mean_token_accuracy": 0.22962358742952346,
      "num_tokens": 140970281.0,
      "step": 67935
    },
    {
      "entropy": 5.777413272857666,
      "epoch": 6.162917271407838,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019690741292134186,
      "loss": 4.8311,
      "mean_token_accuracy": 0.2339954122900963,
      "num_tokens": 140980518.0,
      "step": 67940
    },
    {
      "entropy": 5.77697606086731,
      "epoch": 6.163370827285922,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019687706870697452,
      "loss": 4.8852,
      "mean_token_accuracy": 0.22318308502435685,
      "num_tokens": 140991219.0,
      "step": 67945
    },
    {
      "entropy": 5.842084455490112,
      "epoch": 6.163824383164005,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019684672610820948,
      "loss": 5.0017,
      "mean_token_accuracy": 0.21788234263658524,
      "num_tokens": 141001308.0,
      "step": 67950
    },
    {
      "entropy": 5.767294597625733,
      "epoch": 6.16427793904209,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001968163851256739,
      "loss": 4.8516,
      "mean_token_accuracy": 0.22718295007944106,
      "num_tokens": 141012305.0,
      "step": 67955
    },
    {
      "entropy": 5.768946647644043,
      "epoch": 6.164731494920174,
      "grad_norm": 1.3125,
      "learning_rate": 0.0001967860457599954,
      "loss": 4.7896,
      "mean_token_accuracy": 0.239033280313015,
      "num_tokens": 141023374.0,
      "step": 67960
    },
    {
      "entropy": 5.737782907485962,
      "epoch": 6.165185050798258,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019675570801180149,
      "loss": 4.8449,
      "mean_token_accuracy": 0.23482732623815536,
      "num_tokens": 141035056.0,
      "step": 67965
    },
    {
      "entropy": 5.695316743850708,
      "epoch": 6.1656386066763424,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001967253718817194,
      "loss": 4.7842,
      "mean_token_accuracy": 0.23574114292860032,
      "num_tokens": 141045721.0,
      "step": 67970
    },
    {
      "entropy": 5.760253763198852,
      "epoch": 6.166092162554427,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019669503737037658,
      "loss": 4.8632,
      "mean_token_accuracy": 0.22933317422866822,
      "num_tokens": 141057078.0,
      "step": 67975
    },
    {
      "entropy": 5.810639142990112,
      "epoch": 6.166545718432511,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019666470447840035,
      "loss": 4.8361,
      "mean_token_accuracy": 0.23416619151830673,
      "num_tokens": 141067059.0,
      "step": 67980
    },
    {
      "entropy": 5.893857192993164,
      "epoch": 6.166999274310595,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019663437320641803,
      "loss": 4.9732,
      "mean_token_accuracy": 0.23059966564178466,
      "num_tokens": 141078299.0,
      "step": 67985
    },
    {
      "entropy": 5.790112447738648,
      "epoch": 6.1674528301886795,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001966040435550568,
      "loss": 4.8187,
      "mean_token_accuracy": 0.23141716569662094,
      "num_tokens": 141088455.0,
      "step": 67990
    },
    {
      "entropy": 5.7773419380187985,
      "epoch": 6.167906386066764,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019657371552494397,
      "loss": 4.8569,
      "mean_token_accuracy": 0.22768006920814515,
      "num_tokens": 141098660.0,
      "step": 67995
    },
    {
      "entropy": 5.731306457519532,
      "epoch": 6.168359941944848,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019654338911670666,
      "loss": 4.7628,
      "mean_token_accuracy": 0.23905733227729797,
      "num_tokens": 141109226.0,
      "step": 68000
    },
    {
      "entropy": 5.824625253677368,
      "epoch": 6.168813497822931,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019651306433097205,
      "loss": 4.955,
      "mean_token_accuracy": 0.22798698097467424,
      "num_tokens": 141120702.0,
      "step": 68005
    },
    {
      "entropy": 5.754119205474853,
      "epoch": 6.169267053701016,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019648274116836723,
      "loss": 4.8592,
      "mean_token_accuracy": 0.22700864523649217,
      "num_tokens": 141131562.0,
      "step": 68010
    },
    {
      "entropy": 5.865433549880981,
      "epoch": 6.1697206095791,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019645241962951937,
      "loss": 4.8967,
      "mean_token_accuracy": 0.22951936721801758,
      "num_tokens": 141141552.0,
      "step": 68015
    },
    {
      "entropy": 5.811544179916382,
      "epoch": 6.170174165457184,
      "grad_norm": 1.25,
      "learning_rate": 0.00019642209971505555,
      "loss": 4.8119,
      "mean_token_accuracy": 0.2355509877204895,
      "num_tokens": 141151731.0,
      "step": 68020
    },
    {
      "entropy": 5.671693181991577,
      "epoch": 6.1706277213352685,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001963917814256026,
      "loss": 4.7548,
      "mean_token_accuracy": 0.23711583763360977,
      "num_tokens": 141161654.0,
      "step": 68025
    },
    {
      "entropy": 5.732721757888794,
      "epoch": 6.171081277213353,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019636146476178774,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23366188257932663,
      "num_tokens": 141172305.0,
      "step": 68030
    },
    {
      "entropy": 5.6948380947113035,
      "epoch": 6.171534833091437,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001963311497242378,
      "loss": 4.8582,
      "mean_token_accuracy": 0.22677429169416427,
      "num_tokens": 141184422.0,
      "step": 68035
    },
    {
      "entropy": 5.787738084793091,
      "epoch": 6.171988388969521,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019630083631357975,
      "loss": 4.8564,
      "mean_token_accuracy": 0.2301722288131714,
      "num_tokens": 141194933.0,
      "step": 68040
    },
    {
      "entropy": 5.871438217163086,
      "epoch": 6.1724419448476056,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019627052453044047,
      "loss": 4.8766,
      "mean_token_accuracy": 0.23387590497732164,
      "num_tokens": 141204868.0,
      "step": 68045
    },
    {
      "entropy": 5.746789407730103,
      "epoch": 6.17289550072569,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019624021437544676,
      "loss": 4.7519,
      "mean_token_accuracy": 0.23647409677505493,
      "num_tokens": 141214953.0,
      "step": 68050
    },
    {
      "entropy": 5.757099676132202,
      "epoch": 6.173349056603773,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019620990584922555,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23311498612165452,
      "num_tokens": 141224727.0,
      "step": 68055
    },
    {
      "entropy": 5.767353534698486,
      "epoch": 6.1738026124818575,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019617959895240346,
      "loss": 4.8516,
      "mean_token_accuracy": 0.23784394860267638,
      "num_tokens": 141234967.0,
      "step": 68060
    },
    {
      "entropy": 5.759497356414795,
      "epoch": 6.174256168359942,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019614929368560742,
      "loss": 4.9228,
      "mean_token_accuracy": 0.22231636494398116,
      "num_tokens": 141244993.0,
      "step": 68065
    },
    {
      "entropy": 5.700583267211914,
      "epoch": 6.174709724238026,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001961189900494641,
      "loss": 4.7791,
      "mean_token_accuracy": 0.24134743958711624,
      "num_tokens": 141254785.0,
      "step": 68070
    },
    {
      "entropy": 5.759486675262451,
      "epoch": 6.17516328011611,
      "grad_norm": 1.5,
      "learning_rate": 0.0001960886880446001,
      "loss": 4.8119,
      "mean_token_accuracy": 0.23162134289741515,
      "num_tokens": 141263951.0,
      "step": 68075
    },
    {
      "entropy": 5.7857184410095215,
      "epoch": 6.1756168359941945,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001960583876716422,
      "loss": 4.7945,
      "mean_token_accuracy": 0.23550778180360793,
      "num_tokens": 141274528.0,
      "step": 68080
    },
    {
      "entropy": 5.739968729019165,
      "epoch": 6.176070391872279,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019602808893121694,
      "loss": 4.8317,
      "mean_token_accuracy": 0.230347740650177,
      "num_tokens": 141285188.0,
      "step": 68085
    },
    {
      "entropy": 5.724470233917236,
      "epoch": 6.176523947750363,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019599779182395094,
      "loss": 4.748,
      "mean_token_accuracy": 0.23905672878026962,
      "num_tokens": 141295687.0,
      "step": 68090
    },
    {
      "entropy": 5.750295925140381,
      "epoch": 6.176977503628447,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019596749635047072,
      "loss": 4.8406,
      "mean_token_accuracy": 0.2322825998067856,
      "num_tokens": 141306458.0,
      "step": 68095
    },
    {
      "entropy": 5.758477306365966,
      "epoch": 6.177431059506532,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019593720251140278,
      "loss": 4.8661,
      "mean_token_accuracy": 0.2310992032289505,
      "num_tokens": 141317542.0,
      "step": 68100
    },
    {
      "entropy": 5.714965963363648,
      "epoch": 6.177884615384615,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019590691030737373,
      "loss": 4.8383,
      "mean_token_accuracy": 0.23019066601991653,
      "num_tokens": 141328358.0,
      "step": 68105
    },
    {
      "entropy": 5.824149656295776,
      "epoch": 6.178338171262699,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019587661973900983,
      "loss": 4.91,
      "mean_token_accuracy": 0.22754525542259216,
      "num_tokens": 141338403.0,
      "step": 68110
    },
    {
      "entropy": 5.8324220180511475,
      "epoch": 6.1787917271407835,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019584633080693765,
      "loss": 4.8919,
      "mean_token_accuracy": 0.23576952517032623,
      "num_tokens": 141349601.0,
      "step": 68115
    },
    {
      "entropy": 5.800671768188477,
      "epoch": 6.179245283018868,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001958160435117835,
      "loss": 4.8437,
      "mean_token_accuracy": 0.23587049692869186,
      "num_tokens": 141359505.0,
      "step": 68120
    },
    {
      "entropy": 5.824385404586792,
      "epoch": 6.179698838896952,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0001957857578541738,
      "loss": 4.8343,
      "mean_token_accuracy": 0.2329612746834755,
      "num_tokens": 141368785.0,
      "step": 68125
    },
    {
      "entropy": 5.763974618911743,
      "epoch": 6.180152394775036,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019575547383473475,
      "loss": 4.6854,
      "mean_token_accuracy": 0.2567302107810974,
      "num_tokens": 141378079.0,
      "step": 68130
    },
    {
      "entropy": 5.721750974655151,
      "epoch": 6.180605950653121,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00019572519145409273,
      "loss": 4.7971,
      "mean_token_accuracy": 0.23392157703638078,
      "num_tokens": 141389032.0,
      "step": 68135
    },
    {
      "entropy": 5.769975328445435,
      "epoch": 6.181059506531205,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019569491071287398,
      "loss": 4.8673,
      "mean_token_accuracy": 0.2288856253027916,
      "num_tokens": 141398914.0,
      "step": 68140
    },
    {
      "entropy": 5.750785255432129,
      "epoch": 6.181513062409289,
      "grad_norm": 1.5625,
      "learning_rate": 0.0001956646316117046,
      "loss": 4.791,
      "mean_token_accuracy": 0.23389577269554138,
      "num_tokens": 141408363.0,
      "step": 68145
    },
    {
      "entropy": 5.793417072296142,
      "epoch": 6.181966618287373,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019563435415121088,
      "loss": 4.8517,
      "mean_token_accuracy": 0.2381124749779701,
      "num_tokens": 141418308.0,
      "step": 68150
    },
    {
      "entropy": 5.808640336990356,
      "epoch": 6.182420174165458,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019560407833201903,
      "loss": 4.9464,
      "mean_token_accuracy": 0.2242927759885788,
      "num_tokens": 141428599.0,
      "step": 68155
    },
    {
      "entropy": 5.7849884033203125,
      "epoch": 6.182873730043541,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019557380415475496,
      "loss": 4.8555,
      "mean_token_accuracy": 0.22782624810934066,
      "num_tokens": 141438813.0,
      "step": 68160
    },
    {
      "entropy": 5.787525653839111,
      "epoch": 6.183327285921625,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019554353162004495,
      "loss": 4.8743,
      "mean_token_accuracy": 0.229655659198761,
      "num_tokens": 141449601.0,
      "step": 68165
    },
    {
      "entropy": 5.799880027770996,
      "epoch": 6.1837808417997095,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0001955132607285149,
      "loss": 4.9276,
      "mean_token_accuracy": 0.22181341499090196,
      "num_tokens": 141461275.0,
      "step": 68170
    },
    {
      "entropy": 5.728449583053589,
      "epoch": 6.184234397677794,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001954829914807909,
      "loss": 4.7964,
      "mean_token_accuracy": 0.23095902651548386,
      "num_tokens": 141471897.0,
      "step": 68175
    },
    {
      "entropy": 5.7315877914428714,
      "epoch": 6.184687953555878,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019545272387749886,
      "loss": 4.7774,
      "mean_token_accuracy": 0.23611980974674224,
      "num_tokens": 141481527.0,
      "step": 68180
    },
    {
      "entropy": 5.824347877502442,
      "epoch": 6.185141509433962,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019542245791926476,
      "loss": 4.9748,
      "mean_token_accuracy": 0.21790377348661422,
      "num_tokens": 141492060.0,
      "step": 68185
    },
    {
      "entropy": 5.737039518356323,
      "epoch": 6.185595065312047,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019539219360671455,
      "loss": 4.7917,
      "mean_token_accuracy": 0.23681549727916718,
      "num_tokens": 141502046.0,
      "step": 68190
    },
    {
      "entropy": 5.675017547607422,
      "epoch": 6.186048621190131,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019536193094047406,
      "loss": 4.7094,
      "mean_token_accuracy": 0.24339082688093186,
      "num_tokens": 141511635.0,
      "step": 68195
    },
    {
      "entropy": 5.757182168960571,
      "epoch": 6.186502177068215,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019533166992116913,
      "loss": 4.865,
      "mean_token_accuracy": 0.22742149829864503,
      "num_tokens": 141520947.0,
      "step": 68200
    },
    {
      "entropy": 5.685097169876099,
      "epoch": 6.186955732946299,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019530141054942558,
      "loss": 4.7882,
      "mean_token_accuracy": 0.23673250377178193,
      "num_tokens": 141531098.0,
      "step": 68205
    },
    {
      "entropy": 5.830147838592529,
      "epoch": 6.187409288824383,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00019527115282586916,
      "loss": 4.9749,
      "mean_token_accuracy": 0.22160821706056594,
      "num_tokens": 141542992.0,
      "step": 68210
    },
    {
      "entropy": 5.798212146759033,
      "epoch": 6.187862844702467,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019524089675112554,
      "loss": 4.9029,
      "mean_token_accuracy": 0.2315204694867134,
      "num_tokens": 141553240.0,
      "step": 68215
    },
    {
      "entropy": 5.732491827011108,
      "epoch": 6.188316400580551,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001952106423258206,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23013880103826523,
      "num_tokens": 141564074.0,
      "step": 68220
    },
    {
      "entropy": 5.747028541564942,
      "epoch": 6.188769956458636,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019518038955057987,
      "loss": 4.7873,
      "mean_token_accuracy": 0.23450636267662048,
      "num_tokens": 141574377.0,
      "step": 68225
    },
    {
      "entropy": 5.865301942825317,
      "epoch": 6.18922351233672,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019515013842602902,
      "loss": 4.9497,
      "mean_token_accuracy": 0.21837826073169708,
      "num_tokens": 141584010.0,
      "step": 68230
    },
    {
      "entropy": 5.747813892364502,
      "epoch": 6.189677068214804,
      "grad_norm": 1.625,
      "learning_rate": 0.0001951198889527937,
      "loss": 4.7714,
      "mean_token_accuracy": 0.2372491642832756,
      "num_tokens": 141593820.0,
      "step": 68235
    },
    {
      "entropy": 5.728704738616943,
      "epoch": 6.190130624092888,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019508964113149936,
      "loss": 4.8157,
      "mean_token_accuracy": 0.23683718144893645,
      "num_tokens": 141603977.0,
      "step": 68240
    },
    {
      "entropy": 5.716247081756592,
      "epoch": 6.190584179970973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001950593949627717,
      "loss": 4.7892,
      "mean_token_accuracy": 0.23626225888729097,
      "num_tokens": 141614008.0,
      "step": 68245
    },
    {
      "entropy": 5.734132719039917,
      "epoch": 6.191037735849057,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019502915044723601,
      "loss": 4.8698,
      "mean_token_accuracy": 0.2318076640367508,
      "num_tokens": 141625387.0,
      "step": 68250
    },
    {
      "entropy": 5.7685376644134525,
      "epoch": 6.191491291727141,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001949989075855179,
      "loss": 4.8828,
      "mean_token_accuracy": 0.22420918047428132,
      "num_tokens": 141635904.0,
      "step": 68255
    },
    {
      "entropy": 5.770950365066528,
      "epoch": 6.191944847605225,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019496866637824285,
      "loss": 4.8457,
      "mean_token_accuracy": 0.23832871913909912,
      "num_tokens": 141647506.0,
      "step": 68260
    },
    {
      "entropy": 5.762837696075439,
      "epoch": 6.192398403483309,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019493842682603607,
      "loss": 4.9211,
      "mean_token_accuracy": 0.23122931122779847,
      "num_tokens": 141660070.0,
      "step": 68265
    },
    {
      "entropy": 5.79764609336853,
      "epoch": 6.192851959361393,
      "grad_norm": 1.375,
      "learning_rate": 0.0001949081889295231,
      "loss": 4.8796,
      "mean_token_accuracy": 0.23389183282852172,
      "num_tokens": 141670833.0,
      "step": 68270
    },
    {
      "entropy": 5.7974060535430905,
      "epoch": 6.193305515239477,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019487795268932918,
      "loss": 4.8384,
      "mean_token_accuracy": 0.2298470616340637,
      "num_tokens": 141680426.0,
      "step": 68275
    },
    {
      "entropy": 5.748804712295533,
      "epoch": 6.193759071117562,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019484771810607954,
      "loss": 4.7913,
      "mean_token_accuracy": 0.23568861037492753,
      "num_tokens": 141691064.0,
      "step": 68280
    },
    {
      "entropy": 5.648009014129639,
      "epoch": 6.194212626995646,
      "grad_norm": 1.421875,
      "learning_rate": 0.0001948174851803996,
      "loss": 4.7571,
      "mean_token_accuracy": 0.23604741245508193,
      "num_tokens": 141700852.0,
      "step": 68285
    },
    {
      "entropy": 5.729213619232178,
      "epoch": 6.19466618287373,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019478725391291452,
      "loss": 4.7988,
      "mean_token_accuracy": 0.2366610899567604,
      "num_tokens": 141710686.0,
      "step": 68290
    },
    {
      "entropy": 5.723003292083741,
      "epoch": 6.195119738751814,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00019475702430424941,
      "loss": 4.7671,
      "mean_token_accuracy": 0.24486714452505112,
      "num_tokens": 141721004.0,
      "step": 68295
    },
    {
      "entropy": 5.81302170753479,
      "epoch": 6.195573294629899,
      "grad_norm": 1.265625,
      "learning_rate": 0.0001947267963550295,
      "loss": 4.9123,
      "mean_token_accuracy": 0.22691481858491896,
      "num_tokens": 141732052.0,
      "step": 68300
    },
    {
      "entropy": 5.834182405471802,
      "epoch": 6.196026850507983,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019469657006587987,
      "loss": 4.8837,
      "mean_token_accuracy": 0.23678964078426362,
      "num_tokens": 141743205.0,
      "step": 68305
    },
    {
      "entropy": 5.735522794723511,
      "epoch": 6.196480406386066,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00019466634543742568,
      "loss": 4.7497,
      "mean_token_accuracy": 0.23659525960683822,
      "num_tokens": 141753270.0,
      "step": 68310
    },
    {
      "entropy": 5.735728311538696,
      "epoch": 6.196933962264151,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0001946361224702919,
      "loss": 4.8461,
      "mean_token_accuracy": 0.23258105665445328,
      "num_tokens": 141763421.0,
      "step": 68315
    },
    {
      "entropy": 5.726525974273682,
      "epoch": 6.197387518142235,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019460590116510357,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23676591813564302,
      "num_tokens": 141772916.0,
      "step": 68320
    },
    {
      "entropy": 5.781346845626831,
      "epoch": 6.197841074020319,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019457568152248577,
      "loss": 4.8572,
      "mean_token_accuracy": 0.235866779088974,
      "num_tokens": 141783641.0,
      "step": 68325
    },
    {
      "entropy": 5.743303728103638,
      "epoch": 6.198294629898403,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019454546354306335,
      "loss": 4.8009,
      "mean_token_accuracy": 0.23780644983053206,
      "num_tokens": 141796413.0,
      "step": 68330
    },
    {
      "entropy": 5.677054166793823,
      "epoch": 6.198748185776488,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019451524722746116,
      "loss": 4.7341,
      "mean_token_accuracy": 0.24377057701349258,
      "num_tokens": 141805869.0,
      "step": 68335
    },
    {
      "entropy": 5.766119289398193,
      "epoch": 6.199201741654572,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019448503257630428,
      "loss": 4.84,
      "mean_token_accuracy": 0.2296954944729805,
      "num_tokens": 141814766.0,
      "step": 68340
    },
    {
      "entropy": 5.686682224273682,
      "epoch": 6.199655297532656,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019445481959021744,
      "loss": 4.7968,
      "mean_token_accuracy": 0.22977294474840165,
      "num_tokens": 141825758.0,
      "step": 68345
    },
    {
      "entropy": 5.751473903656006,
      "epoch": 6.2001088534107405,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019442460826982533,
      "loss": 4.8543,
      "mean_token_accuracy": 0.23564538955688477,
      "num_tokens": 141836825.0,
      "step": 68350
    },
    {
      "entropy": 5.728642559051513,
      "epoch": 6.200562409288825,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019439439861575292,
      "loss": 4.8275,
      "mean_token_accuracy": 0.23331257253885268,
      "num_tokens": 141847603.0,
      "step": 68355
    },
    {
      "entropy": 5.7370274543762205,
      "epoch": 6.201015965166908,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019436419062862493,
      "loss": 4.8445,
      "mean_token_accuracy": 0.23082710951566696,
      "num_tokens": 141857582.0,
      "step": 68360
    },
    {
      "entropy": 5.800535249710083,
      "epoch": 6.201469521044992,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019433398430906596,
      "loss": 4.911,
      "mean_token_accuracy": 0.2263369157910347,
      "num_tokens": 141868018.0,
      "step": 68365
    },
    {
      "entropy": 5.77344651222229,
      "epoch": 6.201923076923077,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019430377965770084,
      "loss": 4.9281,
      "mean_token_accuracy": 0.22782595753669738,
      "num_tokens": 141879135.0,
      "step": 68370
    },
    {
      "entropy": 5.787368297576904,
      "epoch": 6.202376632801161,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001942735766751541,
      "loss": 4.8167,
      "mean_token_accuracy": 0.24242521524429322,
      "num_tokens": 141889734.0,
      "step": 68375
    },
    {
      "entropy": 5.703829193115235,
      "epoch": 6.202830188679245,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001942433753620504,
      "loss": 4.7493,
      "mean_token_accuracy": 0.24154142737388612,
      "num_tokens": 141900665.0,
      "step": 68380
    },
    {
      "entropy": 5.771446752548218,
      "epoch": 6.2032837445573294,
      "grad_norm": 1.46875,
      "learning_rate": 0.0001942131757190142,
      "loss": 4.831,
      "mean_token_accuracy": 0.23643974661827089,
      "num_tokens": 141911620.0,
      "step": 68385
    },
    {
      "entropy": 5.813462448120117,
      "epoch": 6.203737300435414,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001941829777466701,
      "loss": 4.8482,
      "mean_token_accuracy": 0.22913027554750443,
      "num_tokens": 141923289.0,
      "step": 68390
    },
    {
      "entropy": 5.746178817749024,
      "epoch": 6.204190856313498,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019415278144564263,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23800437152385712,
      "num_tokens": 141934034.0,
      "step": 68395
    },
    {
      "entropy": 5.715216732025146,
      "epoch": 6.204644412191582,
      "grad_norm": 1.515625,
      "learning_rate": 0.00019412258681655625,
      "loss": 4.7762,
      "mean_token_accuracy": 0.2460413709282875,
      "num_tokens": 141944581.0,
      "step": 68400
    },
    {
      "entropy": 5.666214036941528,
      "epoch": 6.2050979680696665,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019409239386003542,
      "loss": 4.7558,
      "mean_token_accuracy": 0.24233687072992324,
      "num_tokens": 141955825.0,
      "step": 68405
    },
    {
      "entropy": 5.74931492805481,
      "epoch": 6.205551523947751,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019406220257670443,
      "loss": 4.8254,
      "mean_token_accuracy": 0.23065067529678346,
      "num_tokens": 141966818.0,
      "step": 68410
    },
    {
      "entropy": 5.723203182220459,
      "epoch": 6.206005079825834,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019403201296718775,
      "loss": 4.8636,
      "mean_token_accuracy": 0.23184253126382828,
      "num_tokens": 141977211.0,
      "step": 68415
    },
    {
      "entropy": 5.7555022716522215,
      "epoch": 6.206458635703918,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019400182503210973,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23995588421821595,
      "num_tokens": 141988366.0,
      "step": 68420
    },
    {
      "entropy": 5.724486827850342,
      "epoch": 6.206912191582003,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019397163877209455,
      "loss": 4.8147,
      "mean_token_accuracy": 0.23412748128175737,
      "num_tokens": 141997915.0,
      "step": 68425
    },
    {
      "entropy": 5.693898248672485,
      "epoch": 6.207365747460087,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001939414541877665,
      "loss": 4.7819,
      "mean_token_accuracy": 0.23800448030233384,
      "num_tokens": 142009082.0,
      "step": 68430
    },
    {
      "entropy": 5.771804618835449,
      "epoch": 6.207819303338171,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0001939112712797499,
      "loss": 4.8639,
      "mean_token_accuracy": 0.2345823675394058,
      "num_tokens": 142020110.0,
      "step": 68435
    },
    {
      "entropy": 5.762719440460205,
      "epoch": 6.2082728592162555,
      "grad_norm": 1.46875,
      "learning_rate": 0.00019388109004866882,
      "loss": 4.894,
      "mean_token_accuracy": 0.23041084855794908,
      "num_tokens": 142030926.0,
      "step": 68440
    },
    {
      "entropy": 5.758678913116455,
      "epoch": 6.20872641509434,
      "grad_norm": 1.40625,
      "learning_rate": 0.00019385091049514748,
      "loss": 4.7944,
      "mean_token_accuracy": 0.23465597182512282,
      "num_tokens": 142041207.0,
      "step": 68445
    },
    {
      "entropy": 5.788967227935791,
      "epoch": 6.209179970972424,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019382073261981004,
      "loss": 4.846,
      "mean_token_accuracy": 0.22997045516967773,
      "num_tokens": 142051044.0,
      "step": 68450
    },
    {
      "entropy": 5.796521711349487,
      "epoch": 6.209633526850508,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019379055642328052,
      "loss": 4.848,
      "mean_token_accuracy": 0.23569117486476898,
      "num_tokens": 142061303.0,
      "step": 68455
    },
    {
      "entropy": 5.741607618331909,
      "epoch": 6.2100870827285926,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019376038190618294,
      "loss": 4.848,
      "mean_token_accuracy": 0.22628170549869536,
      "num_tokens": 142071120.0,
      "step": 68460
    },
    {
      "entropy": 5.818660402297974,
      "epoch": 6.210540638606676,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001937302090691414,
      "loss": 4.9065,
      "mean_token_accuracy": 0.23222894966602325,
      "num_tokens": 142082400.0,
      "step": 68465
    },
    {
      "entropy": 5.7817198753356935,
      "epoch": 6.21099419448476,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00019370003791277984,
      "loss": 4.8322,
      "mean_token_accuracy": 0.23542119413614274,
      "num_tokens": 142092720.0,
      "step": 68470
    },
    {
      "entropy": 5.714219760894776,
      "epoch": 6.2114477503628445,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019366986843772223,
      "loss": 4.7592,
      "mean_token_accuracy": 0.24331459999084473,
      "num_tokens": 142103739.0,
      "step": 68475
    },
    {
      "entropy": 5.750930118560791,
      "epoch": 6.211901306240929,
      "grad_norm": 1.390625,
      "learning_rate": 0.0001936397006445924,
      "loss": 4.8433,
      "mean_token_accuracy": 0.2326820597052574,
      "num_tokens": 142113781.0,
      "step": 68480
    },
    {
      "entropy": 5.760606384277343,
      "epoch": 6.212354862119013,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019360953453401438,
      "loss": 4.8669,
      "mean_token_accuracy": 0.22821013629436493,
      "num_tokens": 142125551.0,
      "step": 68485
    },
    {
      "entropy": 5.760803604125977,
      "epoch": 6.212808417997097,
      "grad_norm": 1.71875,
      "learning_rate": 0.00019357937010661187,
      "loss": 4.805,
      "mean_token_accuracy": 0.23692142367362976,
      "num_tokens": 142135999.0,
      "step": 68490
    },
    {
      "entropy": 5.770647287368774,
      "epoch": 6.2132619738751815,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019354920736300874,
      "loss": 4.8401,
      "mean_token_accuracy": 0.2358032777905464,
      "num_tokens": 142146245.0,
      "step": 68495
    },
    {
      "entropy": 5.721803665161133,
      "epoch": 6.213715529753266,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019351904630382881,
      "loss": 4.8557,
      "mean_token_accuracy": 0.2256752371788025,
      "num_tokens": 142156189.0,
      "step": 68500
    },
    {
      "entropy": 5.785386991500855,
      "epoch": 6.21416908563135,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019348888692969574,
      "loss": 4.8558,
      "mean_token_accuracy": 0.23299306333065034,
      "num_tokens": 142165997.0,
      "step": 68505
    },
    {
      "entropy": 5.773633337020874,
      "epoch": 6.214622641509434,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019345872924123327,
      "loss": 4.8552,
      "mean_token_accuracy": 0.22459729313850402,
      "num_tokens": 142176976.0,
      "step": 68510
    },
    {
      "entropy": 5.755441808700562,
      "epoch": 6.215076197387518,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019342857323906499,
      "loss": 4.8515,
      "mean_token_accuracy": 0.23145491480827332,
      "num_tokens": 142188010.0,
      "step": 68515
    },
    {
      "entropy": 5.791053295135498,
      "epoch": 6.215529753265602,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019339841892381464,
      "loss": 4.9018,
      "mean_token_accuracy": 0.2262973427772522,
      "num_tokens": 142198170.0,
      "step": 68520
    },
    {
      "entropy": 5.799345636367798,
      "epoch": 6.215983309143686,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019336826629610582,
      "loss": 4.8165,
      "mean_token_accuracy": 0.2301468849182129,
      "num_tokens": 142207534.0,
      "step": 68525
    },
    {
      "entropy": 5.765764999389648,
      "epoch": 6.2164368650217705,
      "grad_norm": 1.453125,
      "learning_rate": 0.000193338115356562,
      "loss": 4.8409,
      "mean_token_accuracy": 0.2299135744571686,
      "num_tokens": 142217752.0,
      "step": 68530
    },
    {
      "entropy": 5.764827728271484,
      "epoch": 6.216890420899855,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001933079661058068,
      "loss": 4.8325,
      "mean_token_accuracy": 0.230502949655056,
      "num_tokens": 142227488.0,
      "step": 68535
    },
    {
      "entropy": 5.725417518615723,
      "epoch": 6.217343976777939,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019327781854446368,
      "loss": 4.8422,
      "mean_token_accuracy": 0.23271271884441375,
      "num_tokens": 142237384.0,
      "step": 68540
    },
    {
      "entropy": 5.757314395904541,
      "epoch": 6.217797532656023,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019324767267315612,
      "loss": 4.881,
      "mean_token_accuracy": 0.23196639567613603,
      "num_tokens": 142248588.0,
      "step": 68545
    },
    {
      "entropy": 5.8229282855987545,
      "epoch": 6.218251088534108,
      "grad_norm": 1.53125,
      "learning_rate": 0.00019321752849250746,
      "loss": 4.8663,
      "mean_token_accuracy": 0.2319558694958687,
      "num_tokens": 142259019.0,
      "step": 68550
    },
    {
      "entropy": 5.806328010559082,
      "epoch": 6.218704644412192,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019318738600314117,
      "loss": 4.8476,
      "mean_token_accuracy": 0.22903346717357637,
      "num_tokens": 142269154.0,
      "step": 68555
    },
    {
      "entropy": 5.760844612121582,
      "epoch": 6.219158200290276,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001931572452056806,
      "loss": 4.8438,
      "mean_token_accuracy": 0.23679348081350327,
      "num_tokens": 142279253.0,
      "step": 68560
    },
    {
      "entropy": 5.7216489791870115,
      "epoch": 6.21961175616836,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000193127106100749,
      "loss": 4.8313,
      "mean_token_accuracy": 0.23151487708091736,
      "num_tokens": 142289671.0,
      "step": 68565
    },
    {
      "entropy": 5.681403684616089,
      "epoch": 6.220065312046444,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001930969686889698,
      "loss": 4.7694,
      "mean_token_accuracy": 0.2409108981490135,
      "num_tokens": 142300716.0,
      "step": 68570
    },
    {
      "entropy": 5.743958473205566,
      "epoch": 6.220518867924528,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019306683297096617,
      "loss": 4.7922,
      "mean_token_accuracy": 0.23076585084199905,
      "num_tokens": 142311144.0,
      "step": 68575
    },
    {
      "entropy": 5.782385396957397,
      "epoch": 6.220972423802612,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0001930366989473612,
      "loss": 4.8583,
      "mean_token_accuracy": 0.23400825262069702,
      "num_tokens": 142320781.0,
      "step": 68580
    },
    {
      "entropy": 5.726973915100098,
      "epoch": 6.2214259796806965,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019300656661877825,
      "loss": 4.8168,
      "mean_token_accuracy": 0.2354642316699028,
      "num_tokens": 142331197.0,
      "step": 68585
    },
    {
      "entropy": 5.7472950458526615,
      "epoch": 6.221879535558781,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019297643598584042,
      "loss": 4.8194,
      "mean_token_accuracy": 0.23120469599962234,
      "num_tokens": 142341593.0,
      "step": 68590
    },
    {
      "entropy": 5.710521936416626,
      "epoch": 6.222333091436865,
      "grad_norm": 1.3125,
      "learning_rate": 0.00019294630704917078,
      "loss": 4.8097,
      "mean_token_accuracy": 0.23163318186998366,
      "num_tokens": 142351920.0,
      "step": 68595
    },
    {
      "entropy": 5.771629047393799,
      "epoch": 6.222786647314949,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019291617980939235,
      "loss": 4.9055,
      "mean_token_accuracy": 0.2215645879507065,
      "num_tokens": 142361889.0,
      "step": 68600
    },
    {
      "entropy": 5.728597688674927,
      "epoch": 6.223240203193034,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00019288605426712836,
      "loss": 4.7689,
      "mean_token_accuracy": 0.23889108896255493,
      "num_tokens": 142372588.0,
      "step": 68605
    },
    {
      "entropy": 5.8259114742279055,
      "epoch": 6.223693759071118,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019285593042300163,
      "loss": 4.945,
      "mean_token_accuracy": 0.22134792804718018,
      "num_tokens": 142382957.0,
      "step": 68610
    },
    {
      "entropy": 5.849714279174805,
      "epoch": 6.224147314949202,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019282580827763518,
      "loss": 4.8742,
      "mean_token_accuracy": 0.22989605963230134,
      "num_tokens": 142392928.0,
      "step": 68615
    },
    {
      "entropy": 5.772914791107178,
      "epoch": 6.2246008708272855,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001927956878316519,
      "loss": 4.8562,
      "mean_token_accuracy": 0.2381088986992836,
      "num_tokens": 142402788.0,
      "step": 68620
    },
    {
      "entropy": 5.786472749710083,
      "epoch": 6.22505442670537,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001927655690856749,
      "loss": 4.8441,
      "mean_token_accuracy": 0.22369681000709535,
      "num_tokens": 142412791.0,
      "step": 68625
    },
    {
      "entropy": 5.728670072555542,
      "epoch": 6.225507982583454,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00019273545204032678,
      "loss": 4.8403,
      "mean_token_accuracy": 0.22451526820659637,
      "num_tokens": 142423499.0,
      "step": 68630
    },
    {
      "entropy": 5.796323490142822,
      "epoch": 6.225961538461538,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019270533669623046,
      "loss": 4.88,
      "mean_token_accuracy": 0.23415573984384536,
      "num_tokens": 142434055.0,
      "step": 68635
    },
    {
      "entropy": 5.7409432888031,
      "epoch": 6.226415094339623,
      "grad_norm": 1.40625,
      "learning_rate": 0.0001926752230540088,
      "loss": 4.818,
      "mean_token_accuracy": 0.23293601870536804,
      "num_tokens": 142444501.0,
      "step": 68640
    },
    {
      "entropy": 5.799442195892334,
      "epoch": 6.226868650217707,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019264511111428446,
      "loss": 4.8855,
      "mean_token_accuracy": 0.22690346390008925,
      "num_tokens": 142454702.0,
      "step": 68645
    },
    {
      "entropy": 5.746590423583984,
      "epoch": 6.227322206095791,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019261500087768024,
      "loss": 4.871,
      "mean_token_accuracy": 0.23152179569005965,
      "num_tokens": 142465415.0,
      "step": 68650
    },
    {
      "entropy": 5.67354383468628,
      "epoch": 6.227775761973875,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001925848923448188,
      "loss": 4.7276,
      "mean_token_accuracy": 0.23490179032087327,
      "num_tokens": 142475681.0,
      "step": 68655
    },
    {
      "entropy": 5.774940729141235,
      "epoch": 6.22822931785196,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019255478551632278,
      "loss": 4.8293,
      "mean_token_accuracy": 0.23320697695016862,
      "num_tokens": 142485120.0,
      "step": 68660
    },
    {
      "entropy": 5.744373559951782,
      "epoch": 6.228682873730044,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019252468039281478,
      "loss": 4.8517,
      "mean_token_accuracy": 0.23123349994421005,
      "num_tokens": 142494580.0,
      "step": 68665
    },
    {
      "entropy": 5.735438966751099,
      "epoch": 6.229136429608127,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019249457697491745,
      "loss": 4.8456,
      "mean_token_accuracy": 0.232622030377388,
      "num_tokens": 142504472.0,
      "step": 68670
    },
    {
      "entropy": 5.773210144042968,
      "epoch": 6.229589985486212,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019246447526325328,
      "loss": 4.8809,
      "mean_token_accuracy": 0.2326679527759552,
      "num_tokens": 142515355.0,
      "step": 68675
    },
    {
      "entropy": 5.715752410888672,
      "epoch": 6.230043541364296,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019243437525844482,
      "loss": 4.7584,
      "mean_token_accuracy": 0.23921316862106323,
      "num_tokens": 142524948.0,
      "step": 68680
    },
    {
      "entropy": 5.776347637176514,
      "epoch": 6.23049709724238,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019240427696111447,
      "loss": 4.8086,
      "mean_token_accuracy": 0.2381380945444107,
      "num_tokens": 142536117.0,
      "step": 68685
    },
    {
      "entropy": 5.729855251312256,
      "epoch": 6.230950653120464,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019237418037188476,
      "loss": 4.7662,
      "mean_token_accuracy": 0.23038151413202285,
      "num_tokens": 142546571.0,
      "step": 68690
    },
    {
      "entropy": 5.739654111862182,
      "epoch": 6.231404208998549,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000192344085491378,
      "loss": 4.8041,
      "mean_token_accuracy": 0.2357732266187668,
      "num_tokens": 142556839.0,
      "step": 68695
    },
    {
      "entropy": 5.7700430393219,
      "epoch": 6.231857764876633,
      "grad_norm": 1.4375,
      "learning_rate": 0.0001923139923202166,
      "loss": 4.7852,
      "mean_token_accuracy": 0.23283611536026,
      "num_tokens": 142566959.0,
      "step": 68700
    },
    {
      "entropy": 5.733782148361206,
      "epoch": 6.232311320754717,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019228390085902303,
      "loss": 4.8641,
      "mean_token_accuracy": 0.23060573041439056,
      "num_tokens": 142576894.0,
      "step": 68705
    },
    {
      "entropy": 5.794253206253051,
      "epoch": 6.232764876632801,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00019225381110841945,
      "loss": 4.7997,
      "mean_token_accuracy": 0.22575515657663345,
      "num_tokens": 142587636.0,
      "step": 68710
    },
    {
      "entropy": 5.756149435043335,
      "epoch": 6.233218432510886,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000192223723069028,
      "loss": 4.8304,
      "mean_token_accuracy": 0.22620662152767182,
      "num_tokens": 142598179.0,
      "step": 68715
    },
    {
      "entropy": 5.740657901763916,
      "epoch": 6.23367198838897,
      "grad_norm": 1.59375,
      "learning_rate": 0.00019219363674147116,
      "loss": 4.7954,
      "mean_token_accuracy": 0.2432243973016739,
      "num_tokens": 142606919.0,
      "step": 68720
    },
    {
      "entropy": 5.707255411148071,
      "epoch": 6.234125544267053,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019216355212637104,
      "loss": 4.7469,
      "mean_token_accuracy": 0.2427612543106079,
      "num_tokens": 142617595.0,
      "step": 68725
    },
    {
      "entropy": 5.7644249439239506,
      "epoch": 6.234579100145138,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001921334692243497,
      "loss": 4.8591,
      "mean_token_accuracy": 0.23263539522886276,
      "num_tokens": 142627948.0,
      "step": 68730
    },
    {
      "entropy": 5.729491472244263,
      "epoch": 6.235032656023222,
      "grad_norm": 1.5,
      "learning_rate": 0.00019210338803602933,
      "loss": 4.7856,
      "mean_token_accuracy": 0.23299750685691833,
      "num_tokens": 142638542.0,
      "step": 68735
    },
    {
      "entropy": 5.753056383132934,
      "epoch": 6.235486211901306,
      "grad_norm": 1.484375,
      "learning_rate": 0.0001920733085620321,
      "loss": 4.7727,
      "mean_token_accuracy": 0.23523242473602296,
      "num_tokens": 142648454.0,
      "step": 68740
    },
    {
      "entropy": 5.676243638992309,
      "epoch": 6.23593976777939,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019204323080297992,
      "loss": 4.7453,
      "mean_token_accuracy": 0.24695050567388535,
      "num_tokens": 142658834.0,
      "step": 68745
    },
    {
      "entropy": 5.689120531082153,
      "epoch": 6.236393323657475,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00019201315475949487,
      "loss": 4.7952,
      "mean_token_accuracy": 0.23241864740848542,
      "num_tokens": 142669667.0,
      "step": 68750
    },
    {
      "entropy": 5.775503253936767,
      "epoch": 6.236846879535559,
      "grad_norm": 1.5625,
      "learning_rate": 0.00019198308043219891,
      "loss": 4.874,
      "mean_token_accuracy": 0.22712098509073259,
      "num_tokens": 142678994.0,
      "step": 68755
    },
    {
      "entropy": 5.728653764724731,
      "epoch": 6.237300435413643,
      "grad_norm": 1.34375,
      "learning_rate": 0.00019195300782171403,
      "loss": 4.8277,
      "mean_token_accuracy": 0.240304796397686,
      "num_tokens": 142690069.0,
      "step": 68760
    },
    {
      "entropy": 5.78955044746399,
      "epoch": 6.2377539912917275,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019192293692866215,
      "loss": 4.8118,
      "mean_token_accuracy": 0.24101473540067672,
      "num_tokens": 142700086.0,
      "step": 68765
    },
    {
      "entropy": 5.774494218826294,
      "epoch": 6.238207547169812,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00019189286775366505,
      "loss": 4.8542,
      "mean_token_accuracy": 0.23951285183429719,
      "num_tokens": 142710035.0,
      "step": 68770
    },
    {
      "entropy": 5.738926649093628,
      "epoch": 6.238661103047895,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0001918628002973446,
      "loss": 4.8284,
      "mean_token_accuracy": 0.23039046823978424,
      "num_tokens": 142720421.0,
      "step": 68775
    },
    {
      "entropy": 5.808580589294434,
      "epoch": 6.239114658925979,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019183273456032269,
      "loss": 4.9208,
      "mean_token_accuracy": 0.22608001977205278,
      "num_tokens": 142731053.0,
      "step": 68780
    },
    {
      "entropy": 5.797839736938476,
      "epoch": 6.239568214804064,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019180267054322097,
      "loss": 4.8156,
      "mean_token_accuracy": 0.23674831688404083,
      "num_tokens": 142741510.0,
      "step": 68785
    },
    {
      "entropy": 5.7959362983703615,
      "epoch": 6.240021770682148,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019177260824666126,
      "loss": 4.8644,
      "mean_token_accuracy": 0.23390918523073195,
      "num_tokens": 142752817.0,
      "step": 68790
    },
    {
      "entropy": 5.731265687942505,
      "epoch": 6.240475326560232,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0001917425476712652,
      "loss": 4.7949,
      "mean_token_accuracy": 0.24029470086097718,
      "num_tokens": 142762742.0,
      "step": 68795
    },
    {
      "entropy": 5.765463209152221,
      "epoch": 6.2409288824383164,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0001917124888176545,
      "loss": 4.8863,
      "mean_token_accuracy": 0.22673682272434234,
      "num_tokens": 142773424.0,
      "step": 68800
    },
    {
      "entropy": 5.653775024414062,
      "epoch": 6.241382438316401,
      "grad_norm": 1.484375,
      "learning_rate": 0.00019168243168645084,
      "loss": 4.7236,
      "mean_token_accuracy": 0.2459185838699341,
      "num_tokens": 142783530.0,
      "step": 68805
    },
    {
      "entropy": 5.742193269729614,
      "epoch": 6.241835994194485,
      "grad_norm": 1.5,
      "learning_rate": 0.00019165237627827557,
      "loss": 4.7989,
      "mean_token_accuracy": 0.23039519637823105,
      "num_tokens": 142793074.0,
      "step": 68810
    },
    {
      "entropy": 5.771000719070434,
      "epoch": 6.242289550072569,
      "grad_norm": 1.609375,
      "learning_rate": 0.00019162232259375045,
      "loss": 4.8,
      "mean_token_accuracy": 0.23299247026443481,
      "num_tokens": 142803127.0,
      "step": 68815
    },
    {
      "entropy": 5.7672034740448,
      "epoch": 6.2427431059506535,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00019159227063349698,
      "loss": 4.8937,
      "mean_token_accuracy": 0.23087007850408553,
      "num_tokens": 142813473.0,
      "step": 68820
    },
    {
      "entropy": 5.763075590133667,
      "epoch": 6.243196661828737,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00019156222039813658,
      "loss": 4.8512,
      "mean_token_accuracy": 0.23829931020736694,
      "num_tokens": 142824117.0,
      "step": 68825
    },
    {
      "entropy": 5.810499095916748,
      "epoch": 6.243650217706821,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001915321718882907,
      "loss": 4.8669,
      "mean_token_accuracy": 0.22940510660409927,
      "num_tokens": 142834801.0,
      "step": 68830
    },
    {
      "entropy": 5.78297610282898,
      "epoch": 6.244103773584905,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019150212510458077,
      "loss": 4.8038,
      "mean_token_accuracy": 0.23705462962388993,
      "num_tokens": 142845185.0,
      "step": 68835
    },
    {
      "entropy": 5.692420482635498,
      "epoch": 6.24455732946299,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019147208004762823,
      "loss": 4.7558,
      "mean_token_accuracy": 0.2394327476620674,
      "num_tokens": 142855341.0,
      "step": 68840
    },
    {
      "entropy": 5.786822032928467,
      "epoch": 6.245010885341074,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00019144203671805433,
      "loss": 4.8456,
      "mean_token_accuracy": 0.236816143989563,
      "num_tokens": 142865022.0,
      "step": 68845
    },
    {
      "entropy": 5.729842329025269,
      "epoch": 6.245464441219158,
      "grad_norm": 1.453125,
      "learning_rate": 0.00019141199511648032,
      "loss": 4.7995,
      "mean_token_accuracy": 0.23446157872676848,
      "num_tokens": 142875409.0,
      "step": 68850
    },
    {
      "entropy": 5.7198157787323,
      "epoch": 6.2459179970972425,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0001913819552435277,
      "loss": 4.753,
      "mean_token_accuracy": 0.23497931957244872,
      "num_tokens": 142885852.0,
      "step": 68855
    },
    {
      "entropy": 5.796526098251343,
      "epoch": 6.246371552975327,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019135191709981748,
      "loss": 4.8658,
      "mean_token_accuracy": 0.23225551545619966,
      "num_tokens": 142894827.0,
      "step": 68860
    },
    {
      "entropy": 5.816075468063355,
      "epoch": 6.246825108853411,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0001913218806859709,
      "loss": 4.9603,
      "mean_token_accuracy": 0.2209920182824135,
      "num_tokens": 142904495.0,
      "step": 68865
    },
    {
      "entropy": 5.719818115234375,
      "epoch": 6.247278664731495,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019129184600260918,
      "loss": 4.7747,
      "mean_token_accuracy": 0.2443758949637413,
      "num_tokens": 142915105.0,
      "step": 68870
    },
    {
      "entropy": 5.745254945755005,
      "epoch": 6.24773222060958,
      "grad_norm": 1.359375,
      "learning_rate": 0.00019126181305035344,
      "loss": 4.8118,
      "mean_token_accuracy": 0.23440636098384857,
      "num_tokens": 142925807.0,
      "step": 68875
    },
    {
      "entropy": 5.731783056259156,
      "epoch": 6.248185776487663,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001912317818298247,
      "loss": 4.7676,
      "mean_token_accuracy": 0.23805595189332962,
      "num_tokens": 142936967.0,
      "step": 68880
    },
    {
      "entropy": 5.753925943374634,
      "epoch": 6.248639332365747,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019120175234164412,
      "loss": 4.8223,
      "mean_token_accuracy": 0.2351837322115898,
      "num_tokens": 142946332.0,
      "step": 68885
    },
    {
      "entropy": 5.73666615486145,
      "epoch": 6.2490928882438315,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00019117172458643262,
      "loss": 4.8015,
      "mean_token_accuracy": 0.2292983889579773,
      "num_tokens": 142956472.0,
      "step": 68890
    },
    {
      "entropy": 5.834034824371338,
      "epoch": 6.249546444121916,
      "grad_norm": 1.5,
      "learning_rate": 0.00019114169856481123,
      "loss": 4.871,
      "mean_token_accuracy": 0.2298872858285904,
      "num_tokens": 142966281.0,
      "step": 68895
    },
    {
      "entropy": 5.744888687133789,
      "epoch": 6.25,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00019111167427740083,
      "loss": 4.8012,
      "mean_token_accuracy": 0.23722619861364364,
      "num_tokens": 142976423.0,
      "step": 68900
    },
    {
      "entropy": 5.679077005386352,
      "epoch": 6.250453555878084,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00019108165172482245,
      "loss": 4.8249,
      "mean_token_accuracy": 0.23733802884817123,
      "num_tokens": 142986711.0,
      "step": 68905
    },
    {
      "entropy": 5.754064273834229,
      "epoch": 6.2509071117561685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0001910516309076969,
      "loss": 4.8852,
      "mean_token_accuracy": 0.2239206150174141,
      "num_tokens": 142996624.0,
      "step": 68910
    },
    {
      "entropy": 5.779891633987427,
      "epoch": 6.251360667634253,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00019102161182664492,
      "loss": 4.8432,
      "mean_token_accuracy": 0.22981505393981932,
      "num_tokens": 143006627.0,
      "step": 68915
    },
    {
      "entropy": 5.792285346984864,
      "epoch": 6.251814223512337,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00019099159448228754,
      "loss": 4.8105,
      "mean_token_accuracy": 0.23682481050491333,
      "num_tokens": 143016920.0,
      "step": 68920
    },
    {
      "entropy": 5.796517848968506,
      "epoch": 6.25226777939042,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00019096157887524528,
      "loss": 4.8354,
      "mean_token_accuracy": 0.23636822700500487,
      "num_tokens": 143026077.0,
      "step": 68925
    },
    {
      "entropy": 5.7402091979980465,
      "epoch": 6.252721335268505,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00019093156500613908,
      "loss": 4.8315,
      "mean_token_accuracy": 0.2297897756099701,
      "num_tokens": 143036923.0,
      "step": 68930
    },
    {
      "entropy": 5.7800201892852785,
      "epoch": 6.253174891146589,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019090155287558946,
      "loss": 4.8828,
      "mean_token_accuracy": 0.23253460824489594,
      "num_tokens": 143047717.0,
      "step": 68935
    },
    {
      "entropy": 5.750406646728516,
      "epoch": 6.253628447024673,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00019087154248421717,
      "loss": 4.8545,
      "mean_token_accuracy": 0.23448456078767776,
      "num_tokens": 143057830.0,
      "step": 68940
    },
    {
      "entropy": 5.777360439300537,
      "epoch": 6.2540820029027575,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00019084153383264286,
      "loss": 4.8771,
      "mean_token_accuracy": 0.22977794706821442,
      "num_tokens": 143067353.0,
      "step": 68945
    },
    {
      "entropy": 5.769514322280884,
      "epoch": 6.254535558780842,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000190811526921487,
      "loss": 4.8427,
      "mean_token_accuracy": 0.23284498751163482,
      "num_tokens": 143077830.0,
      "step": 68950
    },
    {
      "entropy": 5.822436904907226,
      "epoch": 6.254989114658926,
      "grad_norm": 1.546875,
      "learning_rate": 0.00019078152175137038,
      "loss": 4.9227,
      "mean_token_accuracy": 0.22364840507507325,
      "num_tokens": 143088266.0,
      "step": 68955
    },
    {
      "entropy": 5.716494750976563,
      "epoch": 6.25544267053701,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001907515183229132,
      "loss": 4.7868,
      "mean_token_accuracy": 0.23879676759243013,
      "num_tokens": 143098814.0,
      "step": 68960
    },
    {
      "entropy": 5.68832106590271,
      "epoch": 6.255896226415095,
      "grad_norm": 1.390625,
      "learning_rate": 0.00019072151663673608,
      "loss": 4.7759,
      "mean_token_accuracy": 0.23723544031381608,
      "num_tokens": 143109575.0,
      "step": 68965
    },
    {
      "entropy": 5.800169181823731,
      "epoch": 6.256349782293179,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019069151669345958,
      "loss": 4.8644,
      "mean_token_accuracy": 0.24043179005384446,
      "num_tokens": 143119865.0,
      "step": 68970
    },
    {
      "entropy": 5.729971694946289,
      "epoch": 6.256803338171263,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00019066151849370395,
      "loss": 4.8575,
      "mean_token_accuracy": 0.22633883655071257,
      "num_tokens": 143132059.0,
      "step": 68975
    },
    {
      "entropy": 5.70842490196228,
      "epoch": 6.2572568940493465,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001906315220380896,
      "loss": 4.7617,
      "mean_token_accuracy": 0.24761313796043397,
      "num_tokens": 143142172.0,
      "step": 68980
    },
    {
      "entropy": 5.751295280456543,
      "epoch": 6.257710449927431,
      "grad_norm": 1.671875,
      "learning_rate": 0.00019060152732723685,
      "loss": 4.8415,
      "mean_token_accuracy": 0.23209351897239686,
      "num_tokens": 143152017.0,
      "step": 68985
    },
    {
      "entropy": 5.753719615936279,
      "epoch": 6.258164005805515,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001905715343617661,
      "loss": 4.8299,
      "mean_token_accuracy": 0.2386631265282631,
      "num_tokens": 143163482.0,
      "step": 68990
    },
    {
      "entropy": 5.704120635986328,
      "epoch": 6.258617561683599,
      "grad_norm": 1.4375,
      "learning_rate": 0.00019054154314229749,
      "loss": 4.786,
      "mean_token_accuracy": 0.23058185279369353,
      "num_tokens": 143173448.0,
      "step": 68995
    },
    {
      "entropy": 5.695918273925781,
      "epoch": 6.2590711175616836,
      "grad_norm": 1.328125,
      "learning_rate": 0.00019051155366945128,
      "loss": 4.7574,
      "mean_token_accuracy": 0.23691477626562119,
      "num_tokens": 143183770.0,
      "step": 69000
    },
    {
      "epoch": 6.2590711175616836,
      "eval_entropy": 5.472878076911512,
      "eval_loss": 4.935636520385742,
      "eval_mean_token_accuracy": 0.23541381699854838,
      "eval_num_tokens": 143183770.0,
      "eval_runtime": 20.5215,
      "eval_samples_per_second": 1723.074,
      "eval_steps_per_second": 215.384,
      "step": 69000
    }
  ],
  "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": 3.106370828860416e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}