{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9049346879535558,
  "eval_steps": 3000,
  "global_step": 21000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.69149408340454,
      "epoch": 0.00045355587808418,
      "grad_norm": 13.0625,
      "learning_rate": 2e-06,
      "loss": 10.7699,
      "mean_token_accuracy": 0.0,
      "num_tokens": 9661.0,
      "step": 5
    },
    {
      "entropy": 10.691533756256103,
      "epoch": 0.00090711175616836,
      "grad_norm": 12.8125,
      "learning_rate": 4.5e-06,
      "loss": 10.7409,
      "mean_token_accuracy": 0.00011574074160307646,
      "num_tokens": 21028.0,
      "step": 10
    },
    {
      "entropy": 10.691256523132324,
      "epoch": 0.00136066763425254,
      "grad_norm": 9.6875,
      "learning_rate": 7e-06,
      "loss": 10.4719,
      "mean_token_accuracy": 0.019357142923399805,
      "num_tokens": 31794.0,
      "step": 15
    },
    {
      "entropy": 10.684057140350342,
      "epoch": 0.00181422351233672,
      "grad_norm": 5.8125,
      "learning_rate": 9.5e-06,
      "loss": 10.1776,
      "mean_token_accuracy": 0.04135416336357593,
      "num_tokens": 42892.0,
      "step": 20
    },
    {
      "entropy": 10.636651706695556,
      "epoch": 0.0022677793904208998,
      "grad_norm": 4.125,
      "learning_rate": 1.2e-05,
      "loss": 9.8988,
      "mean_token_accuracy": 0.04433486089110374,
      "num_tokens": 53015.0,
      "step": 25
    },
    {
      "entropy": 10.585999393463135,
      "epoch": 0.00272133526850508,
      "grad_norm": 3.15625,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 9.7643,
      "mean_token_accuracy": 0.04109931848943234,
      "num_tokens": 63334.0,
      "step": 30
    },
    {
      "entropy": 10.606278038024902,
      "epoch": 0.00317489114658926,
      "grad_norm": 2.8125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 9.6589,
      "mean_token_accuracy": 0.04485522694885731,
      "num_tokens": 74256.0,
      "step": 35
    },
    {
      "entropy": 10.576828956604004,
      "epoch": 0.00362844702467344,
      "grad_norm": 2.734375,
      "learning_rate": 1.95e-05,
      "loss": 9.6182,
      "mean_token_accuracy": 0.046111271157860755,
      "num_tokens": 84699.0,
      "step": 40
    },
    {
      "entropy": 10.563911628723144,
      "epoch": 0.00408200290275762,
      "grad_norm": 2.46875,
      "learning_rate": 2.2e-05,
      "loss": 9.581,
      "mean_token_accuracy": 0.04264976643025875,
      "num_tokens": 95178.0,
      "step": 45
    },
    {
      "entropy": 10.59095401763916,
      "epoch": 0.0045355587808417995,
      "grad_norm": 2.3125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 9.5471,
      "mean_token_accuracy": 0.04867156893014908,
      "num_tokens": 105438.0,
      "step": 50
    },
    {
      "entropy": 10.587072944641113,
      "epoch": 0.00498911465892598,
      "grad_norm": 2.46875,
      "learning_rate": 2.7e-05,
      "loss": 9.4759,
      "mean_token_accuracy": 0.0453902505338192,
      "num_tokens": 115518.0,
      "step": 55
    },
    {
      "entropy": 10.536315631866454,
      "epoch": 0.00544267053701016,
      "grad_norm": 2.4375,
      "learning_rate": 2.95e-05,
      "loss": 9.377,
      "mean_token_accuracy": 0.053391102328896524,
      "num_tokens": 125636.0,
      "step": 60
    },
    {
      "entropy": 10.496469593048095,
      "epoch": 0.00589622641509434,
      "grad_norm": 2.515625,
      "learning_rate": 3.2e-05,
      "loss": 9.3108,
      "mean_token_accuracy": 0.06191379800438881,
      "num_tokens": 135990.0,
      "step": 65
    },
    {
      "entropy": 10.413052368164063,
      "epoch": 0.00634978229317852,
      "grad_norm": 2.328125,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.2425,
      "mean_token_accuracy": 0.0613830640912056,
      "num_tokens": 146069.0,
      "step": 70
    },
    {
      "entropy": 10.440055561065673,
      "epoch": 0.006803338171262699,
      "grad_norm": 2.265625,
      "learning_rate": 3.7e-05,
      "loss": 9.1547,
      "mean_token_accuracy": 0.07091366611421109,
      "num_tokens": 156335.0,
      "step": 75
    },
    {
      "entropy": 10.438591575622558,
      "epoch": 0.00725689404934688,
      "grad_norm": 2.4375,
      "learning_rate": 3.95e-05,
      "loss": 9.0463,
      "mean_token_accuracy": 0.06796186193823814,
      "num_tokens": 166214.0,
      "step": 80
    },
    {
      "entropy": 10.390478229522705,
      "epoch": 0.007710449927431059,
      "grad_norm": 3.015625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 8.9608,
      "mean_token_accuracy": 0.06832841411232948,
      "num_tokens": 177010.0,
      "step": 85
    },
    {
      "entropy": 10.332666969299316,
      "epoch": 0.00816400580551524,
      "grad_norm": 2.671875,
      "learning_rate": 4.45e-05,
      "loss": 8.7359,
      "mean_token_accuracy": 0.07582402378320693,
      "num_tokens": 187829.0,
      "step": 90
    },
    {
      "entropy": 10.192521572113037,
      "epoch": 0.00861756168359942,
      "grad_norm": 2.609375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 8.6972,
      "mean_token_accuracy": 0.0755643829703331,
      "num_tokens": 198107.0,
      "step": 95
    },
    {
      "entropy": 10.20487937927246,
      "epoch": 0.009071117561683599,
      "grad_norm": 2.15625,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 8.5722,
      "mean_token_accuracy": 0.07500871792435646,
      "num_tokens": 208607.0,
      "step": 100
    },
    {
      "entropy": 10.160844707489014,
      "epoch": 0.009524673439767779,
      "grad_norm": 2.296875,
      "learning_rate": 5.2e-05,
      "loss": 8.5622,
      "mean_token_accuracy": 0.07453125491738319,
      "num_tokens": 219899.0,
      "step": 105
    },
    {
      "entropy": 10.124257946014405,
      "epoch": 0.00997822931785196,
      "grad_norm": 1.890625,
      "learning_rate": 5.45e-05,
      "loss": 8.4797,
      "mean_token_accuracy": 0.07378747165203095,
      "num_tokens": 230625.0,
      "step": 110
    },
    {
      "entropy": 10.023534488677978,
      "epoch": 0.01043178519593614,
      "grad_norm": 2.03125,
      "learning_rate": 5.7e-05,
      "loss": 8.332,
      "mean_token_accuracy": 0.08365851938724518,
      "num_tokens": 241156.0,
      "step": 115
    },
    {
      "entropy": 9.936892032623291,
      "epoch": 0.01088534107402032,
      "grad_norm": 1.8828125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 8.1824,
      "mean_token_accuracy": 0.08069921731948852,
      "num_tokens": 252297.0,
      "step": 120
    },
    {
      "entropy": 9.793073558807373,
      "epoch": 0.011338896952104499,
      "grad_norm": 1.5390625,
      "learning_rate": 6.2e-05,
      "loss": 8.1575,
      "mean_token_accuracy": 0.07586940303444863,
      "num_tokens": 263188.0,
      "step": 125
    },
    {
      "entropy": 9.714948558807373,
      "epoch": 0.01179245283018868,
      "grad_norm": 1.6171875,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9988,
      "mean_token_accuracy": 0.07976540848612786,
      "num_tokens": 274064.0,
      "step": 130
    },
    {
      "entropy": 9.50658073425293,
      "epoch": 0.01224600870827286,
      "grad_norm": 1.90625,
      "learning_rate": 6.7e-05,
      "loss": 7.8083,
      "mean_token_accuracy": 0.08558063134551049,
      "num_tokens": 283736.0,
      "step": 135
    },
    {
      "entropy": 9.259089279174805,
      "epoch": 0.01269956458635704,
      "grad_norm": 1.640625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.7077,
      "mean_token_accuracy": 0.08600766956806183,
      "num_tokens": 294473.0,
      "step": 140
    },
    {
      "entropy": 9.041689109802245,
      "epoch": 0.013153120464441219,
      "grad_norm": 1.390625,
      "learning_rate": 7.2e-05,
      "loss": 7.6574,
      "mean_token_accuracy": 0.08418083190917969,
      "num_tokens": 303849.0,
      "step": 145
    },
    {
      "entropy": 8.879548072814941,
      "epoch": 0.013606676342525399,
      "grad_norm": 1.6015625,
      "learning_rate": 7.45e-05,
      "loss": 7.5414,
      "mean_token_accuracy": 0.08226052820682525,
      "num_tokens": 314521.0,
      "step": 150
    },
    {
      "entropy": 8.658054065704345,
      "epoch": 0.01406023222060958,
      "grad_norm": 1.46875,
      "learning_rate": 7.7e-05,
      "loss": 7.4964,
      "mean_token_accuracy": 0.08540133312344551,
      "num_tokens": 325210.0,
      "step": 155
    },
    {
      "entropy": 8.465118217468262,
      "epoch": 0.01451378809869376,
      "grad_norm": 1.4140625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.4289,
      "mean_token_accuracy": 0.0857605628669262,
      "num_tokens": 335035.0,
      "step": 160
    },
    {
      "entropy": 8.27260684967041,
      "epoch": 0.014967343976777939,
      "grad_norm": 1.3515625,
      "learning_rate": 8.2e-05,
      "loss": 7.4279,
      "mean_token_accuracy": 0.08624853491783142,
      "num_tokens": 345402.0,
      "step": 165
    },
    {
      "entropy": 8.192702579498292,
      "epoch": 0.015420899854862119,
      "grad_norm": 1.3125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.3865,
      "mean_token_accuracy": 0.0875001959502697,
      "num_tokens": 356161.0,
      "step": 170
    },
    {
      "entropy": 8.096983528137207,
      "epoch": 0.0158744557329463,
      "grad_norm": 1.5078125,
      "learning_rate": 8.7e-05,
      "loss": 7.3087,
      "mean_token_accuracy": 0.08894997537136078,
      "num_tokens": 366695.0,
      "step": 175
    },
    {
      "entropy": 8.000302648544311,
      "epoch": 0.01632801161103048,
      "grad_norm": 1.3515625,
      "learning_rate": 8.95e-05,
      "loss": 7.2459,
      "mean_token_accuracy": 0.08826502189040183,
      "num_tokens": 376354.0,
      "step": 180
    },
    {
      "entropy": 7.892614269256592,
      "epoch": 0.01678156748911466,
      "grad_norm": 1.28125,
      "learning_rate": 9.2e-05,
      "loss": 7.2853,
      "mean_token_accuracy": 0.08936820179224014,
      "num_tokens": 387468.0,
      "step": 185
    },
    {
      "entropy": 7.8675158500671385,
      "epoch": 0.01723512336719884,
      "grad_norm": 1.1875,
      "learning_rate": 9.45e-05,
      "loss": 7.2366,
      "mean_token_accuracy": 0.08572344705462456,
      "num_tokens": 397658.0,
      "step": 190
    },
    {
      "entropy": 7.800578165054321,
      "epoch": 0.01768867924528302,
      "grad_norm": 1.2109375,
      "learning_rate": 9.7e-05,
      "loss": 7.2295,
      "mean_token_accuracy": 0.08996806442737579,
      "num_tokens": 408009.0,
      "step": 195
    },
    {
      "entropy": 7.81451735496521,
      "epoch": 0.018142235123367198,
      "grad_norm": 1.8515625,
      "learning_rate": 9.95e-05,
      "loss": 7.1252,
      "mean_token_accuracy": 0.09177512675523758,
      "num_tokens": 417781.0,
      "step": 200
    },
    {
      "entropy": 7.679977512359619,
      "epoch": 0.018595791001451378,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000102,
      "loss": 7.3176,
      "mean_token_accuracy": 0.08960338979959488,
      "num_tokens": 428009.0,
      "step": 205
    },
    {
      "entropy": 7.7656045913696286,
      "epoch": 0.019049346879535557,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.1482,
      "mean_token_accuracy": 0.0907668188214302,
      "num_tokens": 438369.0,
      "step": 210
    },
    {
      "entropy": 7.7085036277771,
      "epoch": 0.01950290275761974,
      "grad_norm": 1.421875,
      "learning_rate": 0.000107,
      "loss": 7.1341,
      "mean_token_accuracy": 0.09216333702206611,
      "num_tokens": 449047.0,
      "step": 215
    },
    {
      "entropy": 7.61162919998169,
      "epoch": 0.01995645863570392,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001095,
      "loss": 7.1272,
      "mean_token_accuracy": 0.093924118578434,
      "num_tokens": 459992.0,
      "step": 220
    },
    {
      "entropy": 7.708517932891846,
      "epoch": 0.0204100145137881,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000112,
      "loss": 7.207,
      "mean_token_accuracy": 0.09182708933949471,
      "num_tokens": 470542.0,
      "step": 225
    },
    {
      "entropy": 7.576424551010132,
      "epoch": 0.02086357039187228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001145,
      "loss": 7.1577,
      "mean_token_accuracy": 0.09969761073589326,
      "num_tokens": 479983.0,
      "step": 230
    },
    {
      "entropy": 7.695700073242188,
      "epoch": 0.02131712626995646,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.1299,
      "mean_token_accuracy": 0.09077642261981964,
      "num_tokens": 489827.0,
      "step": 235
    },
    {
      "entropy": 7.514981985092163,
      "epoch": 0.02177068214804064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.1267,
      "mean_token_accuracy": 0.08924711495637894,
      "num_tokens": 500257.0,
      "step": 240
    },
    {
      "entropy": 7.638169956207276,
      "epoch": 0.022224238026124818,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000122,
      "loss": 7.1507,
      "mean_token_accuracy": 0.08717882186174393,
      "num_tokens": 510227.0,
      "step": 245
    },
    {
      "entropy": 7.506703805923462,
      "epoch": 0.022677793904208998,
      "grad_norm": 1.140625,
      "learning_rate": 0.0001245,
      "loss": 7.0492,
      "mean_token_accuracy": 0.0963970109820366,
      "num_tokens": 520442.0,
      "step": 250
    },
    {
      "entropy": 7.448557662963867,
      "epoch": 0.023131349782293177,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000127,
      "loss": 7.0689,
      "mean_token_accuracy": 0.0880157247185707,
      "num_tokens": 531075.0,
      "step": 255
    },
    {
      "entropy": 7.53116626739502,
      "epoch": 0.02358490566037736,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001295,
      "loss": 7.1308,
      "mean_token_accuracy": 0.08964502289891244,
      "num_tokens": 542576.0,
      "step": 260
    },
    {
      "entropy": 7.471751165390015,
      "epoch": 0.02403846153846154,
      "grad_norm": 1.171875,
      "learning_rate": 0.000132,
      "loss": 7.0485,
      "mean_token_accuracy": 0.09242086634039878,
      "num_tokens": 552380.0,
      "step": 265
    },
    {
      "entropy": 7.469887447357178,
      "epoch": 0.02449201741654572,
      "grad_norm": 1.28125,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.0608,
      "mean_token_accuracy": 0.08920432329177856,
      "num_tokens": 563310.0,
      "step": 270
    },
    {
      "entropy": 7.405267572402954,
      "epoch": 0.0249455732946299,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.9664,
      "mean_token_accuracy": 0.09448141977190971,
      "num_tokens": 574113.0,
      "step": 275
    },
    {
      "entropy": 7.440743923187256,
      "epoch": 0.02539912917271408,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001395,
      "loss": 6.9802,
      "mean_token_accuracy": 0.09543166309595108,
      "num_tokens": 585198.0,
      "step": 280
    },
    {
      "entropy": 7.435253047943116,
      "epoch": 0.025852685050798258,
      "grad_norm": 1.3125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.9734,
      "mean_token_accuracy": 0.0946137085556984,
      "num_tokens": 594837.0,
      "step": 285
    },
    {
      "entropy": 7.340703868865967,
      "epoch": 0.026306240928882438,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0001445,
      "loss": 7.0063,
      "mean_token_accuracy": 0.09453380405902863,
      "num_tokens": 604000.0,
      "step": 290
    },
    {
      "entropy": 7.364510869979858,
      "epoch": 0.026759796806966617,
      "grad_norm": 1.203125,
      "learning_rate": 0.000147,
      "loss": 7.0165,
      "mean_token_accuracy": 0.09476850554347038,
      "num_tokens": 614831.0,
      "step": 295
    },
    {
      "entropy": 7.43275899887085,
      "epoch": 0.027213352685050797,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0001495,
      "loss": 7.0328,
      "mean_token_accuracy": 0.09851854294538498,
      "num_tokens": 625584.0,
      "step": 300
    },
    {
      "entropy": 7.33774585723877,
      "epoch": 0.027666908563134977,
      "grad_norm": 1.328125,
      "learning_rate": 0.000152,
      "loss": 6.9117,
      "mean_token_accuracy": 0.09698478281497955,
      "num_tokens": 636282.0,
      "step": 305
    },
    {
      "entropy": 7.42357873916626,
      "epoch": 0.02812046444121916,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.0307,
      "mean_token_accuracy": 0.09482022523880004,
      "num_tokens": 646556.0,
      "step": 310
    },
    {
      "entropy": 7.372399139404297,
      "epoch": 0.02857402031930334,
      "grad_norm": 1.328125,
      "learning_rate": 0.000157,
      "loss": 6.9899,
      "mean_token_accuracy": 0.09617659002542496,
      "num_tokens": 656983.0,
      "step": 315
    },
    {
      "entropy": 7.231950092315674,
      "epoch": 0.02902757619738752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001595,
      "loss": 6.9155,
      "mean_token_accuracy": 0.1019688904285431,
      "num_tokens": 668148.0,
      "step": 320
    },
    {
      "entropy": 7.4007758617401125,
      "epoch": 0.0294811320754717,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000162,
      "loss": 7.0107,
      "mean_token_accuracy": 0.09772700667381287,
      "num_tokens": 677400.0,
      "step": 325
    },
    {
      "entropy": 7.2482020378112795,
      "epoch": 0.029934687953555878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.896,
      "mean_token_accuracy": 0.09065234810113906,
      "num_tokens": 687519.0,
      "step": 330
    },
    {
      "entropy": 7.264542865753174,
      "epoch": 0.030388243831640058,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8005,
      "mean_token_accuracy": 0.10230840593576432,
      "num_tokens": 698402.0,
      "step": 335
    },
    {
      "entropy": 7.134629440307617,
      "epoch": 0.030841799709724237,
      "grad_norm": 1.078125,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.8627,
      "mean_token_accuracy": 0.10249442160129547,
      "num_tokens": 710628.0,
      "step": 340
    },
    {
      "entropy": 7.336190032958984,
      "epoch": 0.03129535558780842,
      "grad_norm": 1.359375,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.9194,
      "mean_token_accuracy": 0.09371023699641227,
      "num_tokens": 720511.0,
      "step": 345
    },
    {
      "entropy": 7.196484518051148,
      "epoch": 0.0317489114658926,
      "grad_norm": 1.28125,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.9486,
      "mean_token_accuracy": 0.10238764733076096,
      "num_tokens": 731357.0,
      "step": 350
    },
    {
      "entropy": 7.2075560092926025,
      "epoch": 0.032202467343976776,
      "grad_norm": 1.5625,
      "learning_rate": 0.000177,
      "loss": 6.8483,
      "mean_token_accuracy": 0.1011496514081955,
      "num_tokens": 741569.0,
      "step": 355
    },
    {
      "entropy": 7.342852926254272,
      "epoch": 0.03265602322206096,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0001795,
      "loss": 6.8818,
      "mean_token_accuracy": 0.09690657183527947,
      "num_tokens": 751194.0,
      "step": 360
    },
    {
      "entropy": 7.139012861251831,
      "epoch": 0.033109579100145135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000182,
      "loss": 6.8674,
      "mean_token_accuracy": 0.0964200034737587,
      "num_tokens": 762554.0,
      "step": 365
    },
    {
      "entropy": 7.168058109283447,
      "epoch": 0.03356313497822932,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001845,
      "loss": 6.7947,
      "mean_token_accuracy": 0.1081994466483593,
      "num_tokens": 772819.0,
      "step": 370
    },
    {
      "entropy": 7.116481113433838,
      "epoch": 0.034016690856313495,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000187,
      "loss": 6.8457,
      "mean_token_accuracy": 0.09905849322676659,
      "num_tokens": 782717.0,
      "step": 375
    },
    {
      "entropy": 7.275808191299438,
      "epoch": 0.03447024673439768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0001895,
      "loss": 6.9231,
      "mean_token_accuracy": 0.10113033130764962,
      "num_tokens": 792523.0,
      "step": 380
    },
    {
      "entropy": 7.136093997955323,
      "epoch": 0.03492380261248186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000192,
      "loss": 6.852,
      "mean_token_accuracy": 0.10536141693592072,
      "num_tokens": 802696.0,
      "step": 385
    },
    {
      "entropy": 7.199086666107178,
      "epoch": 0.03537735849056604,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0001945,
      "loss": 6.8945,
      "mean_token_accuracy": 0.09893356338143348,
      "num_tokens": 813041.0,
      "step": 390
    },
    {
      "entropy": 7.261086368560791,
      "epoch": 0.03583091436865022,
      "grad_norm": 1.171875,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.8549,
      "mean_token_accuracy": 0.09995201453566552,
      "num_tokens": 823366.0,
      "step": 395
    },
    {
      "entropy": 7.149146509170532,
      "epoch": 0.036284470246734396,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.8201,
      "mean_token_accuracy": 0.100285005569458,
      "num_tokens": 835455.0,
      "step": 400
    },
    {
      "entropy": 7.098216772079468,
      "epoch": 0.03673802612481858,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000202,
      "loss": 6.8581,
      "mean_token_accuracy": 0.10088052004575729,
      "num_tokens": 846190.0,
      "step": 405
    },
    {
      "entropy": 7.215963459014892,
      "epoch": 0.037191582002902755,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.8161,
      "mean_token_accuracy": 0.10391620472073555,
      "num_tokens": 856303.0,
      "step": 410
    },
    {
      "entropy": 7.0431232929229735,
      "epoch": 0.03764513788098694,
      "grad_norm": 1.203125,
      "learning_rate": 0.000207,
      "loss": 6.7939,
      "mean_token_accuracy": 0.09926707297563553,
      "num_tokens": 866240.0,
      "step": 415
    },
    {
      "entropy": 7.138209819793701,
      "epoch": 0.038098693759071114,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0002095,
      "loss": 6.9358,
      "mean_token_accuracy": 0.09860606789588929,
      "num_tokens": 877035.0,
      "step": 420
    },
    {
      "entropy": 7.092493534088135,
      "epoch": 0.0385522496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000212,
      "loss": 6.8158,
      "mean_token_accuracy": 0.09759353399276734,
      "num_tokens": 887344.0,
      "step": 425
    },
    {
      "entropy": 7.141839790344238,
      "epoch": 0.03900580551523948,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002145,
      "loss": 6.8314,
      "mean_token_accuracy": 0.10464526489377021,
      "num_tokens": 897271.0,
      "step": 430
    },
    {
      "entropy": 7.065910148620605,
      "epoch": 0.03945936139332366,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.7285,
      "mean_token_accuracy": 0.1087069682776928,
      "num_tokens": 908700.0,
      "step": 435
    },
    {
      "entropy": 6.984343385696411,
      "epoch": 0.03991291727140784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002195,
      "loss": 6.7459,
      "mean_token_accuracy": 0.10460741296410561,
      "num_tokens": 918980.0,
      "step": 440
    },
    {
      "entropy": 7.071007776260376,
      "epoch": 0.040366473149492016,
      "grad_norm": 1.25,
      "learning_rate": 0.000222,
      "loss": 6.7444,
      "mean_token_accuracy": 0.10659246519207954,
      "num_tokens": 928616.0,
      "step": 445
    },
    {
      "entropy": 7.033933305740357,
      "epoch": 0.0408200290275762,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002245,
      "loss": 6.897,
      "mean_token_accuracy": 0.10052595734596252,
      "num_tokens": 939066.0,
      "step": 450
    },
    {
      "entropy": 7.086242055892944,
      "epoch": 0.041273584905660375,
      "grad_norm": 1.765625,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.7852,
      "mean_token_accuracy": 0.1003576584160328,
      "num_tokens": 949160.0,
      "step": 455
    },
    {
      "entropy": 7.0361896514892575,
      "epoch": 0.04172714078374456,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.8063,
      "mean_token_accuracy": 0.10417447835206986,
      "num_tokens": 959959.0,
      "step": 460
    },
    {
      "entropy": 7.061341714859009,
      "epoch": 0.042180696661828734,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.8092,
      "mean_token_accuracy": 0.10948210060596467,
      "num_tokens": 969752.0,
      "step": 465
    },
    {
      "entropy": 6.854967498779297,
      "epoch": 0.04263425253991292,
      "grad_norm": 1.25,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.8149,
      "mean_token_accuracy": 0.10138045027852058,
      "num_tokens": 979917.0,
      "step": 470
    },
    {
      "entropy": 7.039763593673706,
      "epoch": 0.0430878084179971,
      "grad_norm": 1.234375,
      "learning_rate": 0.000237,
      "loss": 6.8134,
      "mean_token_accuracy": 0.09902644529938698,
      "num_tokens": 990613.0,
      "step": 475
    },
    {
      "entropy": 7.048016119003296,
      "epoch": 0.04354136429608128,
      "grad_norm": 1.53125,
      "learning_rate": 0.0002395,
      "loss": 6.8134,
      "mean_token_accuracy": 0.10063022077083587,
      "num_tokens": 1000753.0,
      "step": 480
    },
    {
      "entropy": 6.978139591217041,
      "epoch": 0.04399492017416546,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000242,
      "loss": 6.7445,
      "mean_token_accuracy": 0.10282852351665497,
      "num_tokens": 1012141.0,
      "step": 485
    },
    {
      "entropy": 7.031801795959472,
      "epoch": 0.044448476052249636,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002445,
      "loss": 6.7459,
      "mean_token_accuracy": 0.10179351791739463,
      "num_tokens": 1023322.0,
      "step": 490
    },
    {
      "entropy": 6.850466966629028,
      "epoch": 0.04490203193033382,
      "grad_norm": 1.15625,
      "learning_rate": 0.000247,
      "loss": 6.6463,
      "mean_token_accuracy": 0.09672334268689156,
      "num_tokens": 1033949.0,
      "step": 495
    },
    {
      "entropy": 6.988512468338013,
      "epoch": 0.045355587808417995,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002495,
      "loss": 6.7366,
      "mean_token_accuracy": 0.10716779977083206,
      "num_tokens": 1043580.0,
      "step": 500
    },
    {
      "entropy": 6.902674722671509,
      "epoch": 0.04580914368650218,
      "grad_norm": 1.296875,
      "learning_rate": 0.000252,
      "loss": 6.656,
      "mean_token_accuracy": 0.11081461235880852,
      "num_tokens": 1052106.0,
      "step": 505
    },
    {
      "entropy": 6.90415506362915,
      "epoch": 0.046262699564586354,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0002545,
      "loss": 6.6971,
      "mean_token_accuracy": 0.1064011238515377,
      "num_tokens": 1061764.0,
      "step": 510
    },
    {
      "entropy": 6.838364171981811,
      "epoch": 0.04671625544267054,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000257,
      "loss": 6.5697,
      "mean_token_accuracy": 0.10854723304510117,
      "num_tokens": 1071164.0,
      "step": 515
    },
    {
      "entropy": 6.968482303619385,
      "epoch": 0.04716981132075472,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002595,
      "loss": 6.8149,
      "mean_token_accuracy": 0.10352605357766151,
      "num_tokens": 1082602.0,
      "step": 520
    },
    {
      "entropy": 6.9721253395080565,
      "epoch": 0.047623367198838897,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000262,
      "loss": 6.7378,
      "mean_token_accuracy": 0.1052236057817936,
      "num_tokens": 1091871.0,
      "step": 525
    },
    {
      "entropy": 6.919319772720337,
      "epoch": 0.04807692307692308,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.7272,
      "mean_token_accuracy": 0.10676647424697876,
      "num_tokens": 1102568.0,
      "step": 530
    },
    {
      "entropy": 6.941032934188843,
      "epoch": 0.048530478955007256,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.6203,
      "mean_token_accuracy": 0.10674264281988144,
      "num_tokens": 1112777.0,
      "step": 535
    },
    {
      "entropy": 6.716324281692505,
      "epoch": 0.04898403483309144,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6119,
      "mean_token_accuracy": 0.10510820895433426,
      "num_tokens": 1123809.0,
      "step": 540
    },
    {
      "entropy": 6.8692456722259525,
      "epoch": 0.049437590711175615,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6644,
      "mean_token_accuracy": 0.10762705281376839,
      "num_tokens": 1133793.0,
      "step": 545
    },
    {
      "entropy": 6.804552936553955,
      "epoch": 0.0498911465892598,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002745,
      "loss": 6.5543,
      "mean_token_accuracy": 0.10652651414275169,
      "num_tokens": 1144478.0,
      "step": 550
    },
    {
      "entropy": 6.81122727394104,
      "epoch": 0.050344702467343974,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000277,
      "loss": 6.595,
      "mean_token_accuracy": 0.10836011320352554,
      "num_tokens": 1155518.0,
      "step": 555
    },
    {
      "entropy": 6.848041963577271,
      "epoch": 0.05079825834542816,
      "grad_norm": 1.125,
      "learning_rate": 0.0002795,
      "loss": 6.6803,
      "mean_token_accuracy": 0.10240246802568435,
      "num_tokens": 1166494.0,
      "step": 560
    },
    {
      "entropy": 6.864998054504395,
      "epoch": 0.05125181422351233,
      "grad_norm": 1.015625,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6603,
      "mean_token_accuracy": 0.10499662458896637,
      "num_tokens": 1177052.0,
      "step": 565
    },
    {
      "entropy": 6.7566328048706055,
      "epoch": 0.051705370101596516,
      "grad_norm": 1.0,
      "learning_rate": 0.0002845,
      "loss": 6.588,
      "mean_token_accuracy": 0.1072057232260704,
      "num_tokens": 1187382.0,
      "step": 570
    },
    {
      "entropy": 6.84862585067749,
      "epoch": 0.0521589259796807,
      "grad_norm": 1.078125,
      "learning_rate": 0.000287,
      "loss": 6.6371,
      "mean_token_accuracy": 0.10280174314975739,
      "num_tokens": 1197535.0,
      "step": 575
    },
    {
      "entropy": 6.854958152770996,
      "epoch": 0.052612481857764876,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002895,
      "loss": 6.6932,
      "mean_token_accuracy": 0.11012415438890458,
      "num_tokens": 1207626.0,
      "step": 580
    },
    {
      "entropy": 6.8038904666900635,
      "epoch": 0.05306603773584906,
      "grad_norm": 1.15625,
      "learning_rate": 0.000292,
      "loss": 6.5812,
      "mean_token_accuracy": 0.10877829939126968,
      "num_tokens": 1218433.0,
      "step": 585
    },
    {
      "entropy": 6.773093318939209,
      "epoch": 0.053519593613933235,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002945,
      "loss": 6.5272,
      "mean_token_accuracy": 0.10834594070911407,
      "num_tokens": 1228698.0,
      "step": 590
    },
    {
      "entropy": 6.839606189727784,
      "epoch": 0.05397314949201742,
      "grad_norm": 1.078125,
      "learning_rate": 0.000297,
      "loss": 6.6608,
      "mean_token_accuracy": 0.11245888248085975,
      "num_tokens": 1239278.0,
      "step": 595
    },
    {
      "entropy": 6.663392496109009,
      "epoch": 0.054426705370101594,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002995,
      "loss": 6.6276,
      "mean_token_accuracy": 0.11012510806322098,
      "num_tokens": 1248976.0,
      "step": 600
    },
    {
      "entropy": 6.770804977416992,
      "epoch": 0.05488026124818578,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000302,
      "loss": 6.5805,
      "mean_token_accuracy": 0.11286153569817543,
      "num_tokens": 1259545.0,
      "step": 605
    },
    {
      "entropy": 6.810527324676514,
      "epoch": 0.05533381712626995,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003045,
      "loss": 6.6029,
      "mean_token_accuracy": 0.10805755332112313,
      "num_tokens": 1269444.0,
      "step": 610
    },
    {
      "entropy": 6.8169921875,
      "epoch": 0.055787373004354136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000307,
      "loss": 6.6931,
      "mean_token_accuracy": 0.10932271853089333,
      "num_tokens": 1279821.0,
      "step": 615
    },
    {
      "entropy": 6.654322671890259,
      "epoch": 0.05624092888243832,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003095,
      "loss": 6.5238,
      "mean_token_accuracy": 0.10867622718214989,
      "num_tokens": 1290398.0,
      "step": 620
    },
    {
      "entropy": 6.814783906936645,
      "epoch": 0.056694484760522496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000312,
      "loss": 6.6259,
      "mean_token_accuracy": 0.10474204868078232,
      "num_tokens": 1300920.0,
      "step": 625
    },
    {
      "entropy": 6.690603113174438,
      "epoch": 0.05714804063860668,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003145,
      "loss": 6.6062,
      "mean_token_accuracy": 0.10462497025728226,
      "num_tokens": 1311928.0,
      "step": 630
    },
    {
      "entropy": 6.83460955619812,
      "epoch": 0.057601596516690855,
      "grad_norm": 1.21875,
      "learning_rate": 0.000317,
      "loss": 6.6113,
      "mean_token_accuracy": 0.10693813562393188,
      "num_tokens": 1322554.0,
      "step": 635
    },
    {
      "entropy": 6.699485635757446,
      "epoch": 0.05805515239477504,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0003195,
      "loss": 6.5474,
      "mean_token_accuracy": 0.11296007037162781,
      "num_tokens": 1333195.0,
      "step": 640
    },
    {
      "entropy": 6.70126633644104,
      "epoch": 0.058508708272859214,
      "grad_norm": 1.046875,
      "learning_rate": 0.000322,
      "loss": 6.6557,
      "mean_token_accuracy": 0.10681698620319366,
      "num_tokens": 1343655.0,
      "step": 645
    },
    {
      "entropy": 6.784696960449219,
      "epoch": 0.0589622641509434,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5197,
      "mean_token_accuracy": 0.11125476583838463,
      "num_tokens": 1352528.0,
      "step": 650
    },
    {
      "entropy": 6.562724494934082,
      "epoch": 0.05941582002902757,
      "grad_norm": 1.015625,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.5629,
      "mean_token_accuracy": 0.11251123249530792,
      "num_tokens": 1363143.0,
      "step": 655
    },
    {
      "entropy": 6.857563257217407,
      "epoch": 0.059869375907111756,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.578,
      "mean_token_accuracy": 0.1062569722533226,
      "num_tokens": 1373049.0,
      "step": 660
    },
    {
      "entropy": 6.634031963348389,
      "epoch": 0.06032293178519594,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.5634,
      "mean_token_accuracy": 0.11504042670130729,
      "num_tokens": 1384040.0,
      "step": 665
    },
    {
      "entropy": 6.730848979949951,
      "epoch": 0.060776487663280115,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.6586,
      "mean_token_accuracy": 0.11258623376488686,
      "num_tokens": 1394511.0,
      "step": 670
    },
    {
      "entropy": 6.641570663452148,
      "epoch": 0.0612300435413643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000337,
      "loss": 6.4682,
      "mean_token_accuracy": 0.11615501791238785,
      "num_tokens": 1405381.0,
      "step": 675
    },
    {
      "entropy": 6.756093549728393,
      "epoch": 0.061683599419448475,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003395,
      "loss": 6.6404,
      "mean_token_accuracy": 0.11007921248674393,
      "num_tokens": 1415840.0,
      "step": 680
    },
    {
      "entropy": 6.7638633251190186,
      "epoch": 0.06213715529753266,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000342,
      "loss": 6.6825,
      "mean_token_accuracy": 0.10886021852493286,
      "num_tokens": 1426634.0,
      "step": 685
    },
    {
      "entropy": 6.609492397308349,
      "epoch": 0.06259071117561683,
      "grad_norm": 1.125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.5622,
      "mean_token_accuracy": 0.11168285235762596,
      "num_tokens": 1436794.0,
      "step": 690
    },
    {
      "entropy": 6.853395700454712,
      "epoch": 0.06304426705370102,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000347,
      "loss": 6.6395,
      "mean_token_accuracy": 0.11344821155071258,
      "num_tokens": 1447114.0,
      "step": 695
    },
    {
      "entropy": 6.582093334197998,
      "epoch": 0.0634978229317852,
      "grad_norm": 1.25,
      "learning_rate": 0.0003495,
      "loss": 6.4899,
      "mean_token_accuracy": 0.11673147901892662,
      "num_tokens": 1456494.0,
      "step": 700
    },
    {
      "entropy": 6.599524879455567,
      "epoch": 0.06395137880986937,
      "grad_norm": 1.171875,
      "learning_rate": 0.000352,
      "loss": 6.4935,
      "mean_token_accuracy": 0.11297031342983246,
      "num_tokens": 1466580.0,
      "step": 705
    },
    {
      "entropy": 6.699067687988281,
      "epoch": 0.06440493468795355,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003545,
      "loss": 6.5999,
      "mean_token_accuracy": 0.10977515503764153,
      "num_tokens": 1476854.0,
      "step": 710
    },
    {
      "entropy": 6.683993911743164,
      "epoch": 0.06485849056603774,
      "grad_norm": 1.078125,
      "learning_rate": 0.000357,
      "loss": 6.5426,
      "mean_token_accuracy": 0.10898664593696594,
      "num_tokens": 1487133.0,
      "step": 715
    },
    {
      "entropy": 6.6332762241363525,
      "epoch": 0.06531204644412192,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003595,
      "loss": 6.563,
      "mean_token_accuracy": 0.11930395513772965,
      "num_tokens": 1496974.0,
      "step": 720
    },
    {
      "entropy": 6.615490865707398,
      "epoch": 0.0657656023222061,
      "grad_norm": 1.0,
      "learning_rate": 0.000362,
      "loss": 6.5058,
      "mean_token_accuracy": 0.11035025268793106,
      "num_tokens": 1507665.0,
      "step": 725
    },
    {
      "entropy": 6.698334789276123,
      "epoch": 0.06621915820029027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003645,
      "loss": 6.4698,
      "mean_token_accuracy": 0.11344846114516258,
      "num_tokens": 1518625.0,
      "step": 730
    },
    {
      "entropy": 6.690861082077026,
      "epoch": 0.06667271407837445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000367,
      "loss": 6.608,
      "mean_token_accuracy": 0.10640608072280884,
      "num_tokens": 1529184.0,
      "step": 735
    },
    {
      "entropy": 6.481381893157959,
      "epoch": 0.06712626995645864,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003695,
      "loss": 6.526,
      "mean_token_accuracy": 0.10988623276352882,
      "num_tokens": 1540130.0,
      "step": 740
    },
    {
      "entropy": 6.694868326187134,
      "epoch": 0.06757982583454282,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000372,
      "loss": 6.5181,
      "mean_token_accuracy": 0.11159003973007202,
      "num_tokens": 1550161.0,
      "step": 745
    },
    {
      "entropy": 6.574037361145019,
      "epoch": 0.06803338171262699,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003745,
      "loss": 6.4826,
      "mean_token_accuracy": 0.11772774383425713,
      "num_tokens": 1560076.0,
      "step": 750
    },
    {
      "entropy": 6.67036771774292,
      "epoch": 0.06848693759071117,
      "grad_norm": 1.140625,
      "learning_rate": 0.000377,
      "loss": 6.4693,
      "mean_token_accuracy": 0.11172557696700096,
      "num_tokens": 1570882.0,
      "step": 755
    },
    {
      "entropy": 6.481294822692871,
      "epoch": 0.06894049346879536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003795,
      "loss": 6.4696,
      "mean_token_accuracy": 0.11403056010603904,
      "num_tokens": 1582227.0,
      "step": 760
    },
    {
      "entropy": 6.62347674369812,
      "epoch": 0.06939404934687954,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000382,
      "loss": 6.5213,
      "mean_token_accuracy": 0.11278930827975273,
      "num_tokens": 1592691.0,
      "step": 765
    },
    {
      "entropy": 6.617785978317261,
      "epoch": 0.06984760522496372,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003845,
      "loss": 6.4965,
      "mean_token_accuracy": 0.11648303642868996,
      "num_tokens": 1602258.0,
      "step": 770
    },
    {
      "entropy": 6.573126649856567,
      "epoch": 0.07030116110304789,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.5672,
      "mean_token_accuracy": 0.11526603251695633,
      "num_tokens": 1611953.0,
      "step": 775
    },
    {
      "entropy": 6.6316383361816404,
      "epoch": 0.07075471698113207,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5492,
      "mean_token_accuracy": 0.11260883659124374,
      "num_tokens": 1621944.0,
      "step": 780
    },
    {
      "entropy": 6.496122121810913,
      "epoch": 0.07120827285921626,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4659,
      "mean_token_accuracy": 0.11800124645233154,
      "num_tokens": 1632702.0,
      "step": 785
    },
    {
      "entropy": 6.6150144100189205,
      "epoch": 0.07166182873730044,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.5214,
      "mean_token_accuracy": 0.11103227436542511,
      "num_tokens": 1642622.0,
      "step": 790
    },
    {
      "entropy": 6.574394226074219,
      "epoch": 0.07211538461538461,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.4592,
      "mean_token_accuracy": 0.11586317420005798,
      "num_tokens": 1652442.0,
      "step": 795
    },
    {
      "entropy": 6.473141813278199,
      "epoch": 0.07256894049346879,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995,
      "loss": 6.4886,
      "mean_token_accuracy": 0.11505223363637924,
      "num_tokens": 1663022.0,
      "step": 800
    },
    {
      "entropy": 6.656652927398682,
      "epoch": 0.07302249637155298,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000402,
      "loss": 6.5181,
      "mean_token_accuracy": 0.11541026830673218,
      "num_tokens": 1673825.0,
      "step": 805
    },
    {
      "entropy": 6.590756893157959,
      "epoch": 0.07347605224963716,
      "grad_norm": 1.25,
      "learning_rate": 0.0004045,
      "loss": 6.4718,
      "mean_token_accuracy": 0.11257318034768105,
      "num_tokens": 1683783.0,
      "step": 810
    },
    {
      "entropy": 6.504452848434449,
      "epoch": 0.07392960812772134,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.5008,
      "mean_token_accuracy": 0.1076662614941597,
      "num_tokens": 1694456.0,
      "step": 815
    },
    {
      "entropy": 6.454553031921387,
      "epoch": 0.07438316400580551,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004095,
      "loss": 6.5053,
      "mean_token_accuracy": 0.11051234900951386,
      "num_tokens": 1706108.0,
      "step": 820
    },
    {
      "entropy": 6.532086324691773,
      "epoch": 0.0748367198838897,
      "grad_norm": 1.0625,
      "learning_rate": 0.000412,
      "loss": 6.3786,
      "mean_token_accuracy": 0.1280963383615017,
      "num_tokens": 1716273.0,
      "step": 825
    },
    {
      "entropy": 6.5029052734375,
      "epoch": 0.07529027576197388,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004145,
      "loss": 6.4237,
      "mean_token_accuracy": 0.11734267845749854,
      "num_tokens": 1727271.0,
      "step": 830
    },
    {
      "entropy": 6.529041481018067,
      "epoch": 0.07574383164005806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000417,
      "loss": 6.4755,
      "mean_token_accuracy": 0.1184304565191269,
      "num_tokens": 1737198.0,
      "step": 835
    },
    {
      "entropy": 6.576714324951172,
      "epoch": 0.07619738751814223,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004195,
      "loss": 6.5197,
      "mean_token_accuracy": 0.11203703507781029,
      "num_tokens": 1747077.0,
      "step": 840
    },
    {
      "entropy": 6.594869232177734,
      "epoch": 0.07665094339622641,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000422,
      "loss": 6.5119,
      "mean_token_accuracy": 0.1193366564810276,
      "num_tokens": 1757626.0,
      "step": 845
    },
    {
      "entropy": 6.527251434326172,
      "epoch": 0.0771044992743106,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004245,
      "loss": 6.5454,
      "mean_token_accuracy": 0.11205081716179847,
      "num_tokens": 1768431.0,
      "step": 850
    },
    {
      "entropy": 6.592037868499756,
      "epoch": 0.07755805515239478,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000427,
      "loss": 6.4819,
      "mean_token_accuracy": 0.11357785761356354,
      "num_tokens": 1779071.0,
      "step": 855
    },
    {
      "entropy": 6.481006050109864,
      "epoch": 0.07801161103047896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004295,
      "loss": 6.504,
      "mean_token_accuracy": 0.11291966959834099,
      "num_tokens": 1788905.0,
      "step": 860
    },
    {
      "entropy": 6.435082483291626,
      "epoch": 0.07846516690856313,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000432,
      "loss": 6.3973,
      "mean_token_accuracy": 0.12037285566329955,
      "num_tokens": 1799003.0,
      "step": 865
    },
    {
      "entropy": 6.53296480178833,
      "epoch": 0.07891872278664731,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004345,
      "loss": 6.436,
      "mean_token_accuracy": 0.11971265748143196,
      "num_tokens": 1809262.0,
      "step": 870
    },
    {
      "entropy": 6.4183008670806885,
      "epoch": 0.0793722786647315,
      "grad_norm": 0.90625,
      "learning_rate": 0.000437,
      "loss": 6.4654,
      "mean_token_accuracy": 0.11073716729879379,
      "num_tokens": 1821059.0,
      "step": 875
    },
    {
      "entropy": 6.492483615875244,
      "epoch": 0.07982583454281568,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004395,
      "loss": 6.3791,
      "mean_token_accuracy": 0.11755654066801072,
      "num_tokens": 1830246.0,
      "step": 880
    },
    {
      "entropy": 6.335828971862793,
      "epoch": 0.08027939042089985,
      "grad_norm": 1.328125,
      "learning_rate": 0.000442,
      "loss": 6.408,
      "mean_token_accuracy": 0.12151500731706619,
      "num_tokens": 1839950.0,
      "step": 885
    },
    {
      "entropy": 6.464663171768189,
      "epoch": 0.08073294629898403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004445,
      "loss": 6.3876,
      "mean_token_accuracy": 0.11592675521969795,
      "num_tokens": 1852026.0,
      "step": 890
    },
    {
      "entropy": 6.560555934906006,
      "epoch": 0.08118650217706821,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447,
      "loss": 6.4635,
      "mean_token_accuracy": 0.11812743693590164,
      "num_tokens": 1861672.0,
      "step": 895
    },
    {
      "entropy": 6.551700782775879,
      "epoch": 0.0816400580551524,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.4836,
      "mean_token_accuracy": 0.11402211338281631,
      "num_tokens": 1872294.0,
      "step": 900
    },
    {
      "entropy": 6.45425329208374,
      "epoch": 0.08209361393323658,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4414,
      "mean_token_accuracy": 0.11440050825476647,
      "num_tokens": 1883049.0,
      "step": 905
    },
    {
      "entropy": 6.345271444320678,
      "epoch": 0.08254716981132075,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.3992,
      "mean_token_accuracy": 0.12349174469709397,
      "num_tokens": 1893026.0,
      "step": 910
    },
    {
      "entropy": 6.415435361862182,
      "epoch": 0.08300072568940493,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.4304,
      "mean_token_accuracy": 0.11467023715376853,
      "num_tokens": 1903987.0,
      "step": 915
    },
    {
      "entropy": 6.59291787147522,
      "epoch": 0.08345428156748912,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4365,
      "mean_token_accuracy": 0.11061537489295006,
      "num_tokens": 1914285.0,
      "step": 920
    },
    {
      "entropy": 6.461905241012573,
      "epoch": 0.0839078374455733,
      "grad_norm": 1.21875,
      "learning_rate": 0.000462,
      "loss": 6.4536,
      "mean_token_accuracy": 0.1159026212990284,
      "num_tokens": 1924841.0,
      "step": 925
    },
    {
      "entropy": 6.357869672775268,
      "epoch": 0.08436139332365747,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004645,
      "loss": 6.3829,
      "mean_token_accuracy": 0.11473607048392295,
      "num_tokens": 1935536.0,
      "step": 930
    },
    {
      "entropy": 6.538698625564575,
      "epoch": 0.08481494920174165,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000467,
      "loss": 6.4852,
      "mean_token_accuracy": 0.11677240207791328,
      "num_tokens": 1946179.0,
      "step": 935
    },
    {
      "entropy": 6.407983589172363,
      "epoch": 0.08526850507982583,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004695,
      "loss": 6.383,
      "mean_token_accuracy": 0.11190493553876876,
      "num_tokens": 1956139.0,
      "step": 940
    },
    {
      "entropy": 6.46036171913147,
      "epoch": 0.08572206095791002,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000472,
      "loss": 6.4659,
      "mean_token_accuracy": 0.11674941927194596,
      "num_tokens": 1966944.0,
      "step": 945
    },
    {
      "entropy": 6.424592447280884,
      "epoch": 0.0861756168359942,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004745,
      "loss": 6.4265,
      "mean_token_accuracy": 0.1166701041162014,
      "num_tokens": 1976910.0,
      "step": 950
    },
    {
      "entropy": 6.57973198890686,
      "epoch": 0.08662917271407837,
      "grad_norm": 1.171875,
      "learning_rate": 0.000477,
      "loss": 6.4712,
      "mean_token_accuracy": 0.11671740487217903,
      "num_tokens": 1987111.0,
      "step": 955
    },
    {
      "entropy": 6.31265082359314,
      "epoch": 0.08708272859216255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004795,
      "loss": 6.3582,
      "mean_token_accuracy": 0.12392476126551628,
      "num_tokens": 1996962.0,
      "step": 960
    },
    {
      "entropy": 6.368039894104004,
      "epoch": 0.08753628447024674,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000482,
      "loss": 6.2983,
      "mean_token_accuracy": 0.11986770704388619,
      "num_tokens": 2006775.0,
      "step": 965
    },
    {
      "entropy": 6.482371377944946,
      "epoch": 0.08798984034833092,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004845,
      "loss": 6.4064,
      "mean_token_accuracy": 0.12042228356003762,
      "num_tokens": 2016453.0,
      "step": 970
    },
    {
      "entropy": 6.435760164260865,
      "epoch": 0.08844339622641509,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487,
      "loss": 6.4685,
      "mean_token_accuracy": 0.11267957463860512,
      "num_tokens": 2029100.0,
      "step": 975
    },
    {
      "entropy": 6.409471607208252,
      "epoch": 0.08889695210449927,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004895,
      "loss": 6.4124,
      "mean_token_accuracy": 0.11172624826431274,
      "num_tokens": 2040041.0,
      "step": 980
    },
    {
      "entropy": 6.419856071472168,
      "epoch": 0.08935050798258345,
      "grad_norm": 1.0625,
      "learning_rate": 0.000492,
      "loss": 6.3962,
      "mean_token_accuracy": 0.11573249772191048,
      "num_tokens": 2050377.0,
      "step": 985
    },
    {
      "entropy": 6.359882593154907,
      "epoch": 0.08980406386066764,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004945,
      "loss": 6.3743,
      "mean_token_accuracy": 0.1218277484178543,
      "num_tokens": 2060081.0,
      "step": 990
    },
    {
      "entropy": 6.496881818771362,
      "epoch": 0.09025761973875182,
      "grad_norm": 1.0625,
      "learning_rate": 0.000497,
      "loss": 6.5431,
      "mean_token_accuracy": 0.1156518079340458,
      "num_tokens": 2070335.0,
      "step": 995
    },
    {
      "entropy": 6.389731550216675,
      "epoch": 0.09071117561683599,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995,
      "loss": 6.3597,
      "mean_token_accuracy": 0.1212635301053524,
      "num_tokens": 2079259.0,
      "step": 1000
    },
    {
      "entropy": 6.374619245529175,
      "epoch": 0.09116473149492017,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999999985110217,
      "loss": 6.4764,
      "mean_token_accuracy": 0.11514603495597839,
      "num_tokens": 2089256.0,
      "step": 1005
    },
    {
      "entropy": 6.3883942604064945,
      "epoch": 0.09161828737300436,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999999924620471,
      "loss": 6.4064,
      "mean_token_accuracy": 0.11673571988940239,
      "num_tokens": 2099141.0,
      "step": 1010
    },
    {
      "entropy": 6.387277984619141,
      "epoch": 0.09207184325108854,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999999817600155,
      "loss": 6.3246,
      "mean_token_accuracy": 0.12680575177073478,
      "num_tokens": 2108934.0,
      "step": 1015
    },
    {
      "entropy": 6.33391375541687,
      "epoch": 0.09252539912917271,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999999664049268,
      "loss": 6.4441,
      "mean_token_accuracy": 0.1187924824655056,
      "num_tokens": 2120302.0,
      "step": 1020
    },
    {
      "entropy": 6.453229522705078,
      "epoch": 0.09297895500725689,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999463967816,
      "loss": 6.4247,
      "mean_token_accuracy": 0.1195537157356739,
      "num_tokens": 2131096.0,
      "step": 1025
    },
    {
      "entropy": 6.416885185241699,
      "epoch": 0.09343251088534107,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00049999992173558,
      "loss": 6.4041,
      "mean_token_accuracy": 0.12701962441205977,
      "num_tokens": 2140603.0,
      "step": 1030
    },
    {
      "entropy": 6.377317237854004,
      "epoch": 0.09388606676342526,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999998924213228,
      "loss": 6.4604,
      "mean_token_accuracy": 0.11771498769521713,
      "num_tokens": 2151150.0,
      "step": 1035
    },
    {
      "entropy": 6.381362867355347,
      "epoch": 0.09433962264150944,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999998584540105,
      "loss": 6.3849,
      "mean_token_accuracy": 0.11927906274795533,
      "num_tokens": 2161237.0,
      "step": 1040
    },
    {
      "entropy": 6.435468864440918,
      "epoch": 0.09479317851959361,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999998198336437,
      "loss": 6.39,
      "mean_token_accuracy": 0.11034503877162934,
      "num_tokens": 2171861.0,
      "step": 1045
    },
    {
      "entropy": 6.281571960449218,
      "epoch": 0.09524673439767779,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999997765602233,
      "loss": 6.3031,
      "mean_token_accuracy": 0.12425279021263122,
      "num_tokens": 2181415.0,
      "step": 1050
    },
    {
      "entropy": 6.323150634765625,
      "epoch": 0.09570029027576198,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999997286337502,
      "loss": 6.3244,
      "mean_token_accuracy": 0.1208683580160141,
      "num_tokens": 2192036.0,
      "step": 1055
    },
    {
      "entropy": 6.358468723297119,
      "epoch": 0.09615384615384616,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999996760542254,
      "loss": 6.4465,
      "mean_token_accuracy": 0.1163475088775158,
      "num_tokens": 2203001.0,
      "step": 1060
    },
    {
      "entropy": 6.384481906890869,
      "epoch": 0.09660740203193033,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00049999961882165,
      "loss": 6.2868,
      "mean_token_accuracy": 0.1244203880429268,
      "num_tokens": 2213310.0,
      "step": 1065
    },
    {
      "entropy": 6.250124216079712,
      "epoch": 0.09706095791001451,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999995569360251,
      "loss": 6.2546,
      "mean_token_accuracy": 0.12180497273802757,
      "num_tokens": 2223909.0,
      "step": 1070
    },
    {
      "entropy": 6.361983823776245,
      "epoch": 0.0975145137880987,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499999490397352,
      "loss": 6.3172,
      "mean_token_accuracy": 0.11828529611229896,
      "num_tokens": 2234547.0,
      "step": 1075
    },
    {
      "entropy": 6.273873519897461,
      "epoch": 0.09796806966618288,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999994192056321,
      "loss": 6.3341,
      "mean_token_accuracy": 0.12068448886275292,
      "num_tokens": 2244493.0,
      "step": 1080
    },
    {
      "entropy": 6.257021093368531,
      "epoch": 0.09842162554426705,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999993433608669,
      "loss": 6.236,
      "mean_token_accuracy": 0.1235893614590168,
      "num_tokens": 2254881.0,
      "step": 1085
    },
    {
      "entropy": 6.276112365722656,
      "epoch": 0.09887518142235123,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999992628630579,
      "loss": 6.3323,
      "mean_token_accuracy": 0.11811816319823265,
      "num_tokens": 2264964.0,
      "step": 1090
    },
    {
      "entropy": 6.486545896530151,
      "epoch": 0.09932873730043541,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999991777122069,
      "loss": 6.4318,
      "mean_token_accuracy": 0.11846130490303039,
      "num_tokens": 2275507.0,
      "step": 1095
    },
    {
      "entropy": 6.27265567779541,
      "epoch": 0.0997822931785196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999990879083156,
      "loss": 6.2819,
      "mean_token_accuracy": 0.12145883291959762,
      "num_tokens": 2284405.0,
      "step": 1100
    },
    {
      "entropy": 6.259899759292603,
      "epoch": 0.10023584905660378,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999989934513857,
      "loss": 6.324,
      "mean_token_accuracy": 0.1184744507074356,
      "num_tokens": 2294082.0,
      "step": 1105
    },
    {
      "entropy": 6.317243003845215,
      "epoch": 0.10068940493468795,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999988943414193,
      "loss": 6.3617,
      "mean_token_accuracy": 0.11865237578749657,
      "num_tokens": 2303805.0,
      "step": 1110
    },
    {
      "entropy": 6.3640618324279785,
      "epoch": 0.10114296081277213,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999987905784184,
      "loss": 6.2915,
      "mean_token_accuracy": 0.11877703964710236,
      "num_tokens": 2314974.0,
      "step": 1115
    },
    {
      "entropy": 6.386012029647827,
      "epoch": 0.10159651669085631,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999986821623853,
      "loss": 6.4138,
      "mean_token_accuracy": 0.1207954816520214,
      "num_tokens": 2326719.0,
      "step": 1120
    },
    {
      "entropy": 6.265592050552368,
      "epoch": 0.1020500725689405,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999985690933219,
      "loss": 6.3131,
      "mean_token_accuracy": 0.11659505292773246,
      "num_tokens": 2337379.0,
      "step": 1125
    },
    {
      "entropy": 6.320196676254272,
      "epoch": 0.10250362844702467,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999984513712311,
      "loss": 6.2983,
      "mean_token_accuracy": 0.12314697355031967,
      "num_tokens": 2347509.0,
      "step": 1130
    },
    {
      "entropy": 6.3434178829193115,
      "epoch": 0.10295718432510885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999983289961146,
      "loss": 6.2882,
      "mean_token_accuracy": 0.11846850141882896,
      "num_tokens": 2357835.0,
      "step": 1135
    },
    {
      "entropy": 6.269227457046509,
      "epoch": 0.10341074020319303,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999982019679755,
      "loss": 6.3579,
      "mean_token_accuracy": 0.12629576474428178,
      "num_tokens": 2368373.0,
      "step": 1140
    },
    {
      "entropy": 6.270691156387329,
      "epoch": 0.10386429608127722,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999980702868164,
      "loss": 6.3287,
      "mean_token_accuracy": 0.12598486691713334,
      "num_tokens": 2379170.0,
      "step": 1145
    },
    {
      "entropy": 6.256811571121216,
      "epoch": 0.1043178519593614,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999979339526396,
      "loss": 6.2471,
      "mean_token_accuracy": 0.12656034380197526,
      "num_tokens": 2389711.0,
      "step": 1150
    },
    {
      "entropy": 6.2689375400543215,
      "epoch": 0.10477140783744557,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999977929654484,
      "loss": 6.2285,
      "mean_token_accuracy": 0.13038562908768653,
      "num_tokens": 2399710.0,
      "step": 1155
    },
    {
      "entropy": 6.235695457458496,
      "epoch": 0.10522496371552975,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999976473252453,
      "loss": 6.3514,
      "mean_token_accuracy": 0.11452672109007836,
      "num_tokens": 2410567.0,
      "step": 1160
    },
    {
      "entropy": 6.344023704528809,
      "epoch": 0.10567851959361393,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999974970320338,
      "loss": 6.307,
      "mean_token_accuracy": 0.12359435632824897,
      "num_tokens": 2420010.0,
      "step": 1165
    },
    {
      "entropy": 6.309238719940185,
      "epoch": 0.10613207547169812,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999973420858165,
      "loss": 6.2661,
      "mean_token_accuracy": 0.1272447846829891,
      "num_tokens": 2429706.0,
      "step": 1170
    },
    {
      "entropy": 6.296723413467407,
      "epoch": 0.10658563134978229,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999971824865968,
      "loss": 6.4111,
      "mean_token_accuracy": 0.11686691716313362,
      "num_tokens": 2441482.0,
      "step": 1175
    },
    {
      "entropy": 6.422159862518311,
      "epoch": 0.10703918722786647,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999970182343782,
      "loss": 6.3682,
      "mean_token_accuracy": 0.11550676226615905,
      "num_tokens": 2452094.0,
      "step": 1180
    },
    {
      "entropy": 6.148933982849121,
      "epoch": 0.10749274310595065,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999968493291638,
      "loss": 6.1916,
      "mean_token_accuracy": 0.12536179572343825,
      "num_tokens": 2462174.0,
      "step": 1185
    },
    {
      "entropy": 6.300602436065674,
      "epoch": 0.10794629898403484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999966757709573,
      "loss": 6.2581,
      "mean_token_accuracy": 0.1321061834692955,
      "num_tokens": 2473294.0,
      "step": 1190
    },
    {
      "entropy": 6.228463363647461,
      "epoch": 0.10839985486211902,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999964975597622,
      "loss": 6.3172,
      "mean_token_accuracy": 0.1278726689517498,
      "num_tokens": 2482954.0,
      "step": 1195
    },
    {
      "entropy": 6.31824312210083,
      "epoch": 0.10885341074020319,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999963146955821,
      "loss": 6.3121,
      "mean_token_accuracy": 0.12523070126771926,
      "num_tokens": 2493894.0,
      "step": 1200
    },
    {
      "entropy": 6.202036809921265,
      "epoch": 0.10930696661828737,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499996127178421,
      "loss": 6.2219,
      "mean_token_accuracy": 0.12660781443119049,
      "num_tokens": 2503983.0,
      "step": 1205
    },
    {
      "entropy": 6.2187714099884035,
      "epoch": 0.10976052249637155,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999959350082825,
      "loss": 6.1808,
      "mean_token_accuracy": 0.12330685779452324,
      "num_tokens": 2513721.0,
      "step": 1210
    },
    {
      "entropy": 6.269692087173462,
      "epoch": 0.11021407837445574,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999957381851709,
      "loss": 6.2502,
      "mean_token_accuracy": 0.12575069963932037,
      "num_tokens": 2523312.0,
      "step": 1215
    },
    {
      "entropy": 6.2597119331359865,
      "epoch": 0.1106676342525399,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999955367090901,
      "loss": 6.1965,
      "mean_token_accuracy": 0.12818494513630868,
      "num_tokens": 2532685.0,
      "step": 1220
    },
    {
      "entropy": 6.278915119171143,
      "epoch": 0.11112119013062409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999953305800441,
      "loss": 6.3135,
      "mean_token_accuracy": 0.12152358293533325,
      "num_tokens": 2543324.0,
      "step": 1225
    },
    {
      "entropy": 6.2923393726348875,
      "epoch": 0.11157474600870827,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999951197980374,
      "loss": 6.2963,
      "mean_token_accuracy": 0.12113744467496872,
      "num_tokens": 2554242.0,
      "step": 1230
    },
    {
      "entropy": 6.273481845855713,
      "epoch": 0.11202830188679246,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999949043630744,
      "loss": 6.2467,
      "mean_token_accuracy": 0.1278710223734379,
      "num_tokens": 2563372.0,
      "step": 1235
    },
    {
      "entropy": 6.234022188186645,
      "epoch": 0.11248185776487664,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999946842751593,
      "loss": 6.3483,
      "mean_token_accuracy": 0.12514663413167,
      "num_tokens": 2574029.0,
      "step": 1240
    },
    {
      "entropy": 6.359460401535034,
      "epoch": 0.11293541364296081,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999944595342968,
      "loss": 6.3691,
      "mean_token_accuracy": 0.1290361575782299,
      "num_tokens": 2583308.0,
      "step": 1245
    },
    {
      "entropy": 6.420801258087158,
      "epoch": 0.11338896952104499,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999942301404916,
      "loss": 6.3833,
      "mean_token_accuracy": 0.11551335826516151,
      "num_tokens": 2595105.0,
      "step": 1250
    },
    {
      "entropy": 6.13080940246582,
      "epoch": 0.11384252539912917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999939960937484,
      "loss": 6.1801,
      "mean_token_accuracy": 0.13008004054427147,
      "num_tokens": 2604937.0,
      "step": 1255
    },
    {
      "entropy": 6.238644075393677,
      "epoch": 0.11429608127721336,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999937573940721,
      "loss": 6.259,
      "mean_token_accuracy": 0.12672679945826532,
      "num_tokens": 2615218.0,
      "step": 1260
    },
    {
      "entropy": 6.200268936157227,
      "epoch": 0.11474963715529753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999935140414674,
      "loss": 6.1979,
      "mean_token_accuracy": 0.12415119707584381,
      "num_tokens": 2625668.0,
      "step": 1265
    },
    {
      "entropy": 6.2111094951629635,
      "epoch": 0.11520319303338171,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999932660359395,
      "loss": 6.2862,
      "mean_token_accuracy": 0.12427769452333451,
      "num_tokens": 2636233.0,
      "step": 1270
    },
    {
      "entropy": 6.263781785964966,
      "epoch": 0.11565674891146589,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999930133774936,
      "loss": 6.2352,
      "mean_token_accuracy": 0.12104817926883697,
      "num_tokens": 2647725.0,
      "step": 1275
    },
    {
      "entropy": 6.226747751235962,
      "epoch": 0.11611030478955008,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999927560661348,
      "loss": 6.2223,
      "mean_token_accuracy": 0.13130831718444824,
      "num_tokens": 2658670.0,
      "step": 1280
    },
    {
      "entropy": 6.139042472839355,
      "epoch": 0.11656386066763426,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999924941018685,
      "loss": 6.2077,
      "mean_token_accuracy": 0.13245302811264992,
      "num_tokens": 2668782.0,
      "step": 1285
    },
    {
      "entropy": 6.23336877822876,
      "epoch": 0.11701741654571843,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999922274847,
      "loss": 6.2498,
      "mean_token_accuracy": 0.12909818440675735,
      "num_tokens": 2677973.0,
      "step": 1290
    },
    {
      "entropy": 6.244420528411865,
      "epoch": 0.11747097242380261,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499991956214635,
      "loss": 6.1934,
      "mean_token_accuracy": 0.12651574686169625,
      "num_tokens": 2687813.0,
      "step": 1295
    },
    {
      "entropy": 6.160448169708252,
      "epoch": 0.1179245283018868,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499991680291679,
      "loss": 6.2207,
      "mean_token_accuracy": 0.1258676081895828,
      "num_tokens": 2698664.0,
      "step": 1300
    },
    {
      "entropy": 6.191532325744629,
      "epoch": 0.11837808417997098,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999913997158378,
      "loss": 6.1572,
      "mean_token_accuracy": 0.13130440413951874,
      "num_tokens": 2708936.0,
      "step": 1305
    },
    {
      "entropy": 6.242334794998169,
      "epoch": 0.11883164005805515,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999911144871169,
      "loss": 6.2468,
      "mean_token_accuracy": 0.1300816871225834,
      "num_tokens": 2718711.0,
      "step": 1310
    },
    {
      "entropy": 6.0896015644073485,
      "epoch": 0.11928519593613933,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999908246055226,
      "loss": 6.1894,
      "mean_token_accuracy": 0.13144304975867271,
      "num_tokens": 2729956.0,
      "step": 1315
    },
    {
      "entropy": 6.332587003707886,
      "epoch": 0.11973875181422351,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999905300710607,
      "loss": 6.3051,
      "mean_token_accuracy": 0.12343701347708702,
      "num_tokens": 2740932.0,
      "step": 1320
    },
    {
      "entropy": 6.29391508102417,
      "epoch": 0.1201923076923077,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999902308837373,
      "loss": 6.3452,
      "mean_token_accuracy": 0.12498589381575584,
      "num_tokens": 2750781.0,
      "step": 1325
    },
    {
      "entropy": 6.286653852462768,
      "epoch": 0.12064586357039188,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999899270435584,
      "loss": 6.2415,
      "mean_token_accuracy": 0.12924491837620736,
      "num_tokens": 2761542.0,
      "step": 1330
    },
    {
      "entropy": 6.169296550750732,
      "epoch": 0.12109941944847605,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999896185505307,
      "loss": 6.1863,
      "mean_token_accuracy": 0.12321913093328477,
      "num_tokens": 2772507.0,
      "step": 1335
    },
    {
      "entropy": 6.211149549484253,
      "epoch": 0.12155297532656023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999893054046603,
      "loss": 6.2009,
      "mean_token_accuracy": 0.12625789046287536,
      "num_tokens": 2782806.0,
      "step": 1340
    },
    {
      "entropy": 6.175951194763184,
      "epoch": 0.12200653120464441,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999889876059537,
      "loss": 6.1991,
      "mean_token_accuracy": 0.12693845182657243,
      "num_tokens": 2793484.0,
      "step": 1345
    },
    {
      "entropy": 6.184288167953492,
      "epoch": 0.1224600870827286,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999886651544176,
      "loss": 6.2731,
      "mean_token_accuracy": 0.1297612629830837,
      "num_tokens": 2804443.0,
      "step": 1350
    },
    {
      "entropy": 6.163767099380493,
      "epoch": 0.12291364296081277,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999883380500584,
      "loss": 6.1485,
      "mean_token_accuracy": 0.1262630842626095,
      "num_tokens": 2814840.0,
      "step": 1355
    },
    {
      "entropy": 6.2059821605682375,
      "epoch": 0.12336719883889695,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999880062928831,
      "loss": 6.2207,
      "mean_token_accuracy": 0.1247759722173214,
      "num_tokens": 2824789.0,
      "step": 1360
    },
    {
      "entropy": 6.2396345138549805,
      "epoch": 0.12382075471698113,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999876698828985,
      "loss": 6.1775,
      "mean_token_accuracy": 0.13116121515631676,
      "num_tokens": 2834501.0,
      "step": 1365
    },
    {
      "entropy": 6.202531719207764,
      "epoch": 0.12427431059506532,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999873288201116,
      "loss": 6.1893,
      "mean_token_accuracy": 0.12950848862528802,
      "num_tokens": 2845399.0,
      "step": 1370
    },
    {
      "entropy": 6.032520198822022,
      "epoch": 0.1247278664731495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999869831045294,
      "loss": 6.0872,
      "mean_token_accuracy": 0.14028632938861846,
      "num_tokens": 2855146.0,
      "step": 1375
    },
    {
      "entropy": 6.3156661033630375,
      "epoch": 0.12518142235123367,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999866327361589,
      "loss": 6.2597,
      "mean_token_accuracy": 0.12432174906134605,
      "num_tokens": 2866438.0,
      "step": 1380
    },
    {
      "entropy": 6.019028043746948,
      "epoch": 0.12563497822931785,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999862777150077,
      "loss": 6.1301,
      "mean_token_accuracy": 0.12995802015066146,
      "num_tokens": 2876802.0,
      "step": 1385
    },
    {
      "entropy": 6.315221118927002,
      "epoch": 0.12608853410740203,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999859180410829,
      "loss": 6.3285,
      "mean_token_accuracy": 0.128648017346859,
      "num_tokens": 2888657.0,
      "step": 1390
    },
    {
      "entropy": 6.177397584915161,
      "epoch": 0.12654208998548622,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999855537143919,
      "loss": 6.1832,
      "mean_token_accuracy": 0.1301404818892479,
      "num_tokens": 2899605.0,
      "step": 1395
    },
    {
      "entropy": 6.116765117645263,
      "epoch": 0.1269956458635704,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999851847349425,
      "loss": 6.1427,
      "mean_token_accuracy": 0.1246509574353695,
      "num_tokens": 2910332.0,
      "step": 1400
    },
    {
      "entropy": 6.126271677017212,
      "epoch": 0.12744920174165458,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999848111027419,
      "loss": 6.1231,
      "mean_token_accuracy": 0.1296907387673855,
      "num_tokens": 2921226.0,
      "step": 1405
    },
    {
      "entropy": 6.1753684997558596,
      "epoch": 0.12790275761973874,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999844328177983,
      "loss": 6.165,
      "mean_token_accuracy": 0.129497080296278,
      "num_tokens": 2930941.0,
      "step": 1410
    },
    {
      "entropy": 6.1793512344360355,
      "epoch": 0.12835631349782292,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999840498801191,
      "loss": 6.1917,
      "mean_token_accuracy": 0.13117415681481362,
      "num_tokens": 2941383.0,
      "step": 1415
    },
    {
      "entropy": 6.207342672348022,
      "epoch": 0.1288098693759071,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999836622897126,
      "loss": 6.1587,
      "mean_token_accuracy": 0.12762442380189895,
      "num_tokens": 2951634.0,
      "step": 1420
    },
    {
      "entropy": 6.203984594345092,
      "epoch": 0.1292634252539913,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999832700465865,
      "loss": 6.2219,
      "mean_token_accuracy": 0.12739467918872832,
      "num_tokens": 2962240.0,
      "step": 1425
    },
    {
      "entropy": 6.187308359146118,
      "epoch": 0.12971698113207547,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999828731507491,
      "loss": 6.1977,
      "mean_token_accuracy": 0.12866248711943626,
      "num_tokens": 2973466.0,
      "step": 1430
    },
    {
      "entropy": 6.248796129226685,
      "epoch": 0.13017053701015965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999824716022085,
      "loss": 6.1989,
      "mean_token_accuracy": 0.13207736387848854,
      "num_tokens": 2982986.0,
      "step": 1435
    },
    {
      "entropy": 6.168663787841797,
      "epoch": 0.13062409288824384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999820654009731,
      "loss": 6.2264,
      "mean_token_accuracy": 0.1301150493323803,
      "num_tokens": 2994151.0,
      "step": 1440
    },
    {
      "entropy": 6.106994342803955,
      "epoch": 0.13107764876632802,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999816545470513,
      "loss": 6.0382,
      "mean_token_accuracy": 0.1328746996819973,
      "num_tokens": 3003481.0,
      "step": 1445
    },
    {
      "entropy": 6.062295961380005,
      "epoch": 0.1315312046444122,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999812390404514,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1257198430597782,
      "num_tokens": 3013835.0,
      "step": 1450
    },
    {
      "entropy": 6.146769189834595,
      "epoch": 0.13198476052249636,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999808188811823,
      "loss": 6.1207,
      "mean_token_accuracy": 0.14100732132792473,
      "num_tokens": 3024292.0,
      "step": 1455
    },
    {
      "entropy": 6.171922969818115,
      "epoch": 0.13243831640058054,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999803940692524,
      "loss": 6.2384,
      "mean_token_accuracy": 0.12596325874328612,
      "num_tokens": 3035114.0,
      "step": 1460
    },
    {
      "entropy": 6.1891436100006105,
      "epoch": 0.13289187227866472,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999799646046707,
      "loss": 6.1387,
      "mean_token_accuracy": 0.13315948471426964,
      "num_tokens": 3045151.0,
      "step": 1465
    },
    {
      "entropy": 6.107039928436279,
      "epoch": 0.1333454281567489,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999795304874461,
      "loss": 6.1193,
      "mean_token_accuracy": 0.13525033965706826,
      "num_tokens": 3054552.0,
      "step": 1470
    },
    {
      "entropy": 6.211303329467773,
      "epoch": 0.1337989840348331,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999790917175873,
      "loss": 6.2495,
      "mean_token_accuracy": 0.12571337893605233,
      "num_tokens": 3065386.0,
      "step": 1475
    },
    {
      "entropy": 6.1838174819946286,
      "epoch": 0.13425253991291727,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999786482951036,
      "loss": 6.0979,
      "mean_token_accuracy": 0.13161127269268036,
      "num_tokens": 3075934.0,
      "step": 1480
    },
    {
      "entropy": 6.044588327407837,
      "epoch": 0.13470609579100146,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999782002200041,
      "loss": 6.0649,
      "mean_token_accuracy": 0.13151389360427856,
      "num_tokens": 3086578.0,
      "step": 1485
    },
    {
      "entropy": 6.235484266281128,
      "epoch": 0.13515965166908564,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999777474922982,
      "loss": 6.2253,
      "mean_token_accuracy": 0.12512897700071335,
      "num_tokens": 3098183.0,
      "step": 1490
    },
    {
      "entropy": 6.0513090133667,
      "epoch": 0.13561320754716982,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999772901119951,
      "loss": 6.0581,
      "mean_token_accuracy": 0.1343822069466114,
      "num_tokens": 3107649.0,
      "step": 1495
    },
    {
      "entropy": 6.221631050109863,
      "epoch": 0.13606676342525398,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999768280791043,
      "loss": 6.1646,
      "mean_token_accuracy": 0.13448878824710847,
      "num_tokens": 3117652.0,
      "step": 1500
    },
    {
      "entropy": 6.061485052108765,
      "epoch": 0.13652031930333816,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999763613936352,
      "loss": 6.1065,
      "mean_token_accuracy": 0.13679977506399155,
      "num_tokens": 3127785.0,
      "step": 1505
    },
    {
      "entropy": 6.083025693893433,
      "epoch": 0.13697387518142234,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999758900555979,
      "loss": 6.0635,
      "mean_token_accuracy": 0.13337264955043793,
      "num_tokens": 3138657.0,
      "step": 1510
    },
    {
      "entropy": 6.140411472320556,
      "epoch": 0.13742743105950653,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999754140650017,
      "loss": 6.1911,
      "mean_token_accuracy": 0.12004235461354255,
      "num_tokens": 3150692.0,
      "step": 1515
    },
    {
      "entropy": 6.23555703163147,
      "epoch": 0.1378809869375907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999749334218567,
      "loss": 6.2595,
      "mean_token_accuracy": 0.12430977448821068,
      "num_tokens": 3161807.0,
      "step": 1520
    },
    {
      "entropy": 6.107507514953613,
      "epoch": 0.1383345428156749,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999744481261726,
      "loss": 6.0703,
      "mean_token_accuracy": 0.1393376462161541,
      "num_tokens": 3172446.0,
      "step": 1525
    },
    {
      "entropy": 6.145687675476074,
      "epoch": 0.13878809869375908,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999739581779597,
      "loss": 6.2071,
      "mean_token_accuracy": 0.1238051250576973,
      "num_tokens": 3183476.0,
      "step": 1530
    },
    {
      "entropy": 6.0691694736480715,
      "epoch": 0.13924165457184326,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499973463577228,
      "loss": 6.0885,
      "mean_token_accuracy": 0.13867690786719322,
      "num_tokens": 3193586.0,
      "step": 1535
    },
    {
      "entropy": 6.1251280307769775,
      "epoch": 0.13969521044992744,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999729643239877,
      "loss": 6.1195,
      "mean_token_accuracy": 0.13339588195085525,
      "num_tokens": 3203341.0,
      "step": 1540
    },
    {
      "entropy": 6.136832141876221,
      "epoch": 0.1401487663280116,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999724604182492,
      "loss": 6.1088,
      "mean_token_accuracy": 0.12823742404580116,
      "num_tokens": 3214177.0,
      "step": 1545
    },
    {
      "entropy": 6.141563987731933,
      "epoch": 0.14060232220609578,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999719518600229,
      "loss": 6.182,
      "mean_token_accuracy": 0.1269642286002636,
      "num_tokens": 3224516.0,
      "step": 1550
    },
    {
      "entropy": 6.098951148986816,
      "epoch": 0.14105587808417996,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999714386493192,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14152515679597855,
      "num_tokens": 3234001.0,
      "step": 1555
    },
    {
      "entropy": 6.196884202957153,
      "epoch": 0.14150943396226415,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999709207861489,
      "loss": 6.1286,
      "mean_token_accuracy": 0.13318521454930304,
      "num_tokens": 3246095.0,
      "step": 1560
    },
    {
      "entropy": 6.135161066055298,
      "epoch": 0.14196298984034833,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999703982705227,
      "loss": 6.177,
      "mean_token_accuracy": 0.12648626491427423,
      "num_tokens": 3255828.0,
      "step": 1565
    },
    {
      "entropy": 6.101465272903442,
      "epoch": 0.1424165457184325,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999698711024513,
      "loss": 6.1378,
      "mean_token_accuracy": 0.13857534304261207,
      "num_tokens": 3265127.0,
      "step": 1570
    },
    {
      "entropy": 6.23589973449707,
      "epoch": 0.1428701015965167,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999693392819456,
      "loss": 6.2083,
      "mean_token_accuracy": 0.12722571119666098,
      "num_tokens": 3277205.0,
      "step": 1575
    },
    {
      "entropy": 6.039449787139892,
      "epoch": 0.14332365747460088,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999688028090166,
      "loss": 6.1109,
      "mean_token_accuracy": 0.13461936116218567,
      "num_tokens": 3287119.0,
      "step": 1580
    },
    {
      "entropy": 6.237564992904663,
      "epoch": 0.14377721335268506,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999682616836755,
      "loss": 6.2073,
      "mean_token_accuracy": 0.12796173691749574,
      "num_tokens": 3297324.0,
      "step": 1585
    },
    {
      "entropy": 6.166965675354004,
      "epoch": 0.14423076923076922,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999677159059334,
      "loss": 6.1568,
      "mean_token_accuracy": 0.12957976162433624,
      "num_tokens": 3308162.0,
      "step": 1590
    },
    {
      "entropy": 6.156198501586914,
      "epoch": 0.1446843251088534,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999671654758016,
      "loss": 6.1587,
      "mean_token_accuracy": 0.12751780077815056,
      "num_tokens": 3319422.0,
      "step": 1595
    },
    {
      "entropy": 6.074300146102905,
      "epoch": 0.14513788098693758,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999666103932915,
      "loss": 6.1303,
      "mean_token_accuracy": 0.13079726696014404,
      "num_tokens": 3328668.0,
      "step": 1600
    },
    {
      "entropy": 6.196269941329956,
      "epoch": 0.14559143686502177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999660506584145,
      "loss": 6.1789,
      "mean_token_accuracy": 0.12973756417632104,
      "num_tokens": 3338563.0,
      "step": 1605
    },
    {
      "entropy": 6.123033428192139,
      "epoch": 0.14604499274310595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999654862711823,
      "loss": 6.1866,
      "mean_token_accuracy": 0.13245568349957465,
      "num_tokens": 3348025.0,
      "step": 1610
    },
    {
      "entropy": 6.125161266326904,
      "epoch": 0.14649854862119013,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999649172316065,
      "loss": 6.0621,
      "mean_token_accuracy": 0.14053588956594468,
      "num_tokens": 3359094.0,
      "step": 1615
    },
    {
      "entropy": 6.113393926620484,
      "epoch": 0.14695210449927432,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499964343539699,
      "loss": 6.1022,
      "mean_token_accuracy": 0.13681961521506308,
      "num_tokens": 3368965.0,
      "step": 1620
    },
    {
      "entropy": 6.058830976486206,
      "epoch": 0.1474056603773585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999637651954714,
      "loss": 6.1126,
      "mean_token_accuracy": 0.13101110011339187,
      "num_tokens": 3378612.0,
      "step": 1625
    },
    {
      "entropy": 6.140189456939697,
      "epoch": 0.14785921625544268,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999631821989358,
      "loss": 6.0731,
      "mean_token_accuracy": 0.13366867899894713,
      "num_tokens": 3389634.0,
      "step": 1630
    },
    {
      "entropy": 6.045487594604492,
      "epoch": 0.14831277213352684,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999625945501043,
      "loss": 6.1454,
      "mean_token_accuracy": 0.13442882299423217,
      "num_tokens": 3399604.0,
      "step": 1635
    },
    {
      "entropy": 6.223248147964478,
      "epoch": 0.14876632801161102,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499962002248989,
      "loss": 6.0949,
      "mean_token_accuracy": 0.13319409787654876,
      "num_tokens": 3410314.0,
      "step": 1640
    },
    {
      "entropy": 6.071327877044678,
      "epoch": 0.1492198838896952,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999614052956024,
      "loss": 6.1164,
      "mean_token_accuracy": 0.13651233538985252,
      "num_tokens": 3420773.0,
      "step": 1645
    },
    {
      "entropy": 6.234537887573242,
      "epoch": 0.1496734397677794,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999608036899563,
      "loss": 6.2406,
      "mean_token_accuracy": 0.1285805732011795,
      "num_tokens": 3431702.0,
      "step": 1650
    },
    {
      "entropy": 6.026836967468261,
      "epoch": 0.15012699564586357,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999601974320636,
      "loss": 6.1376,
      "mean_token_accuracy": 0.133916974067688,
      "num_tokens": 3441925.0,
      "step": 1655
    },
    {
      "entropy": 6.225948667526245,
      "epoch": 0.15058055152394775,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999595865219367,
      "loss": 6.1489,
      "mean_token_accuracy": 0.13318530544638635,
      "num_tokens": 3452391.0,
      "step": 1660
    },
    {
      "entropy": 6.018878602981568,
      "epoch": 0.15103410740203194,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999589709595881,
      "loss": 6.0672,
      "mean_token_accuracy": 0.12874112129211426,
      "num_tokens": 3463028.0,
      "step": 1665
    },
    {
      "entropy": 6.063703966140747,
      "epoch": 0.15148766328011612,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999583507450308,
      "loss": 6.0641,
      "mean_token_accuracy": 0.13683240562677385,
      "num_tokens": 3473470.0,
      "step": 1670
    },
    {
      "entropy": 6.126550817489624,
      "epoch": 0.1519412191582003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999577258782774,
      "loss": 6.1066,
      "mean_token_accuracy": 0.13182729333639145,
      "num_tokens": 3484486.0,
      "step": 1675
    },
    {
      "entropy": 6.084259557723999,
      "epoch": 0.15239477503628446,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499957096359341,
      "loss": 6.0679,
      "mean_token_accuracy": 0.13736046478152275,
      "num_tokens": 3494885.0,
      "step": 1680
    },
    {
      "entropy": 6.07341947555542,
      "epoch": 0.15284833091436864,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999564621882343,
      "loss": 6.079,
      "mean_token_accuracy": 0.13572860211133958,
      "num_tokens": 3506088.0,
      "step": 1685
    },
    {
      "entropy": 6.015376949310303,
      "epoch": 0.15330188679245282,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999558233649708,
      "loss": 6.0114,
      "mean_token_accuracy": 0.13501859828829765,
      "num_tokens": 3516380.0,
      "step": 1690
    },
    {
      "entropy": 6.094719457626343,
      "epoch": 0.153755442670537,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999551798895636,
      "loss": 6.0229,
      "mean_token_accuracy": 0.14153220504522324,
      "num_tokens": 3525965.0,
      "step": 1695
    },
    {
      "entropy": 6.060795879364013,
      "epoch": 0.1542089985486212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999545317620258,
      "loss": 6.04,
      "mean_token_accuracy": 0.13283166736364366,
      "num_tokens": 3536592.0,
      "step": 1700
    },
    {
      "entropy": 6.117245006561279,
      "epoch": 0.15466255442670537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999538789823711,
      "loss": 6.1651,
      "mean_token_accuracy": 0.13683197945356368,
      "num_tokens": 3546001.0,
      "step": 1705
    },
    {
      "entropy": 6.123035430908203,
      "epoch": 0.15511611030478956,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999532215506127,
      "loss": 6.0487,
      "mean_token_accuracy": 0.12888889610767365,
      "num_tokens": 3555575.0,
      "step": 1710
    },
    {
      "entropy": 6.06559419631958,
      "epoch": 0.15556966618287374,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999525594667645,
      "loss": 6.0429,
      "mean_token_accuracy": 0.14311017096042633,
      "num_tokens": 3565685.0,
      "step": 1715
    },
    {
      "entropy": 6.119355535507202,
      "epoch": 0.15602322206095792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049995189273084,
      "loss": 6.0872,
      "mean_token_accuracy": 0.14001662507653237,
      "num_tokens": 3575119.0,
      "step": 1720
    },
    {
      "entropy": 6.048756504058838,
      "epoch": 0.15647677793904208,
      "grad_norm": 1.125,
      "learning_rate": 0.000499951221342853,
      "loss": 6.0393,
      "mean_token_accuracy": 0.13779830187559128,
      "num_tokens": 3584620.0,
      "step": 1725
    },
    {
      "entropy": 6.0873788356781,
      "epoch": 0.15693033381712626,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999505453028174,
      "loss": 6.0783,
      "mean_token_accuracy": 0.13565813079476358,
      "num_tokens": 3595403.0,
      "step": 1730
    },
    {
      "entropy": 6.042232465744019,
      "epoch": 0.15738388969521044,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999498646107472,
      "loss": 6.0665,
      "mean_token_accuracy": 0.135944102704525,
      "num_tokens": 3606355.0,
      "step": 1735
    },
    {
      "entropy": 6.039039087295532,
      "epoch": 0.15783744557329463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999491792666566,
      "loss": 6.0526,
      "mean_token_accuracy": 0.1334886446595192,
      "num_tokens": 3616222.0,
      "step": 1740
    },
    {
      "entropy": 6.117766857147217,
      "epoch": 0.1582910014513788,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999484892705597,
      "loss": 6.0978,
      "mean_token_accuracy": 0.1427782490849495,
      "num_tokens": 3625179.0,
      "step": 1745
    },
    {
      "entropy": 6.066883754730225,
      "epoch": 0.158744557329463,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999477946224705,
      "loss": 6.057,
      "mean_token_accuracy": 0.13318037912249564,
      "num_tokens": 3636256.0,
      "step": 1750
    },
    {
      "entropy": 5.9570455074310305,
      "epoch": 0.15919811320754718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999470953224037,
      "loss": 5.9024,
      "mean_token_accuracy": 0.14086213558912278,
      "num_tokens": 3645919.0,
      "step": 1755
    },
    {
      "entropy": 6.067557430267334,
      "epoch": 0.15965166908563136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999463913703735,
      "loss": 6.1052,
      "mean_token_accuracy": 0.13390027210116387,
      "num_tokens": 3657008.0,
      "step": 1760
    },
    {
      "entropy": 6.014396524429321,
      "epoch": 0.16010522496371554,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999456827663948,
      "loss": 5.9915,
      "mean_token_accuracy": 0.13444581776857376,
      "num_tokens": 3667686.0,
      "step": 1765
    },
    {
      "entropy": 6.022483873367309,
      "epoch": 0.1605587808417997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499944969510482,
      "loss": 5.9907,
      "mean_token_accuracy": 0.13489970788359643,
      "num_tokens": 3678972.0,
      "step": 1770
    },
    {
      "entropy": 6.032540559768677,
      "epoch": 0.16101233671988388,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999442516026498,
      "loss": 5.9973,
      "mean_token_accuracy": 0.13953388929367067,
      "num_tokens": 3688903.0,
      "step": 1775
    },
    {
      "entropy": 6.049791049957276,
      "epoch": 0.16146589259796806,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999435290429133,
      "loss": 6.0542,
      "mean_token_accuracy": 0.13899827301502227,
      "num_tokens": 3697880.0,
      "step": 1780
    },
    {
      "entropy": 6.113507795333862,
      "epoch": 0.16191944847605225,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999428018312872,
      "loss": 6.1146,
      "mean_token_accuracy": 0.13196710646152496,
      "num_tokens": 3708500.0,
      "step": 1785
    },
    {
      "entropy": 5.967461633682251,
      "epoch": 0.16237300435413643,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999420699677867,
      "loss": 5.9456,
      "mean_token_accuracy": 0.14808355420827865,
      "num_tokens": 3718711.0,
      "step": 1790
    },
    {
      "entropy": 6.003653526306152,
      "epoch": 0.1628265602322206,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999413334524269,
      "loss": 6.0784,
      "mean_token_accuracy": 0.13483019024133683,
      "num_tokens": 3729431.0,
      "step": 1795
    },
    {
      "entropy": 6.118337678909302,
      "epoch": 0.1632801161103048,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999405922852229,
      "loss": 6.0157,
      "mean_token_accuracy": 0.13496499881148338,
      "num_tokens": 3741061.0,
      "step": 1800
    },
    {
      "entropy": 5.951924753189087,
      "epoch": 0.16373367198838898,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999398464661901,
      "loss": 6.0049,
      "mean_token_accuracy": 0.14013575315475463,
      "num_tokens": 3751878.0,
      "step": 1805
    },
    {
      "entropy": 6.038108873367309,
      "epoch": 0.16418722786647316,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999390959953441,
      "loss": 6.0101,
      "mean_token_accuracy": 0.13939072266221048,
      "num_tokens": 3762039.0,
      "step": 1810
    },
    {
      "entropy": 5.974385404586792,
      "epoch": 0.16464078374455732,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999383408727002,
      "loss": 5.9441,
      "mean_token_accuracy": 0.1419306755065918,
      "num_tokens": 3770755.0,
      "step": 1815
    },
    {
      "entropy": 6.02751522064209,
      "epoch": 0.1650943396226415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999375810982741,
      "loss": 6.1083,
      "mean_token_accuracy": 0.13299722522497176,
      "num_tokens": 3781698.0,
      "step": 1820
    },
    {
      "entropy": 6.067613887786865,
      "epoch": 0.16554789550072568,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999368166720815,
      "loss": 6.1488,
      "mean_token_accuracy": 0.13327884972095488,
      "num_tokens": 3792508.0,
      "step": 1825
    },
    {
      "entropy": 6.086944532394409,
      "epoch": 0.16600145137880987,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999360475941382,
      "loss": 6.0028,
      "mean_token_accuracy": 0.14385750517249107,
      "num_tokens": 3803065.0,
      "step": 1830
    },
    {
      "entropy": 6.022203254699707,
      "epoch": 0.16645500725689405,
      "grad_norm": 1.0625,
      "learning_rate": 0.00049993527386446,
      "loss": 6.0675,
      "mean_token_accuracy": 0.1315029554069042,
      "num_tokens": 3813696.0,
      "step": 1835
    },
    {
      "entropy": 6.0408408641815186,
      "epoch": 0.16690856313497823,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999344954830633,
      "loss": 6.022,
      "mean_token_accuracy": 0.14210820943117142,
      "num_tokens": 3823786.0,
      "step": 1840
    },
    {
      "entropy": 6.0095936298370365,
      "epoch": 0.16736211901306242,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999337124499638,
      "loss": 5.9917,
      "mean_token_accuracy": 0.13492688685655593,
      "num_tokens": 3835011.0,
      "step": 1845
    },
    {
      "entropy": 6.074155664443969,
      "epoch": 0.1678156748911466,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999329247651777,
      "loss": 6.0408,
      "mean_token_accuracy": 0.13303120210766792,
      "num_tokens": 3846161.0,
      "step": 1850
    },
    {
      "entropy": 5.95374641418457,
      "epoch": 0.16826923076923078,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999321324287213,
      "loss": 5.972,
      "mean_token_accuracy": 0.14118631407618523,
      "num_tokens": 3856013.0,
      "step": 1855
    },
    {
      "entropy": 6.1738697528839115,
      "epoch": 0.16872278664731494,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999313354406114,
      "loss": 6.1002,
      "mean_token_accuracy": 0.1409381054341793,
      "num_tokens": 3865521.0,
      "step": 1860
    },
    {
      "entropy": 5.912585163116455,
      "epoch": 0.16917634252539912,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499930533800864,
      "loss": 5.9034,
      "mean_token_accuracy": 0.14283333346247673,
      "num_tokens": 3876290.0,
      "step": 1865
    },
    {
      "entropy": 5.980841970443725,
      "epoch": 0.1696298984034833,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999297275094958,
      "loss": 6.1048,
      "mean_token_accuracy": 0.13230404928326606,
      "num_tokens": 3886437.0,
      "step": 1870
    },
    {
      "entropy": 6.112230062484741,
      "epoch": 0.1700834542815675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999289165665235,
      "loss": 6.1435,
      "mean_token_accuracy": 0.13440538123250007,
      "num_tokens": 3896501.0,
      "step": 1875
    },
    {
      "entropy": 6.15153021812439,
      "epoch": 0.17053701015965167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999281009719639,
      "loss": 6.0671,
      "mean_token_accuracy": 0.13559964820742607,
      "num_tokens": 3906725.0,
      "step": 1880
    },
    {
      "entropy": 6.02051043510437,
      "epoch": 0.17099056603773585,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999272807258339,
      "loss": 5.9351,
      "mean_token_accuracy": 0.1449750393629074,
      "num_tokens": 3917428.0,
      "step": 1885
    },
    {
      "entropy": 5.859882068634033,
      "epoch": 0.17144412191582004,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999264558281504,
      "loss": 6.0304,
      "mean_token_accuracy": 0.1393364980816841,
      "num_tokens": 3927298.0,
      "step": 1890
    },
    {
      "entropy": 6.047381448745727,
      "epoch": 0.17189767779390422,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999256262789305,
      "loss": 5.9394,
      "mean_token_accuracy": 0.1372090183198452,
      "num_tokens": 3936727.0,
      "step": 1895
    },
    {
      "entropy": 6.062630701065063,
      "epoch": 0.1723512336719884,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999247920781913,
      "loss": 6.0596,
      "mean_token_accuracy": 0.13730003088712692,
      "num_tokens": 3947246.0,
      "step": 1900
    },
    {
      "entropy": 5.838617277145386,
      "epoch": 0.17280478955007256,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999239532259502,
      "loss": 5.9592,
      "mean_token_accuracy": 0.14350975602865218,
      "num_tokens": 3958535.0,
      "step": 1905
    },
    {
      "entropy": 6.161811971664429,
      "epoch": 0.17325834542815674,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999231097222244,
      "loss": 6.0521,
      "mean_token_accuracy": 0.1353868529200554,
      "num_tokens": 3969635.0,
      "step": 1910
    },
    {
      "entropy": 6.025433874130249,
      "epoch": 0.17371190130624092,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999222615670312,
      "loss": 6.024,
      "mean_token_accuracy": 0.1359100967645645,
      "num_tokens": 3980227.0,
      "step": 1915
    },
    {
      "entropy": 6.06690092086792,
      "epoch": 0.1741654571843251,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999214087603885,
      "loss": 6.0263,
      "mean_token_accuracy": 0.14165394008159637,
      "num_tokens": 3990915.0,
      "step": 1920
    },
    {
      "entropy": 5.958260536193848,
      "epoch": 0.1746190130624093,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999205513023135,
      "loss": 5.9216,
      "mean_token_accuracy": 0.13917687460780143,
      "num_tokens": 4001273.0,
      "step": 1925
    },
    {
      "entropy": 6.072769498825073,
      "epoch": 0.17507256894049347,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999196891928245,
      "loss": 5.994,
      "mean_token_accuracy": 0.141386017203331,
      "num_tokens": 4011596.0,
      "step": 1930
    },
    {
      "entropy": 5.960225534439087,
      "epoch": 0.17552612481857766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999188224319388,
      "loss": 6.0366,
      "mean_token_accuracy": 0.13730789497494697,
      "num_tokens": 4022638.0,
      "step": 1935
    },
    {
      "entropy": 6.080376815795899,
      "epoch": 0.17597968069666184,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999179510196746,
      "loss": 6.0156,
      "mean_token_accuracy": 0.13227792233228683,
      "num_tokens": 4032398.0,
      "step": 1940
    },
    {
      "entropy": 5.898893404006958,
      "epoch": 0.17643323657474602,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999170749560498,
      "loss": 5.9724,
      "mean_token_accuracy": 0.1371435798704624,
      "num_tokens": 4042852.0,
      "step": 1945
    },
    {
      "entropy": 6.040456962585449,
      "epoch": 0.17688679245283018,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999161942410827,
      "loss": 6.0056,
      "mean_token_accuracy": 0.1349559858441353,
      "num_tokens": 4053110.0,
      "step": 1950
    },
    {
      "entropy": 6.002525901794433,
      "epoch": 0.17734034833091436,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999153088747913,
      "loss": 6.0212,
      "mean_token_accuracy": 0.1401715062558651,
      "num_tokens": 4064571.0,
      "step": 1955
    },
    {
      "entropy": 5.953612756729126,
      "epoch": 0.17779390420899854,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499914418857194,
      "loss": 5.9906,
      "mean_token_accuracy": 0.13749739825725554,
      "num_tokens": 4075400.0,
      "step": 1960
    },
    {
      "entropy": 6.028836965560913,
      "epoch": 0.17824746008708273,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999135241883092,
      "loss": 5.9439,
      "mean_token_accuracy": 0.14185010865330697,
      "num_tokens": 4085763.0,
      "step": 1965
    },
    {
      "entropy": 5.986079740524292,
      "epoch": 0.1787010159651669,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999126248681555,
      "loss": 6.0109,
      "mean_token_accuracy": 0.13681597337126733,
      "num_tokens": 4095993.0,
      "step": 1970
    },
    {
      "entropy": 5.986701774597168,
      "epoch": 0.1791545718432511,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999117208967513,
      "loss": 5.9225,
      "mean_token_accuracy": 0.1380699433386326,
      "num_tokens": 4105012.0,
      "step": 1975
    },
    {
      "entropy": 5.942403030395508,
      "epoch": 0.17960812772133528,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999108122741154,
      "loss": 5.9692,
      "mean_token_accuracy": 0.1353602610528469,
      "num_tokens": 4115142.0,
      "step": 1980
    },
    {
      "entropy": 6.031651258468628,
      "epoch": 0.18006168359941946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999098990002666,
      "loss": 5.957,
      "mean_token_accuracy": 0.13746799603104592,
      "num_tokens": 4125297.0,
      "step": 1985
    },
    {
      "entropy": 5.977378511428833,
      "epoch": 0.18051523947750364,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999089810752237,
      "loss": 6.0917,
      "mean_token_accuracy": 0.132157414406538,
      "num_tokens": 4136404.0,
      "step": 1990
    },
    {
      "entropy": 6.080759954452515,
      "epoch": 0.1809687953555878,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499908058499006,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14288827329874038,
      "num_tokens": 4146228.0,
      "step": 1995
    },
    {
      "entropy": 5.95260910987854,
      "epoch": 0.18142235123367198,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999071312716321,
      "loss": 5.8708,
      "mean_token_accuracy": 0.1427669905126095,
      "num_tokens": 4156435.0,
      "step": 2000
    },
    {
      "entropy": 5.975965404510498,
      "epoch": 0.18187590711175616,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999061993931215,
      "loss": 6.0042,
      "mean_token_accuracy": 0.1480802610516548,
      "num_tokens": 4166525.0,
      "step": 2005
    },
    {
      "entropy": 6.0720906257629395,
      "epoch": 0.18232946298984035,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004999052628634934,
      "loss": 6.0833,
      "mean_token_accuracy": 0.13352881222963334,
      "num_tokens": 4176831.0,
      "step": 2010
    },
    {
      "entropy": 6.019681024551391,
      "epoch": 0.18278301886792453,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999043216827671,
      "loss": 6.06,
      "mean_token_accuracy": 0.13369136229157447,
      "num_tokens": 4187830.0,
      "step": 2015
    },
    {
      "entropy": 6.006418657302857,
      "epoch": 0.1832365747460087,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999033758509621,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1393364667892456,
      "num_tokens": 4198898.0,
      "step": 2020
    },
    {
      "entropy": 5.97959532737732,
      "epoch": 0.1836901306240929,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499902425368098,
      "loss": 5.9709,
      "mean_token_accuracy": 0.1392282821238041,
      "num_tokens": 4210297.0,
      "step": 2025
    },
    {
      "entropy": 6.037605619430542,
      "epoch": 0.18414368650217708,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004999014702341944,
      "loss": 6.1283,
      "mean_token_accuracy": 0.13221636712551116,
      "num_tokens": 4221469.0,
      "step": 2030
    },
    {
      "entropy": 5.987315082550049,
      "epoch": 0.18459724238026126,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999005104492712,
      "loss": 5.9446,
      "mean_token_accuracy": 0.14053445905447007,
      "num_tokens": 4233046.0,
      "step": 2035
    },
    {
      "entropy": 6.048150110244751,
      "epoch": 0.18505079825834542,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998995460133481,
      "loss": 5.9829,
      "mean_token_accuracy": 0.13854565769433974,
      "num_tokens": 4243233.0,
      "step": 2040
    },
    {
      "entropy": 6.046663284301758,
      "epoch": 0.1855043541364296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998985769264451,
      "loss": 6.0973,
      "mean_token_accuracy": 0.13540371656417846,
      "num_tokens": 4254020.0,
      "step": 2045
    },
    {
      "entropy": 5.959017562866211,
      "epoch": 0.18595791001451378,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998976031885821,
      "loss": 5.8928,
      "mean_token_accuracy": 0.14246483519673347,
      "num_tokens": 4263901.0,
      "step": 2050
    },
    {
      "entropy": 5.890605878829956,
      "epoch": 0.18641146589259797,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998966247997794,
      "loss": 5.9255,
      "mean_token_accuracy": 0.13970548510551453,
      "num_tokens": 4274814.0,
      "step": 2055
    },
    {
      "entropy": 6.063959264755249,
      "epoch": 0.18686502177068215,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998956417600572,
      "loss": 6.0704,
      "mean_token_accuracy": 0.1417410969734192,
      "num_tokens": 4284846.0,
      "step": 2060
    },
    {
      "entropy": 6.033328819274902,
      "epoch": 0.18731857764876633,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998946540694358,
      "loss": 5.9731,
      "mean_token_accuracy": 0.14464445784687996,
      "num_tokens": 4294862.0,
      "step": 2065
    },
    {
      "entropy": 6.068034410476685,
      "epoch": 0.18777213352685052,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998936617279357,
      "loss": 6.0175,
      "mean_token_accuracy": 0.1416530877351761,
      "num_tokens": 4305708.0,
      "step": 2070
    },
    {
      "entropy": 5.995953607559204,
      "epoch": 0.1882256894049347,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998926647355773,
      "loss": 5.9943,
      "mean_token_accuracy": 0.13258325606584548,
      "num_tokens": 4315853.0,
      "step": 2075
    },
    {
      "entropy": 5.9227760314941404,
      "epoch": 0.18867924528301888,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998916630923813,
      "loss": 5.9565,
      "mean_token_accuracy": 0.13823306187987328,
      "num_tokens": 4326607.0,
      "step": 2080
    },
    {
      "entropy": 5.959938955307007,
      "epoch": 0.18913280116110304,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998906567983685,
      "loss": 5.991,
      "mean_token_accuracy": 0.13902745693922042,
      "num_tokens": 4336050.0,
      "step": 2085
    },
    {
      "entropy": 6.056903553009033,
      "epoch": 0.18958635703918722,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998896458535595,
      "loss": 5.9604,
      "mean_token_accuracy": 0.14332208260893822,
      "num_tokens": 4345453.0,
      "step": 2090
    },
    {
      "entropy": 5.951250743865967,
      "epoch": 0.1900399129172714,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998886302579752,
      "loss": 5.9444,
      "mean_token_accuracy": 0.1378558337688446,
      "num_tokens": 4356372.0,
      "step": 2095
    },
    {
      "entropy": 5.958211088180542,
      "epoch": 0.19049346879535559,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499887610011637,
      "loss": 6.0014,
      "mean_token_accuracy": 0.14464358389377593,
      "num_tokens": 4365500.0,
      "step": 2100
    },
    {
      "entropy": 6.067464399337768,
      "epoch": 0.19094702467343977,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998865851145656,
      "loss": 5.9311,
      "mean_token_accuracy": 0.14481838345527648,
      "num_tokens": 4375031.0,
      "step": 2105
    },
    {
      "entropy": 5.901741409301758,
      "epoch": 0.19140058055152395,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998855555667822,
      "loss": 5.972,
      "mean_token_accuracy": 0.1389401637017727,
      "num_tokens": 4386409.0,
      "step": 2110
    },
    {
      "entropy": 6.059998941421509,
      "epoch": 0.19185413642960814,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998845213683084,
      "loss": 5.9942,
      "mean_token_accuracy": 0.13713788390159606,
      "num_tokens": 4397040.0,
      "step": 2115
    },
    {
      "entropy": 5.938760089874267,
      "epoch": 0.19230769230769232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998834825191652,
      "loss": 5.9508,
      "mean_token_accuracy": 0.14106623604893684,
      "num_tokens": 4406723.0,
      "step": 2120
    },
    {
      "entropy": 6.02097454071045,
      "epoch": 0.1927612481857765,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998824390193743,
      "loss": 5.9982,
      "mean_token_accuracy": 0.14898034185171127,
      "num_tokens": 4416013.0,
      "step": 2125
    },
    {
      "entropy": 5.96734733581543,
      "epoch": 0.19321480406386066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998813908689573,
      "loss": 5.9793,
      "mean_token_accuracy": 0.14236808717250823,
      "num_tokens": 4426441.0,
      "step": 2130
    },
    {
      "entropy": 6.063247156143189,
      "epoch": 0.19366835994194484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004998803380679359,
      "loss": 6.0948,
      "mean_token_accuracy": 0.1332187458872795,
      "num_tokens": 4437328.0,
      "step": 2135
    },
    {
      "entropy": 6.089685106277466,
      "epoch": 0.19412191582002902,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998792806163317,
      "loss": 5.9181,
      "mean_token_accuracy": 0.14083896055817605,
      "num_tokens": 4447546.0,
      "step": 2140
    },
    {
      "entropy": 5.88359489440918,
      "epoch": 0.1945754716981132,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998782185141667,
      "loss": 5.9561,
      "mean_token_accuracy": 0.14243533834815025,
      "num_tokens": 4457319.0,
      "step": 2145
    },
    {
      "entropy": 6.066173934936524,
      "epoch": 0.1950290275761974,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998771517614629,
      "loss": 6.0624,
      "mean_token_accuracy": 0.13416438773274422,
      "num_tokens": 4468348.0,
      "step": 2150
    },
    {
      "entropy": 6.009496736526489,
      "epoch": 0.19548258345428157,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998760803582422,
      "loss": 5.9612,
      "mean_token_accuracy": 0.14601074978709222,
      "num_tokens": 4478263.0,
      "step": 2155
    },
    {
      "entropy": 5.905720138549805,
      "epoch": 0.19593613933236576,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004998750043045269,
      "loss": 5.8817,
      "mean_token_accuracy": 0.14295333474874497,
      "num_tokens": 4488772.0,
      "step": 2160
    },
    {
      "entropy": 6.011665964126587,
      "epoch": 0.19638969521044994,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998739236003392,
      "loss": 5.9905,
      "mean_token_accuracy": 0.13327775821089743,
      "num_tokens": 4499443.0,
      "step": 2165
    },
    {
      "entropy": 5.983055496215821,
      "epoch": 0.1968432510885341,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998728382457014,
      "loss": 5.9876,
      "mean_token_accuracy": 0.13847402930259706,
      "num_tokens": 4509969.0,
      "step": 2170
    },
    {
      "entropy": 6.022305631637574,
      "epoch": 0.19729680696661828,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998717482406362,
      "loss": 5.979,
      "mean_token_accuracy": 0.14845634549856185,
      "num_tokens": 4519668.0,
      "step": 2175
    },
    {
      "entropy": 5.9130627632141115,
      "epoch": 0.19775036284470246,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998706535851657,
      "loss": 5.9578,
      "mean_token_accuracy": 0.14601308181881906,
      "num_tokens": 4530303.0,
      "step": 2180
    },
    {
      "entropy": 5.992128801345825,
      "epoch": 0.19820391872278664,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998695542793129,
      "loss": 6.0187,
      "mean_token_accuracy": 0.14158663749694825,
      "num_tokens": 4541801.0,
      "step": 2185
    },
    {
      "entropy": 5.938469648361206,
      "epoch": 0.19865747460087083,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998684503231003,
      "loss": 5.8341,
      "mean_token_accuracy": 0.1426684334874153,
      "num_tokens": 4551346.0,
      "step": 2190
    },
    {
      "entropy": 5.922765111923217,
      "epoch": 0.199111030478955,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499867341716551,
      "loss": 5.915,
      "mean_token_accuracy": 0.14817361980676652,
      "num_tokens": 4561835.0,
      "step": 2195
    },
    {
      "entropy": 5.904284858703614,
      "epoch": 0.1995645863570392,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998662284596878,
      "loss": 5.8667,
      "mean_token_accuracy": 0.145841383934021,
      "num_tokens": 4572760.0,
      "step": 2200
    },
    {
      "entropy": 5.936977243423462,
      "epoch": 0.20001814223512338,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998651105525335,
      "loss": 5.9473,
      "mean_token_accuracy": 0.14006382599473,
      "num_tokens": 4582884.0,
      "step": 2205
    },
    {
      "entropy": 5.961916112899781,
      "epoch": 0.20047169811320756,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998639879951115,
      "loss": 5.9315,
      "mean_token_accuracy": 0.14331716150045395,
      "num_tokens": 4594297.0,
      "step": 2210
    },
    {
      "entropy": 5.988901615142822,
      "epoch": 0.2009252539912917,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499862860787445,
      "loss": 5.9907,
      "mean_token_accuracy": 0.13747334852814674,
      "num_tokens": 4605072.0,
      "step": 2215
    },
    {
      "entropy": 5.911744117736816,
      "epoch": 0.2013788098693759,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998617289295571,
      "loss": 5.8751,
      "mean_token_accuracy": 0.15403711050748825,
      "num_tokens": 4615207.0,
      "step": 2220
    },
    {
      "entropy": 5.916146278381348,
      "epoch": 0.20183236574746008,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998605924214714,
      "loss": 5.9039,
      "mean_token_accuracy": 0.14326297491788864,
      "num_tokens": 4625552.0,
      "step": 2225
    },
    {
      "entropy": 6.068982744216919,
      "epoch": 0.20228592162554426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998594512632114,
      "loss": 6.0052,
      "mean_token_accuracy": 0.13947765678167343,
      "num_tokens": 4635864.0,
      "step": 2230
    },
    {
      "entropy": 5.9241475582122805,
      "epoch": 0.20273947750362845,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998583054548006,
      "loss": 5.9602,
      "mean_token_accuracy": 0.13143666759133338,
      "num_tokens": 4646978.0,
      "step": 2235
    },
    {
      "entropy": 5.949285650253296,
      "epoch": 0.20319303338171263,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998571549962628,
      "loss": 5.8858,
      "mean_token_accuracy": 0.14171508550643921,
      "num_tokens": 4656737.0,
      "step": 2240
    },
    {
      "entropy": 5.886570167541504,
      "epoch": 0.2036465892597968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998559998876216,
      "loss": 6.0072,
      "mean_token_accuracy": 0.14202161729335785,
      "num_tokens": 4667663.0,
      "step": 2245
    },
    {
      "entropy": 5.985239171981812,
      "epoch": 0.204100145137881,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998548401289012,
      "loss": 5.9673,
      "mean_token_accuracy": 0.13756298944354056,
      "num_tokens": 4678814.0,
      "step": 2250
    },
    {
      "entropy": 6.042722129821778,
      "epoch": 0.20455370101596518,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998536757201253,
      "loss": 5.9821,
      "mean_token_accuracy": 0.13774357065558435,
      "num_tokens": 4689446.0,
      "step": 2255
    },
    {
      "entropy": 5.971980857849121,
      "epoch": 0.20500725689404933,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998525066613181,
      "loss": 5.993,
      "mean_token_accuracy": 0.13813588321208953,
      "num_tokens": 4700207.0,
      "step": 2260
    },
    {
      "entropy": 6.003995275497436,
      "epoch": 0.20546081277213352,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998513329525038,
      "loss": 5.9595,
      "mean_token_accuracy": 0.1428474046289921,
      "num_tokens": 4710743.0,
      "step": 2265
    },
    {
      "entropy": 5.981139850616455,
      "epoch": 0.2059143686502177,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998501545937066,
      "loss": 5.9624,
      "mean_token_accuracy": 0.14789393842220305,
      "num_tokens": 4720964.0,
      "step": 2270
    },
    {
      "entropy": 5.965071296691894,
      "epoch": 0.20636792452830188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499848971584951,
      "loss": 6.001,
      "mean_token_accuracy": 0.13834339007735252,
      "num_tokens": 4732022.0,
      "step": 2275
    },
    {
      "entropy": 5.995220470428467,
      "epoch": 0.20682148040638607,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998477839262611,
      "loss": 5.9789,
      "mean_token_accuracy": 0.13753922730684282,
      "num_tokens": 4743428.0,
      "step": 2280
    },
    {
      "entropy": 5.996651983261108,
      "epoch": 0.20727503628447025,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499846591617662,
      "loss": 5.9059,
      "mean_token_accuracy": 0.14165481328964233,
      "num_tokens": 4754530.0,
      "step": 2285
    },
    {
      "entropy": 5.848381853103637,
      "epoch": 0.20772859216255443,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998453946591779,
      "loss": 5.8387,
      "mean_token_accuracy": 0.14721979647874833,
      "num_tokens": 4764453.0,
      "step": 2290
    },
    {
      "entropy": 6.030099821090698,
      "epoch": 0.20818214804063861,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998441930508339,
      "loss": 6.0122,
      "mean_token_accuracy": 0.14176427572965622,
      "num_tokens": 4774788.0,
      "step": 2295
    },
    {
      "entropy": 5.991954565048218,
      "epoch": 0.2086357039187228,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998429867926547,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1358835816383362,
      "num_tokens": 4785827.0,
      "step": 2300
    },
    {
      "entropy": 5.9251960754394535,
      "epoch": 0.20908925979680695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998417758846651,
      "loss": 5.836,
      "mean_token_accuracy": 0.15004699677228928,
      "num_tokens": 4795683.0,
      "step": 2305
    },
    {
      "entropy": 5.960970211029053,
      "epoch": 0.20954281567489114,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998405603268903,
      "loss": 5.9322,
      "mean_token_accuracy": 0.14288516640663146,
      "num_tokens": 4805929.0,
      "step": 2310
    },
    {
      "entropy": 5.915129613876343,
      "epoch": 0.20999637155297532,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998393401193554,
      "loss": 5.9297,
      "mean_token_accuracy": 0.14064979255199433,
      "num_tokens": 4816067.0,
      "step": 2315
    },
    {
      "entropy": 5.940646457672119,
      "epoch": 0.2104499274310595,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998381152620857,
      "loss": 5.9153,
      "mean_token_accuracy": 0.1419698029756546,
      "num_tokens": 4826757.0,
      "step": 2320
    },
    {
      "entropy": 5.9893388748168945,
      "epoch": 0.21090348330914369,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998368857551067,
      "loss": 5.9248,
      "mean_token_accuracy": 0.14424414560198784,
      "num_tokens": 4836548.0,
      "step": 2325
    },
    {
      "entropy": 6.0586730480194095,
      "epoch": 0.21135703918722787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998356515984433,
      "loss": 5.9785,
      "mean_token_accuracy": 0.13741380125284194,
      "num_tokens": 4847274.0,
      "step": 2330
    },
    {
      "entropy": 5.959125995635986,
      "epoch": 0.21181059506531205,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998344127921213,
      "loss": 5.9483,
      "mean_token_accuracy": 0.1400488071143627,
      "num_tokens": 4857506.0,
      "step": 2335
    },
    {
      "entropy": 5.960813760757446,
      "epoch": 0.21226415094339623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998331693361666,
      "loss": 5.9773,
      "mean_token_accuracy": 0.1386716030538082,
      "num_tokens": 4868069.0,
      "step": 2340
    },
    {
      "entropy": 5.946921539306641,
      "epoch": 0.21271770682148042,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998319212306046,
      "loss": 5.8775,
      "mean_token_accuracy": 0.14960167706012725,
      "num_tokens": 4877731.0,
      "step": 2345
    },
    {
      "entropy": 5.923819208145142,
      "epoch": 0.21317126269956457,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499830668475461,
      "loss": 5.9369,
      "mean_token_accuracy": 0.13987276405096055,
      "num_tokens": 4887780.0,
      "step": 2350
    },
    {
      "entropy": 5.958174419403076,
      "epoch": 0.21362481857764876,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499829411070762,
      "loss": 5.9577,
      "mean_token_accuracy": 0.14128173291683196,
      "num_tokens": 4897578.0,
      "step": 2355
    },
    {
      "entropy": 5.871128082275391,
      "epoch": 0.21407837445573294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998281490165335,
      "loss": 5.9053,
      "mean_token_accuracy": 0.13984225764870645,
      "num_tokens": 4908187.0,
      "step": 2360
    },
    {
      "entropy": 6.030574989318848,
      "epoch": 0.21453193033381712,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998268823128016,
      "loss": 5.9289,
      "mean_token_accuracy": 0.14696599096059798,
      "num_tokens": 4918352.0,
      "step": 2365
    },
    {
      "entropy": 5.877168655395508,
      "epoch": 0.2149854862119013,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998256109595923,
      "loss": 5.8947,
      "mean_token_accuracy": 0.14657547250390052,
      "num_tokens": 4928065.0,
      "step": 2370
    },
    {
      "entropy": 5.952869129180908,
      "epoch": 0.2154390420899855,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998243349569323,
      "loss": 5.9084,
      "mean_token_accuracy": 0.14537300616502763,
      "num_tokens": 4939492.0,
      "step": 2375
    },
    {
      "entropy": 5.872677612304687,
      "epoch": 0.21589259796806967,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998230543048477,
      "loss": 5.8546,
      "mean_token_accuracy": 0.14155039489269255,
      "num_tokens": 4949304.0,
      "step": 2380
    },
    {
      "entropy": 5.894457006454468,
      "epoch": 0.21634615384615385,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499821769003365,
      "loss": 5.8853,
      "mean_token_accuracy": 0.1352695107460022,
      "num_tokens": 4959649.0,
      "step": 2385
    },
    {
      "entropy": 5.9166566848754885,
      "epoch": 0.21679970972423804,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998204790525109,
      "loss": 5.9306,
      "mean_token_accuracy": 0.13895038962364198,
      "num_tokens": 4970029.0,
      "step": 2390
    },
    {
      "entropy": 5.966705751419068,
      "epoch": 0.2172532656023222,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998191844523119,
      "loss": 5.8501,
      "mean_token_accuracy": 0.14436866864562034,
      "num_tokens": 4980431.0,
      "step": 2395
    },
    {
      "entropy": 5.862818002700806,
      "epoch": 0.21770682148040638,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499817885202795,
      "loss": 5.9549,
      "mean_token_accuracy": 0.14049403220415116,
      "num_tokens": 4991448.0,
      "step": 2400
    },
    {
      "entropy": 5.938876914978027,
      "epoch": 0.21816037735849056,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998165813039869,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1356552518904209,
      "num_tokens": 5001225.0,
      "step": 2405
    },
    {
      "entropy": 5.924232530593872,
      "epoch": 0.21861393323657474,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998152727559145,
      "loss": 5.8808,
      "mean_token_accuracy": 0.14155457615852357,
      "num_tokens": 5012096.0,
      "step": 2410
    },
    {
      "entropy": 5.954684495925903,
      "epoch": 0.21906748911465893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000499813959558605,
      "loss": 5.9579,
      "mean_token_accuracy": 0.13705262392759324,
      "num_tokens": 5021675.0,
      "step": 2415
    },
    {
      "entropy": 5.930799150466919,
      "epoch": 0.2195210449927431,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998126417120856,
      "loss": 5.9114,
      "mean_token_accuracy": 0.1381124421954155,
      "num_tokens": 5032250.0,
      "step": 2420
    },
    {
      "entropy": 5.906846952438355,
      "epoch": 0.2199746008708273,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998113192163835,
      "loss": 5.9758,
      "mean_token_accuracy": 0.13671192824840545,
      "num_tokens": 5042508.0,
      "step": 2425
    },
    {
      "entropy": 6.031311511993408,
      "epoch": 0.22042815674891147,
      "grad_norm": 1.28125,
      "learning_rate": 0.000499809992071526,
      "loss": 5.8645,
      "mean_token_accuracy": 0.14089929684996605,
      "num_tokens": 5052765.0,
      "step": 2430
    },
    {
      "entropy": 5.894246578216553,
      "epoch": 0.22088171262699566,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998086602775406,
      "loss": 5.8884,
      "mean_token_accuracy": 0.14432070553302764,
      "num_tokens": 5062379.0,
      "step": 2435
    },
    {
      "entropy": 5.954690265655517,
      "epoch": 0.2213352685050798,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004998073238344547,
      "loss": 5.9271,
      "mean_token_accuracy": 0.14122223407030104,
      "num_tokens": 5072267.0,
      "step": 2440
    },
    {
      "entropy": 5.857289552688599,
      "epoch": 0.221788824383164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998059827422962,
      "loss": 5.8285,
      "mean_token_accuracy": 0.1439412847161293,
      "num_tokens": 5082214.0,
      "step": 2445
    },
    {
      "entropy": 5.875279474258423,
      "epoch": 0.22224238026124818,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004998046370010926,
      "loss": 5.8849,
      "mean_token_accuracy": 0.14643458500504494,
      "num_tokens": 5092129.0,
      "step": 2450
    },
    {
      "entropy": 5.914616823196411,
      "epoch": 0.22269593613933236,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998032866108718,
      "loss": 5.903,
      "mean_token_accuracy": 0.13629107251763345,
      "num_tokens": 5102924.0,
      "step": 2455
    },
    {
      "entropy": 5.9880273818969725,
      "epoch": 0.22314949201741655,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998019315716618,
      "loss": 5.8787,
      "mean_token_accuracy": 0.1420026421546936,
      "num_tokens": 5113216.0,
      "step": 2460
    },
    {
      "entropy": 5.792202377319336,
      "epoch": 0.22360304789550073,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004998005718834905,
      "loss": 5.8578,
      "mean_token_accuracy": 0.1403028056025505,
      "num_tokens": 5122978.0,
      "step": 2465
    },
    {
      "entropy": 5.964217710494995,
      "epoch": 0.2240566037735849,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997992075463862,
      "loss": 5.9519,
      "mean_token_accuracy": 0.14088138267397882,
      "num_tokens": 5133167.0,
      "step": 2470
    },
    {
      "entropy": 5.898852062225342,
      "epoch": 0.2245101596516691,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997978385603769,
      "loss": 5.8241,
      "mean_token_accuracy": 0.14743937328457832,
      "num_tokens": 5143848.0,
      "step": 2475
    },
    {
      "entropy": 5.9200560569763185,
      "epoch": 0.22496371552975328,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997964649254911,
      "loss": 5.8611,
      "mean_token_accuracy": 0.15123707801103592,
      "num_tokens": 5154172.0,
      "step": 2480
    },
    {
      "entropy": 5.81275691986084,
      "epoch": 0.22541727140783743,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997950866417572,
      "loss": 5.8186,
      "mean_token_accuracy": 0.1486186221241951,
      "num_tokens": 5164459.0,
      "step": 2485
    },
    {
      "entropy": 5.979988098144531,
      "epoch": 0.22587082728592162,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997937037092034,
      "loss": 5.9206,
      "mean_token_accuracy": 0.13727832362055778,
      "num_tokens": 5175751.0,
      "step": 2490
    },
    {
      "entropy": 5.904782056808472,
      "epoch": 0.2263243831640058,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997923161278587,
      "loss": 5.836,
      "mean_token_accuracy": 0.1466525003314018,
      "num_tokens": 5186004.0,
      "step": 2495
    },
    {
      "entropy": 5.8142805099487305,
      "epoch": 0.22677793904208998,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997909238977514,
      "loss": 5.7535,
      "mean_token_accuracy": 0.1469535768032074,
      "num_tokens": 5195610.0,
      "step": 2500
    },
    {
      "entropy": 5.915460872650146,
      "epoch": 0.22723149492017417,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004997895270189108,
      "loss": 5.9106,
      "mean_token_accuracy": 0.1398490011692047,
      "num_tokens": 5205607.0,
      "step": 2505
    },
    {
      "entropy": 5.935828447341919,
      "epoch": 0.22768505079825835,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997881254913653,
      "loss": 5.8928,
      "mean_token_accuracy": 0.1415001295506954,
      "num_tokens": 5217372.0,
      "step": 2510
    },
    {
      "entropy": 5.895864343643188,
      "epoch": 0.22813860667634253,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997867193151441,
      "loss": 5.8553,
      "mean_token_accuracy": 0.14375398457050323,
      "num_tokens": 5226782.0,
      "step": 2515
    },
    {
      "entropy": 5.875649499893188,
      "epoch": 0.22859216255442671,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004997853084902763,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1496797136962414,
      "num_tokens": 5236209.0,
      "step": 2520
    },
    {
      "entropy": 5.870350313186646,
      "epoch": 0.2290457184325109,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997838930167911,
      "loss": 5.8964,
      "mean_token_accuracy": 0.14222536683082582,
      "num_tokens": 5246041.0,
      "step": 2525
    },
    {
      "entropy": 5.970703125,
      "epoch": 0.22949927431059505,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997824728947176,
      "loss": 5.8557,
      "mean_token_accuracy": 0.14660589545965194,
      "num_tokens": 5256902.0,
      "step": 2530
    },
    {
      "entropy": 5.928110027313233,
      "epoch": 0.22995283018867924,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997810481240853,
      "loss": 5.8547,
      "mean_token_accuracy": 0.14946367517113684,
      "num_tokens": 5266180.0,
      "step": 2535
    },
    {
      "entropy": 5.936565351486206,
      "epoch": 0.23040638606676342,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997796187049238,
      "loss": 6.0483,
      "mean_token_accuracy": 0.13646556437015533,
      "num_tokens": 5277475.0,
      "step": 2540
    },
    {
      "entropy": 6.029916381835937,
      "epoch": 0.2308599419448476,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997781846372623,
      "loss": 5.8693,
      "mean_token_accuracy": 0.14439473375678064,
      "num_tokens": 5287108.0,
      "step": 2545
    },
    {
      "entropy": 5.854767751693726,
      "epoch": 0.23131349782293179,
      "grad_norm": 1.25,
      "learning_rate": 0.0004997767459211307,
      "loss": 5.9654,
      "mean_token_accuracy": 0.13685449138283728,
      "num_tokens": 5297598.0,
      "step": 2550
    },
    {
      "entropy": 6.062006139755249,
      "epoch": 0.23176705370101597,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997753025565587,
      "loss": 5.9442,
      "mean_token_accuracy": 0.1358692966401577,
      "num_tokens": 5306953.0,
      "step": 2555
    },
    {
      "entropy": 5.887425899505615,
      "epoch": 0.23222060957910015,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997738545435763,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15026361793279647,
      "num_tokens": 5317608.0,
      "step": 2560
    },
    {
      "entropy": 5.862730932235718,
      "epoch": 0.23267416545718433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997724018822133,
      "loss": 5.8058,
      "mean_token_accuracy": 0.13242100477218627,
      "num_tokens": 5328503.0,
      "step": 2565
    },
    {
      "entropy": 5.914726114273071,
      "epoch": 0.23312772133526852,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997709445724996,
      "loss": 5.883,
      "mean_token_accuracy": 0.14427884221076964,
      "num_tokens": 5338108.0,
      "step": 2570
    },
    {
      "entropy": 5.95159878730774,
      "epoch": 0.23358127721335267,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997694826144656,
      "loss": 6.04,
      "mean_token_accuracy": 0.13466350510716438,
      "num_tokens": 5349890.0,
      "step": 2575
    },
    {
      "entropy": 6.051620054244995,
      "epoch": 0.23403483309143686,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997680160081414,
      "loss": 6.0118,
      "mean_token_accuracy": 0.1407109409570694,
      "num_tokens": 5361319.0,
      "step": 2580
    },
    {
      "entropy": 5.919955205917359,
      "epoch": 0.23448838896952104,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997665447535573,
      "loss": 5.9219,
      "mean_token_accuracy": 0.14855802059173584,
      "num_tokens": 5370932.0,
      "step": 2585
    },
    {
      "entropy": 5.9046718120574955,
      "epoch": 0.23494194484760522,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004997650688507438,
      "loss": 5.843,
      "mean_token_accuracy": 0.14306004494428634,
      "num_tokens": 5381661.0,
      "step": 2590
    },
    {
      "entropy": 5.818372058868408,
      "epoch": 0.2353955007256894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004997635882997314,
      "loss": 5.8143,
      "mean_token_accuracy": 0.14963757395744323,
      "num_tokens": 5390999.0,
      "step": 2595
    },
    {
      "entropy": 5.943971872329712,
      "epoch": 0.2358490566037736,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004997621031005507,
      "loss": 5.9085,
      "mean_token_accuracy": 0.14275056049227713,
      "num_tokens": 5401175.0,
      "step": 2600
    },
    {
      "entropy": 5.954121685028076,
      "epoch": 0.23630261248185777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997606132532325,
      "loss": 5.8747,
      "mean_token_accuracy": 0.14469726160168647,
      "num_tokens": 5411644.0,
      "step": 2605
    },
    {
      "entropy": 5.812447595596313,
      "epoch": 0.23675616835994195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997591187578075,
      "loss": 5.8496,
      "mean_token_accuracy": 0.14814821630716324,
      "num_tokens": 5420923.0,
      "step": 2610
    },
    {
      "entropy": 5.941586923599243,
      "epoch": 0.23720972423802614,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997576196143067,
      "loss": 5.8974,
      "mean_token_accuracy": 0.14263342320919037,
      "num_tokens": 5431475.0,
      "step": 2615
    },
    {
      "entropy": 5.9011571407318115,
      "epoch": 0.2376632801161103,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499756115822761,
      "loss": 5.8093,
      "mean_token_accuracy": 0.14849388152360915,
      "num_tokens": 5441822.0,
      "step": 2620
    },
    {
      "entropy": 5.8005060195922855,
      "epoch": 0.23811683599419448,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997546073832015,
      "loss": 5.8219,
      "mean_token_accuracy": 0.13821850046515466,
      "num_tokens": 5453266.0,
      "step": 2625
    },
    {
      "entropy": 5.917859411239624,
      "epoch": 0.23857039187227866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997530942956595,
      "loss": 5.8634,
      "mean_token_accuracy": 0.14885092675685882,
      "num_tokens": 5463023.0,
      "step": 2630
    },
    {
      "entropy": 5.934016132354737,
      "epoch": 0.23902394775036284,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997515765601662,
      "loss": 5.8566,
      "mean_token_accuracy": 0.14655044227838515,
      "num_tokens": 5472345.0,
      "step": 2635
    },
    {
      "entropy": 5.923844289779663,
      "epoch": 0.23947750362844702,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997500541767531,
      "loss": 5.9059,
      "mean_token_accuracy": 0.1498401679098606,
      "num_tokens": 5481468.0,
      "step": 2640
    },
    {
      "entropy": 5.773250436782837,
      "epoch": 0.2399310595065312,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997485271454516,
      "loss": 5.8461,
      "mean_token_accuracy": 0.14630547761917115,
      "num_tokens": 5492107.0,
      "step": 2645
    },
    {
      "entropy": 5.871908712387085,
      "epoch": 0.2403846153846154,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004997469954662933,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1383388437330723,
      "num_tokens": 5502196.0,
      "step": 2650
    },
    {
      "entropy": 5.904657554626465,
      "epoch": 0.24083817126269957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997454591393097,
      "loss": 5.8693,
      "mean_token_accuracy": 0.1385428823530674,
      "num_tokens": 5512859.0,
      "step": 2655
    },
    {
      "entropy": 5.975746059417725,
      "epoch": 0.24129172714078376,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997439181645329,
      "loss": 5.8498,
      "mean_token_accuracy": 0.14605885297060012,
      "num_tokens": 5523041.0,
      "step": 2660
    },
    {
      "entropy": 5.764354801177978,
      "epoch": 0.2417452830188679,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997423725419945,
      "loss": 5.761,
      "mean_token_accuracy": 0.15253512784838677,
      "num_tokens": 5533967.0,
      "step": 2665
    },
    {
      "entropy": 5.947396564483642,
      "epoch": 0.2421988388969521,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997408222717267,
      "loss": 5.8664,
      "mean_token_accuracy": 0.14548483043909072,
      "num_tokens": 5543460.0,
      "step": 2670
    },
    {
      "entropy": 5.93350305557251,
      "epoch": 0.24265239477503628,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004997392673537612,
      "loss": 5.8592,
      "mean_token_accuracy": 0.1520221307873726,
      "num_tokens": 5553139.0,
      "step": 2675
    },
    {
      "entropy": 5.793046188354492,
      "epoch": 0.24310595065312046,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997377077881306,
      "loss": 5.8541,
      "mean_token_accuracy": 0.1441502094268799,
      "num_tokens": 5564550.0,
      "step": 2680
    },
    {
      "entropy": 5.9918012619018555,
      "epoch": 0.24355950653120464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997361435748668,
      "loss": 5.9546,
      "mean_token_accuracy": 0.1459794595837593,
      "num_tokens": 5574393.0,
      "step": 2685
    },
    {
      "entropy": 5.940083932876587,
      "epoch": 0.24401306240928883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004997345747140024,
      "loss": 5.8813,
      "mean_token_accuracy": 0.14294293597340585,
      "num_tokens": 5584206.0,
      "step": 2690
    },
    {
      "entropy": 5.877155923843384,
      "epoch": 0.244466618287373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997330012055695,
      "loss": 5.8873,
      "mean_token_accuracy": 0.1417349547147751,
      "num_tokens": 5594946.0,
      "step": 2695
    },
    {
      "entropy": 5.921493816375732,
      "epoch": 0.2449201741654572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997314230496011,
      "loss": 5.7744,
      "mean_token_accuracy": 0.14902272820472717,
      "num_tokens": 5605984.0,
      "step": 2700
    },
    {
      "entropy": 5.776959466934204,
      "epoch": 0.24537373004354138,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997298402461296,
      "loss": 5.8131,
      "mean_token_accuracy": 0.1488720089197159,
      "num_tokens": 5616227.0,
      "step": 2705
    },
    {
      "entropy": 5.935484266281128,
      "epoch": 0.24582728592162553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997282527951876,
      "loss": 5.9032,
      "mean_token_accuracy": 0.14310448318719865,
      "num_tokens": 5626592.0,
      "step": 2710
    },
    {
      "entropy": 6.0017767429351805,
      "epoch": 0.24628084179970972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004997266606968081,
      "loss": 5.8379,
      "mean_token_accuracy": 0.14523325935006143,
      "num_tokens": 5636486.0,
      "step": 2715
    },
    {
      "entropy": 5.846785545349121,
      "epoch": 0.2467343976777939,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004997250639510241,
      "loss": 5.9128,
      "mean_token_accuracy": 0.14552611336112023,
      "num_tokens": 5646025.0,
      "step": 2720
    },
    {
      "entropy": 5.828639936447144,
      "epoch": 0.24718795355587808,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004997234625578684,
      "loss": 5.7968,
      "mean_token_accuracy": 0.1471302717924118,
      "num_tokens": 5656523.0,
      "step": 2725
    },
    {
      "entropy": 5.8145918369293215,
      "epoch": 0.24764150943396226,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997218565173742,
      "loss": 5.8303,
      "mean_token_accuracy": 0.14437200129032135,
      "num_tokens": 5667598.0,
      "step": 2730
    },
    {
      "entropy": 5.928363561630249,
      "epoch": 0.24809506531204645,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997202458295748,
      "loss": 5.8606,
      "mean_token_accuracy": 0.14459095299243926,
      "num_tokens": 5677721.0,
      "step": 2735
    },
    {
      "entropy": 5.847347164154053,
      "epoch": 0.24854862119013063,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997186304945034,
      "loss": 5.8416,
      "mean_token_accuracy": 0.1417038306593895,
      "num_tokens": 5688450.0,
      "step": 2740
    },
    {
      "entropy": 5.865340757369995,
      "epoch": 0.24900217706821481,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997170105121935,
      "loss": 5.8208,
      "mean_token_accuracy": 0.1495169907808304,
      "num_tokens": 5698142.0,
      "step": 2745
    },
    {
      "entropy": 5.833406114578247,
      "epoch": 0.249455732946299,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997153858826786,
      "loss": 5.8201,
      "mean_token_accuracy": 0.1465844228863716,
      "num_tokens": 5708122.0,
      "step": 2750
    },
    {
      "entropy": 5.837479496002198,
      "epoch": 0.24990928882438315,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997137566059922,
      "loss": 5.8107,
      "mean_token_accuracy": 0.14600400999188423,
      "num_tokens": 5718671.0,
      "step": 2755
    },
    {
      "entropy": 5.8692303657531735,
      "epoch": 0.25036284470246734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004997121226821681,
      "loss": 5.8171,
      "mean_token_accuracy": 0.14405162036418914,
      "num_tokens": 5729463.0,
      "step": 2760
    },
    {
      "entropy": 5.922516870498657,
      "epoch": 0.25081640058055155,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049971048411124,
      "loss": 5.8874,
      "mean_token_accuracy": 0.14321258664131165,
      "num_tokens": 5739212.0,
      "step": 2765
    },
    {
      "entropy": 5.901452112197876,
      "epoch": 0.2512699564586357,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004997088408932418,
      "loss": 5.8029,
      "mean_token_accuracy": 0.14497288316488266,
      "num_tokens": 5749159.0,
      "step": 2770
    },
    {
      "entropy": 5.932657766342163,
      "epoch": 0.25172351233671986,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997071930282076,
      "loss": 5.993,
      "mean_token_accuracy": 0.13827754259109498,
      "num_tokens": 5759711.0,
      "step": 2775
    },
    {
      "entropy": 6.124020576477051,
      "epoch": 0.25217706821480407,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004997055405161712,
      "loss": 6.0043,
      "mean_token_accuracy": 0.1372769333422184,
      "num_tokens": 5771210.0,
      "step": 2780
    },
    {
      "entropy": 5.816952276229858,
      "epoch": 0.2526306240928882,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997038833571671,
      "loss": 5.8426,
      "mean_token_accuracy": 0.14726067781448365,
      "num_tokens": 5780960.0,
      "step": 2785
    },
    {
      "entropy": 5.935285472869873,
      "epoch": 0.25308417997097243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997022215512295,
      "loss": 5.885,
      "mean_token_accuracy": 0.14095496237277985,
      "num_tokens": 5790891.0,
      "step": 2790
    },
    {
      "entropy": 5.894222116470337,
      "epoch": 0.2535377358490566,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997005550983925,
      "loss": 5.8396,
      "mean_token_accuracy": 0.14395667985081673,
      "num_tokens": 5801078.0,
      "step": 2795
    },
    {
      "entropy": 5.885415887832641,
      "epoch": 0.2539912917271408,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000499698883998691,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1393453024327755,
      "num_tokens": 5811025.0,
      "step": 2800
    },
    {
      "entropy": 5.909905290603637,
      "epoch": 0.25444484760522496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996972082521591,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1448357567191124,
      "num_tokens": 5821242.0,
      "step": 2805
    },
    {
      "entropy": 5.845452737808228,
      "epoch": 0.25489840348330917,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996955278588319,
      "loss": 5.8592,
      "mean_token_accuracy": 0.14133516177535058,
      "num_tokens": 5832039.0,
      "step": 2810
    },
    {
      "entropy": 5.914500284194946,
      "epoch": 0.2553519593613933,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996938428187437,
      "loss": 5.8397,
      "mean_token_accuracy": 0.1460021823644638,
      "num_tokens": 5841723.0,
      "step": 2815
    },
    {
      "entropy": 5.9476793766021725,
      "epoch": 0.2558055152394775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004996921531319297,
      "loss": 5.8873,
      "mean_token_accuracy": 0.14610021114349364,
      "num_tokens": 5852814.0,
      "step": 2820
    },
    {
      "entropy": 5.826781702041626,
      "epoch": 0.2562590711175617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004996904587984246,
      "loss": 5.8323,
      "mean_token_accuracy": 0.14333641827106475,
      "num_tokens": 5863174.0,
      "step": 2825
    },
    {
      "entropy": 5.853058385848999,
      "epoch": 0.25671262699564584,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996887598182636,
      "loss": 5.7837,
      "mean_token_accuracy": 0.14737775102257727,
      "num_tokens": 5873698.0,
      "step": 2830
    },
    {
      "entropy": 5.760656118392944,
      "epoch": 0.25716618287373005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996870561914819,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1456883355975151,
      "num_tokens": 5883982.0,
      "step": 2835
    },
    {
      "entropy": 5.894684791564941,
      "epoch": 0.2576197387518142,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996853479181145,
      "loss": 5.7891,
      "mean_token_accuracy": 0.14917163103818892,
      "num_tokens": 5894454.0,
      "step": 2840
    },
    {
      "entropy": 5.844020223617553,
      "epoch": 0.2580732946298984,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996836349981969,
      "loss": 5.8806,
      "mean_token_accuracy": 0.1445062406361103,
      "num_tokens": 5904739.0,
      "step": 2845
    },
    {
      "entropy": 5.931279706954956,
      "epoch": 0.2585268505079826,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004996819174317645,
      "loss": 5.8601,
      "mean_token_accuracy": 0.14103500097990035,
      "num_tokens": 5914578.0,
      "step": 2850
    },
    {
      "entropy": 5.924049043655396,
      "epoch": 0.2589804063860668,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004996801952188528,
      "loss": 5.8417,
      "mean_token_accuracy": 0.1458780586719513,
      "num_tokens": 5923918.0,
      "step": 2855
    },
    {
      "entropy": 5.938410520553589,
      "epoch": 0.25943396226415094,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996784683594973,
      "loss": 5.8868,
      "mean_token_accuracy": 0.14107644706964492,
      "num_tokens": 5934096.0,
      "step": 2860
    },
    {
      "entropy": 5.885492944717408,
      "epoch": 0.2598875181422351,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996767368537338,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1389304779469967,
      "num_tokens": 5945416.0,
      "step": 2865
    },
    {
      "entropy": 5.875432109832763,
      "epoch": 0.2603410740203193,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004996750007015983,
      "loss": 5.8406,
      "mean_token_accuracy": 0.14503862857818603,
      "num_tokens": 5956296.0,
      "step": 2870
    },
    {
      "entropy": 5.864486980438232,
      "epoch": 0.26079462989840346,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996732599031264,
      "loss": 5.733,
      "mean_token_accuracy": 0.15628641992807388,
      "num_tokens": 5966147.0,
      "step": 2875
    },
    {
      "entropy": 5.841119432449341,
      "epoch": 0.2612481857764877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004996715144583543,
      "loss": 5.8385,
      "mean_token_accuracy": 0.15209657400846482,
      "num_tokens": 5976650.0,
      "step": 2880
    },
    {
      "entropy": 5.918705177307129,
      "epoch": 0.26170174165457183,
      "grad_norm": 1.203125,
      "learning_rate": 0.000499669764367318,
      "loss": 5.9027,
      "mean_token_accuracy": 0.14683974608778955,
      "num_tokens": 5986643.0,
      "step": 2885
    },
    {
      "entropy": 5.891564416885376,
      "epoch": 0.26215529753265604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004996680096300537,
      "loss": 5.8697,
      "mean_token_accuracy": 0.14015998244285582,
      "num_tokens": 5997991.0,
      "step": 2890
    },
    {
      "entropy": 5.783881521224975,
      "epoch": 0.2626088534107402,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996662502465978,
      "loss": 5.6945,
      "mean_token_accuracy": 0.15556951612234116,
      "num_tokens": 6007179.0,
      "step": 2895
    },
    {
      "entropy": 5.919160652160644,
      "epoch": 0.2630624092888244,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996644862169864,
      "loss": 5.8789,
      "mean_token_accuracy": 0.14572433680295943,
      "num_tokens": 6017786.0,
      "step": 2900
    },
    {
      "entropy": 5.915647411346436,
      "epoch": 0.26351596516690856,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004996627175412563,
      "loss": 5.9004,
      "mean_token_accuracy": 0.14329932779073715,
      "num_tokens": 6027751.0,
      "step": 2905
    },
    {
      "entropy": 5.8271812915802,
      "epoch": 0.2639695210449927,
      "grad_norm": 1.390625,
      "learning_rate": 0.000499660944219444,
      "loss": 5.7774,
      "mean_token_accuracy": 0.14923276603221894,
      "num_tokens": 6037136.0,
      "step": 2910
    },
    {
      "entropy": 5.790592765808105,
      "epoch": 0.2644230769230769,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996591662515861,
      "loss": 5.7695,
      "mean_token_accuracy": 0.14568121135234832,
      "num_tokens": 6047341.0,
      "step": 2915
    },
    {
      "entropy": 5.903992128372193,
      "epoch": 0.2648766328011611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996573836377193,
      "loss": 5.9207,
      "mean_token_accuracy": 0.14674188792705536,
      "num_tokens": 6058337.0,
      "step": 2920
    },
    {
      "entropy": 5.892431545257568,
      "epoch": 0.2653301886792453,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996555963778806,
      "loss": 5.8044,
      "mean_token_accuracy": 0.14313996583223343,
      "num_tokens": 6068566.0,
      "step": 2925
    },
    {
      "entropy": 5.918859958648682,
      "epoch": 0.26578374455732945,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499653804472107,
      "loss": 5.8215,
      "mean_token_accuracy": 0.15070558488368987,
      "num_tokens": 6078813.0,
      "step": 2930
    },
    {
      "entropy": 5.816320133209229,
      "epoch": 0.26623730043541366,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996520079204353,
      "loss": 5.8376,
      "mean_token_accuracy": 0.15071689635515212,
      "num_tokens": 6090007.0,
      "step": 2935
    },
    {
      "entropy": 5.924297523498535,
      "epoch": 0.2666908563134978,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499650206722903,
      "loss": 5.9844,
      "mean_token_accuracy": 0.1373060829937458,
      "num_tokens": 6100917.0,
      "step": 2940
    },
    {
      "entropy": 6.015955924987793,
      "epoch": 0.267144412191582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996484008795471,
      "loss": 5.93,
      "mean_token_accuracy": 0.147200308740139,
      "num_tokens": 6110958.0,
      "step": 2945
    },
    {
      "entropy": 5.904809999465942,
      "epoch": 0.2675979680696662,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499646590390405,
      "loss": 5.8316,
      "mean_token_accuracy": 0.14588915333151817,
      "num_tokens": 6120819.0,
      "step": 2950
    },
    {
      "entropy": 5.836217546463013,
      "epoch": 0.26805152394775034,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004996447752555142,
      "loss": 5.8205,
      "mean_token_accuracy": 0.14316449761390687,
      "num_tokens": 6131253.0,
      "step": 2955
    },
    {
      "entropy": 5.898743295669556,
      "epoch": 0.26850507982583455,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996429554749122,
      "loss": 5.8484,
      "mean_token_accuracy": 0.148350540548563,
      "num_tokens": 6141338.0,
      "step": 2960
    },
    {
      "entropy": 5.846032381057739,
      "epoch": 0.2689586357039187,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996411310486367,
      "loss": 5.8693,
      "mean_token_accuracy": 0.14200861528515815,
      "num_tokens": 6151081.0,
      "step": 2965
    },
    {
      "entropy": 5.9468964576721195,
      "epoch": 0.2694121915820029,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004996393019767253,
      "loss": 5.8606,
      "mean_token_accuracy": 0.13914820924401283,
      "num_tokens": 6160997.0,
      "step": 2970
    },
    {
      "entropy": 5.878990554809571,
      "epoch": 0.26986574746008707,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004996374682592158,
      "loss": 5.7723,
      "mean_token_accuracy": 0.14771230965852739,
      "num_tokens": 6171131.0,
      "step": 2975
    },
    {
      "entropy": 5.830885028839111,
      "epoch": 0.2703193033381713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004996356298961463,
      "loss": 5.8911,
      "mean_token_accuracy": 0.14644451886415483,
      "num_tokens": 6181307.0,
      "step": 2980
    },
    {
      "entropy": 5.935413074493408,
      "epoch": 0.27077285921625543,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004996337868875549,
      "loss": 5.7345,
      "mean_token_accuracy": 0.15274717211723327,
      "num_tokens": 6191689.0,
      "step": 2985
    },
    {
      "entropy": 5.780738258361817,
      "epoch": 0.27122641509433965,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996319392334792,
      "loss": 5.8039,
      "mean_token_accuracy": 0.14987841248512268,
      "num_tokens": 6201183.0,
      "step": 2990
    },
    {
      "entropy": 5.876131439208985,
      "epoch": 0.2716799709724238,
      "grad_norm": 1.453125,
      "learning_rate": 0.000499630086933958,
      "loss": 5.897,
      "mean_token_accuracy": 0.14560322165489198,
      "num_tokens": 6211507.0,
      "step": 2995
    },
    {
      "entropy": 5.882185268402099,
      "epoch": 0.27213352685050796,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996282299890292,
      "loss": 5.758,
      "mean_token_accuracy": 0.15807864367961882,
      "num_tokens": 6220899.0,
      "step": 3000
    },
    {
      "epoch": 0.27213352685050796,
      "eval_entropy": 5.707925649789663,
      "eval_loss": 5.863901615142822,
      "eval_mean_token_accuracy": 0.15150013840356968,
      "eval_num_tokens": 6220899.0,
      "eval_runtime": 26.8067,
      "eval_samples_per_second": 1319.074,
      "eval_steps_per_second": 164.884,
      "step": 3000
    },
    {
      "entropy": 5.864054012298584,
      "epoch": 0.27258708272859217,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996263683987316,
      "loss": 5.8838,
      "mean_token_accuracy": 0.13928473219275475,
      "num_tokens": 6231591.0,
      "step": 3005
    },
    {
      "entropy": 5.858609437942505,
      "epoch": 0.2730406386066763,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996245021631033,
      "loss": 5.8046,
      "mean_token_accuracy": 0.14948550909757613,
      "num_tokens": 6240991.0,
      "step": 3010
    },
    {
      "entropy": 5.891993761062622,
      "epoch": 0.27349419448476053,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499622631282183,
      "loss": 5.8438,
      "mean_token_accuracy": 0.14875294119119645,
      "num_tokens": 6251081.0,
      "step": 3015
    },
    {
      "entropy": 5.850441408157349,
      "epoch": 0.2739477503628447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004996207557560095,
      "loss": 5.7878,
      "mean_token_accuracy": 0.15141250863671302,
      "num_tokens": 6260557.0,
      "step": 3020
    },
    {
      "entropy": 5.869618844985962,
      "epoch": 0.2744013062409289,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004996188755846216,
      "loss": 5.8934,
      "mean_token_accuracy": 0.14358878880739212,
      "num_tokens": 6272325.0,
      "step": 3025
    },
    {
      "entropy": 5.905278873443604,
      "epoch": 0.27485486211901305,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499616990768058,
      "loss": 5.8365,
      "mean_token_accuracy": 0.145791694521904,
      "num_tokens": 6281672.0,
      "step": 3030
    },
    {
      "entropy": 5.862424087524414,
      "epoch": 0.27530841799709727,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004996151013063579,
      "loss": 5.8855,
      "mean_token_accuracy": 0.14248670935630797,
      "num_tokens": 6292908.0,
      "step": 3035
    },
    {
      "entropy": 5.848897933959961,
      "epoch": 0.2757619738751814,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004996132071995601,
      "loss": 5.8133,
      "mean_token_accuracy": 0.14569919556379318,
      "num_tokens": 6303244.0,
      "step": 3040
    },
    {
      "entropy": 5.873018789291382,
      "epoch": 0.2762155297532656,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996113084477042,
      "loss": 5.9411,
      "mean_token_accuracy": 0.1410228006541729,
      "num_tokens": 6313644.0,
      "step": 3045
    },
    {
      "entropy": 5.883062839508057,
      "epoch": 0.2766690856313498,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004996094050508291,
      "loss": 5.8493,
      "mean_token_accuracy": 0.1453947678208351,
      "num_tokens": 6323498.0,
      "step": 3050
    },
    {
      "entropy": 5.905374526977539,
      "epoch": 0.27712264150943394,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004996074970089741,
      "loss": 5.7756,
      "mean_token_accuracy": 0.14867558106780052,
      "num_tokens": 6333253.0,
      "step": 3055
    },
    {
      "entropy": 5.8678144931793215,
      "epoch": 0.27757619738751815,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004996055843221789,
      "loss": 5.786,
      "mean_token_accuracy": 0.15047635659575462,
      "num_tokens": 6343321.0,
      "step": 3060
    },
    {
      "entropy": 5.950027418136597,
      "epoch": 0.2780297532656023,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996036669904831,
      "loss": 5.9411,
      "mean_token_accuracy": 0.13945802450180053,
      "num_tokens": 6353344.0,
      "step": 3065
    },
    {
      "entropy": 5.920107841491699,
      "epoch": 0.2784833091436865,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996017450139261,
      "loss": 5.8937,
      "mean_token_accuracy": 0.14309242218732834,
      "num_tokens": 6364082.0,
      "step": 3070
    },
    {
      "entropy": 5.91571273803711,
      "epoch": 0.2789368650217707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995998183925478,
      "loss": 5.8514,
      "mean_token_accuracy": 0.14139717817306519,
      "num_tokens": 6375985.0,
      "step": 3075
    },
    {
      "entropy": 5.894878768920899,
      "epoch": 0.2793904208998549,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995978871263881,
      "loss": 5.8078,
      "mean_token_accuracy": 0.14690662622451783,
      "num_tokens": 6386065.0,
      "step": 3080
    },
    {
      "entropy": 5.843625783920288,
      "epoch": 0.27984397677793904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995959512154868,
      "loss": 5.8471,
      "mean_token_accuracy": 0.14574623256921768,
      "num_tokens": 6396732.0,
      "step": 3085
    },
    {
      "entropy": 5.851778793334961,
      "epoch": 0.2802975326560232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499594010659884,
      "loss": 5.9152,
      "mean_token_accuracy": 0.13981067016720772,
      "num_tokens": 6407416.0,
      "step": 3090
    },
    {
      "entropy": 5.908939695358276,
      "epoch": 0.2807510885341074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004995920654596199,
      "loss": 5.7993,
      "mean_token_accuracy": 0.14782450050115586,
      "num_tokens": 6417673.0,
      "step": 3095
    },
    {
      "entropy": 5.937315511703491,
      "epoch": 0.28120464441219156,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004995901156147345,
      "loss": 5.8587,
      "mean_token_accuracy": 0.15051635801792146,
      "num_tokens": 6426771.0,
      "step": 3100
    },
    {
      "entropy": 5.831364393234253,
      "epoch": 0.2816582002902758,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995881611252684,
      "loss": 5.7023,
      "mean_token_accuracy": 0.15109678208827973,
      "num_tokens": 6436906.0,
      "step": 3105
    },
    {
      "entropy": 5.869474411010742,
      "epoch": 0.28211175616835993,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995862019912619,
      "loss": 5.9014,
      "mean_token_accuracy": 0.1423293836414814,
      "num_tokens": 6447253.0,
      "step": 3110
    },
    {
      "entropy": 5.922127056121826,
      "epoch": 0.28256531204644414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995842382127555,
      "loss": 5.8923,
      "mean_token_accuracy": 0.14200060367584227,
      "num_tokens": 6458486.0,
      "step": 3115
    },
    {
      "entropy": 5.880311489105225,
      "epoch": 0.2830188679245283,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004995822697897897,
      "loss": 5.7997,
      "mean_token_accuracy": 0.14626570492982865,
      "num_tokens": 6467614.0,
      "step": 3120
    },
    {
      "entropy": 5.846430110931396,
      "epoch": 0.2834724238026125,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995802967224056,
      "loss": 5.7535,
      "mean_token_accuracy": 0.15461970269680023,
      "num_tokens": 6477857.0,
      "step": 3125
    },
    {
      "entropy": 5.848233366012574,
      "epoch": 0.28392597968069666,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995783190106434,
      "loss": 5.8375,
      "mean_token_accuracy": 0.1500306785106659,
      "num_tokens": 6488091.0,
      "step": 3130
    },
    {
      "entropy": 5.835888004302978,
      "epoch": 0.2843795355587808,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995763366545445,
      "loss": 5.7958,
      "mean_token_accuracy": 0.14885595068335533,
      "num_tokens": 6498725.0,
      "step": 3135
    },
    {
      "entropy": 5.856668138504029,
      "epoch": 0.284833091436865,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004995743496541497,
      "loss": 5.8384,
      "mean_token_accuracy": 0.1368933156132698,
      "num_tokens": 6509850.0,
      "step": 3140
    },
    {
      "entropy": 5.8681940078735355,
      "epoch": 0.2852866473149492,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995723580095001,
      "loss": 5.7962,
      "mean_token_accuracy": 0.1502167746424675,
      "num_tokens": 6519758.0,
      "step": 3145
    },
    {
      "entropy": 5.899721384048462,
      "epoch": 0.2857402031930334,
      "grad_norm": 1.59375,
      "learning_rate": 0.000499570361720637,
      "loss": 5.8057,
      "mean_token_accuracy": 0.1445355698466301,
      "num_tokens": 6530422.0,
      "step": 3150
    },
    {
      "entropy": 5.885447835922241,
      "epoch": 0.28619375907111755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995683607876015,
      "loss": 5.9207,
      "mean_token_accuracy": 0.14514559730887414,
      "num_tokens": 6540785.0,
      "step": 3155
    },
    {
      "entropy": 5.865033054351807,
      "epoch": 0.28664731494920176,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995663552104352,
      "loss": 5.7687,
      "mean_token_accuracy": 0.15479573905467986,
      "num_tokens": 6550830.0,
      "step": 3160
    },
    {
      "entropy": 5.857868719100952,
      "epoch": 0.2871008708272859,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995643449891793,
      "loss": 5.7433,
      "mean_token_accuracy": 0.15413329899311065,
      "num_tokens": 6560913.0,
      "step": 3165
    },
    {
      "entropy": 5.842754697799682,
      "epoch": 0.2875544267053701,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004995623301238757,
      "loss": 5.7869,
      "mean_token_accuracy": 0.14893174916505814,
      "num_tokens": 6570596.0,
      "step": 3170
    },
    {
      "entropy": 5.851306915283203,
      "epoch": 0.2880079825834543,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004995603106145657,
      "loss": 5.7612,
      "mean_token_accuracy": 0.15055156946182252,
      "num_tokens": 6581307.0,
      "step": 3175
    },
    {
      "entropy": 5.850955390930176,
      "epoch": 0.28846153846153844,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004995582864612913,
      "loss": 5.7849,
      "mean_token_accuracy": 0.14817112386226655,
      "num_tokens": 6591917.0,
      "step": 3180
    },
    {
      "entropy": 5.81335563659668,
      "epoch": 0.28891509433962265,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004995562576640943,
      "loss": 5.8041,
      "mean_token_accuracy": 0.14926037788391114,
      "num_tokens": 6602917.0,
      "step": 3185
    },
    {
      "entropy": 5.88020806312561,
      "epoch": 0.2893686502177068,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004995542242230167,
      "loss": 5.767,
      "mean_token_accuracy": 0.14859842211008073,
      "num_tokens": 6612761.0,
      "step": 3190
    },
    {
      "entropy": 5.8096456050872805,
      "epoch": 0.289822206095791,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995521861381006,
      "loss": 5.7917,
      "mean_token_accuracy": 0.15249492973089218,
      "num_tokens": 6623563.0,
      "step": 3195
    },
    {
      "entropy": 5.811558675765991,
      "epoch": 0.29027576197387517,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499550143409388,
      "loss": 5.7817,
      "mean_token_accuracy": 0.15435880571603774,
      "num_tokens": 6633201.0,
      "step": 3200
    },
    {
      "entropy": 5.940117883682251,
      "epoch": 0.2907293178519594,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995480960369211,
      "loss": 5.8436,
      "mean_token_accuracy": 0.14822371006011964,
      "num_tokens": 6643738.0,
      "step": 3205
    },
    {
      "entropy": 5.8637003898620605,
      "epoch": 0.29118287373004353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995460440207425,
      "loss": 5.7337,
      "mean_token_accuracy": 0.15387465804815292,
      "num_tokens": 6654053.0,
      "step": 3210
    },
    {
      "entropy": 5.793497085571289,
      "epoch": 0.29163642960812775,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004995439873608945,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1479906916618347,
      "num_tokens": 6665107.0,
      "step": 3215
    },
    {
      "entropy": 5.8725996017456055,
      "epoch": 0.2920899854862119,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004995419260574196,
      "loss": 5.8963,
      "mean_token_accuracy": 0.14110559597611427,
      "num_tokens": 6675605.0,
      "step": 3220
    },
    {
      "entropy": 5.827268457412719,
      "epoch": 0.29254354136429606,
      "grad_norm": 1.375,
      "learning_rate": 0.0004995398601103603,
      "loss": 5.6143,
      "mean_token_accuracy": 0.1541922315955162,
      "num_tokens": 6684936.0,
      "step": 3225
    },
    {
      "entropy": 5.767840433120727,
      "epoch": 0.29299709724238027,
      "grad_norm": 1.25,
      "learning_rate": 0.0004995377895197597,
      "loss": 5.8262,
      "mean_token_accuracy": 0.14804428666830063,
      "num_tokens": 6694695.0,
      "step": 3230
    },
    {
      "entropy": 5.8415192604064945,
      "epoch": 0.2934506531204644,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004995357142856602,
      "loss": 5.7217,
      "mean_token_accuracy": 0.15065886825323105,
      "num_tokens": 6704862.0,
      "step": 3235
    },
    {
      "entropy": 5.764634227752685,
      "epoch": 0.29390420899854863,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499533634408105,
      "loss": 5.6824,
      "mean_token_accuracy": 0.15840475410223007,
      "num_tokens": 6713879.0,
      "step": 3240
    },
    {
      "entropy": 5.820394563674927,
      "epoch": 0.2943577648766328,
      "grad_norm": 1.359375,
      "learning_rate": 0.000499531549887137,
      "loss": 5.8413,
      "mean_token_accuracy": 0.1459912046790123,
      "num_tokens": 6723435.0,
      "step": 3245
    },
    {
      "entropy": 5.909898471832276,
      "epoch": 0.294811320754717,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995294607227993,
      "loss": 5.7703,
      "mean_token_accuracy": 0.15389380604028702,
      "num_tokens": 6732739.0,
      "step": 3250
    },
    {
      "entropy": 5.78162317276001,
      "epoch": 0.29526487663280115,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004995273669151351,
      "loss": 5.7835,
      "mean_token_accuracy": 0.1508500948548317,
      "num_tokens": 6743044.0,
      "step": 3255
    },
    {
      "entropy": 5.810116577148437,
      "epoch": 0.29571843251088537,
      "grad_norm": 1.328125,
      "learning_rate": 0.000499525268464188,
      "loss": 5.7469,
      "mean_token_accuracy": 0.1529194064438343,
      "num_tokens": 6752319.0,
      "step": 3260
    },
    {
      "entropy": 5.876325845718384,
      "epoch": 0.2961719883889695,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004995231653700009,
      "loss": 5.7262,
      "mean_token_accuracy": 0.14905451089143754,
      "num_tokens": 6762458.0,
      "step": 3265
    },
    {
      "entropy": 5.813838768005371,
      "epoch": 0.2966255442670537,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004995210576326176,
      "loss": 5.8253,
      "mean_token_accuracy": 0.15428584665060044,
      "num_tokens": 6772328.0,
      "step": 3270
    },
    {
      "entropy": 5.903802537918091,
      "epoch": 0.2970791001451379,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004995189452520816,
      "loss": 5.8764,
      "mean_token_accuracy": 0.14373027980327607,
      "num_tokens": 6782754.0,
      "step": 3275
    },
    {
      "entropy": 5.917711305618286,
      "epoch": 0.29753265602322204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995168282284366,
      "loss": 5.8203,
      "mean_token_accuracy": 0.14201931580901145,
      "num_tokens": 6793197.0,
      "step": 3280
    },
    {
      "entropy": 5.823143529891968,
      "epoch": 0.29798621190130625,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995147065617263,
      "loss": 5.7498,
      "mean_token_accuracy": 0.15240444839000702,
      "num_tokens": 6802321.0,
      "step": 3285
    },
    {
      "entropy": 5.773812198638916,
      "epoch": 0.2984397677793904,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004995125802519946,
      "loss": 5.7738,
      "mean_token_accuracy": 0.14767244458198547,
      "num_tokens": 6813541.0,
      "step": 3290
    },
    {
      "entropy": 5.831119203567505,
      "epoch": 0.2988933236574746,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004995104492992857,
      "loss": 5.7783,
      "mean_token_accuracy": 0.1455256797373295,
      "num_tokens": 6824119.0,
      "step": 3295
    },
    {
      "entropy": 5.8984941959381105,
      "epoch": 0.2993468795355588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995083137036434,
      "loss": 5.8166,
      "mean_token_accuracy": 0.154207181930542,
      "num_tokens": 6834315.0,
      "step": 3300
    },
    {
      "entropy": 5.8669695377349855,
      "epoch": 0.299800435413643,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004995061734651121,
      "loss": 5.7905,
      "mean_token_accuracy": 0.14483988136053086,
      "num_tokens": 6844773.0,
      "step": 3305
    },
    {
      "entropy": 5.827662134170533,
      "epoch": 0.30025399129172714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004995040285837358,
      "loss": 5.774,
      "mean_token_accuracy": 0.1427043065428734,
      "num_tokens": 6855865.0,
      "step": 3310
    },
    {
      "entropy": 5.8488105773925785,
      "epoch": 0.3007075471698113,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004995018790595589,
      "loss": 5.7489,
      "mean_token_accuracy": 0.1507802575826645,
      "num_tokens": 6865142.0,
      "step": 3315
    },
    {
      "entropy": 5.828653955459595,
      "epoch": 0.3011611030478955,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994997248926261,
      "loss": 5.7642,
      "mean_token_accuracy": 0.14814960435032845,
      "num_tokens": 6876178.0,
      "step": 3320
    },
    {
      "entropy": 5.855875825881958,
      "epoch": 0.30161465892597966,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994975660829817,
      "loss": 5.886,
      "mean_token_accuracy": 0.14564612358808518,
      "num_tokens": 6887391.0,
      "step": 3325
    },
    {
      "entropy": 5.866506433486938,
      "epoch": 0.3020682148040639,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994954026306705,
      "loss": 5.7831,
      "mean_token_accuracy": 0.14443814307451247,
      "num_tokens": 6897764.0,
      "step": 3330
    },
    {
      "entropy": 5.84368929862976,
      "epoch": 0.30252177068214803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994932345357371,
      "loss": 5.8402,
      "mean_token_accuracy": 0.1517421394586563,
      "num_tokens": 6909251.0,
      "step": 3335
    },
    {
      "entropy": 5.848319673538208,
      "epoch": 0.30297532656023224,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004994910617982264,
      "loss": 5.7666,
      "mean_token_accuracy": 0.1507476307451725,
      "num_tokens": 6919654.0,
      "step": 3340
    },
    {
      "entropy": 5.835614204406738,
      "epoch": 0.3034288824383164,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994888844181833,
      "loss": 5.8576,
      "mean_token_accuracy": 0.13817089200019836,
      "num_tokens": 6929673.0,
      "step": 3345
    },
    {
      "entropy": 5.9251823902130125,
      "epoch": 0.3038824383164006,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994867023956529,
      "loss": 5.8579,
      "mean_token_accuracy": 0.14975906163454056,
      "num_tokens": 6938994.0,
      "step": 3350
    },
    {
      "entropy": 5.901224994659424,
      "epoch": 0.30433599419448476,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994845157306803,
      "loss": 5.9223,
      "mean_token_accuracy": 0.14631777852773667,
      "num_tokens": 6950444.0,
      "step": 3355
    },
    {
      "entropy": 5.947768449783325,
      "epoch": 0.3047895500725689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994823244233107,
      "loss": 5.8255,
      "mean_token_accuracy": 0.14204518795013427,
      "num_tokens": 6961702.0,
      "step": 3360
    },
    {
      "entropy": 5.8486240863800045,
      "epoch": 0.3052431059506531,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004994801284735895,
      "loss": 5.8097,
      "mean_token_accuracy": 0.14448109567165374,
      "num_tokens": 6972428.0,
      "step": 3365
    },
    {
      "entropy": 5.752938604354858,
      "epoch": 0.3056966618287373,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994779278815619,
      "loss": 5.7787,
      "mean_token_accuracy": 0.14913846850395202,
      "num_tokens": 6982364.0,
      "step": 3370
    },
    {
      "entropy": 5.852268695831299,
      "epoch": 0.3061502177068215,
      "grad_norm": 1.375,
      "learning_rate": 0.0004994757226472736,
      "loss": 5.7757,
      "mean_token_accuracy": 0.14681893065571786,
      "num_tokens": 6992235.0,
      "step": 3375
    },
    {
      "entropy": 5.895943021774292,
      "epoch": 0.30660377358490565,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004994735127707702,
      "loss": 5.722,
      "mean_token_accuracy": 0.15772044211626052,
      "num_tokens": 7003079.0,
      "step": 3380
    },
    {
      "entropy": 5.8661706924438475,
      "epoch": 0.30705732946298986,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994712982520972,
      "loss": 5.9052,
      "mean_token_accuracy": 0.1421538583934307,
      "num_tokens": 7013501.0,
      "step": 3385
    },
    {
      "entropy": 5.889679431915283,
      "epoch": 0.307510885341074,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004994690790913008,
      "loss": 5.7738,
      "mean_token_accuracy": 0.15045519173145294,
      "num_tokens": 7023978.0,
      "step": 3390
    },
    {
      "entropy": 5.774082851409912,
      "epoch": 0.3079644412191582,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004994668552884265,
      "loss": 5.7155,
      "mean_token_accuracy": 0.14592754542827607,
      "num_tokens": 7032999.0,
      "step": 3395
    },
    {
      "entropy": 5.909044790267944,
      "epoch": 0.3084179970972424,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004994646268435204,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1387397527694702,
      "num_tokens": 7044776.0,
      "step": 3400
    },
    {
      "entropy": 5.976453256607056,
      "epoch": 0.30887155297532654,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004994623937566286,
      "loss": 5.8137,
      "mean_token_accuracy": 0.14521186649799347,
      "num_tokens": 7054667.0,
      "step": 3405
    },
    {
      "entropy": 5.72452940940857,
      "epoch": 0.30932510885341075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994601560277974,
      "loss": 5.7793,
      "mean_token_accuracy": 0.14921418875455855,
      "num_tokens": 7065477.0,
      "step": 3410
    },
    {
      "entropy": 5.865390348434448,
      "epoch": 0.3097786647314949,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499457913657073,
      "loss": 5.8831,
      "mean_token_accuracy": 0.15155483782291412,
      "num_tokens": 7075704.0,
      "step": 3415
    },
    {
      "entropy": 5.871505212783814,
      "epoch": 0.3102322206095791,
      "grad_norm": 1.25,
      "learning_rate": 0.0004994556666445016,
      "loss": 5.9229,
      "mean_token_accuracy": 0.14383582025766373,
      "num_tokens": 7086371.0,
      "step": 3420
    },
    {
      "entropy": 5.814038896560669,
      "epoch": 0.31068577648766327,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994534149901299,
      "loss": 5.7946,
      "mean_token_accuracy": 0.14878375381231307,
      "num_tokens": 7097588.0,
      "step": 3425
    },
    {
      "entropy": 5.889452600479126,
      "epoch": 0.3111393323657475,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004994511586940044,
      "loss": 5.7666,
      "mean_token_accuracy": 0.14635036289691924,
      "num_tokens": 7107650.0,
      "step": 3430
    },
    {
      "entropy": 5.833066129684449,
      "epoch": 0.31159288824383163,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994488977561717,
      "loss": 5.808,
      "mean_token_accuracy": 0.14654033184051513,
      "num_tokens": 7118202.0,
      "step": 3435
    },
    {
      "entropy": 5.783775854110718,
      "epoch": 0.31204644412191584,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994466321766786,
      "loss": 5.7569,
      "mean_token_accuracy": 0.15153429359197618,
      "num_tokens": 7128834.0,
      "step": 3440
    },
    {
      "entropy": 5.866316509246826,
      "epoch": 0.3125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000499444361955572,
      "loss": 5.7893,
      "mean_token_accuracy": 0.1500387281179428,
      "num_tokens": 7138625.0,
      "step": 3445
    },
    {
      "entropy": 5.806637716293335,
      "epoch": 0.31295355587808416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994420870928988,
      "loss": 5.7667,
      "mean_token_accuracy": 0.15101150274276734,
      "num_tokens": 7148815.0,
      "step": 3450
    },
    {
      "entropy": 5.821873950958252,
      "epoch": 0.31340711175616837,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004994398075887059,
      "loss": 5.8138,
      "mean_token_accuracy": 0.14523909389972686,
      "num_tokens": 7158962.0,
      "step": 3455
    },
    {
      "entropy": 5.866718482971192,
      "epoch": 0.3138606676342525,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004994375234430407,
      "loss": 5.8271,
      "mean_token_accuracy": 0.1487603962421417,
      "num_tokens": 7169281.0,
      "step": 3460
    },
    {
      "entropy": 5.858565902709961,
      "epoch": 0.31431422351233673,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004994352346559504,
      "loss": 5.8205,
      "mean_token_accuracy": 0.1458689272403717,
      "num_tokens": 7179934.0,
      "step": 3465
    },
    {
      "entropy": 5.853692817687988,
      "epoch": 0.3147677793904209,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994329412274821,
      "loss": 5.7938,
      "mean_token_accuracy": 0.14809443354606627,
      "num_tokens": 7190203.0,
      "step": 3470
    },
    {
      "entropy": 5.871757459640503,
      "epoch": 0.3152213352685051,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004994306431576835,
      "loss": 5.7194,
      "mean_token_accuracy": 0.14831513911485672,
      "num_tokens": 7200327.0,
      "step": 3475
    },
    {
      "entropy": 5.730972814559936,
      "epoch": 0.31567489114658925,
      "grad_norm": 1.21875,
      "learning_rate": 0.000499428340446602,
      "loss": 5.7515,
      "mean_token_accuracy": 0.14930091053247452,
      "num_tokens": 7211006.0,
      "step": 3480
    },
    {
      "entropy": 5.785858917236328,
      "epoch": 0.31612844702467346,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004994260330942851,
      "loss": 5.7559,
      "mean_token_accuracy": 0.15045426189899444,
      "num_tokens": 7221249.0,
      "step": 3485
    },
    {
      "entropy": 5.857178020477295,
      "epoch": 0.3165820029027576,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994237211007808,
      "loss": 5.7426,
      "mean_token_accuracy": 0.15272777229547502,
      "num_tokens": 7232534.0,
      "step": 3490
    },
    {
      "entropy": 5.88194055557251,
      "epoch": 0.3170355587808418,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004994214044661367,
      "loss": 5.87,
      "mean_token_accuracy": 0.14889724403619767,
      "num_tokens": 7242776.0,
      "step": 3495
    },
    {
      "entropy": 5.828200054168701,
      "epoch": 0.317489114658926,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994190831904007,
      "loss": 5.8039,
      "mean_token_accuracy": 0.14493839144706727,
      "num_tokens": 7253070.0,
      "step": 3500
    },
    {
      "entropy": 5.842406272888184,
      "epoch": 0.31794267053701014,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000499416757273621,
      "loss": 5.7517,
      "mean_token_accuracy": 0.14610901474952698,
      "num_tokens": 7263398.0,
      "step": 3505
    },
    {
      "entropy": 5.83699779510498,
      "epoch": 0.31839622641509435,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004994144267158454,
      "loss": 5.7957,
      "mean_token_accuracy": 0.1461636394262314,
      "num_tokens": 7273584.0,
      "step": 3510
    },
    {
      "entropy": 5.865672397613525,
      "epoch": 0.3188497822931785,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004994120915171224,
      "loss": 5.7336,
      "mean_token_accuracy": 0.15144022703170776,
      "num_tokens": 7283403.0,
      "step": 3515
    },
    {
      "entropy": 5.699055528640747,
      "epoch": 0.3193033381712627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004994097516775002,
      "loss": 5.6233,
      "mean_token_accuracy": 0.15417083352804184,
      "num_tokens": 7293738.0,
      "step": 3520
    },
    {
      "entropy": 5.834370374679565,
      "epoch": 0.3197568940493469,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004994074071970271,
      "loss": 5.8359,
      "mean_token_accuracy": 0.14510649293661118,
      "num_tokens": 7304069.0,
      "step": 3525
    },
    {
      "entropy": 5.931438398361206,
      "epoch": 0.3202104499274311,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004994050580757515,
      "loss": 5.8113,
      "mean_token_accuracy": 0.15269186496734619,
      "num_tokens": 7314433.0,
      "step": 3530
    },
    {
      "entropy": 5.783717203140259,
      "epoch": 0.32066400580551524,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004994027043137223,
      "loss": 5.7187,
      "mean_token_accuracy": 0.1535136118531227,
      "num_tokens": 7324516.0,
      "step": 3535
    },
    {
      "entropy": 5.7936512470245365,
      "epoch": 0.3211175616835994,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004994003459109879,
      "loss": 5.8297,
      "mean_token_accuracy": 0.14828921854496002,
      "num_tokens": 7333746.0,
      "step": 3540
    },
    {
      "entropy": 5.862206029891968,
      "epoch": 0.3215711175616836,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004993979828675972,
      "loss": 5.8186,
      "mean_token_accuracy": 0.14887310564517975,
      "num_tokens": 7344165.0,
      "step": 3545
    },
    {
      "entropy": 5.889583778381348,
      "epoch": 0.32202467343976776,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000499395615183599,
      "loss": 5.7609,
      "mean_token_accuracy": 0.15327379256486892,
      "num_tokens": 7353544.0,
      "step": 3550
    },
    {
      "entropy": 5.809433460235596,
      "epoch": 0.32247822931785197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004993932428590422,
      "loss": 5.722,
      "mean_token_accuracy": 0.15682561993598937,
      "num_tokens": 7363387.0,
      "step": 3555
    },
    {
      "entropy": 5.836307859420776,
      "epoch": 0.3229317851959361,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004993908658939762,
      "loss": 5.7268,
      "mean_token_accuracy": 0.15479713678359985,
      "num_tokens": 7373738.0,
      "step": 3560
    },
    {
      "entropy": 5.846750164031983,
      "epoch": 0.32338534107402034,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004993884842884496,
      "loss": 5.822,
      "mean_token_accuracy": 0.14990484118461608,
      "num_tokens": 7383580.0,
      "step": 3565
    },
    {
      "entropy": 5.804448747634888,
      "epoch": 0.3238388969521045,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993860980425121,
      "loss": 5.7424,
      "mean_token_accuracy": 0.1557959645986557,
      "num_tokens": 7394047.0,
      "step": 3570
    },
    {
      "entropy": 5.866255474090576,
      "epoch": 0.3242924528301887,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004993837071562131,
      "loss": 5.8729,
      "mean_token_accuracy": 0.14274352118372918,
      "num_tokens": 7404834.0,
      "step": 3575
    },
    {
      "entropy": 5.947095918655395,
      "epoch": 0.32474600870827286,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004993813116296017,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1503795638680458,
      "num_tokens": 7416075.0,
      "step": 3580
    },
    {
      "entropy": 5.782832288742066,
      "epoch": 0.325199564586357,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993789114627275,
      "loss": 5.775,
      "mean_token_accuracy": 0.14573703706264496,
      "num_tokens": 7426049.0,
      "step": 3585
    },
    {
      "entropy": 5.8114879608154295,
      "epoch": 0.3256531204644412,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004993765066556403,
      "loss": 5.7557,
      "mean_token_accuracy": 0.15304522216320038,
      "num_tokens": 7436719.0,
      "step": 3590
    },
    {
      "entropy": 5.752126789093017,
      "epoch": 0.3261066763425254,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004993740972083899,
      "loss": 5.7798,
      "mean_token_accuracy": 0.149585048109293,
      "num_tokens": 7447215.0,
      "step": 3595
    },
    {
      "entropy": 5.785154199600219,
      "epoch": 0.3265602322206096,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004993716831210259,
      "loss": 5.8566,
      "mean_token_accuracy": 0.14159765243530273,
      "num_tokens": 7457602.0,
      "step": 3600
    },
    {
      "entropy": 5.8140160083770756,
      "epoch": 0.32701378809869375,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004993692643935983,
      "loss": 5.6691,
      "mean_token_accuracy": 0.1545745313167572,
      "num_tokens": 7468392.0,
      "step": 3605
    },
    {
      "entropy": 5.876634788513184,
      "epoch": 0.32746734397677796,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004993668410261571,
      "loss": 5.7523,
      "mean_token_accuracy": 0.15066123604774476,
      "num_tokens": 7478320.0,
      "step": 3610
    },
    {
      "entropy": 5.794388294219971,
      "epoch": 0.3279208998548621,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004993644130187525,
      "loss": 5.7769,
      "mean_token_accuracy": 0.1582022950053215,
      "num_tokens": 7488557.0,
      "step": 3615
    },
    {
      "entropy": 5.816341876983643,
      "epoch": 0.3283744557329463,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004993619803714347,
      "loss": 5.7516,
      "mean_token_accuracy": 0.1480060413479805,
      "num_tokens": 7499108.0,
      "step": 3620
    },
    {
      "entropy": 5.820813465118408,
      "epoch": 0.3288280116110305,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004993595430842539,
      "loss": 5.7514,
      "mean_token_accuracy": 0.15406939685344695,
      "num_tokens": 7508770.0,
      "step": 3625
    },
    {
      "entropy": 5.77728796005249,
      "epoch": 0.32928156748911463,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004993571011572606,
      "loss": 5.7319,
      "mean_token_accuracy": 0.150125253200531,
      "num_tokens": 7519633.0,
      "step": 3630
    },
    {
      "entropy": 5.9409888744354244,
      "epoch": 0.32973512336719885,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993546545905053,
      "loss": 5.9125,
      "mean_token_accuracy": 0.14048471823334693,
      "num_tokens": 7531634.0,
      "step": 3635
    },
    {
      "entropy": 5.80763258934021,
      "epoch": 0.330188679245283,
      "grad_norm": 1.375,
      "learning_rate": 0.0004993522033840386,
      "loss": 5.7227,
      "mean_token_accuracy": 0.14995568990707397,
      "num_tokens": 7541992.0,
      "step": 3640
    },
    {
      "entropy": 5.824553918838501,
      "epoch": 0.3306422351233672,
      "grad_norm": 1.4375,
      "learning_rate": 0.000499349747537911,
      "loss": 5.7843,
      "mean_token_accuracy": 0.1440193310379982,
      "num_tokens": 7552302.0,
      "step": 3645
    },
    {
      "entropy": 5.8929589748382565,
      "epoch": 0.33109579100145137,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004993472870521736,
      "loss": 5.7662,
      "mean_token_accuracy": 0.14555299133062363,
      "num_tokens": 7562806.0,
      "step": 3650
    },
    {
      "entropy": 5.79813232421875,
      "epoch": 0.3315493468795356,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004993448219268771,
      "loss": 5.7439,
      "mean_token_accuracy": 0.14852455332875253,
      "num_tokens": 7572811.0,
      "step": 3655
    },
    {
      "entropy": 5.759467267990113,
      "epoch": 0.33200290275761973,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004993423521620726,
      "loss": 5.7378,
      "mean_token_accuracy": 0.1495088294148445,
      "num_tokens": 7583458.0,
      "step": 3660
    },
    {
      "entropy": 5.940987539291382,
      "epoch": 0.33245645863570394,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004993398777578109,
      "loss": 5.8731,
      "mean_token_accuracy": 0.1426645740866661,
      "num_tokens": 7595227.0,
      "step": 3665
    },
    {
      "entropy": 5.8508298873901365,
      "epoch": 0.3329100145137881,
      "grad_norm": 1.25,
      "learning_rate": 0.0004993373987141435,
      "loss": 5.8156,
      "mean_token_accuracy": 0.14546542689204217,
      "num_tokens": 7605446.0,
      "step": 3670
    },
    {
      "entropy": 5.858935737609864,
      "epoch": 0.33336357039187225,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993349150311214,
      "loss": 5.796,
      "mean_token_accuracy": 0.14503608867526055,
      "num_tokens": 7616001.0,
      "step": 3675
    },
    {
      "entropy": 5.826646518707276,
      "epoch": 0.33381712626995647,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993324267087962,
      "loss": 5.7583,
      "mean_token_accuracy": 0.15043987184762955,
      "num_tokens": 7626747.0,
      "step": 3680
    },
    {
      "entropy": 5.828233861923218,
      "epoch": 0.3342706821480406,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004993299337472192,
      "loss": 5.7352,
      "mean_token_accuracy": 0.15164113491773606,
      "num_tokens": 7636687.0,
      "step": 3685
    },
    {
      "entropy": 5.7816657543182375,
      "epoch": 0.33472423802612483,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000499327436146442,
      "loss": 5.694,
      "mean_token_accuracy": 0.16283626556396485,
      "num_tokens": 7647271.0,
      "step": 3690
    },
    {
      "entropy": 5.789823007583618,
      "epoch": 0.335177793904209,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004993249339065162,
      "loss": 5.7246,
      "mean_token_accuracy": 0.14822520613670348,
      "num_tokens": 7657034.0,
      "step": 3695
    },
    {
      "entropy": 5.874470567703247,
      "epoch": 0.3356313497822932,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993224270274937,
      "loss": 5.7758,
      "mean_token_accuracy": 0.14884890168905257,
      "num_tokens": 7667833.0,
      "step": 3700
    },
    {
      "entropy": 5.901823997497559,
      "epoch": 0.33608490566037735,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004993199155094263,
      "loss": 5.8724,
      "mean_token_accuracy": 0.14083609431982042,
      "num_tokens": 7678247.0,
      "step": 3705
    },
    {
      "entropy": 5.87119836807251,
      "epoch": 0.33653846153846156,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004993173993523658,
      "loss": 5.7453,
      "mean_token_accuracy": 0.1406989276409149,
      "num_tokens": 7689293.0,
      "step": 3710
    },
    {
      "entropy": 5.767151117324829,
      "epoch": 0.3369920174165457,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004993148785563643,
      "loss": 5.8204,
      "mean_token_accuracy": 0.14787752479314803,
      "num_tokens": 7699462.0,
      "step": 3715
    },
    {
      "entropy": 5.892066431045532,
      "epoch": 0.3374455732946299,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000499312353121474,
      "loss": 5.8631,
      "mean_token_accuracy": 0.14873588383197783,
      "num_tokens": 7709855.0,
      "step": 3720
    },
    {
      "entropy": 5.925738763809204,
      "epoch": 0.3378991291727141,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004993098230477471,
      "loss": 5.7697,
      "mean_token_accuracy": 0.14335086718201637,
      "num_tokens": 7720922.0,
      "step": 3725
    },
    {
      "entropy": 5.912420415878296,
      "epoch": 0.33835268505079824,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004993072883352358,
      "loss": 5.7878,
      "mean_token_accuracy": 0.1521473616361618,
      "num_tokens": 7730656.0,
      "step": 3730
    },
    {
      "entropy": 5.783479452133179,
      "epoch": 0.33880624092888245,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004993047489839927,
      "loss": 5.6906,
      "mean_token_accuracy": 0.15430137664079666,
      "num_tokens": 7739769.0,
      "step": 3735
    },
    {
      "entropy": 5.798482847213745,
      "epoch": 0.3392597968069666,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004993022049940702,
      "loss": 5.7794,
      "mean_token_accuracy": 0.14408639296889306,
      "num_tokens": 7749882.0,
      "step": 3740
    },
    {
      "entropy": 5.883006191253662,
      "epoch": 0.3397133526850508,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004992996563655209,
      "loss": 5.7714,
      "mean_token_accuracy": 0.14932019859552384,
      "num_tokens": 7760675.0,
      "step": 3745
    },
    {
      "entropy": 5.821650838851928,
      "epoch": 0.340166908563135,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992971030983975,
      "loss": 5.8316,
      "mean_token_accuracy": 0.14970973134040833,
      "num_tokens": 7771023.0,
      "step": 3750
    },
    {
      "entropy": 5.843093347549439,
      "epoch": 0.3406204644412192,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992945451927531,
      "loss": 5.8182,
      "mean_token_accuracy": 0.14141693264245986,
      "num_tokens": 7782014.0,
      "step": 3755
    },
    {
      "entropy": 5.882049036026001,
      "epoch": 0.34107402031930334,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992919826486402,
      "loss": 5.7741,
      "mean_token_accuracy": 0.14652178436517715,
      "num_tokens": 7792041.0,
      "step": 3760
    },
    {
      "entropy": 5.8244024276733395,
      "epoch": 0.3415275761973875,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992894154661118,
      "loss": 5.6397,
      "mean_token_accuracy": 0.154453644156456,
      "num_tokens": 7802027.0,
      "step": 3765
    },
    {
      "entropy": 5.750616121292114,
      "epoch": 0.3419811320754717,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992868436452213,
      "loss": 5.7673,
      "mean_token_accuracy": 0.14698233753442763,
      "num_tokens": 7812442.0,
      "step": 3770
    },
    {
      "entropy": 5.783560419082642,
      "epoch": 0.34243468795355586,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004992842671860217,
      "loss": 5.7998,
      "mean_token_accuracy": 0.14647551104426385,
      "num_tokens": 7823261.0,
      "step": 3775
    },
    {
      "entropy": 5.845564603805542,
      "epoch": 0.34288824383164007,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992816860885662,
      "loss": 5.7127,
      "mean_token_accuracy": 0.1526141047477722,
      "num_tokens": 7833190.0,
      "step": 3780
    },
    {
      "entropy": 5.778737258911133,
      "epoch": 0.3433417997097242,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004992791003529084,
      "loss": 5.6905,
      "mean_token_accuracy": 0.15017810985445976,
      "num_tokens": 7842607.0,
      "step": 3785
    },
    {
      "entropy": 5.844851398468018,
      "epoch": 0.34379535558780844,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004992765099791017,
      "loss": 5.7411,
      "mean_token_accuracy": 0.15202188193798066,
      "num_tokens": 7852607.0,
      "step": 3790
    },
    {
      "entropy": 5.735257244110107,
      "epoch": 0.3442489114658926,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004992739149671995,
      "loss": 5.6719,
      "mean_token_accuracy": 0.15777993947267532,
      "num_tokens": 7863500.0,
      "step": 3795
    },
    {
      "entropy": 5.822976922988891,
      "epoch": 0.3447024673439768,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004992713153172556,
      "loss": 5.8321,
      "mean_token_accuracy": 0.1504603460431099,
      "num_tokens": 7874403.0,
      "step": 3800
    },
    {
      "entropy": 5.9864270210266115,
      "epoch": 0.34515602322206096,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004992687110293238,
      "loss": 5.872,
      "mean_token_accuracy": 0.14370461404323578,
      "num_tokens": 7885342.0,
      "step": 3805
    },
    {
      "entropy": 5.819399309158325,
      "epoch": 0.3456095791001451,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004992661021034579,
      "loss": 5.7533,
      "mean_token_accuracy": 0.1517915889620781,
      "num_tokens": 7895905.0,
      "step": 3810
    },
    {
      "entropy": 5.808559989929199,
      "epoch": 0.3460631349782293,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004992634885397118,
      "loss": 5.8912,
      "mean_token_accuracy": 0.14670953154563904,
      "num_tokens": 7905293.0,
      "step": 3815
    },
    {
      "entropy": 5.8831467628479,
      "epoch": 0.3465166908563135,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004992608703381396,
      "loss": 5.7349,
      "mean_token_accuracy": 0.1525384560227394,
      "num_tokens": 7915996.0,
      "step": 3820
    },
    {
      "entropy": 5.855575037002564,
      "epoch": 0.3469702467343977,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992582474987955,
      "loss": 5.7408,
      "mean_token_accuracy": 0.15123631060123444,
      "num_tokens": 7926739.0,
      "step": 3825
    },
    {
      "entropy": 5.799056148529052,
      "epoch": 0.34742380261248185,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992556200217337,
      "loss": 5.7734,
      "mean_token_accuracy": 0.14914042353630066,
      "num_tokens": 7937109.0,
      "step": 3830
    },
    {
      "entropy": 5.815071058273316,
      "epoch": 0.34787735849056606,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004992529879070085,
      "loss": 5.7226,
      "mean_token_accuracy": 0.15417693704366683,
      "num_tokens": 7947562.0,
      "step": 3835
    },
    {
      "entropy": 5.822039318084717,
      "epoch": 0.3483309143686502,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004992503511546745,
      "loss": 5.7967,
      "mean_token_accuracy": 0.15639951974153518,
      "num_tokens": 7957400.0,
      "step": 3840
    },
    {
      "entropy": 5.814460420608521,
      "epoch": 0.3487844702467344,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992477097647859,
      "loss": 5.726,
      "mean_token_accuracy": 0.1585880383849144,
      "num_tokens": 7967600.0,
      "step": 3845
    },
    {
      "entropy": 5.8076441287994385,
      "epoch": 0.3492380261248186,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004992450637373976,
      "loss": 5.7054,
      "mean_token_accuracy": 0.1458221599459648,
      "num_tokens": 7978012.0,
      "step": 3850
    },
    {
      "entropy": 5.746403789520263,
      "epoch": 0.34969158200290273,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004992424130725642,
      "loss": 5.7534,
      "mean_token_accuracy": 0.14770088642835616,
      "num_tokens": 7988495.0,
      "step": 3855
    },
    {
      "entropy": 5.911701726913452,
      "epoch": 0.35014513788098695,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004992397577703406,
      "loss": 5.7616,
      "mean_token_accuracy": 0.14998696744441986,
      "num_tokens": 7998125.0,
      "step": 3860
    },
    {
      "entropy": 5.80164270401001,
      "epoch": 0.3505986937590711,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992370978307816,
      "loss": 5.6729,
      "mean_token_accuracy": 0.15145281702280045,
      "num_tokens": 8009151.0,
      "step": 3865
    },
    {
      "entropy": 5.736188745498657,
      "epoch": 0.3510522496371553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004992344332539424,
      "loss": 5.6916,
      "mean_token_accuracy": 0.15478172451257705,
      "num_tokens": 8019501.0,
      "step": 3870
    },
    {
      "entropy": 5.775134611129761,
      "epoch": 0.35150580551523947,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992317640398779,
      "loss": 5.7821,
      "mean_token_accuracy": 0.15074608698487282,
      "num_tokens": 8030362.0,
      "step": 3875
    },
    {
      "entropy": 5.773638105392456,
      "epoch": 0.3519593613933237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992290901886434,
      "loss": 5.6885,
      "mean_token_accuracy": 0.14943585619330407,
      "num_tokens": 8041671.0,
      "step": 3880
    },
    {
      "entropy": 5.83705472946167,
      "epoch": 0.35241291727140783,
      "grad_norm": 1.34375,
      "learning_rate": 0.000499226411700294,
      "loss": 5.7949,
      "mean_token_accuracy": 0.14793721884489058,
      "num_tokens": 8052609.0,
      "step": 3885
    },
    {
      "entropy": 5.854422664642334,
      "epoch": 0.35286647314949204,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004992237285748855,
      "loss": 5.7463,
      "mean_token_accuracy": 0.15416623651981354,
      "num_tokens": 8062611.0,
      "step": 3890
    },
    {
      "entropy": 5.8708113670349125,
      "epoch": 0.3533200290275762,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004992210408124731,
      "loss": 5.7815,
      "mean_token_accuracy": 0.15578939616680146,
      "num_tokens": 8073542.0,
      "step": 3895
    },
    {
      "entropy": 5.816879558563232,
      "epoch": 0.35377358490566035,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992183484131123,
      "loss": 5.6918,
      "mean_token_accuracy": 0.1484497994184494,
      "num_tokens": 8083634.0,
      "step": 3900
    },
    {
      "entropy": 5.8308281898498535,
      "epoch": 0.35422714078374457,
      "grad_norm": 1.625,
      "learning_rate": 0.0004992156513768591,
      "loss": 5.767,
      "mean_token_accuracy": 0.15284653156995773,
      "num_tokens": 8094710.0,
      "step": 3905
    },
    {
      "entropy": 5.889791870117188,
      "epoch": 0.3546806966618287,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000499212949703769,
      "loss": 5.908,
      "mean_token_accuracy": 0.14653956666588783,
      "num_tokens": 8104191.0,
      "step": 3910
    },
    {
      "entropy": 5.7595172882080075,
      "epoch": 0.35513425253991293,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004992102433938981,
      "loss": 5.7,
      "mean_token_accuracy": 0.15048497617244722,
      "num_tokens": 8114744.0,
      "step": 3915
    },
    {
      "entropy": 5.854955530166626,
      "epoch": 0.3555878084179971,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004992075324473021,
      "loss": 5.8059,
      "mean_token_accuracy": 0.14852231740951538,
      "num_tokens": 8124267.0,
      "step": 3920
    },
    {
      "entropy": 5.880210924148559,
      "epoch": 0.3560413642960813,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004992048168640371,
      "loss": 5.8157,
      "mean_token_accuracy": 0.14342952743172646,
      "num_tokens": 8134383.0,
      "step": 3925
    },
    {
      "entropy": 5.803681993484497,
      "epoch": 0.35649492017416545,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992020966441595,
      "loss": 5.7152,
      "mean_token_accuracy": 0.15491905063390732,
      "num_tokens": 8144131.0,
      "step": 3930
    },
    {
      "entropy": 5.750734806060791,
      "epoch": 0.35694847605224966,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004991993717877254,
      "loss": 5.6509,
      "mean_token_accuracy": 0.15241849273443223,
      "num_tokens": 8154947.0,
      "step": 3935
    },
    {
      "entropy": 5.78973536491394,
      "epoch": 0.3574020319303338,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004991966422947911,
      "loss": 5.6998,
      "mean_token_accuracy": 0.14700091183185576,
      "num_tokens": 8165692.0,
      "step": 3940
    },
    {
      "entropy": 5.8159383773803714,
      "epoch": 0.357855587808418,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991939081654132,
      "loss": 5.8189,
      "mean_token_accuracy": 0.14284562468528747,
      "num_tokens": 8177402.0,
      "step": 3945
    },
    {
      "entropy": 5.765165567398071,
      "epoch": 0.3583091436865022,
      "grad_norm": 1.3125,
      "learning_rate": 0.000499191169399648,
      "loss": 5.5947,
      "mean_token_accuracy": 0.15420890748500823,
      "num_tokens": 8187545.0,
      "step": 3950
    },
    {
      "entropy": 5.743610429763794,
      "epoch": 0.35876269956458634,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004991884259975525,
      "loss": 5.6825,
      "mean_token_accuracy": 0.15421922504901886,
      "num_tokens": 8198142.0,
      "step": 3955
    },
    {
      "entropy": 5.767709255218506,
      "epoch": 0.35921625544267055,
      "grad_norm": 2.484375,
      "learning_rate": 0.000499185677959183,
      "loss": 5.6975,
      "mean_token_accuracy": 0.15191758424043655,
      "num_tokens": 8208131.0,
      "step": 3960
    },
    {
      "entropy": 5.7967020034790036,
      "epoch": 0.3596698113207547,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004991829252845966,
      "loss": 5.7845,
      "mean_token_accuracy": 0.14716573655605317,
      "num_tokens": 8218091.0,
      "step": 3965
    },
    {
      "entropy": 5.77277889251709,
      "epoch": 0.3601233671988389,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004991801679738502,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15576295107603072,
      "num_tokens": 8228392.0,
      "step": 3970
    },
    {
      "entropy": 5.767637395858765,
      "epoch": 0.3605769230769231,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004991774060270008,
      "loss": 5.7269,
      "mean_token_accuracy": 0.15118133872747422,
      "num_tokens": 8238272.0,
      "step": 3975
    },
    {
      "entropy": 5.802407693862915,
      "epoch": 0.3610304789550073,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004991746394441055,
      "loss": 5.7679,
      "mean_token_accuracy": 0.147048220038414,
      "num_tokens": 8248255.0,
      "step": 3980
    },
    {
      "entropy": 5.91586012840271,
      "epoch": 0.36148403483309144,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991718682252216,
      "loss": 5.8726,
      "mean_token_accuracy": 0.1428057424724102,
      "num_tokens": 8258839.0,
      "step": 3985
    },
    {
      "entropy": 5.771537828445434,
      "epoch": 0.3619375907111756,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004991690923704063,
      "loss": 5.7673,
      "mean_token_accuracy": 0.1511303447186947,
      "num_tokens": 8269400.0,
      "step": 3990
    },
    {
      "entropy": 5.940103054046631,
      "epoch": 0.3623911465892598,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991663118797171,
      "loss": 5.7763,
      "mean_token_accuracy": 0.1525319442152977,
      "num_tokens": 8279468.0,
      "step": 3995
    },
    {
      "entropy": 5.80928750038147,
      "epoch": 0.36284470246734396,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004991635267532113,
      "loss": 5.6949,
      "mean_token_accuracy": 0.15478259027004243,
      "num_tokens": 8289207.0,
      "step": 4000
    },
    {
      "entropy": 5.637155389785766,
      "epoch": 0.36329825834542817,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004991607369909468,
      "loss": 5.739,
      "mean_token_accuracy": 0.15483498573303223,
      "num_tokens": 8299819.0,
      "step": 4005
    },
    {
      "entropy": 5.815872097015381,
      "epoch": 0.3637518142235123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004991579425929811,
      "loss": 5.6595,
      "mean_token_accuracy": 0.15534015893936157,
      "num_tokens": 8310216.0,
      "step": 4010
    },
    {
      "entropy": 5.781091022491455,
      "epoch": 0.36420537010159654,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000499155143559372,
      "loss": 5.6879,
      "mean_token_accuracy": 0.15090734809637069,
      "num_tokens": 8319967.0,
      "step": 4015
    },
    {
      "entropy": 5.850452327728272,
      "epoch": 0.3646589259796807,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004991523398901773,
      "loss": 5.7414,
      "mean_token_accuracy": 0.1563695937395096,
      "num_tokens": 8329271.0,
      "step": 4020
    },
    {
      "entropy": 5.788423871994018,
      "epoch": 0.3651124818577649,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991495315854553,
      "loss": 5.7322,
      "mean_token_accuracy": 0.15406767427921295,
      "num_tokens": 8340042.0,
      "step": 4025
    },
    {
      "entropy": 5.86919207572937,
      "epoch": 0.36556603773584906,
      "grad_norm": 1.625,
      "learning_rate": 0.0004991467186452637,
      "loss": 5.8773,
      "mean_token_accuracy": 0.14352630823850632,
      "num_tokens": 8351005.0,
      "step": 4030
    },
    {
      "entropy": 5.810650253295899,
      "epoch": 0.3660195936139332,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004991439010696609,
      "loss": 5.6873,
      "mean_token_accuracy": 0.15218240916728973,
      "num_tokens": 8360964.0,
      "step": 4035
    },
    {
      "entropy": 5.818238067626953,
      "epoch": 0.3664731494920174,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004991410788587051,
      "loss": 5.7632,
      "mean_token_accuracy": 0.14673216938972472,
      "num_tokens": 8370933.0,
      "step": 4040
    },
    {
      "entropy": 5.8133055686950685,
      "epoch": 0.3669267053701016,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991382520124547,
      "loss": 5.7941,
      "mean_token_accuracy": 0.15076968371868132,
      "num_tokens": 8382422.0,
      "step": 4045
    },
    {
      "entropy": 5.8689398765563965,
      "epoch": 0.3673802612481858,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004991354205309681,
      "loss": 5.7331,
      "mean_token_accuracy": 0.14591655284166336,
      "num_tokens": 8392162.0,
      "step": 4050
    },
    {
      "entropy": 5.831794023513794,
      "epoch": 0.36783381712626995,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000499132584414304,
      "loss": 5.7765,
      "mean_token_accuracy": 0.14830462336540223,
      "num_tokens": 8402145.0,
      "step": 4055
    },
    {
      "entropy": 5.808120536804199,
      "epoch": 0.36828737300435416,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004991297436625207,
      "loss": 5.7002,
      "mean_token_accuracy": 0.15320413708686828,
      "num_tokens": 8413117.0,
      "step": 4060
    },
    {
      "entropy": 5.78600697517395,
      "epoch": 0.3687409288824383,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004991268982756774,
      "loss": 5.7367,
      "mean_token_accuracy": 0.15311348736286162,
      "num_tokens": 8422997.0,
      "step": 4065
    },
    {
      "entropy": 5.80166392326355,
      "epoch": 0.3691944847605225,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004991240482538327,
      "loss": 5.6844,
      "mean_token_accuracy": 0.15038370192050934,
      "num_tokens": 8432718.0,
      "step": 4070
    },
    {
      "entropy": 5.836380958557129,
      "epoch": 0.3696480406386067,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004991211935970455,
      "loss": 5.753,
      "mean_token_accuracy": 0.14487562626600264,
      "num_tokens": 8442808.0,
      "step": 4075
    },
    {
      "entropy": 5.758193063735962,
      "epoch": 0.37010159651669083,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991183343053748,
      "loss": 5.748,
      "mean_token_accuracy": 0.15124204605817795,
      "num_tokens": 8451919.0,
      "step": 4080
    },
    {
      "entropy": 5.852297353744507,
      "epoch": 0.37055515239477504,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00049911547037888,
      "loss": 5.7785,
      "mean_token_accuracy": 0.14427137970924378,
      "num_tokens": 8463593.0,
      "step": 4085
    },
    {
      "entropy": 5.772682237625122,
      "epoch": 0.3710087082728592,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004991126018176202,
      "loss": 5.748,
      "mean_token_accuracy": 0.14865192174911498,
      "num_tokens": 8474025.0,
      "step": 4090
    },
    {
      "entropy": 5.847655487060547,
      "epoch": 0.3714622641509434,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004991097286216545,
      "loss": 5.8365,
      "mean_token_accuracy": 0.14862846881151198,
      "num_tokens": 8484225.0,
      "step": 4095
    },
    {
      "entropy": 5.888778352737427,
      "epoch": 0.37191582002902757,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004991068507910427,
      "loss": 5.6863,
      "mean_token_accuracy": 0.15229319036006927,
      "num_tokens": 8494173.0,
      "step": 4100
    },
    {
      "entropy": 5.863654279708863,
      "epoch": 0.3723693759071118,
      "grad_norm": 1.390625,
      "learning_rate": 0.000499103968325844,
      "loss": 5.7619,
      "mean_token_accuracy": 0.15187404453754424,
      "num_tokens": 8504731.0,
      "step": 4105
    },
    {
      "entropy": 5.6381782531738285,
      "epoch": 0.37282293178519593,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991010812261181,
      "loss": 5.5973,
      "mean_token_accuracy": 0.15621297359466552,
      "num_tokens": 8514814.0,
      "step": 4110
    },
    {
      "entropy": 5.740231275558472,
      "epoch": 0.37327648766328014,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004990981894919248,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1475978210568428,
      "num_tokens": 8525486.0,
      "step": 4115
    },
    {
      "entropy": 5.844956254959106,
      "epoch": 0.3737300435413643,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004990952931233238,
      "loss": 5.6066,
      "mean_token_accuracy": 0.15925588607788085,
      "num_tokens": 8535290.0,
      "step": 4120
    },
    {
      "entropy": 5.746306133270264,
      "epoch": 0.37418359941944845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000499092392120375,
      "loss": 5.7572,
      "mean_token_accuracy": 0.15315198004245759,
      "num_tokens": 8545506.0,
      "step": 4125
    },
    {
      "entropy": 5.788954257965088,
      "epoch": 0.37463715529753266,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004990894864831385,
      "loss": 5.7196,
      "mean_token_accuracy": 0.15347600281238555,
      "num_tokens": 8555360.0,
      "step": 4130
    },
    {
      "entropy": 5.856559896469117,
      "epoch": 0.3750907111756168,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004990865762116743,
      "loss": 5.7694,
      "mean_token_accuracy": 0.1511617124080658,
      "num_tokens": 8565582.0,
      "step": 4135
    },
    {
      "entropy": 5.810130071640015,
      "epoch": 0.37554426705370103,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004990836613060426,
      "loss": 5.7718,
      "mean_token_accuracy": 0.1501256100833416,
      "num_tokens": 8576043.0,
      "step": 4140
    },
    {
      "entropy": 5.712186050415039,
      "epoch": 0.3759978229317852,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004990807417663036,
      "loss": 5.7407,
      "mean_token_accuracy": 0.15447329580783845,
      "num_tokens": 8586538.0,
      "step": 4145
    },
    {
      "entropy": 5.796316289901734,
      "epoch": 0.3764513788098694,
      "grad_norm": 2.125,
      "learning_rate": 0.0004990778175925179,
      "loss": 5.6664,
      "mean_token_accuracy": 0.16033863127231598,
      "num_tokens": 8596622.0,
      "step": 4150
    },
    {
      "entropy": 5.796439552307129,
      "epoch": 0.37690493468795355,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004990748887847457,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15430489629507066,
      "num_tokens": 8605945.0,
      "step": 4155
    },
    {
      "entropy": 5.768828010559082,
      "epoch": 0.37735849056603776,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004990719553430477,
      "loss": 5.7323,
      "mean_token_accuracy": 0.14899329990148544,
      "num_tokens": 8615841.0,
      "step": 4160
    },
    {
      "entropy": 5.858073425292969,
      "epoch": 0.3778120464441219,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004990690172674847,
      "loss": 5.7988,
      "mean_token_accuracy": 0.15224681198596954,
      "num_tokens": 8627228.0,
      "step": 4165
    },
    {
      "entropy": 5.792933702468872,
      "epoch": 0.3782656023222061,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004990660745581172,
      "loss": 5.7821,
      "mean_token_accuracy": 0.15444036573171616,
      "num_tokens": 8637848.0,
      "step": 4170
    },
    {
      "entropy": 5.786196756362915,
      "epoch": 0.3787191582002903,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004990631272150062,
      "loss": 5.6905,
      "mean_token_accuracy": 0.14932020604610444,
      "num_tokens": 8648147.0,
      "step": 4175
    },
    {
      "entropy": 5.8579583168029785,
      "epoch": 0.37917271407837444,
      "grad_norm": 1.5,
      "learning_rate": 0.0004990601752382127,
      "loss": 5.799,
      "mean_token_accuracy": 0.14665968865156173,
      "num_tokens": 8659464.0,
      "step": 4180
    },
    {
      "entropy": 5.8483274459838865,
      "epoch": 0.37962626995645865,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004990572186277975,
      "loss": 5.6859,
      "mean_token_accuracy": 0.15550817549228668,
      "num_tokens": 8670435.0,
      "step": 4185
    },
    {
      "entropy": 5.801906681060791,
      "epoch": 0.3800798258345428,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990542573838221,
      "loss": 5.7537,
      "mean_token_accuracy": 0.14839256703853607,
      "num_tokens": 8681193.0,
      "step": 4190
    },
    {
      "entropy": 5.90633544921875,
      "epoch": 0.380533381712627,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004990512915063475,
      "loss": 5.8735,
      "mean_token_accuracy": 0.14305422231554984,
      "num_tokens": 8692191.0,
      "step": 4195
    },
    {
      "entropy": 5.803619432449341,
      "epoch": 0.38098693759071117,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004990483209954352,
      "loss": 5.6579,
      "mean_token_accuracy": 0.15340645760297775,
      "num_tokens": 8703199.0,
      "step": 4200
    },
    {
      "entropy": 5.755993366241455,
      "epoch": 0.3814404934687954,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004990453458511464,
      "loss": 5.7134,
      "mean_token_accuracy": 0.14746004939079285,
      "num_tokens": 8713850.0,
      "step": 4205
    },
    {
      "entropy": 5.801353931427002,
      "epoch": 0.38189404934687954,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004990423660735428,
      "loss": 5.7671,
      "mean_token_accuracy": 0.1433395117521286,
      "num_tokens": 8724045.0,
      "step": 4210
    },
    {
      "entropy": 5.807311630249023,
      "epoch": 0.3823476052249637,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000499039381662686,
      "loss": 5.7119,
      "mean_token_accuracy": 0.15545451492071152,
      "num_tokens": 8734187.0,
      "step": 4215
    },
    {
      "entropy": 5.832239580154419,
      "epoch": 0.3828011611030479,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990363926186377,
      "loss": 5.7582,
      "mean_token_accuracy": 0.15328917652368546,
      "num_tokens": 8744553.0,
      "step": 4220
    },
    {
      "entropy": 5.762201309204102,
      "epoch": 0.38325471698113206,
      "grad_norm": 1.75,
      "learning_rate": 0.0004990333989414597,
      "loss": 5.6831,
      "mean_token_accuracy": 0.15205525010824203,
      "num_tokens": 8755209.0,
      "step": 4225
    },
    {
      "entropy": 5.778427696228027,
      "epoch": 0.38370827285921627,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004990304006312138,
      "loss": 5.733,
      "mean_token_accuracy": 0.147205713391304,
      "num_tokens": 8765552.0,
      "step": 4230
    },
    {
      "entropy": 5.735978126525879,
      "epoch": 0.3841618287373004,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004990273976879622,
      "loss": 5.6384,
      "mean_token_accuracy": 0.15794021636247635,
      "num_tokens": 8776118.0,
      "step": 4235
    },
    {
      "entropy": 5.715007734298706,
      "epoch": 0.38461538461538464,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004990243901117669,
      "loss": 5.7085,
      "mean_token_accuracy": 0.15574559867382048,
      "num_tokens": 8785927.0,
      "step": 4240
    },
    {
      "entropy": 5.883077335357666,
      "epoch": 0.3850689404934688,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004990213779026903,
      "loss": 5.8626,
      "mean_token_accuracy": 0.14674270004034043,
      "num_tokens": 8797257.0,
      "step": 4245
    },
    {
      "entropy": 5.878513336181641,
      "epoch": 0.385522496371553,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004990183610607944,
      "loss": 5.7492,
      "mean_token_accuracy": 0.14894410818815232,
      "num_tokens": 8808266.0,
      "step": 4250
    },
    {
      "entropy": 5.755019950866699,
      "epoch": 0.38597605224963716,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004990153395861415,
      "loss": 5.6602,
      "mean_token_accuracy": 0.15247720181941987,
      "num_tokens": 8818989.0,
      "step": 4255
    },
    {
      "entropy": 5.732574892044068,
      "epoch": 0.3864296081277213,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004990123134787946,
      "loss": 5.7704,
      "mean_token_accuracy": 0.1495862066745758,
      "num_tokens": 8828352.0,
      "step": 4260
    },
    {
      "entropy": 5.792064380645752,
      "epoch": 0.3868831640058055,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004990092827388159,
      "loss": 5.6837,
      "mean_token_accuracy": 0.15442929714918135,
      "num_tokens": 8838743.0,
      "step": 4265
    },
    {
      "entropy": 5.794605159759522,
      "epoch": 0.3873367198838897,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004990062473662681,
      "loss": 5.7208,
      "mean_token_accuracy": 0.14651462882757188,
      "num_tokens": 8848865.0,
      "step": 4270
    },
    {
      "entropy": 5.7486231327056885,
      "epoch": 0.3877902757619739,
      "grad_norm": 1.78125,
      "learning_rate": 0.000499003207361214,
      "loss": 5.6718,
      "mean_token_accuracy": 0.15159275084733964,
      "num_tokens": 8859156.0,
      "step": 4275
    },
    {
      "entropy": 5.886171054840088,
      "epoch": 0.38824383164005805,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004990001627237165,
      "loss": 5.7152,
      "mean_token_accuracy": 0.15245936065912247,
      "num_tokens": 8868860.0,
      "step": 4280
    },
    {
      "entropy": 5.804189872741699,
      "epoch": 0.38869738751814226,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004989971134538386,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1453443519771099,
      "num_tokens": 8879355.0,
      "step": 4285
    },
    {
      "entropy": 5.769974565505981,
      "epoch": 0.3891509433962264,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004989940595516432,
      "loss": 5.7946,
      "mean_token_accuracy": 0.14543934911489487,
      "num_tokens": 8890116.0,
      "step": 4290
    },
    {
      "entropy": 5.792919635772705,
      "epoch": 0.3896044992743106,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004989910010171938,
      "loss": 5.7411,
      "mean_token_accuracy": 0.14830279499292373,
      "num_tokens": 8901243.0,
      "step": 4295
    },
    {
      "entropy": 5.892558193206787,
      "epoch": 0.3900580551523948,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004989879378505532,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1488331064581871,
      "num_tokens": 8911064.0,
      "step": 4300
    },
    {
      "entropy": 5.760198783874512,
      "epoch": 0.39051161103047893,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004989848700517849,
      "loss": 5.677,
      "mean_token_accuracy": 0.15363716781139375,
      "num_tokens": 8921659.0,
      "step": 4305
    },
    {
      "entropy": 5.769484138488769,
      "epoch": 0.39096516690856314,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004989817976209526,
      "loss": 5.6531,
      "mean_token_accuracy": 0.15186847597360612,
      "num_tokens": 8932424.0,
      "step": 4310
    },
    {
      "entropy": 5.780123567581176,
      "epoch": 0.3914187227866473,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004989787205581196,
      "loss": 5.7394,
      "mean_token_accuracy": 0.1513989195227623,
      "num_tokens": 8943097.0,
      "step": 4315
    },
    {
      "entropy": 5.7439628601074215,
      "epoch": 0.3918722786647315,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004989756388633496,
      "loss": 5.7086,
      "mean_token_accuracy": 0.15355851948261262,
      "num_tokens": 8953817.0,
      "step": 4320
    },
    {
      "entropy": 5.824839019775391,
      "epoch": 0.39232583454281567,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004989725525367063,
      "loss": 5.7412,
      "mean_token_accuracy": 0.15554174035787582,
      "num_tokens": 8963731.0,
      "step": 4325
    },
    {
      "entropy": 5.878753566741944,
      "epoch": 0.3927793904208999,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004989694615782534,
      "loss": 5.8384,
      "mean_token_accuracy": 0.14490067809820176,
      "num_tokens": 8974063.0,
      "step": 4330
    },
    {
      "entropy": 5.821165657043457,
      "epoch": 0.39323294629898403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004989663659880552,
      "loss": 5.642,
      "mean_token_accuracy": 0.15681983530521393,
      "num_tokens": 8984153.0,
      "step": 4335
    },
    {
      "entropy": 5.798787260055542,
      "epoch": 0.3936865021770682,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004989632657661754,
      "loss": 5.6806,
      "mean_token_accuracy": 0.15686829388141632,
      "num_tokens": 8994043.0,
      "step": 4340
    },
    {
      "entropy": 5.813419961929322,
      "epoch": 0.3941400580551524,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000498960160912678,
      "loss": 5.79,
      "mean_token_accuracy": 0.14800060093402861,
      "num_tokens": 9004852.0,
      "step": 4345
    },
    {
      "entropy": 5.790053844451904,
      "epoch": 0.39459361393323655,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004989570514276276,
      "loss": 5.6228,
      "mean_token_accuracy": 0.15587036162614823,
      "num_tokens": 9015314.0,
      "step": 4350
    },
    {
      "entropy": 5.7702381134033205,
      "epoch": 0.39504716981132076,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989539373110883,
      "loss": 5.709,
      "mean_token_accuracy": 0.15190639197826386,
      "num_tokens": 9026191.0,
      "step": 4355
    },
    {
      "entropy": 5.803680753707885,
      "epoch": 0.3955007256894049,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004989508185631245,
      "loss": 5.7418,
      "mean_token_accuracy": 0.15162685811519622,
      "num_tokens": 9037160.0,
      "step": 4360
    },
    {
      "entropy": 5.825860404968262,
      "epoch": 0.39595428156748913,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004989476951838006,
      "loss": 5.8152,
      "mean_token_accuracy": 0.14626863971352577,
      "num_tokens": 9047397.0,
      "step": 4365
    },
    {
      "entropy": 5.858776044845581,
      "epoch": 0.3964078374455733,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004989445671731814,
      "loss": 5.7194,
      "mean_token_accuracy": 0.14717085361480714,
      "num_tokens": 9057416.0,
      "step": 4370
    },
    {
      "entropy": 5.869134092330933,
      "epoch": 0.3968613933236575,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989414345313315,
      "loss": 5.746,
      "mean_token_accuracy": 0.15247844010591508,
      "num_tokens": 9068428.0,
      "step": 4375
    },
    {
      "entropy": 5.849895668029785,
      "epoch": 0.39731494920174165,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004989382972583156,
      "loss": 5.7836,
      "mean_token_accuracy": 0.1457201287150383,
      "num_tokens": 9079443.0,
      "step": 4380
    },
    {
      "entropy": 5.742780542373657,
      "epoch": 0.3977685050798258,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004989351553541986,
      "loss": 5.6747,
      "mean_token_accuracy": 0.1511688381433487,
      "num_tokens": 9089856.0,
      "step": 4385
    },
    {
      "entropy": 5.7647631645202635,
      "epoch": 0.39822206095791,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004989320088190456,
      "loss": 5.7425,
      "mean_token_accuracy": 0.15218283236026764,
      "num_tokens": 9100893.0,
      "step": 4390
    },
    {
      "entropy": 5.862979936599731,
      "epoch": 0.3986756168359942,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004989288576529216,
      "loss": 5.7928,
      "mean_token_accuracy": 0.14252141565084459,
      "num_tokens": 9112308.0,
      "step": 4395
    },
    {
      "entropy": 5.785729789733887,
      "epoch": 0.3991291727140784,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004989257018558917,
      "loss": 5.7633,
      "mean_token_accuracy": 0.14584366753697395,
      "num_tokens": 9123052.0,
      "step": 4400
    },
    {
      "entropy": 5.81339020729065,
      "epoch": 0.39958272859216254,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004989225414280213,
      "loss": 5.7341,
      "mean_token_accuracy": 0.15271426737308502,
      "num_tokens": 9134117.0,
      "step": 4405
    },
    {
      "entropy": 5.829674625396729,
      "epoch": 0.40003628447024675,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989193763693756,
      "loss": 5.8163,
      "mean_token_accuracy": 0.150883536785841,
      "num_tokens": 9145737.0,
      "step": 4410
    },
    {
      "entropy": 5.812243700027466,
      "epoch": 0.4004898403483309,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004989162066800202,
      "loss": 5.689,
      "mean_token_accuracy": 0.1514497235417366,
      "num_tokens": 9156443.0,
      "step": 4415
    },
    {
      "entropy": 5.752085208892822,
      "epoch": 0.4009433962264151,
      "grad_norm": 1.375,
      "learning_rate": 0.0004989130323600205,
      "loss": 5.727,
      "mean_token_accuracy": 0.15138549357652664,
      "num_tokens": 9166657.0,
      "step": 4420
    },
    {
      "entropy": 5.812001085281372,
      "epoch": 0.40139695210449927,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004989098534094424,
      "loss": 5.7012,
      "mean_token_accuracy": 0.14933129101991655,
      "num_tokens": 9177475.0,
      "step": 4425
    },
    {
      "entropy": 5.821232891082763,
      "epoch": 0.4018505079825834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004989066698283512,
      "loss": 5.7191,
      "mean_token_accuracy": 0.15208391845226288,
      "num_tokens": 9187250.0,
      "step": 4430
    },
    {
      "entropy": 5.799831676483154,
      "epoch": 0.40230406386066764,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004989034816168132,
      "loss": 5.7489,
      "mean_token_accuracy": 0.14922597855329514,
      "num_tokens": 9198071.0,
      "step": 4435
    },
    {
      "entropy": 5.785824966430664,
      "epoch": 0.4027576197387518,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004989002887748941,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1565877303481102,
      "num_tokens": 9208102.0,
      "step": 4440
    },
    {
      "entropy": 5.717708539962769,
      "epoch": 0.403211175616836,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004988970913026601,
      "loss": 5.6304,
      "mean_token_accuracy": 0.1590559408068657,
      "num_tokens": 9217564.0,
      "step": 4445
    },
    {
      "entropy": 5.740492057800293,
      "epoch": 0.40366473149492016,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004988938892001771,
      "loss": 5.6798,
      "mean_token_accuracy": 0.15244835019111633,
      "num_tokens": 9228027.0,
      "step": 4450
    },
    {
      "entropy": 5.846228504180909,
      "epoch": 0.40411828737300437,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004988906824675113,
      "loss": 5.7565,
      "mean_token_accuracy": 0.14298200979828835,
      "num_tokens": 9239073.0,
      "step": 4455
    },
    {
      "entropy": 5.837181520462036,
      "epoch": 0.4045718432510885,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004988874711047293,
      "loss": 5.7421,
      "mean_token_accuracy": 0.15341956913471222,
      "num_tokens": 9249936.0,
      "step": 4460
    },
    {
      "entropy": 5.770998907089234,
      "epoch": 0.40502539912917274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004988842551118972,
      "loss": 5.7247,
      "mean_token_accuracy": 0.15381136983633042,
      "num_tokens": 9259758.0,
      "step": 4465
    },
    {
      "entropy": 5.762871217727661,
      "epoch": 0.4054789550072569,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004988810344890818,
      "loss": 5.668,
      "mean_token_accuracy": 0.15472375750541686,
      "num_tokens": 9270731.0,
      "step": 4470
    },
    {
      "entropy": 5.7881018161773685,
      "epoch": 0.40593251088534105,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004988778092363494,
      "loss": 5.6447,
      "mean_token_accuracy": 0.14924555122852326,
      "num_tokens": 9281377.0,
      "step": 4475
    },
    {
      "entropy": 5.760777235031128,
      "epoch": 0.40638606676342526,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988745793537671,
      "loss": 5.7338,
      "mean_token_accuracy": 0.15252517759799958,
      "num_tokens": 9290477.0,
      "step": 4480
    },
    {
      "entropy": 5.755960464477539,
      "epoch": 0.4068396226415094,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004988713448414011,
      "loss": 5.6228,
      "mean_token_accuracy": 0.15938864648342133,
      "num_tokens": 9300129.0,
      "step": 4485
    },
    {
      "entropy": 5.784805202484131,
      "epoch": 0.4072931785195936,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000498868105699319,
      "loss": 5.6727,
      "mean_token_accuracy": 0.1487440623342991,
      "num_tokens": 9310734.0,
      "step": 4490
    },
    {
      "entropy": 5.710954809188843,
      "epoch": 0.4077467343976778,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004988648619275872,
      "loss": 5.7175,
      "mean_token_accuracy": 0.1490918666124344,
      "num_tokens": 9320470.0,
      "step": 4495
    },
    {
      "entropy": 5.719701671600342,
      "epoch": 0.408200290275762,
      "grad_norm": 1.5,
      "learning_rate": 0.0004988616135262729,
      "loss": 5.6566,
      "mean_token_accuracy": 0.15467289835214615,
      "num_tokens": 9330480.0,
      "step": 4500
    },
    {
      "entropy": 5.829697036743164,
      "epoch": 0.40865384615384615,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004988583604954436,
      "loss": 5.6992,
      "mean_token_accuracy": 0.1557617023587227,
      "num_tokens": 9342101.0,
      "step": 4505
    },
    {
      "entropy": 5.828498792648316,
      "epoch": 0.40910740203193036,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004988551028351663,
      "loss": 5.7287,
      "mean_token_accuracy": 0.15712219029664992,
      "num_tokens": 9352792.0,
      "step": 4510
    },
    {
      "entropy": 5.765654420852661,
      "epoch": 0.4095609579100145,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004988518405455084,
      "loss": 5.69,
      "mean_token_accuracy": 0.15821951627731323,
      "num_tokens": 9363883.0,
      "step": 4515
    },
    {
      "entropy": 5.798408794403076,
      "epoch": 0.41001451378809867,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004988485736265374,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1483925238251686,
      "num_tokens": 9375434.0,
      "step": 4520
    },
    {
      "entropy": 5.761251735687256,
      "epoch": 0.4104680696661829,
      "grad_norm": 1.359375,
      "learning_rate": 0.000498845302078321,
      "loss": 5.6998,
      "mean_token_accuracy": 0.14736573696136473,
      "num_tokens": 9386430.0,
      "step": 4525
    },
    {
      "entropy": 5.8544196605682375,
      "epoch": 0.41092162554426703,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004988420259009266,
      "loss": 5.7617,
      "mean_token_accuracy": 0.15411313623189926,
      "num_tokens": 9396043.0,
      "step": 4530
    },
    {
      "entropy": 5.7836181640625,
      "epoch": 0.41137518142235124,
      "grad_norm": 1.328125,
      "learning_rate": 0.000498838745094422,
      "loss": 5.6943,
      "mean_token_accuracy": 0.15280985534191133,
      "num_tokens": 9406346.0,
      "step": 4535
    },
    {
      "entropy": 5.783379983901978,
      "epoch": 0.4118287373004354,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004988354596588751,
      "loss": 5.664,
      "mean_token_accuracy": 0.1566889375448227,
      "num_tokens": 9416171.0,
      "step": 4540
    },
    {
      "entropy": 5.761880350112915,
      "epoch": 0.4122822931785196,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004988321695943539,
      "loss": 5.6857,
      "mean_token_accuracy": 0.1549982585012913,
      "num_tokens": 9426123.0,
      "step": 4545
    },
    {
      "entropy": 5.72822413444519,
      "epoch": 0.41273584905660377,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004988288749009264,
      "loss": 5.5885,
      "mean_token_accuracy": 0.15963514298200607,
      "num_tokens": 9437173.0,
      "step": 4550
    },
    {
      "entropy": 5.712554788589477,
      "epoch": 0.413189404934688,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004988255755786608,
      "loss": 5.7518,
      "mean_token_accuracy": 0.14644763842225075,
      "num_tokens": 9446928.0,
      "step": 4555
    },
    {
      "entropy": 5.7979991912841795,
      "epoch": 0.41364296081277213,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004988222716276251,
      "loss": 5.6679,
      "mean_token_accuracy": 0.15122124552726746,
      "num_tokens": 9458180.0,
      "step": 4560
    },
    {
      "entropy": 5.84049243927002,
      "epoch": 0.4140965166908563,
      "grad_norm": 1.3125,
      "learning_rate": 0.000498818963047888,
      "loss": 5.7025,
      "mean_token_accuracy": 0.15793658643960953,
      "num_tokens": 9467791.0,
      "step": 4565
    },
    {
      "entropy": 5.761186265945435,
      "epoch": 0.4145500725689405,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004988156498395176,
      "loss": 5.7532,
      "mean_token_accuracy": 0.14440502598881721,
      "num_tokens": 9478567.0,
      "step": 4570
    },
    {
      "entropy": 5.799375820159912,
      "epoch": 0.41500362844702465,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004988123320025825,
      "loss": 5.7404,
      "mean_token_accuracy": 0.14579950124025345,
      "num_tokens": 9489226.0,
      "step": 4575
    },
    {
      "entropy": 5.769663286209107,
      "epoch": 0.41545718432510886,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004988090095371514,
      "loss": 5.7103,
      "mean_token_accuracy": 0.14600292295217515,
      "num_tokens": 9499598.0,
      "step": 4580
    },
    {
      "entropy": 5.799290132522583,
      "epoch": 0.415910740203193,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004988056824432928,
      "loss": 5.7884,
      "mean_token_accuracy": 0.1523233249783516,
      "num_tokens": 9509128.0,
      "step": 4585
    },
    {
      "entropy": 5.859962844848633,
      "epoch": 0.41636429608127723,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004988023507210758,
      "loss": 5.6945,
      "mean_token_accuracy": 0.15456788688898088,
      "num_tokens": 9519564.0,
      "step": 4590
    },
    {
      "entropy": 5.826257753372192,
      "epoch": 0.4168178519593614,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004987990143705692,
      "loss": 5.748,
      "mean_token_accuracy": 0.15228485465049743,
      "num_tokens": 9530393.0,
      "step": 4595
    },
    {
      "entropy": 5.765258502960205,
      "epoch": 0.4172714078374456,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004987956733918418,
      "loss": 5.7572,
      "mean_token_accuracy": 0.1483850434422493,
      "num_tokens": 9540746.0,
      "step": 4600
    },
    {
      "entropy": 5.794782209396362,
      "epoch": 0.41772496371552975,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004987923277849629,
      "loss": 5.7277,
      "mean_token_accuracy": 0.1539371058344841,
      "num_tokens": 9550611.0,
      "step": 4605
    },
    {
      "entropy": 5.948241472244263,
      "epoch": 0.4181785195936139,
      "grad_norm": 1.375,
      "learning_rate": 0.0004987889775500017,
      "loss": 5.8003,
      "mean_token_accuracy": 0.14910096898674965,
      "num_tokens": 9560273.0,
      "step": 4610
    },
    {
      "entropy": 5.828892946243286,
      "epoch": 0.4186320754716981,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004987856226870272,
      "loss": 5.7715,
      "mean_token_accuracy": 0.15116819590330124,
      "num_tokens": 9571199.0,
      "step": 4615
    },
    {
      "entropy": 5.7489643573760985,
      "epoch": 0.4190856313497823,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004987822631961092,
      "loss": 5.6641,
      "mean_token_accuracy": 0.1496698245406151,
      "num_tokens": 9580638.0,
      "step": 4620
    },
    {
      "entropy": 5.81583309173584,
      "epoch": 0.4195391872278665,
      "grad_norm": 1.5,
      "learning_rate": 0.000498778899077317,
      "loss": 5.7711,
      "mean_token_accuracy": 0.1492114171385765,
      "num_tokens": 9592067.0,
      "step": 4625
    },
    {
      "entropy": 5.745096969604492,
      "epoch": 0.41999274310595064,
      "grad_norm": 1.4375,
      "learning_rate": 0.00049877553033072,
      "loss": 5.6742,
      "mean_token_accuracy": 0.15567420721054076,
      "num_tokens": 9603151.0,
      "step": 4630
    },
    {
      "entropy": 5.748169183731079,
      "epoch": 0.42044629898403485,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004987721569563882,
      "loss": 5.704,
      "mean_token_accuracy": 0.1470457211136818,
      "num_tokens": 9613651.0,
      "step": 4635
    },
    {
      "entropy": 5.869939565658569,
      "epoch": 0.420899854862119,
      "grad_norm": 1.75,
      "learning_rate": 0.000498768778954391,
      "loss": 5.6987,
      "mean_token_accuracy": 0.1548364281654358,
      "num_tokens": 9623705.0,
      "step": 4640
    },
    {
      "entropy": 5.814824962615967,
      "epoch": 0.4213534107402032,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004987653963247985,
      "loss": 5.7549,
      "mean_token_accuracy": 0.14578017443418503,
      "num_tokens": 9634663.0,
      "step": 4645
    },
    {
      "entropy": 5.807049417495728,
      "epoch": 0.42180696661828737,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004987620090676805,
      "loss": 5.7646,
      "mean_token_accuracy": 0.15146553814411162,
      "num_tokens": 9645071.0,
      "step": 4650
    },
    {
      "entropy": 5.720735311508179,
      "epoch": 0.4222605224963715,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004987586171831072,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15271254628896713,
      "num_tokens": 9654942.0,
      "step": 4655
    },
    {
      "entropy": 5.712645530700684,
      "epoch": 0.42271407837445574,
      "grad_norm": 1.625,
      "learning_rate": 0.0004987552206711487,
      "loss": 5.6622,
      "mean_token_accuracy": 0.15510569512844086,
      "num_tokens": 9665379.0,
      "step": 4660
    },
    {
      "entropy": 5.8077527523040775,
      "epoch": 0.4231676342525399,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004987518195318752,
      "loss": 5.694,
      "mean_token_accuracy": 0.1477640151977539,
      "num_tokens": 9674812.0,
      "step": 4665
    },
    {
      "entropy": 5.798747587203979,
      "epoch": 0.4236211901306241,
      "grad_norm": 1.5,
      "learning_rate": 0.000498748413765357,
      "loss": 5.708,
      "mean_token_accuracy": 0.1464133784174919,
      "num_tokens": 9685765.0,
      "step": 4670
    },
    {
      "entropy": 5.836048650741577,
      "epoch": 0.42407474600870826,
      "grad_norm": 1.625,
      "learning_rate": 0.0004987450033716646,
      "loss": 5.7455,
      "mean_token_accuracy": 0.14856940880417824,
      "num_tokens": 9696747.0,
      "step": 4675
    },
    {
      "entropy": 5.759535217285157,
      "epoch": 0.42452830188679247,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004987415883508686,
      "loss": 5.7376,
      "mean_token_accuracy": 0.1529595822095871,
      "num_tokens": 9707431.0,
      "step": 4680
    },
    {
      "entropy": 5.876302242279053,
      "epoch": 0.4249818577648766,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004987381687030394,
      "loss": 5.7691,
      "mean_token_accuracy": 0.14749758690595627,
      "num_tokens": 9717454.0,
      "step": 4685
    },
    {
      "entropy": 5.872333002090454,
      "epoch": 0.42543541364296084,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004987347444282479,
      "loss": 5.756,
      "mean_token_accuracy": 0.15242287069559096,
      "num_tokens": 9726803.0,
      "step": 4690
    },
    {
      "entropy": 5.770662260055542,
      "epoch": 0.425888969521045,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004987313155265649,
      "loss": 5.7407,
      "mean_token_accuracy": 0.14523900747299195,
      "num_tokens": 9737089.0,
      "step": 4695
    },
    {
      "entropy": 5.7127138614654545,
      "epoch": 0.42634252539912915,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004987278819980613,
      "loss": 5.6293,
      "mean_token_accuracy": 0.1546218603849411,
      "num_tokens": 9748698.0,
      "step": 4700
    },
    {
      "entropy": 5.7299487590789795,
      "epoch": 0.42679608127721336,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004987244438428081,
      "loss": 5.5827,
      "mean_token_accuracy": 0.15895820856094361,
      "num_tokens": 9758940.0,
      "step": 4705
    },
    {
      "entropy": 5.66409649848938,
      "epoch": 0.4272496371552975,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004987210010608764,
      "loss": 5.6831,
      "mean_token_accuracy": 0.15151547640562057,
      "num_tokens": 9769051.0,
      "step": 4710
    },
    {
      "entropy": 5.661192941665649,
      "epoch": 0.4277031930333817,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004987175536523373,
      "loss": 5.6012,
      "mean_token_accuracy": 0.1603727474808693,
      "num_tokens": 9779421.0,
      "step": 4715
    },
    {
      "entropy": 5.846398544311524,
      "epoch": 0.4281567489114659,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004987141016172622,
      "loss": 5.6113,
      "mean_token_accuracy": 0.15543566793203353,
      "num_tokens": 9790003.0,
      "step": 4720
    },
    {
      "entropy": 5.727094078063965,
      "epoch": 0.4286103047895501,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004987106449557225,
      "loss": 5.6259,
      "mean_token_accuracy": 0.15138356983661652,
      "num_tokens": 9800085.0,
      "step": 4725
    },
    {
      "entropy": 5.669862222671509,
      "epoch": 0.42906386066763424,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004987071836677896,
      "loss": 5.6815,
      "mean_token_accuracy": 0.1575853556394577,
      "num_tokens": 9809873.0,
      "step": 4730
    },
    {
      "entropy": 5.779130697250366,
      "epoch": 0.42951741654571846,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004987037177535353,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1511673741042614,
      "num_tokens": 9820137.0,
      "step": 4735
    },
    {
      "entropy": 5.789821195602417,
      "epoch": 0.4299709724238026,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004987002472130309,
      "loss": 5.7161,
      "mean_token_accuracy": 0.15311633199453353,
      "num_tokens": 9830758.0,
      "step": 4740
    },
    {
      "entropy": 5.760760021209717,
      "epoch": 0.43042452830188677,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004986967720463485,
      "loss": 5.6183,
      "mean_token_accuracy": 0.1549585849046707,
      "num_tokens": 9841594.0,
      "step": 4745
    },
    {
      "entropy": 5.763367462158203,
      "epoch": 0.430878084179971,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004986932922535598,
      "loss": 5.7617,
      "mean_token_accuracy": 0.1477864995598793,
      "num_tokens": 9853640.0,
      "step": 4750
    },
    {
      "entropy": 5.851624822616577,
      "epoch": 0.43133164005805513,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004986898078347368,
      "loss": 5.7814,
      "mean_token_accuracy": 0.15149234533309935,
      "num_tokens": 9864550.0,
      "step": 4755
    },
    {
      "entropy": 5.867263603210449,
      "epoch": 0.43178519593613934,
      "grad_norm": 1.625,
      "learning_rate": 0.0004986863187899516,
      "loss": 5.7169,
      "mean_token_accuracy": 0.1487038627266884,
      "num_tokens": 9875544.0,
      "step": 4760
    },
    {
      "entropy": 5.818501996994018,
      "epoch": 0.4322387518142235,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004986828251192764,
      "loss": 5.7372,
      "mean_token_accuracy": 0.15489247292280198,
      "num_tokens": 9886387.0,
      "step": 4765
    },
    {
      "entropy": 5.733272361755371,
      "epoch": 0.4326923076923077,
      "grad_norm": 1.375,
      "learning_rate": 0.0004986793268227833,
      "loss": 5.7354,
      "mean_token_accuracy": 0.15493058413267136,
      "num_tokens": 9896340.0,
      "step": 4770
    },
    {
      "entropy": 5.778052186965942,
      "epoch": 0.43314586357039186,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004986758239005448,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1477018713951111,
      "num_tokens": 9907587.0,
      "step": 4775
    },
    {
      "entropy": 5.782829904556275,
      "epoch": 0.4335994194484761,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004986723163526331,
      "loss": 5.6101,
      "mean_token_accuracy": 0.15842603147029877,
      "num_tokens": 9917477.0,
      "step": 4780
    },
    {
      "entropy": 5.7354288578033445,
      "epoch": 0.43405297532656023,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004986688041791209,
      "loss": 5.6787,
      "mean_token_accuracy": 0.15261090695858,
      "num_tokens": 9927808.0,
      "step": 4785
    },
    {
      "entropy": 5.717248201370239,
      "epoch": 0.4345065312046444,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000498665287380081,
      "loss": 5.5854,
      "mean_token_accuracy": 0.1633790224790573,
      "num_tokens": 9937322.0,
      "step": 4790
    },
    {
      "entropy": 5.747660779953003,
      "epoch": 0.4349600870827286,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004986617659555859,
      "loss": 5.7422,
      "mean_token_accuracy": 0.15461928248405457,
      "num_tokens": 9947259.0,
      "step": 4795
    },
    {
      "entropy": 5.845655250549316,
      "epoch": 0.43541364296081275,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004986582399057085,
      "loss": 5.7223,
      "mean_token_accuracy": 0.14935103803873062,
      "num_tokens": 9957660.0,
      "step": 4800
    },
    {
      "entropy": 5.751298189163208,
      "epoch": 0.43586719883889696,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004986547092305216,
      "loss": 5.6079,
      "mean_token_accuracy": 0.15811071246862413,
      "num_tokens": 9967460.0,
      "step": 4805
    },
    {
      "entropy": 5.832250690460205,
      "epoch": 0.4363207547169811,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004986511739300984,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1497562661767006,
      "num_tokens": 9979176.0,
      "step": 4810
    },
    {
      "entropy": 5.822063636779785,
      "epoch": 0.43677431059506533,
      "grad_norm": 3.0,
      "learning_rate": 0.0004986476340045119,
      "loss": 5.6101,
      "mean_token_accuracy": 0.1614718809723854,
      "num_tokens": 9989025.0,
      "step": 4815
    },
    {
      "entropy": 5.800371837615967,
      "epoch": 0.4372278664731495,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004986440894538354,
      "loss": 5.7132,
      "mean_token_accuracy": 0.14864002764225007,
      "num_tokens": 9998887.0,
      "step": 4820
    },
    {
      "entropy": 5.843149709701538,
      "epoch": 0.4376814223512337,
      "grad_norm": 1.5,
      "learning_rate": 0.0004986405402781421,
      "loss": 5.8243,
      "mean_token_accuracy": 0.14749663099646568,
      "num_tokens": 10009425.0,
      "step": 4825
    },
    {
      "entropy": 5.774194431304932,
      "epoch": 0.43813497822931785,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004986369864775054,
      "loss": 5.6857,
      "mean_token_accuracy": 0.1558217227458954,
      "num_tokens": 10019733.0,
      "step": 4830
    },
    {
      "entropy": 5.798313045501709,
      "epoch": 0.438588534107402,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004986334280519989,
      "loss": 5.6947,
      "mean_token_accuracy": 0.14939245581626892,
      "num_tokens": 10029829.0,
      "step": 4835
    },
    {
      "entropy": 5.727769613265991,
      "epoch": 0.4390420899854862,
      "grad_norm": 1.4375,
      "learning_rate": 0.000498629865001696,
      "loss": 5.5467,
      "mean_token_accuracy": 0.15010637938976287,
      "num_tokens": 10040699.0,
      "step": 4840
    },
    {
      "entropy": 5.744697904586792,
      "epoch": 0.43949564586357037,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004986262973266706,
      "loss": 5.649,
      "mean_token_accuracy": 0.15142616853117943,
      "num_tokens": 10051956.0,
      "step": 4845
    },
    {
      "entropy": 5.784964275360108,
      "epoch": 0.4399492017416546,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004986227250269964,
      "loss": 5.6888,
      "mean_token_accuracy": 0.15234601348638535,
      "num_tokens": 10062300.0,
      "step": 4850
    },
    {
      "entropy": 5.750584316253662,
      "epoch": 0.44040275761973874,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004986191481027472,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1578789621591568,
      "num_tokens": 10072577.0,
      "step": 4855
    },
    {
      "entropy": 5.6763123035430905,
      "epoch": 0.44085631349782295,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004986155665539971,
      "loss": 5.6243,
      "mean_token_accuracy": 0.15474955439567567,
      "num_tokens": 10083041.0,
      "step": 4860
    },
    {
      "entropy": 5.720007038116455,
      "epoch": 0.4413098693759071,
      "grad_norm": 1.59375,
      "learning_rate": 0.00049861198038082,
      "loss": 5.7336,
      "mean_token_accuracy": 0.15288010090589524,
      "num_tokens": 10092927.0,
      "step": 4865
    },
    {
      "entropy": 5.853694152832031,
      "epoch": 0.4417634252539913,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004986083895832903,
      "loss": 5.7635,
      "mean_token_accuracy": 0.15163737162947655,
      "num_tokens": 10104033.0,
      "step": 4870
    },
    {
      "entropy": 5.834793519973755,
      "epoch": 0.44221698113207547,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004986047941614821,
      "loss": 5.6622,
      "mean_token_accuracy": 0.15353917330503464,
      "num_tokens": 10113696.0,
      "step": 4875
    },
    {
      "entropy": 5.733519506454468,
      "epoch": 0.4426705370101596,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004986011941154696,
      "loss": 5.7311,
      "mean_token_accuracy": 0.1490173175930977,
      "num_tokens": 10124725.0,
      "step": 4880
    },
    {
      "entropy": 5.866247892379761,
      "epoch": 0.44312409288824384,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004985975894453275,
      "loss": 5.8738,
      "mean_token_accuracy": 0.13853485733270646,
      "num_tokens": 10135079.0,
      "step": 4885
    },
    {
      "entropy": 5.86741828918457,
      "epoch": 0.443577648766328,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004985939801511304,
      "loss": 5.6795,
      "mean_token_accuracy": 0.15153541415929794,
      "num_tokens": 10146518.0,
      "step": 4890
    },
    {
      "entropy": 5.655025863647461,
      "epoch": 0.4440312046444122,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004985903662329527,
      "loss": 5.5469,
      "mean_token_accuracy": 0.15998917520046235,
      "num_tokens": 10156329.0,
      "step": 4895
    },
    {
      "entropy": 5.750478982925415,
      "epoch": 0.44448476052249636,
      "grad_norm": 1.375,
      "learning_rate": 0.0004985867476908693,
      "loss": 5.7559,
      "mean_token_accuracy": 0.14420461431145667,
      "num_tokens": 10167375.0,
      "step": 4900
    },
    {
      "entropy": 5.782648181915283,
      "epoch": 0.44493831640058057,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004985831245249549,
      "loss": 5.703,
      "mean_token_accuracy": 0.14595248699188232,
      "num_tokens": 10177465.0,
      "step": 4905
    },
    {
      "entropy": 5.838851451873779,
      "epoch": 0.4453918722786647,
      "grad_norm": 1.625,
      "learning_rate": 0.0004985794967352846,
      "loss": 5.7494,
      "mean_token_accuracy": 0.15075748264789582,
      "num_tokens": 10187728.0,
      "step": 4910
    },
    {
      "entropy": 5.80021333694458,
      "epoch": 0.44584542815674894,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004985758643219334,
      "loss": 5.7565,
      "mean_token_accuracy": 0.1465687334537506,
      "num_tokens": 10199289.0,
      "step": 4915
    },
    {
      "entropy": 5.745902061462402,
      "epoch": 0.4462989840348331,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004985722272849762,
      "loss": 5.6708,
      "mean_token_accuracy": 0.15045904517173767,
      "num_tokens": 10209073.0,
      "step": 4920
    },
    {
      "entropy": 5.78634090423584,
      "epoch": 0.44675253991291725,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985685856244884,
      "loss": 5.7942,
      "mean_token_accuracy": 0.13826809749007224,
      "num_tokens": 10220238.0,
      "step": 4925
    },
    {
      "entropy": 5.9017843246459964,
      "epoch": 0.44720609579100146,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004985649393405453,
      "loss": 5.796,
      "mean_token_accuracy": 0.145417732745409,
      "num_tokens": 10230734.0,
      "step": 4930
    },
    {
      "entropy": 5.810334587097168,
      "epoch": 0.4476596516690856,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004985612884332223,
      "loss": 5.6383,
      "mean_token_accuracy": 0.15715192556381224,
      "num_tokens": 10240342.0,
      "step": 4935
    },
    {
      "entropy": 5.726405286788941,
      "epoch": 0.4481132075471698,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004985576329025949,
      "loss": 5.6906,
      "mean_token_accuracy": 0.15500271767377855,
      "num_tokens": 10250418.0,
      "step": 4940
    },
    {
      "entropy": 5.758911609649658,
      "epoch": 0.448566763425254,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985539727487385,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1564280316233635,
      "num_tokens": 10260654.0,
      "step": 4945
    },
    {
      "entropy": 5.7844610691070555,
      "epoch": 0.4490203193033382,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004985503079717292,
      "loss": 5.5911,
      "mean_token_accuracy": 0.15968757420778273,
      "num_tokens": 10270535.0,
      "step": 4950
    },
    {
      "entropy": 5.795976734161377,
      "epoch": 0.44947387518142234,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004985466385716425,
      "loss": 5.7361,
      "mean_token_accuracy": 0.15215763300657273,
      "num_tokens": 10280724.0,
      "step": 4955
    },
    {
      "entropy": 5.7172832012176515,
      "epoch": 0.44992743105950656,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004985429645485543,
      "loss": 5.6148,
      "mean_token_accuracy": 0.15808341056108474,
      "num_tokens": 10290795.0,
      "step": 4960
    },
    {
      "entropy": 5.7305827140808105,
      "epoch": 0.4503809869375907,
      "grad_norm": 2.25,
      "learning_rate": 0.0004985392859025407,
      "loss": 5.7419,
      "mean_token_accuracy": 0.15167693346738814,
      "num_tokens": 10302223.0,
      "step": 4965
    },
    {
      "entropy": 5.719541883468628,
      "epoch": 0.45083454281567487,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004985356026336776,
      "loss": 5.658,
      "mean_token_accuracy": 0.16071119904518127,
      "num_tokens": 10312733.0,
      "step": 4970
    },
    {
      "entropy": 5.861865758895874,
      "epoch": 0.4512880986937591,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004985319147420414,
      "loss": 5.7486,
      "mean_token_accuracy": 0.14967662692070008,
      "num_tokens": 10323315.0,
      "step": 4975
    },
    {
      "entropy": 5.841590309143067,
      "epoch": 0.45174165457184323,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004985282222277082,
      "loss": 5.6615,
      "mean_token_accuracy": 0.1541038677096367,
      "num_tokens": 10333633.0,
      "step": 4980
    },
    {
      "entropy": 5.7527015209198,
      "epoch": 0.45219521044992744,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004985245250907544,
      "loss": 5.7756,
      "mean_token_accuracy": 0.14761256128549577,
      "num_tokens": 10343410.0,
      "step": 4985
    },
    {
      "entropy": 5.879261684417725,
      "epoch": 0.4526487663280116,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004985208233312564,
      "loss": 5.7913,
      "mean_token_accuracy": 0.14990729242563247,
      "num_tokens": 10354029.0,
      "step": 4990
    },
    {
      "entropy": 5.8652578830719,
      "epoch": 0.4531023222060958,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004985171169492908,
      "loss": 5.7681,
      "mean_token_accuracy": 0.14575513750314711,
      "num_tokens": 10364875.0,
      "step": 4995
    },
    {
      "entropy": 5.772519302368164,
      "epoch": 0.45355587808417996,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004985134059449344,
      "loss": 5.7248,
      "mean_token_accuracy": 0.15330636352300644,
      "num_tokens": 10374127.0,
      "step": 5000
    },
    {
      "entropy": 5.794177436828614,
      "epoch": 0.4540094339622642,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004985096903182637,
      "loss": 5.6552,
      "mean_token_accuracy": 0.15620183944702148,
      "num_tokens": 10384583.0,
      "step": 5005
    },
    {
      "entropy": 5.832924699783325,
      "epoch": 0.45446298984034833,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004985059700693557,
      "loss": 5.7785,
      "mean_token_accuracy": 0.14847754538059235,
      "num_tokens": 10394476.0,
      "step": 5010
    },
    {
      "entropy": 5.807150220870971,
      "epoch": 0.4549165457184325,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004985022451982872,
      "loss": 5.7655,
      "mean_token_accuracy": 0.14492400586605073,
      "num_tokens": 10405558.0,
      "step": 5015
    },
    {
      "entropy": 5.819724702835083,
      "epoch": 0.4553701015965167,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004984985157051354,
      "loss": 5.6792,
      "mean_token_accuracy": 0.15750367045402527,
      "num_tokens": 10415746.0,
      "step": 5020
    },
    {
      "entropy": 5.732080554962158,
      "epoch": 0.45582365747460085,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004984947815899774,
      "loss": 5.6187,
      "mean_token_accuracy": 0.15581783503293992,
      "num_tokens": 10424824.0,
      "step": 5025
    },
    {
      "entropy": 5.7898101806640625,
      "epoch": 0.45627721335268506,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004984910428528902,
      "loss": 5.7281,
      "mean_token_accuracy": 0.15406316369771958,
      "num_tokens": 10434175.0,
      "step": 5030
    },
    {
      "entropy": 5.794450139999389,
      "epoch": 0.4567307692307692,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004984872994939514,
      "loss": 5.6427,
      "mean_token_accuracy": 0.15420141369104384,
      "num_tokens": 10444614.0,
      "step": 5035
    },
    {
      "entropy": 5.8216742992401125,
      "epoch": 0.45718432510885343,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004984835515132382,
      "loss": 5.7407,
      "mean_token_accuracy": 0.14913367331027985,
      "num_tokens": 10455605.0,
      "step": 5040
    },
    {
      "entropy": 5.80116376876831,
      "epoch": 0.4576378809869376,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004984797989108282,
      "loss": 5.7201,
      "mean_token_accuracy": 0.148945339769125,
      "num_tokens": 10466912.0,
      "step": 5045
    },
    {
      "entropy": 5.818288421630859,
      "epoch": 0.4580914368650218,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004984760416867991,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1489511951804161,
      "num_tokens": 10477762.0,
      "step": 5050
    },
    {
      "entropy": 5.873276567459106,
      "epoch": 0.45854499274310595,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004984722798412286,
      "loss": 5.7312,
      "mean_token_accuracy": 0.14897401630878448,
      "num_tokens": 10488203.0,
      "step": 5055
    },
    {
      "entropy": 5.770886850357056,
      "epoch": 0.4589985486211901,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004984685133741941,
      "loss": 5.6375,
      "mean_token_accuracy": 0.15436122566461563,
      "num_tokens": 10498174.0,
      "step": 5060
    },
    {
      "entropy": 5.735772180557251,
      "epoch": 0.4594521044992743,
      "grad_norm": 2.0625,
      "learning_rate": 0.000498464742285774,
      "loss": 5.7066,
      "mean_token_accuracy": 0.14874908924102784,
      "num_tokens": 10508591.0,
      "step": 5065
    },
    {
      "entropy": 5.72156777381897,
      "epoch": 0.45990566037735847,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000498460966576046,
      "loss": 5.5856,
      "mean_token_accuracy": 0.15871575921773912,
      "num_tokens": 10518364.0,
      "step": 5070
    },
    {
      "entropy": 5.797338390350342,
      "epoch": 0.4603592162554427,
      "grad_norm": 2.0,
      "learning_rate": 0.0004984571862450884,
      "loss": 5.6794,
      "mean_token_accuracy": 0.1542142927646637,
      "num_tokens": 10528646.0,
      "step": 5075
    },
    {
      "entropy": 5.673431062698365,
      "epoch": 0.46081277213352684,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004984534012929791,
      "loss": 5.6494,
      "mean_token_accuracy": 0.14974019974470137,
      "num_tokens": 10539282.0,
      "step": 5080
    },
    {
      "entropy": 5.775971269607544,
      "epoch": 0.46126632801161105,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004984496117197967,
      "loss": 5.7077,
      "mean_token_accuracy": 0.14850226193666458,
      "num_tokens": 10549713.0,
      "step": 5085
    },
    {
      "entropy": 5.795759105682373,
      "epoch": 0.4617198838896952,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004984458175256193,
      "loss": 5.6496,
      "mean_token_accuracy": 0.1567995548248291,
      "num_tokens": 10559068.0,
      "step": 5090
    },
    {
      "entropy": 5.7790666103363035,
      "epoch": 0.4621734397677794,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004984420187105253,
      "loss": 5.6574,
      "mean_token_accuracy": 0.16136553436517714,
      "num_tokens": 10569633.0,
      "step": 5095
    },
    {
      "entropy": 5.74684739112854,
      "epoch": 0.46262699564586357,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004984382152745936,
      "loss": 5.7028,
      "mean_token_accuracy": 0.15542976260185243,
      "num_tokens": 10579833.0,
      "step": 5100
    },
    {
      "entropy": 5.7991992950439455,
      "epoch": 0.4630805515239477,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004984344072179026,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1511756382882595,
      "num_tokens": 10590864.0,
      "step": 5105
    },
    {
      "entropy": 5.815202140808106,
      "epoch": 0.46353410740203194,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004984305945405312,
      "loss": 5.6984,
      "mean_token_accuracy": 0.1536037266254425,
      "num_tokens": 10601180.0,
      "step": 5110
    },
    {
      "entropy": 5.73317403793335,
      "epoch": 0.4639876632801161,
      "grad_norm": 2.078125,
      "learning_rate": 0.000498426777242558,
      "loss": 5.7378,
      "mean_token_accuracy": 0.15111695379018783,
      "num_tokens": 10611760.0,
      "step": 5115
    },
    {
      "entropy": 5.7392219543457035,
      "epoch": 0.4644412191582003,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004984229553240623,
      "loss": 5.5703,
      "mean_token_accuracy": 0.16025138944387435,
      "num_tokens": 10621926.0,
      "step": 5120
    },
    {
      "entropy": 5.762211799621582,
      "epoch": 0.46489477503628446,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004984191287851228,
      "loss": 5.7128,
      "mean_token_accuracy": 0.15515776872634887,
      "num_tokens": 10632890.0,
      "step": 5125
    },
    {
      "entropy": 5.688881206512451,
      "epoch": 0.46534833091436867,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004984152976258188,
      "loss": 5.5915,
      "mean_token_accuracy": 0.15708402097225188,
      "num_tokens": 10642572.0,
      "step": 5130
    },
    {
      "entropy": 5.825290489196777,
      "epoch": 0.4658018867924528,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004984114618462296,
      "loss": 5.7265,
      "mean_token_accuracy": 0.14965202361345292,
      "num_tokens": 10653732.0,
      "step": 5135
    },
    {
      "entropy": 5.750692415237427,
      "epoch": 0.46625544267053703,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004984076214464343,
      "loss": 5.6035,
      "mean_token_accuracy": 0.15400146842002868,
      "num_tokens": 10664089.0,
      "step": 5140
    },
    {
      "entropy": 5.721158790588379,
      "epoch": 0.4667089985486212,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004984037764265126,
      "loss": 5.6441,
      "mean_token_accuracy": 0.15426179766654968,
      "num_tokens": 10674584.0,
      "step": 5145
    },
    {
      "entropy": 5.777506303787232,
      "epoch": 0.46716255442670535,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004983999267865438,
      "loss": 5.6952,
      "mean_token_accuracy": 0.15464306250214577,
      "num_tokens": 10686455.0,
      "step": 5150
    },
    {
      "entropy": 5.876712417602539,
      "epoch": 0.46761611030478956,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004983960725266078,
      "loss": 5.8241,
      "mean_token_accuracy": 0.15148961544036865,
      "num_tokens": 10697124.0,
      "step": 5155
    },
    {
      "entropy": 5.800732183456421,
      "epoch": 0.4680696661828737,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004983922136467838,
      "loss": 5.6129,
      "mean_token_accuracy": 0.1556944265961647,
      "num_tokens": 10707599.0,
      "step": 5160
    },
    {
      "entropy": 5.733756446838379,
      "epoch": 0.4685232220609579,
      "grad_norm": 2.125,
      "learning_rate": 0.000498388350147152,
      "loss": 5.5655,
      "mean_token_accuracy": 0.15888160169124604,
      "num_tokens": 10717866.0,
      "step": 5165
    },
    {
      "entropy": 5.662814092636109,
      "epoch": 0.4689767779390421,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004983844820277922,
      "loss": 5.646,
      "mean_token_accuracy": 0.15497540682554245,
      "num_tokens": 10728482.0,
      "step": 5170
    },
    {
      "entropy": 5.770923233032226,
      "epoch": 0.4694303338171263,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004983806092887843,
      "loss": 5.7529,
      "mean_token_accuracy": 0.14423309415578842,
      "num_tokens": 10738978.0,
      "step": 5175
    },
    {
      "entropy": 5.77985143661499,
      "epoch": 0.46988388969521044,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004983767319302086,
      "loss": 5.6545,
      "mean_token_accuracy": 0.15800116062164307,
      "num_tokens": 10749618.0,
      "step": 5180
    },
    {
      "entropy": 5.8202619552612305,
      "epoch": 0.47033744557329465,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004983728499521451,
      "loss": 5.6797,
      "mean_token_accuracy": 0.15779803246259688,
      "num_tokens": 10759799.0,
      "step": 5185
    },
    {
      "entropy": 5.807422113418579,
      "epoch": 0.4707910014513788,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004983689633546741,
      "loss": 5.5587,
      "mean_token_accuracy": 0.15637572407722472,
      "num_tokens": 10769807.0,
      "step": 5190
    },
    {
      "entropy": 5.704042100906372,
      "epoch": 0.47124455732946297,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004983650721378761,
      "loss": 5.6368,
      "mean_token_accuracy": 0.15547890961170197,
      "num_tokens": 10780185.0,
      "step": 5195
    },
    {
      "entropy": 5.692027235031128,
      "epoch": 0.4716981132075472,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004983611763018314,
      "loss": 5.7035,
      "mean_token_accuracy": 0.1588817447423935,
      "num_tokens": 10791383.0,
      "step": 5200
    },
    {
      "entropy": 5.7197267532348635,
      "epoch": 0.47215166908563133,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004983572758466208,
      "loss": 5.6834,
      "mean_token_accuracy": 0.15821144729852676,
      "num_tokens": 10801624.0,
      "step": 5205
    },
    {
      "entropy": 5.77974762916565,
      "epoch": 0.47260522496371554,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004983533707723247,
      "loss": 5.7224,
      "mean_token_accuracy": 0.1483788952231407,
      "num_tokens": 10813293.0,
      "step": 5210
    },
    {
      "entropy": 5.776114368438721,
      "epoch": 0.4730587808417997,
      "grad_norm": 2.125,
      "learning_rate": 0.0004983494610790241,
      "loss": 5.5876,
      "mean_token_accuracy": 0.1604739934206009,
      "num_tokens": 10822560.0,
      "step": 5215
    },
    {
      "entropy": 5.70116868019104,
      "epoch": 0.4735123367198839,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004983455467667996,
      "loss": 5.6599,
      "mean_token_accuracy": 0.15497548133134842,
      "num_tokens": 10833360.0,
      "step": 5220
    },
    {
      "entropy": 5.777570104598999,
      "epoch": 0.47396589259796806,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004983416278357322,
      "loss": 5.7556,
      "mean_token_accuracy": 0.14502900540828706,
      "num_tokens": 10844189.0,
      "step": 5225
    },
    {
      "entropy": 5.829122018814087,
      "epoch": 0.4744194484760523,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004983377042859032,
      "loss": 5.6667,
      "mean_token_accuracy": 0.14976238906383516,
      "num_tokens": 10855478.0,
      "step": 5230
    },
    {
      "entropy": 5.74685378074646,
      "epoch": 0.47487300435413643,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004983337761173936,
      "loss": 5.6505,
      "mean_token_accuracy": 0.15890851616859436,
      "num_tokens": 10866853.0,
      "step": 5235
    },
    {
      "entropy": 5.838531494140625,
      "epoch": 0.4753265602322206,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004983298433302843,
      "loss": 5.7633,
      "mean_token_accuracy": 0.14805135056376456,
      "num_tokens": 10877194.0,
      "step": 5240
    },
    {
      "entropy": 5.716617250442505,
      "epoch": 0.4757801161103048,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004983259059246572,
      "loss": 5.6271,
      "mean_token_accuracy": 0.15317686796188354,
      "num_tokens": 10887210.0,
      "step": 5245
    },
    {
      "entropy": 5.771576976776123,
      "epoch": 0.47623367198838895,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004983219639005934,
      "loss": 5.7641,
      "mean_token_accuracy": 0.14675695747137069,
      "num_tokens": 10897742.0,
      "step": 5250
    },
    {
      "entropy": 5.771367502212525,
      "epoch": 0.47668722786647316,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004983180172581745,
      "loss": 5.5752,
      "mean_token_accuracy": 0.15986000001430511,
      "num_tokens": 10907863.0,
      "step": 5255
    },
    {
      "entropy": 5.753606605529785,
      "epoch": 0.4771407837445573,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004983140659974819,
      "loss": 5.7599,
      "mean_token_accuracy": 0.14626236334443093,
      "num_tokens": 10918917.0,
      "step": 5260
    },
    {
      "entropy": 5.780586099624633,
      "epoch": 0.47759433962264153,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004983101101185977,
      "loss": 5.7255,
      "mean_token_accuracy": 0.1551913544535637,
      "num_tokens": 10928614.0,
      "step": 5265
    },
    {
      "entropy": 5.856214570999145,
      "epoch": 0.4780478955007257,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004983061496216035,
      "loss": 5.6736,
      "mean_token_accuracy": 0.151229590177536,
      "num_tokens": 10939093.0,
      "step": 5270
    },
    {
      "entropy": 5.854237842559814,
      "epoch": 0.4785014513788099,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004983021845065812,
      "loss": 5.6972,
      "mean_token_accuracy": 0.15564595758914948,
      "num_tokens": 10949339.0,
      "step": 5275
    },
    {
      "entropy": 5.681471300125122,
      "epoch": 0.47895500725689405,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004982982147736128,
      "loss": 5.6857,
      "mean_token_accuracy": 0.15276624113321305,
      "num_tokens": 10959482.0,
      "step": 5280
    },
    {
      "entropy": 5.7495543479919435,
      "epoch": 0.4794085631349782,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004982942404227805,
      "loss": 5.7244,
      "mean_token_accuracy": 0.15499285459518433,
      "num_tokens": 10970158.0,
      "step": 5285
    },
    {
      "entropy": 5.805167818069458,
      "epoch": 0.4798621190130624,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004982902614541662,
      "loss": 5.6769,
      "mean_token_accuracy": 0.14199863746762276,
      "num_tokens": 10982399.0,
      "step": 5290
    },
    {
      "entropy": 5.797307109832763,
      "epoch": 0.48031567489114657,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004982862778678526,
      "loss": 5.7078,
      "mean_token_accuracy": 0.14736454486846923,
      "num_tokens": 10992319.0,
      "step": 5295
    },
    {
      "entropy": 5.838883543014527,
      "epoch": 0.4807692307692308,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004982822896639218,
      "loss": 5.6662,
      "mean_token_accuracy": 0.14749011844396592,
      "num_tokens": 11001828.0,
      "step": 5300
    },
    {
      "entropy": 5.852616786956787,
      "epoch": 0.48122278664731494,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004982782968424563,
      "loss": 5.7372,
      "mean_token_accuracy": 0.15126340687274933,
      "num_tokens": 11011059.0,
      "step": 5305
    },
    {
      "entropy": 5.8091569423675535,
      "epoch": 0.48167634252539915,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004982742994035388,
      "loss": 5.7193,
      "mean_token_accuracy": 0.15086904913187027,
      "num_tokens": 11021621.0,
      "step": 5310
    },
    {
      "entropy": 5.729885625839233,
      "epoch": 0.4821298984034833,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004982702973472518,
      "loss": 5.6457,
      "mean_token_accuracy": 0.1491677612066269,
      "num_tokens": 11032763.0,
      "step": 5315
    },
    {
      "entropy": 5.83299446105957,
      "epoch": 0.4825834542815675,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004982662906736783,
      "loss": 5.6633,
      "mean_token_accuracy": 0.16112029701471328,
      "num_tokens": 11042568.0,
      "step": 5320
    },
    {
      "entropy": 5.766739845275879,
      "epoch": 0.48303701015965167,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004982622793829009,
      "loss": 5.6218,
      "mean_token_accuracy": 0.1589931309223175,
      "num_tokens": 11052475.0,
      "step": 5325
    },
    {
      "entropy": 5.688635396957397,
      "epoch": 0.4834905660377358,
      "grad_norm": 2.125,
      "learning_rate": 0.0004982582634750027,
      "loss": 5.6375,
      "mean_token_accuracy": 0.160682412981987,
      "num_tokens": 11062417.0,
      "step": 5330
    },
    {
      "entropy": 5.717385482788086,
      "epoch": 0.48394412191582004,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004982542429500667,
      "loss": 5.6501,
      "mean_token_accuracy": 0.15277392864227296,
      "num_tokens": 11072282.0,
      "step": 5335
    },
    {
      "entropy": 5.879451084136963,
      "epoch": 0.4843976777939042,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004982502178081761,
      "loss": 5.6763,
      "mean_token_accuracy": 0.15068399012088776,
      "num_tokens": 11083333.0,
      "step": 5340
    },
    {
      "entropy": 5.78740553855896,
      "epoch": 0.4848512336719884,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004982461880494141,
      "loss": 5.6372,
      "mean_token_accuracy": 0.1521876037120819,
      "num_tokens": 11094094.0,
      "step": 5345
    },
    {
      "entropy": 5.790024709701538,
      "epoch": 0.48530478955007256,
      "grad_norm": 1.6484375,
      "learning_rate": 0.000498242153673864,
      "loss": 5.7623,
      "mean_token_accuracy": 0.1470420092344284,
      "num_tokens": 11103999.0,
      "step": 5350
    },
    {
      "entropy": 5.805356550216675,
      "epoch": 0.48575834542815677,
      "grad_norm": 1.75,
      "learning_rate": 0.0004982381146816094,
      "loss": 5.6724,
      "mean_token_accuracy": 0.15599775463342666,
      "num_tokens": 11113459.0,
      "step": 5355
    },
    {
      "entropy": 5.768607950210571,
      "epoch": 0.4862119013062409,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004982340710727336,
      "loss": 5.6776,
      "mean_token_accuracy": 0.1512758806347847,
      "num_tokens": 11124743.0,
      "step": 5360
    },
    {
      "entropy": 5.793423318862915,
      "epoch": 0.48666545718432513,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004982300228473203,
      "loss": 5.6624,
      "mean_token_accuracy": 0.15485051274299622,
      "num_tokens": 11134953.0,
      "step": 5365
    },
    {
      "entropy": 5.773527050018311,
      "epoch": 0.4871190130624093,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004982259700054533,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15058848708868028,
      "num_tokens": 11145275.0,
      "step": 5370
    },
    {
      "entropy": 5.723573064804077,
      "epoch": 0.48757256894049344,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004982219125472163,
      "loss": 5.665,
      "mean_token_accuracy": 0.15825372338294982,
      "num_tokens": 11155155.0,
      "step": 5375
    },
    {
      "entropy": 5.732001161575317,
      "epoch": 0.48802612481857766,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004982178504726933,
      "loss": 5.648,
      "mean_token_accuracy": 0.1569226548075676,
      "num_tokens": 11165269.0,
      "step": 5380
    },
    {
      "entropy": 5.879735612869263,
      "epoch": 0.4884796806966618,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004982137837819682,
      "loss": 5.7668,
      "mean_token_accuracy": 0.14773210883140564,
      "num_tokens": 11175499.0,
      "step": 5385
    },
    {
      "entropy": 5.779443979263306,
      "epoch": 0.488933236574746,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004982097124751252,
      "loss": 5.713,
      "mean_token_accuracy": 0.14961470663547516,
      "num_tokens": 11185662.0,
      "step": 5390
    },
    {
      "entropy": 5.7722508907318115,
      "epoch": 0.4893867924528302,
      "grad_norm": 1.625,
      "learning_rate": 0.0004982056365522486,
      "loss": 5.5986,
      "mean_token_accuracy": 0.15671695321798323,
      "num_tokens": 11196070.0,
      "step": 5395
    },
    {
      "entropy": 5.871857929229736,
      "epoch": 0.4898403483309144,
      "grad_norm": 1.625,
      "learning_rate": 0.0004982015560134225,
      "loss": 5.8257,
      "mean_token_accuracy": 0.14437031298875808,
      "num_tokens": 11209101.0,
      "step": 5400
    },
    {
      "entropy": 5.787049436569214,
      "epoch": 0.49029390420899854,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004981974708587312,
      "loss": 5.7053,
      "mean_token_accuracy": 0.1474752552807331,
      "num_tokens": 11219828.0,
      "step": 5405
    },
    {
      "entropy": 5.812083530426025,
      "epoch": 0.49074746008708275,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004981933810882594,
      "loss": 5.6507,
      "mean_token_accuracy": 0.15424271821975707,
      "num_tokens": 11230681.0,
      "step": 5410
    },
    {
      "entropy": 5.779886293411255,
      "epoch": 0.4912010159651669,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004981892867020917,
      "loss": 5.757,
      "mean_token_accuracy": 0.14833626002073289,
      "num_tokens": 11241963.0,
      "step": 5415
    },
    {
      "entropy": 5.785884189605713,
      "epoch": 0.49165457184325106,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004981851877003125,
      "loss": 5.6863,
      "mean_token_accuracy": 0.14686497002840043,
      "num_tokens": 11252950.0,
      "step": 5420
    },
    {
      "entropy": 5.7913336753845215,
      "epoch": 0.4921081277213353,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004981810840830068,
      "loss": 5.6888,
      "mean_token_accuracy": 0.1526445686817169,
      "num_tokens": 11262486.0,
      "step": 5425
    },
    {
      "entropy": 5.761009216308594,
      "epoch": 0.49256168359941943,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004981769758502596,
      "loss": 5.6525,
      "mean_token_accuracy": 0.15411523282527922,
      "num_tokens": 11272950.0,
      "step": 5430
    },
    {
      "entropy": 5.753162050247193,
      "epoch": 0.49301523947750364,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004981728630021555,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1500825673341751,
      "num_tokens": 11284775.0,
      "step": 5435
    },
    {
      "entropy": 5.908806085586548,
      "epoch": 0.4934687953555878,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004981687455387797,
      "loss": 5.6757,
      "mean_token_accuracy": 0.15840519815683365,
      "num_tokens": 11294810.0,
      "step": 5440
    },
    {
      "entropy": 5.816017150878906,
      "epoch": 0.493922351233672,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004981646234602173,
      "loss": 5.6983,
      "mean_token_accuracy": 0.15476290434598922,
      "num_tokens": 11306728.0,
      "step": 5445
    },
    {
      "entropy": 5.753983831405639,
      "epoch": 0.49437590711175616,
      "grad_norm": 1.625,
      "learning_rate": 0.0004981604967665539,
      "loss": 5.6993,
      "mean_token_accuracy": 0.1476845324039459,
      "num_tokens": 11317929.0,
      "step": 5450
    },
    {
      "entropy": 5.791211032867432,
      "epoch": 0.4948294629898404,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004981563654578743,
      "loss": 5.6145,
      "mean_token_accuracy": 0.15718991905450821,
      "num_tokens": 11327864.0,
      "step": 5455
    },
    {
      "entropy": 5.767108821868897,
      "epoch": 0.49528301886792453,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004981522295342643,
      "loss": 5.6644,
      "mean_token_accuracy": 0.15040411800146103,
      "num_tokens": 11337965.0,
      "step": 5460
    },
    {
      "entropy": 5.6784430027008055,
      "epoch": 0.4957365747460087,
      "grad_norm": 1.5,
      "learning_rate": 0.0004981480889958092,
      "loss": 5.6421,
      "mean_token_accuracy": 0.15154165476560594,
      "num_tokens": 11348942.0,
      "step": 5465
    },
    {
      "entropy": 5.779708576202393,
      "epoch": 0.4961901306240929,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004981439438425949,
      "loss": 5.6401,
      "mean_token_accuracy": 0.15476672053337098,
      "num_tokens": 11359702.0,
      "step": 5470
    },
    {
      "entropy": 5.912676811218262,
      "epoch": 0.49664368650217705,
      "grad_norm": 1.421875,
      "learning_rate": 0.000498139794074707,
      "loss": 5.8274,
      "mean_token_accuracy": 0.14696455523371696,
      "num_tokens": 11370238.0,
      "step": 5475
    },
    {
      "entropy": 5.7378565788269045,
      "epoch": 0.49709724238026126,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004981356396922313,
      "loss": 5.6272,
      "mean_token_accuracy": 0.15095601379871368,
      "num_tokens": 11380822.0,
      "step": 5480
    },
    {
      "entropy": 5.7243176937103275,
      "epoch": 0.4975507982583454,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004981314806952535,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1527250073850155,
      "num_tokens": 11391858.0,
      "step": 5485
    },
    {
      "entropy": 5.761184549331665,
      "epoch": 0.49800435413642963,
      "grad_norm": 1.84375,
      "learning_rate": 0.00049812731708386,
      "loss": 5.6119,
      "mean_token_accuracy": 0.16142940223217012,
      "num_tokens": 11402036.0,
      "step": 5490
    },
    {
      "entropy": 5.7568670272827145,
      "epoch": 0.4984579100145138,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004981231488581367,
      "loss": 5.7373,
      "mean_token_accuracy": 0.14541575014591218,
      "num_tokens": 11413242.0,
      "step": 5495
    },
    {
      "entropy": 5.818656730651855,
      "epoch": 0.498911465892598,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004981189760181699,
      "loss": 5.6752,
      "mean_token_accuracy": 0.15886023342609407,
      "num_tokens": 11422927.0,
      "step": 5500
    },
    {
      "entropy": 5.790579175949096,
      "epoch": 0.49936502177068215,
      "grad_norm": 1.625,
      "learning_rate": 0.0004981147985640456,
      "loss": 5.6868,
      "mean_token_accuracy": 0.15862198173999786,
      "num_tokens": 11432669.0,
      "step": 5505
    },
    {
      "entropy": 5.771609878540039,
      "epoch": 0.4998185776487663,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004981106164958506,
      "loss": 5.7047,
      "mean_token_accuracy": 0.152798593044281,
      "num_tokens": 11442823.0,
      "step": 5510
    },
    {
      "entropy": 5.74314432144165,
      "epoch": 0.5002721335268505,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004981064298136712,
      "loss": 5.6744,
      "mean_token_accuracy": 0.15345294028520584,
      "num_tokens": 11453583.0,
      "step": 5515
    },
    {
      "entropy": 5.6962987899780275,
      "epoch": 0.5007256894049347,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004981022385175939,
      "loss": 5.644,
      "mean_token_accuracy": 0.1497638240456581,
      "num_tokens": 11463883.0,
      "step": 5520
    },
    {
      "entropy": 5.7916618347167965,
      "epoch": 0.5011792452830188,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004980980426077054,
      "loss": 5.6823,
      "mean_token_accuracy": 0.14812253117561341,
      "num_tokens": 11473966.0,
      "step": 5525
    },
    {
      "entropy": 5.681837892532348,
      "epoch": 0.5016328011611031,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004980938420840926,
      "loss": 5.5864,
      "mean_token_accuracy": 0.1510275721549988,
      "num_tokens": 11483898.0,
      "step": 5530
    },
    {
      "entropy": 5.684164094924927,
      "epoch": 0.5020863570391872,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004980896369468423,
      "loss": 5.6115,
      "mean_token_accuracy": 0.15556384176015853,
      "num_tokens": 11494190.0,
      "step": 5535
    },
    {
      "entropy": 5.841269779205322,
      "epoch": 0.5025399129172714,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004980854271960415,
      "loss": 5.7433,
      "mean_token_accuracy": 0.1519545204937458,
      "num_tokens": 11505339.0,
      "step": 5540
    },
    {
      "entropy": 5.82388334274292,
      "epoch": 0.5029934687953556,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004980812128317772,
      "loss": 5.7438,
      "mean_token_accuracy": 0.15373214781284333,
      "num_tokens": 11515892.0,
      "step": 5545
    },
    {
      "entropy": 5.760100889205932,
      "epoch": 0.5034470246734397,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004980769938541364,
      "loss": 5.6819,
      "mean_token_accuracy": 0.15700885728001596,
      "num_tokens": 11526540.0,
      "step": 5550
    },
    {
      "entropy": 5.811464118957519,
      "epoch": 0.503900580551524,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004980727702632066,
      "loss": 5.6881,
      "mean_token_accuracy": 0.15401896983385086,
      "num_tokens": 11536604.0,
      "step": 5555
    },
    {
      "entropy": 5.7957395076751705,
      "epoch": 0.5043541364296081,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004980685420590752,
      "loss": 5.7418,
      "mean_token_accuracy": 0.15740367621183396,
      "num_tokens": 11546764.0,
      "step": 5560
    },
    {
      "entropy": 5.724511289596558,
      "epoch": 0.5048076923076923,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004980643092418294,
      "loss": 5.6585,
      "mean_token_accuracy": 0.15534376949071885,
      "num_tokens": 11556984.0,
      "step": 5565
    },
    {
      "entropy": 5.773011779785156,
      "epoch": 0.5052612481857764,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004980600718115568,
      "loss": 5.6122,
      "mean_token_accuracy": 0.16305769681930543,
      "num_tokens": 11567046.0,
      "step": 5570
    },
    {
      "entropy": 5.753465414047241,
      "epoch": 0.5057148040638607,
      "grad_norm": 2.0,
      "learning_rate": 0.0004980558297683452,
      "loss": 5.6727,
      "mean_token_accuracy": 0.14825989305973053,
      "num_tokens": 11577575.0,
      "step": 5575
    },
    {
      "entropy": 5.703629446029663,
      "epoch": 0.5061683599419449,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004980515831122822,
      "loss": 5.5632,
      "mean_token_accuracy": 0.15776239931583405,
      "num_tokens": 11587322.0,
      "step": 5580
    },
    {
      "entropy": 5.697418451309204,
      "epoch": 0.506621915820029,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004980473318434555,
      "loss": 5.5973,
      "mean_token_accuracy": 0.16032545566558837,
      "num_tokens": 11597544.0,
      "step": 5585
    },
    {
      "entropy": 5.7857341289520265,
      "epoch": 0.5070754716981132,
      "grad_norm": 1.75,
      "learning_rate": 0.0004980430759619532,
      "loss": 5.6118,
      "mean_token_accuracy": 0.15705069452524184,
      "num_tokens": 11607538.0,
      "step": 5590
    },
    {
      "entropy": 5.769230747222901,
      "epoch": 0.5075290275761973,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004980388154678633,
      "loss": 5.6813,
      "mean_token_accuracy": 0.15760140269994735,
      "num_tokens": 11617082.0,
      "step": 5595
    },
    {
      "entropy": 5.685442399978638,
      "epoch": 0.5079825834542816,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004980345503612738,
      "loss": 5.5784,
      "mean_token_accuracy": 0.1552373379468918,
      "num_tokens": 11627233.0,
      "step": 5600
    },
    {
      "entropy": 5.776123332977295,
      "epoch": 0.5084361393323658,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004980302806422729,
      "loss": 5.637,
      "mean_token_accuracy": 0.1540546879172325,
      "num_tokens": 11637520.0,
      "step": 5605
    },
    {
      "entropy": 5.741995143890381,
      "epoch": 0.5088896952104499,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004980260063109491,
      "loss": 5.6899,
      "mean_token_accuracy": 0.1525841623544693,
      "num_tokens": 11647525.0,
      "step": 5610
    },
    {
      "entropy": 5.774639463424682,
      "epoch": 0.5093432510885341,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004980217273673906,
      "loss": 5.6436,
      "mean_token_accuracy": 0.15127140432596206,
      "num_tokens": 11657996.0,
      "step": 5615
    },
    {
      "entropy": 5.770473194122315,
      "epoch": 0.5097968069666183,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004980174438116859,
      "loss": 5.5951,
      "mean_token_accuracy": 0.15702126920223236,
      "num_tokens": 11668752.0,
      "step": 5620
    },
    {
      "entropy": 5.724298429489136,
      "epoch": 0.5102503628447025,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004980131556439236,
      "loss": 5.7007,
      "mean_token_accuracy": 0.1514161080121994,
      "num_tokens": 11678989.0,
      "step": 5625
    },
    {
      "entropy": 5.748479270935059,
      "epoch": 0.5107039187227866,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004980088628641925,
      "loss": 5.6303,
      "mean_token_accuracy": 0.1639922246336937,
      "num_tokens": 11687907.0,
      "step": 5630
    },
    {
      "entropy": 5.693242740631104,
      "epoch": 0.5111574746008708,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004980045654725812,
      "loss": 5.6601,
      "mean_token_accuracy": 0.1565706640481949,
      "num_tokens": 11698293.0,
      "step": 5635
    },
    {
      "entropy": 5.8007673740386965,
      "epoch": 0.511611030478955,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004980002634691787,
      "loss": 5.8063,
      "mean_token_accuracy": 0.14153425469994546,
      "num_tokens": 11708142.0,
      "step": 5640
    },
    {
      "entropy": 5.827743053436279,
      "epoch": 0.5120645863570392,
      "grad_norm": 1.859375,
      "learning_rate": 0.000497995956854074,
      "loss": 5.8451,
      "mean_token_accuracy": 0.14463640600442887,
      "num_tokens": 11719472.0,
      "step": 5645
    },
    {
      "entropy": 5.9134735584259035,
      "epoch": 0.5125181422351234,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004979916456273561,
      "loss": 5.7033,
      "mean_token_accuracy": 0.153484570980072,
      "num_tokens": 11729689.0,
      "step": 5650
    },
    {
      "entropy": 5.771580600738526,
      "epoch": 0.5129716981132075,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004979873297891141,
      "loss": 5.5789,
      "mean_token_accuracy": 0.15919430255889894,
      "num_tokens": 11739222.0,
      "step": 5655
    },
    {
      "entropy": 5.761745595932007,
      "epoch": 0.5134252539912917,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004979830093394372,
      "loss": 5.6735,
      "mean_token_accuracy": 0.15483780205249786,
      "num_tokens": 11749043.0,
      "step": 5660
    },
    {
      "entropy": 5.756465101242066,
      "epoch": 0.513878809869376,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004979786842784149,
      "loss": 5.716,
      "mean_token_accuracy": 0.15349172800779343,
      "num_tokens": 11759231.0,
      "step": 5665
    },
    {
      "entropy": 5.753206396102906,
      "epoch": 0.5143323657474601,
      "grad_norm": 2.125,
      "learning_rate": 0.0004979743546061366,
      "loss": 5.5745,
      "mean_token_accuracy": 0.16228333860635757,
      "num_tokens": 11768380.0,
      "step": 5670
    },
    {
      "entropy": 5.758825826644897,
      "epoch": 0.5147859216255443,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004979700203226918,
      "loss": 5.7021,
      "mean_token_accuracy": 0.14961437582969667,
      "num_tokens": 11780101.0,
      "step": 5675
    },
    {
      "entropy": 5.7383798122406,
      "epoch": 0.5152394775036284,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004979656814281702,
      "loss": 5.6494,
      "mean_token_accuracy": 0.1545555830001831,
      "num_tokens": 11790524.0,
      "step": 5680
    },
    {
      "entropy": 5.727444219589233,
      "epoch": 0.5156930333817126,
      "grad_norm": 2.625,
      "learning_rate": 0.0004979613379226615,
      "loss": 5.6245,
      "mean_token_accuracy": 0.15359244793653487,
      "num_tokens": 11801393.0,
      "step": 5685
    },
    {
      "entropy": 5.752847623825073,
      "epoch": 0.5161465892597968,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004979569898062554,
      "loss": 5.5693,
      "mean_token_accuracy": 0.16035476624965667,
      "num_tokens": 11811117.0,
      "step": 5690
    },
    {
      "entropy": 5.701114749908447,
      "epoch": 0.516600145137881,
      "grad_norm": 1.6875,
      "learning_rate": 0.000497952637079042,
      "loss": 5.5533,
      "mean_token_accuracy": 0.160872420668602,
      "num_tokens": 11821389.0,
      "step": 5695
    },
    {
      "entropy": 5.75704665184021,
      "epoch": 0.5170537010159652,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004979482797411114,
      "loss": 5.6375,
      "mean_token_accuracy": 0.16205394864082337,
      "num_tokens": 11831082.0,
      "step": 5700
    },
    {
      "entropy": 5.7467203617095945,
      "epoch": 0.5175072568940493,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004979439177925534,
      "loss": 5.6264,
      "mean_token_accuracy": 0.15308979004621506,
      "num_tokens": 11841617.0,
      "step": 5705
    },
    {
      "entropy": 5.701697063446045,
      "epoch": 0.5179608127721336,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004979395512334584,
      "loss": 5.5911,
      "mean_token_accuracy": 0.1621819481253624,
      "num_tokens": 11851643.0,
      "step": 5710
    },
    {
      "entropy": 5.769807004928589,
      "epoch": 0.5184143686502177,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004979351800639166,
      "loss": 5.7193,
      "mean_token_accuracy": 0.144927167147398,
      "num_tokens": 11862619.0,
      "step": 5715
    },
    {
      "entropy": 5.7512876987457275,
      "epoch": 0.5188679245283019,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004979308042840186,
      "loss": 5.646,
      "mean_token_accuracy": 0.15768488347530366,
      "num_tokens": 11873574.0,
      "step": 5720
    },
    {
      "entropy": 5.751497602462768,
      "epoch": 0.519321480406386,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004979264238938547,
      "loss": 5.5771,
      "mean_token_accuracy": 0.15962153673171997,
      "num_tokens": 11883472.0,
      "step": 5725
    },
    {
      "entropy": 5.769440221786499,
      "epoch": 0.5197750362844702,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004979220388935155,
      "loss": 5.6521,
      "mean_token_accuracy": 0.15895968824625015,
      "num_tokens": 11894531.0,
      "step": 5730
    },
    {
      "entropy": 5.696563959121704,
      "epoch": 0.5202285921625545,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004979176492830918,
      "loss": 5.6944,
      "mean_token_accuracy": 0.14857350289821625,
      "num_tokens": 11905141.0,
      "step": 5735
    },
    {
      "entropy": 5.85057373046875,
      "epoch": 0.5206821480406386,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004979132550626742,
      "loss": 5.68,
      "mean_token_accuracy": 0.15559621453285216,
      "num_tokens": 11916687.0,
      "step": 5740
    },
    {
      "entropy": 5.7491716861724855,
      "epoch": 0.5211357039187228,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004979088562323538,
      "loss": 5.6465,
      "mean_token_accuracy": 0.1574707433581352,
      "num_tokens": 11926722.0,
      "step": 5745
    },
    {
      "entropy": 5.713038301467895,
      "epoch": 0.5215892597968069,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004979044527922214,
      "loss": 5.6995,
      "mean_token_accuracy": 0.1492026314139366,
      "num_tokens": 11937268.0,
      "step": 5750
    },
    {
      "entropy": 5.701345777511596,
      "epoch": 0.5220428156748912,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004979000447423681,
      "loss": 5.5537,
      "mean_token_accuracy": 0.1611389324069023,
      "num_tokens": 11947018.0,
      "step": 5755
    },
    {
      "entropy": 5.72505750656128,
      "epoch": 0.5224963715529753,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004978956320828851,
      "loss": 5.5965,
      "mean_token_accuracy": 0.1561354696750641,
      "num_tokens": 11956468.0,
      "step": 5760
    },
    {
      "entropy": 5.6504326343536375,
      "epoch": 0.5229499274310595,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004978912148138637,
      "loss": 5.5881,
      "mean_token_accuracy": 0.1537265494465828,
      "num_tokens": 11968206.0,
      "step": 5765
    },
    {
      "entropy": 5.73058066368103,
      "epoch": 0.5234034833091437,
      "grad_norm": 1.7421875,
      "learning_rate": 0.000497886792935395,
      "loss": 5.6603,
      "mean_token_accuracy": 0.15298595577478408,
      "num_tokens": 11978513.0,
      "step": 5770
    },
    {
      "entropy": 5.773151540756226,
      "epoch": 0.5238570391872278,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004978823664475707,
      "loss": 5.5659,
      "mean_token_accuracy": 0.15904878526926042,
      "num_tokens": 11987995.0,
      "step": 5775
    },
    {
      "entropy": 5.708344984054565,
      "epoch": 0.5243105950653121,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004978779353504824,
      "loss": 5.6095,
      "mean_token_accuracy": 0.1586873337626457,
      "num_tokens": 11998142.0,
      "step": 5780
    },
    {
      "entropy": 5.680955839157105,
      "epoch": 0.5247641509433962,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004978734996442214,
      "loss": 5.5659,
      "mean_token_accuracy": 0.1532874196767807,
      "num_tokens": 12008640.0,
      "step": 5785
    },
    {
      "entropy": 5.74573163986206,
      "epoch": 0.5252177068214804,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004978690593288798,
      "loss": 5.6922,
      "mean_token_accuracy": 0.15569562315940857,
      "num_tokens": 12018481.0,
      "step": 5790
    },
    {
      "entropy": 5.796635341644287,
      "epoch": 0.5256712626995645,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004978646144045491,
      "loss": 5.7112,
      "mean_token_accuracy": 0.14771561920642853,
      "num_tokens": 12029230.0,
      "step": 5795
    },
    {
      "entropy": 5.70799150466919,
      "epoch": 0.5261248185776488,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004978601648713215,
      "loss": 5.5875,
      "mean_token_accuracy": 0.15939762741327285,
      "num_tokens": 12039194.0,
      "step": 5800
    },
    {
      "entropy": 5.741345453262329,
      "epoch": 0.526578374455733,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004978557107292889,
      "loss": 5.7766,
      "mean_token_accuracy": 0.1500698059797287,
      "num_tokens": 12049946.0,
      "step": 5805
    },
    {
      "entropy": 5.745180892944336,
      "epoch": 0.5270319303338171,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004978512519785433,
      "loss": 5.6339,
      "mean_token_accuracy": 0.158860544860363,
      "num_tokens": 12059961.0,
      "step": 5810
    },
    {
      "entropy": 5.831522703170776,
      "epoch": 0.5274854862119013,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000497846788619177,
      "loss": 5.7097,
      "mean_token_accuracy": 0.14821714609861375,
      "num_tokens": 12072787.0,
      "step": 5815
    },
    {
      "entropy": 5.755320024490357,
      "epoch": 0.5279390420899854,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004978423206512825,
      "loss": 5.6483,
      "mean_token_accuracy": 0.15618030428886415,
      "num_tokens": 12082658.0,
      "step": 5820
    },
    {
      "entropy": 5.598514795303345,
      "epoch": 0.5283925979680697,
      "grad_norm": 1.7265625,
      "learning_rate": 0.000497837848074952,
      "loss": 5.4811,
      "mean_token_accuracy": 0.16532057225704194,
      "num_tokens": 12092906.0,
      "step": 5825
    },
    {
      "entropy": 5.697569561004639,
      "epoch": 0.5288461538461539,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000497833370890278,
      "loss": 5.7018,
      "mean_token_accuracy": 0.15202219635248185,
      "num_tokens": 12103659.0,
      "step": 5830
    },
    {
      "entropy": 5.728899145126343,
      "epoch": 0.529299709724238,
      "grad_norm": 1.75,
      "learning_rate": 0.0004978288890973529,
      "loss": 5.543,
      "mean_token_accuracy": 0.1601557657122612,
      "num_tokens": 12113202.0,
      "step": 5835
    },
    {
      "entropy": 5.752268695831299,
      "epoch": 0.5297532656023222,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004978244026962698,
      "loss": 5.6816,
      "mean_token_accuracy": 0.1585050493478775,
      "num_tokens": 12123066.0,
      "step": 5840
    },
    {
      "entropy": 5.8051200866699215,
      "epoch": 0.5302068214804064,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004978199116871212,
      "loss": 5.719,
      "mean_token_accuracy": 0.14660870134830475,
      "num_tokens": 12134288.0,
      "step": 5845
    },
    {
      "entropy": 5.806707048416138,
      "epoch": 0.5306603773584906,
      "grad_norm": 1.578125,
      "learning_rate": 0.00049781541607,
      "loss": 5.6691,
      "mean_token_accuracy": 0.1543032556772232,
      "num_tokens": 12146080.0,
      "step": 5850
    },
    {
      "entropy": 5.827581071853638,
      "epoch": 0.5311139332365747,
      "grad_norm": 1.75,
      "learning_rate": 0.0004978109158449993,
      "loss": 5.6961,
      "mean_token_accuracy": 0.1560826063156128,
      "num_tokens": 12156670.0,
      "step": 5855
    },
    {
      "entropy": 5.715680170059204,
      "epoch": 0.5315674891146589,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004978064110122121,
      "loss": 5.6449,
      "mean_token_accuracy": 0.14724010676145555,
      "num_tokens": 12168004.0,
      "step": 5860
    },
    {
      "entropy": 5.7901379585266115,
      "epoch": 0.532021044992743,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004978019015717314,
      "loss": 5.5872,
      "mean_token_accuracy": 0.15376669615507127,
      "num_tokens": 12178833.0,
      "step": 5865
    },
    {
      "entropy": 5.7304534912109375,
      "epoch": 0.5324746008708273,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004977973875236508,
      "loss": 5.6556,
      "mean_token_accuracy": 0.1504141353070736,
      "num_tokens": 12189357.0,
      "step": 5870
    },
    {
      "entropy": 5.748328971862793,
      "epoch": 0.5329281567489115,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004977928688680634,
      "loss": 5.66,
      "mean_token_accuracy": 0.1567172586917877,
      "num_tokens": 12199023.0,
      "step": 5875
    },
    {
      "entropy": 5.787635803222656,
      "epoch": 0.5333817126269956,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004977883456050626,
      "loss": 5.6975,
      "mean_token_accuracy": 0.14763975739479065,
      "num_tokens": 12208789.0,
      "step": 5880
    },
    {
      "entropy": 5.852936601638794,
      "epoch": 0.5338352685050798,
      "grad_norm": 2.484375,
      "learning_rate": 0.000497783817734742,
      "loss": 5.663,
      "mean_token_accuracy": 0.15545896589756011,
      "num_tokens": 12219483.0,
      "step": 5885
    },
    {
      "entropy": 5.809611368179321,
      "epoch": 0.534288824383164,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004977792852571954,
      "loss": 5.7254,
      "mean_token_accuracy": 0.1538556322455406,
      "num_tokens": 12229552.0,
      "step": 5890
    },
    {
      "entropy": 5.725105571746826,
      "epoch": 0.5347423802612482,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004977747481725164,
      "loss": 5.623,
      "mean_token_accuracy": 0.16032094955444337,
      "num_tokens": 12239378.0,
      "step": 5895
    },
    {
      "entropy": 5.750143003463745,
      "epoch": 0.5351959361393324,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004977702064807987,
      "loss": 5.6852,
      "mean_token_accuracy": 0.15214105546474457,
      "num_tokens": 12249496.0,
      "step": 5900
    },
    {
      "entropy": 5.839286422729492,
      "epoch": 0.5356494920174165,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004977656601821366,
      "loss": 5.7695,
      "mean_token_accuracy": 0.14929212182760238,
      "num_tokens": 12259099.0,
      "step": 5905
    },
    {
      "entropy": 5.7886841773986815,
      "epoch": 0.5361030478955007,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004977611092766238,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1530821517109871,
      "num_tokens": 12269578.0,
      "step": 5910
    },
    {
      "entropy": 5.820439577102661,
      "epoch": 0.5365566037735849,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004977565537643544,
      "loss": 5.6987,
      "mean_token_accuracy": 0.14760151132941246,
      "num_tokens": 12280210.0,
      "step": 5915
    },
    {
      "entropy": 5.659420204162598,
      "epoch": 0.5370101596516691,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004977519936454227,
      "loss": 5.5469,
      "mean_token_accuracy": 0.16010452359914779,
      "num_tokens": 12290830.0,
      "step": 5920
    },
    {
      "entropy": 5.735601472854614,
      "epoch": 0.5374637155297532,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004977474289199231,
      "loss": 5.6103,
      "mean_token_accuracy": 0.1563941493630409,
      "num_tokens": 12300233.0,
      "step": 5925
    },
    {
      "entropy": 5.7761067867279055,
      "epoch": 0.5379172714078374,
      "grad_norm": 2.75,
      "learning_rate": 0.0004977428595879499,
      "loss": 5.6908,
      "mean_token_accuracy": 0.15967603921890258,
      "num_tokens": 12310578.0,
      "step": 5930
    },
    {
      "entropy": 5.757687091827393,
      "epoch": 0.5383708272859217,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004977382856495975,
      "loss": 5.6082,
      "mean_token_accuracy": 0.15673843175172805,
      "num_tokens": 12320484.0,
      "step": 5935
    },
    {
      "entropy": 5.719855117797851,
      "epoch": 0.5388243831640058,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004977337071049606,
      "loss": 5.7179,
      "mean_token_accuracy": 0.14939573258161545,
      "num_tokens": 12331518.0,
      "step": 5940
    },
    {
      "entropy": 5.722897338867187,
      "epoch": 0.53927793904209,
      "grad_norm": 2.859375,
      "learning_rate": 0.000497729123954134,
      "loss": 5.6223,
      "mean_token_accuracy": 0.1570252627134323,
      "num_tokens": 12342478.0,
      "step": 5945
    },
    {
      "entropy": 5.740108633041382,
      "epoch": 0.5397314949201741,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004977245361972123,
      "loss": 5.6553,
      "mean_token_accuracy": 0.1535571664571762,
      "num_tokens": 12352811.0,
      "step": 5950
    },
    {
      "entropy": 5.745780086517334,
      "epoch": 0.5401850507982583,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004977199438342904,
      "loss": 5.6519,
      "mean_token_accuracy": 0.15965785086154938,
      "num_tokens": 12363948.0,
      "step": 5955
    },
    {
      "entropy": 5.808036136627197,
      "epoch": 0.5406386066763426,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004977153468654634,
      "loss": 5.6653,
      "mean_token_accuracy": 0.15435378551483153,
      "num_tokens": 12373910.0,
      "step": 5960
    },
    {
      "entropy": 5.761662864685059,
      "epoch": 0.5410921625544267,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004977107452908262,
      "loss": 5.6322,
      "mean_token_accuracy": 0.15638811737298966,
      "num_tokens": 12385038.0,
      "step": 5965
    },
    {
      "entropy": 5.729135322570801,
      "epoch": 0.5415457184325109,
      "grad_norm": 1.8125,
      "learning_rate": 0.000497706139110474,
      "loss": 5.6166,
      "mean_token_accuracy": 0.15953594893217088,
      "num_tokens": 12395500.0,
      "step": 5970
    },
    {
      "entropy": 5.771009159088135,
      "epoch": 0.541999274310595,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004977015283245021,
      "loss": 5.7092,
      "mean_token_accuracy": 0.1496681362390518,
      "num_tokens": 12406249.0,
      "step": 5975
    },
    {
      "entropy": 5.748795080184936,
      "epoch": 0.5424528301886793,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004976969129330059,
      "loss": 5.6159,
      "mean_token_accuracy": 0.15583853870630265,
      "num_tokens": 12416685.0,
      "step": 5980
    },
    {
      "entropy": 5.776485395431519,
      "epoch": 0.5429063860667634,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004976922929360808,
      "loss": 5.6401,
      "mean_token_accuracy": 0.15829548090696335,
      "num_tokens": 12426794.0,
      "step": 5985
    },
    {
      "entropy": 5.757869243621826,
      "epoch": 0.5433599419448476,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004976876683338222,
      "loss": 5.6968,
      "mean_token_accuracy": 0.15154862105846406,
      "num_tokens": 12437115.0,
      "step": 5990
    },
    {
      "entropy": 5.823219347000122,
      "epoch": 0.5438134978229318,
      "grad_norm": 1.921875,
      "learning_rate": 0.000497683039126326,
      "loss": 5.6525,
      "mean_token_accuracy": 0.15575422197580338,
      "num_tokens": 12446955.0,
      "step": 5995
    },
    {
      "entropy": 5.754256916046143,
      "epoch": 0.5442670537010159,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004976784053136878,
      "loss": 5.6072,
      "mean_token_accuracy": 0.16049512177705766,
      "num_tokens": 12456924.0,
      "step": 6000
    },
    {
      "epoch": 0.5442670537010159,
      "eval_entropy": 5.567713360225453,
      "eval_loss": 5.68149471282959,
      "eval_mean_token_accuracy": 0.16022676651142706,
      "eval_num_tokens": 12456924.0,
      "eval_runtime": 26.6019,
      "eval_samples_per_second": 1329.227,
      "eval_steps_per_second": 166.153,
      "step": 6000
    },
    {
      "entropy": 5.656931829452515,
      "epoch": 0.5447206095791002,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004976737668960034,
      "loss": 5.5292,
      "mean_token_accuracy": 0.16199954599142075,
      "num_tokens": 12467712.0,
      "step": 6005
    },
    {
      "entropy": 5.799897336959839,
      "epoch": 0.5451741654571843,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004976691238733688,
      "loss": 5.6414,
      "mean_token_accuracy": 0.15725044012069703,
      "num_tokens": 12477374.0,
      "step": 6010
    },
    {
      "entropy": 5.818873882293701,
      "epoch": 0.5456277213352685,
      "grad_norm": 2.6875,
      "learning_rate": 0.00049766447624588,
      "loss": 5.7068,
      "mean_token_accuracy": 0.15485058575868607,
      "num_tokens": 12488038.0,
      "step": 6015
    },
    {
      "entropy": 5.812593078613281,
      "epoch": 0.5460812772133526,
      "grad_norm": 1.9375,
      "learning_rate": 0.000497659824013633,
      "loss": 5.6108,
      "mean_token_accuracy": 0.16121876537799834,
      "num_tokens": 12498393.0,
      "step": 6020
    },
    {
      "entropy": 5.676734733581543,
      "epoch": 0.5465348330914369,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004976551671767241,
      "loss": 5.5836,
      "mean_token_accuracy": 0.15978361070156097,
      "num_tokens": 12507926.0,
      "step": 6025
    },
    {
      "entropy": 5.705279302597046,
      "epoch": 0.5469883889695211,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004976505057352496,
      "loss": 5.6098,
      "mean_token_accuracy": 0.1612431913614273,
      "num_tokens": 12517835.0,
      "step": 6030
    },
    {
      "entropy": 5.756604337692261,
      "epoch": 0.5474419448476052,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004976458396893059,
      "loss": 5.6734,
      "mean_token_accuracy": 0.14765057265758513,
      "num_tokens": 12528067.0,
      "step": 6035
    },
    {
      "entropy": 5.684308671951294,
      "epoch": 0.5478955007256894,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004976411690389896,
      "loss": 5.6005,
      "mean_token_accuracy": 0.15649594962596894,
      "num_tokens": 12537583.0,
      "step": 6040
    },
    {
      "entropy": 5.780689907073975,
      "epoch": 0.5483490566037735,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000497636493784397,
      "loss": 5.6417,
      "mean_token_accuracy": 0.1508394107222557,
      "num_tokens": 12548750.0,
      "step": 6045
    },
    {
      "entropy": 5.810310411453247,
      "epoch": 0.5488026124818578,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004976318139256251,
      "loss": 5.715,
      "mean_token_accuracy": 0.1568623587489128,
      "num_tokens": 12560197.0,
      "step": 6050
    },
    {
      "entropy": 5.8327185153961185,
      "epoch": 0.549256168359942,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004976271294627706,
      "loss": 5.7512,
      "mean_token_accuracy": 0.15411142259836197,
      "num_tokens": 12570958.0,
      "step": 6055
    },
    {
      "entropy": 5.764525985717773,
      "epoch": 0.5497097242380261,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004976224403959303,
      "loss": 5.6113,
      "mean_token_accuracy": 0.14998921155929565,
      "num_tokens": 12581525.0,
      "step": 6060
    },
    {
      "entropy": 5.765351819992065,
      "epoch": 0.5501632801161103,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004976177467252012,
      "loss": 5.5793,
      "mean_token_accuracy": 0.15715673416852952,
      "num_tokens": 12591765.0,
      "step": 6065
    },
    {
      "entropy": 5.695242023468017,
      "epoch": 0.5506168359941945,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004976130484506804,
      "loss": 5.692,
      "mean_token_accuracy": 0.16310595273971557,
      "num_tokens": 12602471.0,
      "step": 6070
    },
    {
      "entropy": 5.756214284896851,
      "epoch": 0.5510703918722787,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004976083455724649,
      "loss": 5.651,
      "mean_token_accuracy": 0.15674155652523042,
      "num_tokens": 12613656.0,
      "step": 6075
    },
    {
      "entropy": 5.754758644104004,
      "epoch": 0.5515239477503628,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004976036380906523,
      "loss": 5.5769,
      "mean_token_accuracy": 0.153932623565197,
      "num_tokens": 12623922.0,
      "step": 6080
    },
    {
      "entropy": 5.739032125473022,
      "epoch": 0.551977503628447,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004975989260053395,
      "loss": 5.6532,
      "mean_token_accuracy": 0.14958422929048537,
      "num_tokens": 12634213.0,
      "step": 6085
    },
    {
      "entropy": 5.745124387741089,
      "epoch": 0.5524310595065312,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004975942093166242,
      "loss": 5.6214,
      "mean_token_accuracy": 0.15570274442434312,
      "num_tokens": 12643154.0,
      "step": 6090
    },
    {
      "entropy": 5.684883737564087,
      "epoch": 0.5528846153846154,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004975894880246041,
      "loss": 5.6252,
      "mean_token_accuracy": 0.15252370536327362,
      "num_tokens": 12654533.0,
      "step": 6095
    },
    {
      "entropy": 5.738387107849121,
      "epoch": 0.5533381712626996,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004975847621293766,
      "loss": 5.6101,
      "mean_token_accuracy": 0.15831225514411926,
      "num_tokens": 12663417.0,
      "step": 6100
    },
    {
      "entropy": 5.799705648422242,
      "epoch": 0.5537917271407837,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004975800316310394,
      "loss": 5.6172,
      "mean_token_accuracy": 0.16197383552789688,
      "num_tokens": 12673301.0,
      "step": 6105
    },
    {
      "entropy": 5.688973093032837,
      "epoch": 0.5542452830188679,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004975752965296906,
      "loss": 5.6366,
      "mean_token_accuracy": 0.1532282665371895,
      "num_tokens": 12684343.0,
      "step": 6110
    },
    {
      "entropy": 5.712090110778808,
      "epoch": 0.5546988388969522,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004975705568254278,
      "loss": 5.556,
      "mean_token_accuracy": 0.15773797631263733,
      "num_tokens": 12694513.0,
      "step": 6115
    },
    {
      "entropy": 5.777080440521241,
      "epoch": 0.5551523947750363,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004975658125183493,
      "loss": 5.6484,
      "mean_token_accuracy": 0.1541421577334404,
      "num_tokens": 12704778.0,
      "step": 6120
    },
    {
      "entropy": 5.689303159713745,
      "epoch": 0.5556059506531205,
      "grad_norm": 1.453125,
      "learning_rate": 0.000497561063608553,
      "loss": 5.6179,
      "mean_token_accuracy": 0.16233552545309066,
      "num_tokens": 12714171.0,
      "step": 6125
    },
    {
      "entropy": 5.755212259292603,
      "epoch": 0.5560595065312046,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004975563100961372,
      "loss": 5.6747,
      "mean_token_accuracy": 0.1541602686047554,
      "num_tokens": 12724792.0,
      "step": 6130
    },
    {
      "entropy": 5.836889219284058,
      "epoch": 0.5565130624092888,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004975515519812002,
      "loss": 5.6838,
      "mean_token_accuracy": 0.15421807914972305,
      "num_tokens": 12735737.0,
      "step": 6135
    },
    {
      "entropy": 5.7207098484039305,
      "epoch": 0.556966618287373,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004975467892638402,
      "loss": 5.567,
      "mean_token_accuracy": 0.15962569713592528,
      "num_tokens": 12745640.0,
      "step": 6140
    },
    {
      "entropy": 5.719398260116577,
      "epoch": 0.5574201741654572,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004975420219441561,
      "loss": 5.6856,
      "mean_token_accuracy": 0.15408140271902085,
      "num_tokens": 12756374.0,
      "step": 6145
    },
    {
      "entropy": 5.6853748798370365,
      "epoch": 0.5578737300435413,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004975372500222462,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1519828200340271,
      "num_tokens": 12767399.0,
      "step": 6150
    },
    {
      "entropy": 5.724702787399292,
      "epoch": 0.5583272859216255,
      "grad_norm": 1.625,
      "learning_rate": 0.0004975324734982094,
      "loss": 5.5738,
      "mean_token_accuracy": 0.15775190591812133,
      "num_tokens": 12777890.0,
      "step": 6155
    },
    {
      "entropy": 5.760837507247925,
      "epoch": 0.5587808417997098,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004975276923721442,
      "loss": 5.6121,
      "mean_token_accuracy": 0.1617153435945511,
      "num_tokens": 12788094.0,
      "step": 6160
    },
    {
      "entropy": 5.804059791564941,
      "epoch": 0.5592343976777939,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004975229066441496,
      "loss": 5.6613,
      "mean_token_accuracy": 0.1464136704802513,
      "num_tokens": 12799365.0,
      "step": 6165
    },
    {
      "entropy": 5.7211967468261715,
      "epoch": 0.5596879535558781,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004975181163143247,
      "loss": 5.5372,
      "mean_token_accuracy": 0.16188412457704543,
      "num_tokens": 12808856.0,
      "step": 6170
    },
    {
      "entropy": 5.753667974472046,
      "epoch": 0.5601415094339622,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004975133213827684,
      "loss": 5.6342,
      "mean_token_accuracy": 0.1628005862236023,
      "num_tokens": 12818354.0,
      "step": 6175
    },
    {
      "entropy": 5.742711210250855,
      "epoch": 0.5605950653120464,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004975085218495798,
      "loss": 5.6906,
      "mean_token_accuracy": 0.15329492092132568,
      "num_tokens": 12829263.0,
      "step": 6180
    },
    {
      "entropy": 5.8331944942474365,
      "epoch": 0.5610486211901307,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004975037177148585,
      "loss": 5.7152,
      "mean_token_accuracy": 0.15159093886613845,
      "num_tokens": 12840000.0,
      "step": 6185
    },
    {
      "entropy": 5.762791109085083,
      "epoch": 0.5615021770682148,
      "grad_norm": 1.5,
      "learning_rate": 0.0004974989089787035,
      "loss": 5.6953,
      "mean_token_accuracy": 0.14673184007406234,
      "num_tokens": 12850632.0,
      "step": 6190
    },
    {
      "entropy": 5.7715616703033445,
      "epoch": 0.561955732946299,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004974940956412143,
      "loss": 5.6124,
      "mean_token_accuracy": 0.15445255488157272,
      "num_tokens": 12860834.0,
      "step": 6195
    },
    {
      "entropy": 5.722901725769043,
      "epoch": 0.5624092888243831,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004974892777024905,
      "loss": 5.5645,
      "mean_token_accuracy": 0.16279343515634537,
      "num_tokens": 12871267.0,
      "step": 6200
    },
    {
      "entropy": 5.717096328735352,
      "epoch": 0.5628628447024674,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004974844551626318,
      "loss": 5.6041,
      "mean_token_accuracy": 0.15960216373205185,
      "num_tokens": 12881601.0,
      "step": 6205
    },
    {
      "entropy": 5.685572910308838,
      "epoch": 0.5633164005805515,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004974796280217379,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1577148512005806,
      "num_tokens": 12891872.0,
      "step": 6210
    },
    {
      "entropy": 5.670578336715698,
      "epoch": 0.5637699564586357,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004974747962799086,
      "loss": 5.5702,
      "mean_token_accuracy": 0.15749485790729523,
      "num_tokens": 12901408.0,
      "step": 6215
    },
    {
      "entropy": 5.73646183013916,
      "epoch": 0.5642235123367199,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004974699599372437,
      "loss": 5.6834,
      "mean_token_accuracy": 0.14751680940389633,
      "num_tokens": 12911830.0,
      "step": 6220
    },
    {
      "entropy": 5.7733705043792725,
      "epoch": 0.564677068214804,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004974651189938434,
      "loss": 5.5948,
      "mean_token_accuracy": 0.16175662279129027,
      "num_tokens": 12921974.0,
      "step": 6225
    },
    {
      "entropy": 5.6916648864746096,
      "epoch": 0.5651306240928883,
      "grad_norm": 2.75,
      "learning_rate": 0.0004974602734498079,
      "loss": 5.6233,
      "mean_token_accuracy": 0.1593949794769287,
      "num_tokens": 12932504.0,
      "step": 6230
    },
    {
      "entropy": 5.777950477600098,
      "epoch": 0.5655841799709724,
      "grad_norm": 1.90625,
      "learning_rate": 0.000497455423305237,
      "loss": 5.7615,
      "mean_token_accuracy": 0.15056953430175782,
      "num_tokens": 12942800.0,
      "step": 6235
    },
    {
      "entropy": 5.796639394760132,
      "epoch": 0.5660377358490566,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004974505685602314,
      "loss": 5.6271,
      "mean_token_accuracy": 0.15575093477964402,
      "num_tokens": 12952718.0,
      "step": 6240
    },
    {
      "entropy": 5.704430627822876,
      "epoch": 0.5664912917271407,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004974457092148912,
      "loss": 5.4995,
      "mean_token_accuracy": 0.15887019783258438,
      "num_tokens": 12962233.0,
      "step": 6245
    },
    {
      "entropy": 5.628510761260986,
      "epoch": 0.566944847605225,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004974408452693172,
      "loss": 5.5752,
      "mean_token_accuracy": 0.1609184190630913,
      "num_tokens": 12972177.0,
      "step": 6250
    },
    {
      "entropy": 5.69896502494812,
      "epoch": 0.5673984034833092,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004974359767236099,
      "loss": 5.6516,
      "mean_token_accuracy": 0.15473490655422212,
      "num_tokens": 12984282.0,
      "step": 6255
    },
    {
      "entropy": 5.79969048500061,
      "epoch": 0.5678519593613933,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004974311035778698,
      "loss": 5.6191,
      "mean_token_accuracy": 0.1635651931166649,
      "num_tokens": 12994541.0,
      "step": 6260
    },
    {
      "entropy": 5.725707530975342,
      "epoch": 0.5683055152394775,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004974262258321979,
      "loss": 5.703,
      "mean_token_accuracy": 0.15191208347678184,
      "num_tokens": 13005417.0,
      "step": 6265
    },
    {
      "entropy": 5.8123057842254635,
      "epoch": 0.5687590711175616,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004974213434866949,
      "loss": 5.8124,
      "mean_token_accuracy": 0.14900366365909576,
      "num_tokens": 13015665.0,
      "step": 6270
    },
    {
      "entropy": 5.83261456489563,
      "epoch": 0.5692126269956459,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004974164565414617,
      "loss": 5.6126,
      "mean_token_accuracy": 0.15868498533964157,
      "num_tokens": 13025677.0,
      "step": 6275
    },
    {
      "entropy": 5.756546258926392,
      "epoch": 0.56966618287373,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004974115649965998,
      "loss": 5.6632,
      "mean_token_accuracy": 0.1580974504351616,
      "num_tokens": 13037539.0,
      "step": 6280
    },
    {
      "entropy": 5.594256162643433,
      "epoch": 0.5701197387518142,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004974066688522098,
      "loss": 5.5937,
      "mean_token_accuracy": 0.14920975863933564,
      "num_tokens": 13048520.0,
      "step": 6285
    },
    {
      "entropy": 5.690824413299561,
      "epoch": 0.5705732946298984,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004974017681083934,
      "loss": 5.5721,
      "mean_token_accuracy": 0.15977879911661147,
      "num_tokens": 13058751.0,
      "step": 6290
    },
    {
      "entropy": 5.809173583984375,
      "epoch": 0.5710268505079826,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004973968627652516,
      "loss": 5.6722,
      "mean_token_accuracy": 0.15209317505359649,
      "num_tokens": 13068808.0,
      "step": 6295
    },
    {
      "entropy": 5.700589227676391,
      "epoch": 0.5714804063860668,
      "grad_norm": 2.0625,
      "learning_rate": 0.000497391952822886,
      "loss": 5.6744,
      "mean_token_accuracy": 0.15689846128225327,
      "num_tokens": 13078602.0,
      "step": 6300
    },
    {
      "entropy": 5.748346757888794,
      "epoch": 0.5719339622641509,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004973870382813983,
      "loss": 5.6577,
      "mean_token_accuracy": 0.15334153771400452,
      "num_tokens": 13088628.0,
      "step": 6305
    },
    {
      "entropy": 5.744019412994385,
      "epoch": 0.5723875181422351,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004973821191408899,
      "loss": 5.4881,
      "mean_token_accuracy": 0.16003611981868743,
      "num_tokens": 13097938.0,
      "step": 6310
    },
    {
      "entropy": 5.781156444549561,
      "epoch": 0.5728410740203193,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004973771954014626,
      "loss": 5.6609,
      "mean_token_accuracy": 0.15123680084943772,
      "num_tokens": 13107574.0,
      "step": 6315
    },
    {
      "entropy": 5.764164829254151,
      "epoch": 0.5732946298984035,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004973722670632182,
      "loss": 5.6299,
      "mean_token_accuracy": 0.15441531240940093,
      "num_tokens": 13117617.0,
      "step": 6320
    },
    {
      "entropy": 5.778414916992188,
      "epoch": 0.5737481857764877,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004973673341262587,
      "loss": 5.62,
      "mean_token_accuracy": 0.1492793843150139,
      "num_tokens": 13128580.0,
      "step": 6325
    },
    {
      "entropy": 5.809693336486816,
      "epoch": 0.5742017416545718,
      "grad_norm": 2.671875,
      "learning_rate": 0.000497362396590686,
      "loss": 5.7638,
      "mean_token_accuracy": 0.15329949483275412,
      "num_tokens": 13139114.0,
      "step": 6330
    },
    {
      "entropy": 5.781783246994019,
      "epoch": 0.574655297532656,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004973574544566024,
      "loss": 5.6758,
      "mean_token_accuracy": 0.15328806936740874,
      "num_tokens": 13148195.0,
      "step": 6335
    },
    {
      "entropy": 5.687791872024536,
      "epoch": 0.5751088534107403,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004973525077241098,
      "loss": 5.6102,
      "mean_token_accuracy": 0.15689401775598527,
      "num_tokens": 13159018.0,
      "step": 6340
    },
    {
      "entropy": 5.711198997497559,
      "epoch": 0.5755624092888244,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004973475563933109,
      "loss": 5.5889,
      "mean_token_accuracy": 0.16024598926305772,
      "num_tokens": 13169257.0,
      "step": 6345
    },
    {
      "entropy": 5.754970455169678,
      "epoch": 0.5760159651669086,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004973426004643077,
      "loss": 5.6241,
      "mean_token_accuracy": 0.1545379042625427,
      "num_tokens": 13180009.0,
      "step": 6350
    },
    {
      "entropy": 5.736854362487793,
      "epoch": 0.5764695210449927,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004973376399372029,
      "loss": 5.6299,
      "mean_token_accuracy": 0.15409892797470093,
      "num_tokens": 13190828.0,
      "step": 6355
    },
    {
      "entropy": 5.788819360733032,
      "epoch": 0.5769230769230769,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004973326748120991,
      "loss": 5.7162,
      "mean_token_accuracy": 0.15394369661808013,
      "num_tokens": 13202037.0,
      "step": 6360
    },
    {
      "entropy": 5.778576374053955,
      "epoch": 0.5773766328011611,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004973277050890988,
      "loss": 5.6793,
      "mean_token_accuracy": 0.15594272464513778,
      "num_tokens": 13212105.0,
      "step": 6365
    },
    {
      "entropy": 5.913066387176514,
      "epoch": 0.5778301886792453,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004973227307683049,
      "loss": 5.8369,
      "mean_token_accuracy": 0.14454605802893639,
      "num_tokens": 13222643.0,
      "step": 6370
    },
    {
      "entropy": 5.719876337051391,
      "epoch": 0.5782837445573294,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004973177518498204,
      "loss": 5.6244,
      "mean_token_accuracy": 0.15246522575616836,
      "num_tokens": 13232009.0,
      "step": 6375
    },
    {
      "entropy": 5.771259593963623,
      "epoch": 0.5787373004354136,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004973127683337482,
      "loss": 5.7108,
      "mean_token_accuracy": 0.14938376992940902,
      "num_tokens": 13243084.0,
      "step": 6380
    },
    {
      "entropy": 5.844303846359253,
      "epoch": 0.5791908563134979,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004973077802201911,
      "loss": 5.7628,
      "mean_token_accuracy": 0.14356016516685485,
      "num_tokens": 13254065.0,
      "step": 6385
    },
    {
      "entropy": 5.796073770523071,
      "epoch": 0.579644412191582,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004973027875092526,
      "loss": 5.6515,
      "mean_token_accuracy": 0.1564701795578003,
      "num_tokens": 13263239.0,
      "step": 6390
    },
    {
      "entropy": 5.830417013168335,
      "epoch": 0.5800979680696662,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004972977902010357,
      "loss": 5.6982,
      "mean_token_accuracy": 0.15366180315613748,
      "num_tokens": 13273917.0,
      "step": 6395
    },
    {
      "entropy": 5.757380867004395,
      "epoch": 0.5805515239477503,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004972927882956439,
      "loss": 5.6446,
      "mean_token_accuracy": 0.16490886062383653,
      "num_tokens": 13283695.0,
      "step": 6400
    },
    {
      "entropy": 5.76453332901001,
      "epoch": 0.5810050798258345,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004972877817931807,
      "loss": 5.6334,
      "mean_token_accuracy": 0.16031342446804048,
      "num_tokens": 13293691.0,
      "step": 6405
    },
    {
      "entropy": 5.77696533203125,
      "epoch": 0.5814586357039188,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004972827706937493,
      "loss": 5.7093,
      "mean_token_accuracy": 0.15516743883490564,
      "num_tokens": 13304973.0,
      "step": 6410
    },
    {
      "entropy": 5.698088932037353,
      "epoch": 0.5819121915820029,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004972777549974537,
      "loss": 5.5858,
      "mean_token_accuracy": 0.1618632659316063,
      "num_tokens": 13313968.0,
      "step": 6415
    },
    {
      "entropy": 5.772373199462891,
      "epoch": 0.5823657474600871,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004972727347043975,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1542307913303375,
      "num_tokens": 13325086.0,
      "step": 6420
    },
    {
      "entropy": 5.764491415023803,
      "epoch": 0.5828193033381712,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004972677098146845,
      "loss": 5.5734,
      "mean_token_accuracy": 0.15369413495063783,
      "num_tokens": 13335937.0,
      "step": 6425
    },
    {
      "entropy": 5.808954524993896,
      "epoch": 0.5832728592162555,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004972626803284185,
      "loss": 5.635,
      "mean_token_accuracy": 0.15450548529624938,
      "num_tokens": 13346227.0,
      "step": 6430
    },
    {
      "entropy": 5.73987832069397,
      "epoch": 0.5837264150943396,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004972576462457038,
      "loss": 5.5942,
      "mean_token_accuracy": 0.1589360609650612,
      "num_tokens": 13355536.0,
      "step": 6435
    },
    {
      "entropy": 5.779764604568482,
      "epoch": 0.5841799709724238,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004972526075666442,
      "loss": 5.6934,
      "mean_token_accuracy": 0.1520060807466507,
      "num_tokens": 13367027.0,
      "step": 6440
    },
    {
      "entropy": 5.719965982437134,
      "epoch": 0.584633526850508,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004972475642913442,
      "loss": 5.5322,
      "mean_token_accuracy": 0.15796410590410231,
      "num_tokens": 13377079.0,
      "step": 6445
    },
    {
      "entropy": 5.715342378616333,
      "epoch": 0.5850870827285921,
      "grad_norm": 2.125,
      "learning_rate": 0.0004972425164199078,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16331557035446168,
      "num_tokens": 13387189.0,
      "step": 6450
    },
    {
      "entropy": 5.784458303451538,
      "epoch": 0.5855406386066764,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004972374639524396,
      "loss": 5.6798,
      "mean_token_accuracy": 0.149605430662632,
      "num_tokens": 13399446.0,
      "step": 6455
    },
    {
      "entropy": 5.708137845993042,
      "epoch": 0.5859941944847605,
      "grad_norm": 2.546875,
      "learning_rate": 0.000497232406889044,
      "loss": 5.6547,
      "mean_token_accuracy": 0.15906617194414138,
      "num_tokens": 13410320.0,
      "step": 6460
    },
    {
      "entropy": 5.732756042480469,
      "epoch": 0.5864477503628447,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004972273452298255,
      "loss": 5.5782,
      "mean_token_accuracy": 0.15970333516597748,
      "num_tokens": 13421458.0,
      "step": 6465
    },
    {
      "entropy": 5.789757585525512,
      "epoch": 0.5869013062409288,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004972222789748888,
      "loss": 5.6644,
      "mean_token_accuracy": 0.15411718487739562,
      "num_tokens": 13431815.0,
      "step": 6470
    },
    {
      "entropy": 5.804319143295288,
      "epoch": 0.5873548621190131,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004972172081243389,
      "loss": 5.69,
      "mean_token_accuracy": 0.1476941503584385,
      "num_tokens": 13442263.0,
      "step": 6475
    },
    {
      "entropy": 5.7650422096252445,
      "epoch": 0.5878084179970973,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004972121326782806,
      "loss": 5.6136,
      "mean_token_accuracy": 0.15558512806892394,
      "num_tokens": 13452360.0,
      "step": 6480
    },
    {
      "entropy": 5.68228588104248,
      "epoch": 0.5882619738751814,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004972070526368186,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16493213623762132,
      "num_tokens": 13461329.0,
      "step": 6485
    },
    {
      "entropy": 5.768403005599976,
      "epoch": 0.5887155297532656,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004972019680000582,
      "loss": 5.6524,
      "mean_token_accuracy": 0.15866607874631883,
      "num_tokens": 13472042.0,
      "step": 6490
    },
    {
      "entropy": 5.682829475402832,
      "epoch": 0.5891690856313497,
      "grad_norm": 2.375,
      "learning_rate": 0.0004971968787681044,
      "loss": 5.5544,
      "mean_token_accuracy": 0.1655348628759384,
      "num_tokens": 13482190.0,
      "step": 6495
    },
    {
      "entropy": 5.666314172744751,
      "epoch": 0.589622641509434,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004971917849410627,
      "loss": 5.662,
      "mean_token_accuracy": 0.15198497027158736,
      "num_tokens": 13493053.0,
      "step": 6500
    },
    {
      "entropy": 5.810998582839966,
      "epoch": 0.5900761973875182,
      "grad_norm": 2.25,
      "learning_rate": 0.0004971866865190381,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15123215317726135,
      "num_tokens": 13503629.0,
      "step": 6505
    },
    {
      "entropy": 5.829623794555664,
      "epoch": 0.5905297532656023,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004971815835021362,
      "loss": 5.6119,
      "mean_token_accuracy": 0.15927340015769004,
      "num_tokens": 13513776.0,
      "step": 6510
    },
    {
      "entropy": 5.787638854980469,
      "epoch": 0.5909833091436865,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004971764758904626,
      "loss": 5.6682,
      "mean_token_accuracy": 0.15258469730615615,
      "num_tokens": 13524705.0,
      "step": 6515
    },
    {
      "entropy": 5.681888914108276,
      "epoch": 0.5914368650217707,
      "grad_norm": 3.125,
      "learning_rate": 0.0004971713636841229,
      "loss": 5.6382,
      "mean_token_accuracy": 0.15930082648992538,
      "num_tokens": 13535799.0,
      "step": 6520
    },
    {
      "entropy": 5.808771467208862,
      "epoch": 0.5918904208998549,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004971662468832228,
      "loss": 5.6886,
      "mean_token_accuracy": 0.1564719334244728,
      "num_tokens": 13545067.0,
      "step": 6525
    },
    {
      "entropy": 5.733100175857544,
      "epoch": 0.592343976777939,
      "grad_norm": 1.796875,
      "learning_rate": 0.000497161125487868,
      "loss": 5.5617,
      "mean_token_accuracy": 0.1602954849600792,
      "num_tokens": 13555227.0,
      "step": 6530
    },
    {
      "entropy": 5.634074878692627,
      "epoch": 0.5927975326560232,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004971559994981645,
      "loss": 5.5131,
      "mean_token_accuracy": 0.16385299414396287,
      "num_tokens": 13565559.0,
      "step": 6535
    },
    {
      "entropy": 5.695637798309326,
      "epoch": 0.5932510885341074,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004971508689142184,
      "loss": 5.5492,
      "mean_token_accuracy": 0.16074319779872895,
      "num_tokens": 13575313.0,
      "step": 6540
    },
    {
      "entropy": 5.783664608001709,
      "epoch": 0.5937046444121916,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004971457337361356,
      "loss": 5.6206,
      "mean_token_accuracy": 0.15695622861385344,
      "num_tokens": 13584876.0,
      "step": 6545
    },
    {
      "entropy": 5.6620879650115965,
      "epoch": 0.5941582002902758,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004971405939640225,
      "loss": 5.6039,
      "mean_token_accuracy": 0.16317103505134584,
      "num_tokens": 13594811.0,
      "step": 6550
    },
    {
      "entropy": 5.703128814697266,
      "epoch": 0.5946117561683599,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004971354495979854,
      "loss": 5.6353,
      "mean_token_accuracy": 0.15475962609052657,
      "num_tokens": 13605806.0,
      "step": 6555
    },
    {
      "entropy": 5.799729585647583,
      "epoch": 0.5950653120464441,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004971303006381305,
      "loss": 5.699,
      "mean_token_accuracy": 0.15541718900203705,
      "num_tokens": 13616267.0,
      "step": 6560
    },
    {
      "entropy": 5.854510402679443,
      "epoch": 0.5955188679245284,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004971251470845643,
      "loss": 5.6818,
      "mean_token_accuracy": 0.15333730578422547,
      "num_tokens": 13625692.0,
      "step": 6565
    },
    {
      "entropy": 5.799213027954101,
      "epoch": 0.5959724238026125,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004971199889373935,
      "loss": 5.7124,
      "mean_token_accuracy": 0.1410965643823147,
      "num_tokens": 13636832.0,
      "step": 6570
    },
    {
      "entropy": 5.8232988834381105,
      "epoch": 0.5964259796806967,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004971148261967249,
      "loss": 5.6247,
      "mean_token_accuracy": 0.15864029079675673,
      "num_tokens": 13646657.0,
      "step": 6575
    },
    {
      "entropy": 5.697556734085083,
      "epoch": 0.5968795355587808,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004971096588626649,
      "loss": 5.5439,
      "mean_token_accuracy": 0.16281556636095046,
      "num_tokens": 13657612.0,
      "step": 6580
    },
    {
      "entropy": 5.752542018890381,
      "epoch": 0.597333091436865,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004971044869353206,
      "loss": 5.6575,
      "mean_token_accuracy": 0.15236000567674637,
      "num_tokens": 13668263.0,
      "step": 6585
    },
    {
      "entropy": 5.754230117797851,
      "epoch": 0.5977866473149492,
      "grad_norm": 2.140625,
      "learning_rate": 0.000497099310414799,
      "loss": 5.6275,
      "mean_token_accuracy": 0.1575901046395302,
      "num_tokens": 13678298.0,
      "step": 6590
    },
    {
      "entropy": 5.782630681991577,
      "epoch": 0.5982402031930334,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004970941293012069,
      "loss": 5.6138,
      "mean_token_accuracy": 0.1556815966963768,
      "num_tokens": 13688981.0,
      "step": 6595
    },
    {
      "entropy": 5.75571575164795,
      "epoch": 0.5986937590711175,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004970889435946518,
      "loss": 5.5976,
      "mean_token_accuracy": 0.1614097535610199,
      "num_tokens": 13699318.0,
      "step": 6600
    },
    {
      "entropy": 5.660605525970459,
      "epoch": 0.5991473149492017,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004970837532952406,
      "loss": 5.58,
      "mean_token_accuracy": 0.15405092388391495,
      "num_tokens": 13709708.0,
      "step": 6605
    },
    {
      "entropy": 5.737271451950074,
      "epoch": 0.599600870827286,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004970785584030808,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1567952439188957,
      "num_tokens": 13719480.0,
      "step": 6610
    },
    {
      "entropy": 5.748212814331055,
      "epoch": 0.6000544267053701,
      "grad_norm": 2.125,
      "learning_rate": 0.0004970733589182799,
      "loss": 5.6305,
      "mean_token_accuracy": 0.15508442372083664,
      "num_tokens": 13729090.0,
      "step": 6615
    },
    {
      "entropy": 5.683683729171753,
      "epoch": 0.6005079825834543,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004970681548409452,
      "loss": 5.5561,
      "mean_token_accuracy": 0.16317861676216125,
      "num_tokens": 13739094.0,
      "step": 6620
    },
    {
      "entropy": 5.714485454559326,
      "epoch": 0.6009615384615384,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004970629461711846,
      "loss": 5.5818,
      "mean_token_accuracy": 0.1589211955666542,
      "num_tokens": 13749646.0,
      "step": 6625
    },
    {
      "entropy": 5.841981315612793,
      "epoch": 0.6014150943396226,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004970577329091057,
      "loss": 5.703,
      "mean_token_accuracy": 0.1531035877764225,
      "num_tokens": 13760331.0,
      "step": 6630
    },
    {
      "entropy": 5.843902492523194,
      "epoch": 0.6018686502177069,
      "grad_norm": 2.0,
      "learning_rate": 0.0004970525150548162,
      "loss": 5.7305,
      "mean_token_accuracy": 0.1472798280417919,
      "num_tokens": 13771034.0,
      "step": 6635
    },
    {
      "entropy": 5.775877141952515,
      "epoch": 0.602322206095791,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004970472926084242,
      "loss": 5.5965,
      "mean_token_accuracy": 0.1525774747133255,
      "num_tokens": 13782816.0,
      "step": 6640
    },
    {
      "entropy": 5.647054958343506,
      "epoch": 0.6027757619738752,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004970420655700375,
      "loss": 5.5375,
      "mean_token_accuracy": 0.16223322451114655,
      "num_tokens": 13792626.0,
      "step": 6645
    },
    {
      "entropy": 5.694581604003906,
      "epoch": 0.6032293178519593,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004970368339397643,
      "loss": 5.5518,
      "mean_token_accuracy": 0.16072896718978882,
      "num_tokens": 13802863.0,
      "step": 6650
    },
    {
      "entropy": 5.853788375854492,
      "epoch": 0.6036828737300436,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004970315977177129,
      "loss": 5.6377,
      "mean_token_accuracy": 0.1542016237974167,
      "num_tokens": 13812714.0,
      "step": 6655
    },
    {
      "entropy": 5.775206995010376,
      "epoch": 0.6041364296081277,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004970263569039914,
      "loss": 5.6497,
      "mean_token_accuracy": 0.15461187213659286,
      "num_tokens": 13822634.0,
      "step": 6660
    },
    {
      "entropy": 5.716327905654907,
      "epoch": 0.6045899854862119,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004970211114987083,
      "loss": 5.611,
      "mean_token_accuracy": 0.1629389226436615,
      "num_tokens": 13833291.0,
      "step": 6665
    },
    {
      "entropy": 5.710997629165649,
      "epoch": 0.6050435413642961,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004970158615019719,
      "loss": 5.6274,
      "mean_token_accuracy": 0.16321856826543807,
      "num_tokens": 13843882.0,
      "step": 6670
    },
    {
      "entropy": 5.720949459075928,
      "epoch": 0.6054970972423802,
      "grad_norm": 2.75,
      "learning_rate": 0.0004970106069138912,
      "loss": 5.569,
      "mean_token_accuracy": 0.15597841441631316,
      "num_tokens": 13853883.0,
      "step": 6675
    },
    {
      "entropy": 5.796688604354858,
      "epoch": 0.6059506531204645,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004970053477345743,
      "loss": 5.6388,
      "mean_token_accuracy": 0.1547415718436241,
      "num_tokens": 13864440.0,
      "step": 6680
    },
    {
      "entropy": 5.750747394561768,
      "epoch": 0.6064042089985486,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004970000839641304,
      "loss": 5.6915,
      "mean_token_accuracy": 0.15220630168914795,
      "num_tokens": 13875902.0,
      "step": 6685
    },
    {
      "entropy": 5.720804738998413,
      "epoch": 0.6068577648766328,
      "grad_norm": 1.625,
      "learning_rate": 0.0004969948156026681,
      "loss": 5.6677,
      "mean_token_accuracy": 0.15204009115695954,
      "num_tokens": 13886252.0,
      "step": 6690
    },
    {
      "entropy": 5.732796478271484,
      "epoch": 0.6073113207547169,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004969895426502965,
      "loss": 5.5866,
      "mean_token_accuracy": 0.16024616360664368,
      "num_tokens": 13896641.0,
      "step": 6695
    },
    {
      "entropy": 5.797425937652588,
      "epoch": 0.6077648766328012,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004969842651071246,
      "loss": 5.6409,
      "mean_token_accuracy": 0.15383680015802384,
      "num_tokens": 13907504.0,
      "step": 6700
    },
    {
      "entropy": 5.783607053756714,
      "epoch": 0.6082184325108854,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004969789829732615,
      "loss": 5.6327,
      "mean_token_accuracy": 0.15362875834107398,
      "num_tokens": 13917743.0,
      "step": 6705
    },
    {
      "entropy": 5.6884692192077635,
      "epoch": 0.6086719883889695,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004969736962488164,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1540693961083889,
      "num_tokens": 13927626.0,
      "step": 6710
    },
    {
      "entropy": 5.808768081665039,
      "epoch": 0.6091255442670537,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004969684049338987,
      "loss": 5.6496,
      "mean_token_accuracy": 0.15174802988767624,
      "num_tokens": 13938863.0,
      "step": 6715
    },
    {
      "entropy": 5.797508382797242,
      "epoch": 0.6095791001451378,
      "grad_norm": 1.78125,
      "learning_rate": 0.000496963109028618,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1577400431036949,
      "num_tokens": 13948699.0,
      "step": 6720
    },
    {
      "entropy": 5.698787689208984,
      "epoch": 0.6100326560232221,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004969578085330834,
      "loss": 5.5013,
      "mean_token_accuracy": 0.16206391006708146,
      "num_tokens": 13958880.0,
      "step": 6725
    },
    {
      "entropy": 5.684061765670776,
      "epoch": 0.6104862119013063,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004969525034474049,
      "loss": 5.6561,
      "mean_token_accuracy": 0.14983780905604363,
      "num_tokens": 13968982.0,
      "step": 6730
    },
    {
      "entropy": 5.791654348373413,
      "epoch": 0.6109397677793904,
      "grad_norm": 2.171875,
      "learning_rate": 0.000496947193771692,
      "loss": 5.6594,
      "mean_token_accuracy": 0.15571676939725876,
      "num_tokens": 13978298.0,
      "step": 6735
    },
    {
      "entropy": 5.81650333404541,
      "epoch": 0.6113933236574746,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004969418795060548,
      "loss": 5.6118,
      "mean_token_accuracy": 0.15728780776262283,
      "num_tokens": 13988239.0,
      "step": 6740
    },
    {
      "entropy": 5.666800928115845,
      "epoch": 0.6118468795355588,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004969365606506027,
      "loss": 5.5409,
      "mean_token_accuracy": 0.16210765689611434,
      "num_tokens": 13997352.0,
      "step": 6745
    },
    {
      "entropy": 5.769309043884277,
      "epoch": 0.612300435413643,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004969312372054461,
      "loss": 5.7674,
      "mean_token_accuracy": 0.1455084577202797,
      "num_tokens": 14007662.0,
      "step": 6750
    },
    {
      "entropy": 5.798659944534302,
      "epoch": 0.6127539912917271,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004969259091706948,
      "loss": 5.6767,
      "mean_token_accuracy": 0.15299142152071,
      "num_tokens": 14017078.0,
      "step": 6755
    },
    {
      "entropy": 5.724830627441406,
      "epoch": 0.6132075471698113,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004969205765464593,
      "loss": 5.6833,
      "mean_token_accuracy": 0.15806994438171387,
      "num_tokens": 14028050.0,
      "step": 6760
    },
    {
      "entropy": 5.763033485412597,
      "epoch": 0.6136611030478955,
      "grad_norm": 2.25,
      "learning_rate": 0.0004969152393328497,
      "loss": 5.6985,
      "mean_token_accuracy": 0.14964941665530204,
      "num_tokens": 14037792.0,
      "step": 6765
    },
    {
      "entropy": 5.815360116958618,
      "epoch": 0.6141146589259797,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004969098975299764,
      "loss": 5.5908,
      "mean_token_accuracy": 0.15568995475769043,
      "num_tokens": 14047780.0,
      "step": 6770
    },
    {
      "entropy": 5.777581834793091,
      "epoch": 0.6145682148040639,
      "grad_norm": 2.5,
      "learning_rate": 0.0004969045511379499,
      "loss": 5.611,
      "mean_token_accuracy": 0.1585965633392334,
      "num_tokens": 14058076.0,
      "step": 6775
    },
    {
      "entropy": 5.751511955261231,
      "epoch": 0.615021770682148,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004968992001568806,
      "loss": 5.732,
      "mean_token_accuracy": 0.1500725895166397,
      "num_tokens": 14068082.0,
      "step": 6780
    },
    {
      "entropy": 5.764723587036133,
      "epoch": 0.6154753265602322,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004968938445868795,
      "loss": 5.6708,
      "mean_token_accuracy": 0.15336300283670426,
      "num_tokens": 14077905.0,
      "step": 6785
    },
    {
      "entropy": 5.799530506134033,
      "epoch": 0.6159288824383164,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004968884844280569,
      "loss": 5.6548,
      "mean_token_accuracy": 0.1536657318472862,
      "num_tokens": 14087113.0,
      "step": 6790
    },
    {
      "entropy": 5.8303056240081785,
      "epoch": 0.6163824383164006,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004968831196805242,
      "loss": 5.6767,
      "mean_token_accuracy": 0.15169122070074081,
      "num_tokens": 14097684.0,
      "step": 6795
    },
    {
      "entropy": 5.560121488571167,
      "epoch": 0.6168359941944848,
      "grad_norm": 2.125,
      "learning_rate": 0.0004968777503443918,
      "loss": 5.4729,
      "mean_token_accuracy": 0.161190202832222,
      "num_tokens": 14106871.0,
      "step": 6800
    },
    {
      "entropy": 5.70657000541687,
      "epoch": 0.6172895500725689,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004968723764197711,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1636699452996254,
      "num_tokens": 14117480.0,
      "step": 6805
    },
    {
      "entropy": 5.843404817581177,
      "epoch": 0.6177431059506531,
      "grad_norm": 3.140625,
      "learning_rate": 0.000496866997906773,
      "loss": 5.6291,
      "mean_token_accuracy": 0.16347150653600692,
      "num_tokens": 14128153.0,
      "step": 6810
    },
    {
      "entropy": 5.701880931854248,
      "epoch": 0.6181966618287373,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004968616148055089,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1551259309053421,
      "num_tokens": 14137920.0,
      "step": 6815
    },
    {
      "entropy": 5.715606927871704,
      "epoch": 0.6186502177068215,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004968562271160901,
      "loss": 5.5829,
      "mean_token_accuracy": 0.16260792315006256,
      "num_tokens": 14148160.0,
      "step": 6820
    },
    {
      "entropy": 5.7415777206420895,
      "epoch": 0.6191037735849056,
      "grad_norm": 2.984375,
      "learning_rate": 0.000496850834838628,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1547149032354355,
      "num_tokens": 14158646.0,
      "step": 6825
    },
    {
      "entropy": 5.684873151779175,
      "epoch": 0.6195573294629898,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004968454379732342,
      "loss": 5.6343,
      "mean_token_accuracy": 0.15305895507335662,
      "num_tokens": 14168719.0,
      "step": 6830
    },
    {
      "entropy": 5.805060291290284,
      "epoch": 0.6200108853410741,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00049684003652002,
      "loss": 5.7552,
      "mean_token_accuracy": 0.15023768320679665,
      "num_tokens": 14179859.0,
      "step": 6835
    },
    {
      "entropy": 5.772373104095459,
      "epoch": 0.6204644412191582,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004968346304790973,
      "loss": 5.6109,
      "mean_token_accuracy": 0.15124039947986603,
      "num_tokens": 14190413.0,
      "step": 6840
    },
    {
      "entropy": 5.786475276947021,
      "epoch": 0.6209179970972424,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004968292198505781,
      "loss": 5.7079,
      "mean_token_accuracy": 0.1573897510766983,
      "num_tokens": 14200337.0,
      "step": 6845
    },
    {
      "entropy": 5.827665328979492,
      "epoch": 0.6213715529753265,
      "grad_norm": 1.828125,
      "learning_rate": 0.000496823804634574,
      "loss": 5.7395,
      "mean_token_accuracy": 0.15085174515843391,
      "num_tokens": 14210848.0,
      "step": 6850
    },
    {
      "entropy": 5.803124332427979,
      "epoch": 0.6218251088534107,
      "grad_norm": 3.0,
      "learning_rate": 0.0004968183848311972,
      "loss": 5.7075,
      "mean_token_accuracy": 0.14558184519410133,
      "num_tokens": 14221358.0,
      "step": 6855
    },
    {
      "entropy": 5.849875497817993,
      "epoch": 0.622278664731495,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004968129604405596,
      "loss": 5.7384,
      "mean_token_accuracy": 0.15821532607078553,
      "num_tokens": 14232580.0,
      "step": 6860
    },
    {
      "entropy": 5.735620450973511,
      "epoch": 0.6227322206095791,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004968075314627734,
      "loss": 5.7144,
      "mean_token_accuracy": 0.14964599311351776,
      "num_tokens": 14243339.0,
      "step": 6865
    },
    {
      "entropy": 5.699955081939697,
      "epoch": 0.6231857764876633,
      "grad_norm": 2.078125,
      "learning_rate": 0.000496802097897951,
      "loss": 5.5328,
      "mean_token_accuracy": 0.1629217967391014,
      "num_tokens": 14254121.0,
      "step": 6870
    },
    {
      "entropy": 5.7439658641815186,
      "epoch": 0.6236393323657474,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004967966597462046,
      "loss": 5.5701,
      "mean_token_accuracy": 0.1599903240799904,
      "num_tokens": 14264196.0,
      "step": 6875
    },
    {
      "entropy": 5.756141281127929,
      "epoch": 0.6240928882438317,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004967912170076469,
      "loss": 5.585,
      "mean_token_accuracy": 0.1576563149690628,
      "num_tokens": 14274610.0,
      "step": 6880
    },
    {
      "entropy": 5.682001352310181,
      "epoch": 0.6245464441219158,
      "grad_norm": 5.96875,
      "learning_rate": 0.0004967857696823901,
      "loss": 5.5856,
      "mean_token_accuracy": 0.15772237703204156,
      "num_tokens": 14286355.0,
      "step": 6885
    },
    {
      "entropy": 5.726107454299926,
      "epoch": 0.625,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004967803177705472,
      "loss": 5.6177,
      "mean_token_accuracy": 0.15741511732339858,
      "num_tokens": 14296155.0,
      "step": 6890
    },
    {
      "entropy": 5.737909030914307,
      "epoch": 0.6254535558780842,
      "grad_norm": 2.0,
      "learning_rate": 0.0004967748612722308,
      "loss": 5.5918,
      "mean_token_accuracy": 0.15493713021278382,
      "num_tokens": 14305073.0,
      "step": 6895
    },
    {
      "entropy": 5.770492029190064,
      "epoch": 0.6259071117561683,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004967694001875538,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1596616804599762,
      "num_tokens": 14315646.0,
      "step": 6900
    },
    {
      "entropy": 5.777948188781738,
      "epoch": 0.6263606676342526,
      "grad_norm": 3.78125,
      "learning_rate": 0.000496763934516629,
      "loss": 5.6887,
      "mean_token_accuracy": 0.15220129936933519,
      "num_tokens": 14327058.0,
      "step": 6905
    },
    {
      "entropy": 5.656092357635498,
      "epoch": 0.6268142235123367,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004967584642595696,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16333331912755966,
      "num_tokens": 14336661.0,
      "step": 6910
    },
    {
      "entropy": 5.767233991622925,
      "epoch": 0.6272677793904209,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004967529894164887,
      "loss": 5.5932,
      "mean_token_accuracy": 0.1614171862602234,
      "num_tokens": 14346840.0,
      "step": 6915
    },
    {
      "entropy": 5.7081750392913815,
      "epoch": 0.627721335268505,
      "grad_norm": 2.5,
      "learning_rate": 0.0004967475099874992,
      "loss": 5.5466,
      "mean_token_accuracy": 0.1632200375199318,
      "num_tokens": 14356342.0,
      "step": 6920
    },
    {
      "entropy": 5.696488571166992,
      "epoch": 0.6281748911465893,
      "grad_norm": 3.09375,
      "learning_rate": 0.000496742025972715,
      "loss": 5.5584,
      "mean_token_accuracy": 0.16221774816513063,
      "num_tokens": 14365301.0,
      "step": 6925
    },
    {
      "entropy": 5.713243579864502,
      "epoch": 0.6286284470246735,
      "grad_norm": 2.3125,
      "learning_rate": 0.000496736537372249,
      "loss": 5.6484,
      "mean_token_accuracy": 0.15672291368246077,
      "num_tokens": 14376497.0,
      "step": 6930
    },
    {
      "entropy": 5.74711127281189,
      "epoch": 0.6290820029027576,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004967310441862151,
      "loss": 5.6033,
      "mean_token_accuracy": 0.16105988919734954,
      "num_tokens": 14386982.0,
      "step": 6935
    },
    {
      "entropy": 5.778274869918823,
      "epoch": 0.6295355587808418,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004967255464147265,
      "loss": 5.5992,
      "mean_token_accuracy": 0.15858430564403533,
      "num_tokens": 14397944.0,
      "step": 6940
    },
    {
      "entropy": 5.689877605438232,
      "epoch": 0.6299891146589259,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004967200440578972,
      "loss": 5.5592,
      "mean_token_accuracy": 0.16407246887683868,
      "num_tokens": 14408875.0,
      "step": 6945
    },
    {
      "entropy": 5.654283905029297,
      "epoch": 0.6304426705370102,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004967145371158408,
      "loss": 5.5104,
      "mean_token_accuracy": 0.1663491263985634,
      "num_tokens": 14418719.0,
      "step": 6950
    },
    {
      "entropy": 5.779173421859741,
      "epoch": 0.6308962264150944,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004967090255886714,
      "loss": 5.6971,
      "mean_token_accuracy": 0.15254574865102768,
      "num_tokens": 14429333.0,
      "step": 6955
    },
    {
      "entropy": 5.743547868728638,
      "epoch": 0.6313497822931785,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004967035094765028,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1506338968873024,
      "num_tokens": 14439787.0,
      "step": 6960
    },
    {
      "entropy": 5.786348295211792,
      "epoch": 0.6318033381712627,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004966979887794491,
      "loss": 5.6134,
      "mean_token_accuracy": 0.16431703865528108,
      "num_tokens": 14449711.0,
      "step": 6965
    },
    {
      "entropy": 5.736811542510987,
      "epoch": 0.6322568940493469,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004966924634976245,
      "loss": 5.5312,
      "mean_token_accuracy": 0.16069522500038147,
      "num_tokens": 14458750.0,
      "step": 6970
    },
    {
      "entropy": 5.7178483486175535,
      "epoch": 0.6327104499274311,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004966869336311433,
      "loss": 5.6298,
      "mean_token_accuracy": 0.1585947498679161,
      "num_tokens": 14468517.0,
      "step": 6975
    },
    {
      "entropy": 5.704970598220825,
      "epoch": 0.6331640058055152,
      "grad_norm": 2.125,
      "learning_rate": 0.0004966813991801197,
      "loss": 5.6509,
      "mean_token_accuracy": 0.15538498759269714,
      "num_tokens": 14478666.0,
      "step": 6980
    },
    {
      "entropy": 5.865767621994019,
      "epoch": 0.6336175616835994,
      "grad_norm": 3.0,
      "learning_rate": 0.0004966758601446684,
      "loss": 5.6455,
      "mean_token_accuracy": 0.1615960031747818,
      "num_tokens": 14489227.0,
      "step": 6985
    },
    {
      "entropy": 5.835824489593506,
      "epoch": 0.6340711175616836,
      "grad_norm": 2.125,
      "learning_rate": 0.0004966703165249038,
      "loss": 5.7582,
      "mean_token_accuracy": 0.14500172287225724,
      "num_tokens": 14498955.0,
      "step": 6990
    },
    {
      "entropy": 5.853192472457886,
      "epoch": 0.6345246734397678,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004966647683209406,
      "loss": 5.6736,
      "mean_token_accuracy": 0.15712972432374955,
      "num_tokens": 14510565.0,
      "step": 6995
    },
    {
      "entropy": 5.708524417877197,
      "epoch": 0.634978229317852,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004966592155328935,
      "loss": 5.5528,
      "mean_token_accuracy": 0.15350938439369202,
      "num_tokens": 14521367.0,
      "step": 7000
    },
    {
      "entropy": 5.751900768280029,
      "epoch": 0.6354317851959361,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004966536581608771,
      "loss": 5.6512,
      "mean_token_accuracy": 0.1562518835067749,
      "num_tokens": 14531198.0,
      "step": 7005
    },
    {
      "entropy": 5.774559879302979,
      "epoch": 0.6358853410740203,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004966480962050069,
      "loss": 5.6263,
      "mean_token_accuracy": 0.15649291574954988,
      "num_tokens": 14542603.0,
      "step": 7010
    },
    {
      "entropy": 5.839611196517945,
      "epoch": 0.6363388969521045,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004966425296653974,
      "loss": 5.7026,
      "mean_token_accuracy": 0.15144390612840652,
      "num_tokens": 14553431.0,
      "step": 7015
    },
    {
      "entropy": 5.765762996673584,
      "epoch": 0.6367924528301887,
      "grad_norm": 2.296875,
      "learning_rate": 0.000496636958542164,
      "loss": 5.704,
      "mean_token_accuracy": 0.15698604434728622,
      "num_tokens": 14563334.0,
      "step": 7020
    },
    {
      "entropy": 5.729667615890503,
      "epoch": 0.6372460087082729,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004966313828354218,
      "loss": 5.5169,
      "mean_token_accuracy": 0.1549726501107216,
      "num_tokens": 14573330.0,
      "step": 7025
    },
    {
      "entropy": 5.790589046478272,
      "epoch": 0.637699564586357,
      "grad_norm": 2.390625,
      "learning_rate": 0.000496625802545286,
      "loss": 5.7186,
      "mean_token_accuracy": 0.1527685448527336,
      "num_tokens": 14584994.0,
      "step": 7030
    },
    {
      "entropy": 5.837193155288697,
      "epoch": 0.6381531204644412,
      "grad_norm": 5.5,
      "learning_rate": 0.0004966202176718723,
      "loss": 5.5931,
      "mean_token_accuracy": 0.15974610447883605,
      "num_tokens": 14596324.0,
      "step": 7035
    },
    {
      "entropy": 5.807248115539551,
      "epoch": 0.6386066763425254,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004966146282152959,
      "loss": 5.6255,
      "mean_token_accuracy": 0.15620894283056258,
      "num_tokens": 14606159.0,
      "step": 7040
    },
    {
      "entropy": 5.863836669921875,
      "epoch": 0.6390602322206096,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004966090341756726,
      "loss": 5.7335,
      "mean_token_accuracy": 0.14900389909744263,
      "num_tokens": 14616851.0,
      "step": 7045
    },
    {
      "entropy": 5.695793914794922,
      "epoch": 0.6395137880986937,
      "grad_norm": 2.375,
      "learning_rate": 0.000496603435553118,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16355661302804947,
      "num_tokens": 14627729.0,
      "step": 7050
    },
    {
      "entropy": 5.7164922714233395,
      "epoch": 0.6399673439767779,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004965978323477478,
      "loss": 5.5522,
      "mean_token_accuracy": 0.1598214849829674,
      "num_tokens": 14637591.0,
      "step": 7055
    },
    {
      "entropy": 5.759249687194824,
      "epoch": 0.6404208998548622,
      "grad_norm": 3.328125,
      "learning_rate": 0.000496592224559678,
      "loss": 5.6002,
      "mean_token_accuracy": 0.158864888548851,
      "num_tokens": 14647507.0,
      "step": 7060
    },
    {
      "entropy": 5.6821815967559814,
      "epoch": 0.6408744557329463,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004965866121890245,
      "loss": 5.5573,
      "mean_token_accuracy": 0.1597239226102829,
      "num_tokens": 14657339.0,
      "step": 7065
    },
    {
      "entropy": 5.70623517036438,
      "epoch": 0.6413280116110305,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004965809952359034,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16201654821634293,
      "num_tokens": 14668621.0,
      "step": 7070
    },
    {
      "entropy": 5.767104053497315,
      "epoch": 0.6417815674891146,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004965753737004309,
      "loss": 5.6587,
      "mean_token_accuracy": 0.1593472570180893,
      "num_tokens": 14678530.0,
      "step": 7075
    },
    {
      "entropy": 5.714475059509278,
      "epoch": 0.6422351233671988,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004965697475827231,
      "loss": 5.4932,
      "mean_token_accuracy": 0.16647264808416368,
      "num_tokens": 14688303.0,
      "step": 7080
    },
    {
      "entropy": 5.7584463596344,
      "epoch": 0.6426886792452831,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004965641168828965,
      "loss": 5.6249,
      "mean_token_accuracy": 0.1570249542593956,
      "num_tokens": 14698697.0,
      "step": 7085
    },
    {
      "entropy": 5.656041288375855,
      "epoch": 0.6431422351233672,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004965584816010676,
      "loss": 5.5214,
      "mean_token_accuracy": 0.165100035071373,
      "num_tokens": 14708541.0,
      "step": 7090
    },
    {
      "entropy": 5.7682939052581785,
      "epoch": 0.6435957910014514,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004965528417373528,
      "loss": 5.6979,
      "mean_token_accuracy": 0.14949500411748887,
      "num_tokens": 14720123.0,
      "step": 7095
    },
    {
      "entropy": 5.793861341476441,
      "epoch": 0.6440493468795355,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004965471972918688,
      "loss": 5.616,
      "mean_token_accuracy": 0.15569858253002167,
      "num_tokens": 14731439.0,
      "step": 7100
    },
    {
      "entropy": 5.753037643432617,
      "epoch": 0.6445029027576198,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004965415482647322,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15246612578630447,
      "num_tokens": 14742243.0,
      "step": 7105
    },
    {
      "entropy": 5.741925859451294,
      "epoch": 0.6449564586357039,
      "grad_norm": 2.421875,
      "learning_rate": 0.00049653589465606,
      "loss": 5.6013,
      "mean_token_accuracy": 0.1528177335858345,
      "num_tokens": 14751997.0,
      "step": 7110
    },
    {
      "entropy": 5.77686767578125,
      "epoch": 0.6454100145137881,
      "grad_norm": 2.78125,
      "learning_rate": 0.000496530236465969,
      "loss": 5.5737,
      "mean_token_accuracy": 0.15989938303828238,
      "num_tokens": 14762293.0,
      "step": 7115
    },
    {
      "entropy": 5.820109605789185,
      "epoch": 0.6458635703918723,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004965245736945763,
      "loss": 5.6293,
      "mean_token_accuracy": 0.15774548649787903,
      "num_tokens": 14771332.0,
      "step": 7120
    },
    {
      "entropy": 5.7409556865692135,
      "epoch": 0.6463171262699564,
      "grad_norm": 4.125,
      "learning_rate": 0.0004965189063419989,
      "loss": 5.5986,
      "mean_token_accuracy": 0.15244969725608826,
      "num_tokens": 14781757.0,
      "step": 7125
    },
    {
      "entropy": 5.74225435256958,
      "epoch": 0.6467706821480407,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004965132344083541,
      "loss": 5.614,
      "mean_token_accuracy": 0.15569283962249755,
      "num_tokens": 14791338.0,
      "step": 7130
    },
    {
      "entropy": 5.740928411483765,
      "epoch": 0.6472242380261248,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004965075578937593,
      "loss": 5.7476,
      "mean_token_accuracy": 0.1521914705634117,
      "num_tokens": 14802033.0,
      "step": 7135
    },
    {
      "entropy": 5.70343713760376,
      "epoch": 0.647677793904209,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004965018767983316,
      "loss": 5.5736,
      "mean_token_accuracy": 0.16235738694667817,
      "num_tokens": 14812341.0,
      "step": 7140
    },
    {
      "entropy": 5.829333400726318,
      "epoch": 0.6481313497822931,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004964961911221886,
      "loss": 5.7448,
      "mean_token_accuracy": 0.1459219217300415,
      "num_tokens": 14824334.0,
      "step": 7145
    },
    {
      "entropy": 5.769096899032593,
      "epoch": 0.6485849056603774,
      "grad_norm": 2.359375,
      "learning_rate": 0.000496490500865448,
      "loss": 5.5908,
      "mean_token_accuracy": 0.15827197432518006,
      "num_tokens": 14834200.0,
      "step": 7150
    },
    {
      "entropy": 5.7518878936767575,
      "epoch": 0.6490384615384616,
      "grad_norm": 3.125,
      "learning_rate": 0.0004964848060282275,
      "loss": 5.5887,
      "mean_token_accuracy": 0.1599769726395607,
      "num_tokens": 14845425.0,
      "step": 7155
    },
    {
      "entropy": 5.8009261131286625,
      "epoch": 0.6494920174165457,
      "grad_norm": 2.25,
      "learning_rate": 0.0004964791066106447,
      "loss": 5.6985,
      "mean_token_accuracy": 0.15522825419902803,
      "num_tokens": 14856029.0,
      "step": 7160
    },
    {
      "entropy": 5.7412573337554935,
      "epoch": 0.6499455732946299,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004964734026128175,
      "loss": 5.6052,
      "mean_token_accuracy": 0.15963159501552582,
      "num_tokens": 14865481.0,
      "step": 7165
    },
    {
      "entropy": 5.7057243347167965,
      "epoch": 0.650399129172714,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004964676940348638,
      "loss": 5.6088,
      "mean_token_accuracy": 0.15293650478124618,
      "num_tokens": 14875109.0,
      "step": 7170
    },
    {
      "entropy": 5.813092994689941,
      "epoch": 0.6508526850507983,
      "grad_norm": 2.109375,
      "learning_rate": 0.000496461980876902,
      "loss": 5.5766,
      "mean_token_accuracy": 0.16277963519096375,
      "num_tokens": 14885302.0,
      "step": 7175
    },
    {
      "entropy": 5.786996841430664,
      "epoch": 0.6513062409288825,
      "grad_norm": 11.375,
      "learning_rate": 0.0004964562631390498,
      "loss": 5.5822,
      "mean_token_accuracy": 0.1609741762280464,
      "num_tokens": 14894418.0,
      "step": 7180
    },
    {
      "entropy": 5.749535274505615,
      "epoch": 0.6517597968069666,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004964505408214257,
      "loss": 5.5895,
      "mean_token_accuracy": 0.16198379248380662,
      "num_tokens": 14904738.0,
      "step": 7185
    },
    {
      "entropy": 5.735393428802491,
      "epoch": 0.6522133526850508,
      "grad_norm": 4.78125,
      "learning_rate": 0.000496444813924148,
      "loss": 5.5706,
      "mean_token_accuracy": 0.15685581415891647,
      "num_tokens": 14916012.0,
      "step": 7190
    },
    {
      "entropy": 5.778084564208984,
      "epoch": 0.652666908563135,
      "grad_norm": 5.59375,
      "learning_rate": 0.0004964390824473351,
      "loss": 5.6738,
      "mean_token_accuracy": 0.15826205611228944,
      "num_tokens": 14927239.0,
      "step": 7195
    },
    {
      "entropy": 5.801933240890503,
      "epoch": 0.6531204644412192,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004964333463911056,
      "loss": 5.5959,
      "mean_token_accuracy": 0.1508314147591591,
      "num_tokens": 14937915.0,
      "step": 7200
    },
    {
      "entropy": 5.7537510871887205,
      "epoch": 0.6535740203193033,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004964276057555779,
      "loss": 5.6958,
      "mean_token_accuracy": 0.15559779703617097,
      "num_tokens": 14948360.0,
      "step": 7205
    },
    {
      "entropy": 5.762653732299805,
      "epoch": 0.6540275761973875,
      "grad_norm": 3.359375,
      "learning_rate": 0.000496421860540871,
      "loss": 5.7211,
      "mean_token_accuracy": 0.14721062853932382,
      "num_tokens": 14958838.0,
      "step": 7210
    },
    {
      "entropy": 5.84288854598999,
      "epoch": 0.6544811320754716,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004964161107471036,
      "loss": 5.7113,
      "mean_token_accuracy": 0.15069502517580985,
      "num_tokens": 14968960.0,
      "step": 7215
    },
    {
      "entropy": 5.786521244049072,
      "epoch": 0.6549346879535559,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004964103563743946,
      "loss": 5.6103,
      "mean_token_accuracy": 0.164300574362278,
      "num_tokens": 14978550.0,
      "step": 7220
    },
    {
      "entropy": 5.754683971405029,
      "epoch": 0.6553882438316401,
      "grad_norm": 2.53125,
      "learning_rate": 0.000496404597422863,
      "loss": 5.6869,
      "mean_token_accuracy": 0.15014653727412225,
      "num_tokens": 14990390.0,
      "step": 7225
    },
    {
      "entropy": 5.76729564666748,
      "epoch": 0.6558417997097242,
      "grad_norm": 3.078125,
      "learning_rate": 0.000496398833892628,
      "loss": 5.6295,
      "mean_token_accuracy": 0.15480599403381348,
      "num_tokens": 15001563.0,
      "step": 7230
    },
    {
      "entropy": 5.853767395019531,
      "epoch": 0.6562953555878084,
      "grad_norm": 2.25,
      "learning_rate": 0.0004963930657838086,
      "loss": 5.7008,
      "mean_token_accuracy": 0.15056831240653992,
      "num_tokens": 15012740.0,
      "step": 7235
    },
    {
      "entropy": 5.868458700180054,
      "epoch": 0.6567489114658926,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004963872930965242,
      "loss": 5.7208,
      "mean_token_accuracy": 0.1523576334118843,
      "num_tokens": 15024082.0,
      "step": 7240
    },
    {
      "entropy": 5.750052881240845,
      "epoch": 0.6572024673439768,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004963815158308941,
      "loss": 5.6424,
      "mean_token_accuracy": 0.15408817678689957,
      "num_tokens": 15034300.0,
      "step": 7245
    },
    {
      "entropy": 5.751996660232544,
      "epoch": 0.657656023222061,
      "grad_norm": 4.5,
      "learning_rate": 0.0004963757339870379,
      "loss": 5.5751,
      "mean_token_accuracy": 0.16331602334976197,
      "num_tokens": 15043277.0,
      "step": 7250
    },
    {
      "entropy": 5.784009408950806,
      "epoch": 0.6581095791001451,
      "grad_norm": 2.5,
      "learning_rate": 0.0004963699475650751,
      "loss": 5.6267,
      "mean_token_accuracy": 0.15816349685192108,
      "num_tokens": 15053686.0,
      "step": 7255
    },
    {
      "entropy": 5.7810920715332035,
      "epoch": 0.6585631349782293,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004963641565651255,
      "loss": 5.5729,
      "mean_token_accuracy": 0.1629945904016495,
      "num_tokens": 15063144.0,
      "step": 7260
    },
    {
      "entropy": 5.741094255447388,
      "epoch": 0.6590166908563135,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004963583609873084,
      "loss": 5.6733,
      "mean_token_accuracy": 0.1572900578379631,
      "num_tokens": 15072564.0,
      "step": 7265
    },
    {
      "entropy": 5.760691785812378,
      "epoch": 0.6594702467343977,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004963525608317443,
      "loss": 5.6627,
      "mean_token_accuracy": 0.15416491776704788,
      "num_tokens": 15082273.0,
      "step": 7270
    },
    {
      "entropy": 5.754504013061523,
      "epoch": 0.6599238026124818,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004963467560985527,
      "loss": 5.6003,
      "mean_token_accuracy": 0.15689957588911058,
      "num_tokens": 15091662.0,
      "step": 7275
    },
    {
      "entropy": 5.782180547714233,
      "epoch": 0.660377358490566,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004963409467878538,
      "loss": 5.5826,
      "mean_token_accuracy": 0.16143929287791253,
      "num_tokens": 15101188.0,
      "step": 7280
    },
    {
      "entropy": 5.780669927597046,
      "epoch": 0.6608309143686503,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004963351328997678,
      "loss": 5.629,
      "mean_token_accuracy": 0.1538732334971428,
      "num_tokens": 15111352.0,
      "step": 7285
    },
    {
      "entropy": 5.881281089782715,
      "epoch": 0.6612844702467344,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004963293144344148,
      "loss": 5.7161,
      "mean_token_accuracy": 0.1468665912747383,
      "num_tokens": 15122225.0,
      "step": 7290
    },
    {
      "entropy": 5.771633052825928,
      "epoch": 0.6617380261248186,
      "grad_norm": 3.875,
      "learning_rate": 0.0004963234913919151,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1548286035656929,
      "num_tokens": 15132807.0,
      "step": 7295
    },
    {
      "entropy": 5.7575364112854,
      "epoch": 0.6621915820029027,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004963176637723893,
      "loss": 5.6775,
      "mean_token_accuracy": 0.14549415707588195,
      "num_tokens": 15143664.0,
      "step": 7300
    },
    {
      "entropy": 5.808767509460449,
      "epoch": 0.6626451378809869,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004963118315759578,
      "loss": 5.6371,
      "mean_token_accuracy": 0.15712959915399552,
      "num_tokens": 15153977.0,
      "step": 7305
    },
    {
      "entropy": 5.759644079208374,
      "epoch": 0.6630986937590712,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004963059948027411,
      "loss": 5.5384,
      "mean_token_accuracy": 0.16429376006126403,
      "num_tokens": 15163128.0,
      "step": 7310
    },
    {
      "entropy": 5.832194375991821,
      "epoch": 0.6635522496371553,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004963001534528601,
      "loss": 5.7649,
      "mean_token_accuracy": 0.15502927601337432,
      "num_tokens": 15173516.0,
      "step": 7315
    },
    {
      "entropy": 5.745299291610718,
      "epoch": 0.6640058055152395,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004962943075264355,
      "loss": 5.62,
      "mean_token_accuracy": 0.15902740955352784,
      "num_tokens": 15183872.0,
      "step": 7320
    },
    {
      "entropy": 5.77251558303833,
      "epoch": 0.6644593613933236,
      "grad_norm": 7.5,
      "learning_rate": 0.0004962884570235884,
      "loss": 5.6383,
      "mean_token_accuracy": 0.15652472153306007,
      "num_tokens": 15193830.0,
      "step": 7325
    },
    {
      "entropy": 5.676307487487793,
      "epoch": 0.6649129172714079,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004962826019444394,
      "loss": 5.5226,
      "mean_token_accuracy": 0.1557547003030777,
      "num_tokens": 15204021.0,
      "step": 7330
    },
    {
      "entropy": 5.72134370803833,
      "epoch": 0.665366473149492,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004962767422891099,
      "loss": 5.6038,
      "mean_token_accuracy": 0.16044427454471588,
      "num_tokens": 15213820.0,
      "step": 7335
    },
    {
      "entropy": 5.777765989303589,
      "epoch": 0.6658200290275762,
      "grad_norm": 2.953125,
      "learning_rate": 0.000496270878057721,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15643776059150696,
      "num_tokens": 15223285.0,
      "step": 7340
    },
    {
      "entropy": 5.8106386184692385,
      "epoch": 0.6662735849056604,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004962650092503939,
      "loss": 5.5957,
      "mean_token_accuracy": 0.16323304921388626,
      "num_tokens": 15233759.0,
      "step": 7345
    },
    {
      "entropy": 5.796981859207153,
      "epoch": 0.6667271407837445,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004962591358672502,
      "loss": 5.6812,
      "mean_token_accuracy": 0.1529715485870838,
      "num_tokens": 15244633.0,
      "step": 7350
    },
    {
      "entropy": 5.76887469291687,
      "epoch": 0.6671806966618288,
      "grad_norm": 2.75,
      "learning_rate": 0.000496253257908411,
      "loss": 5.684,
      "mean_token_accuracy": 0.15396681278944016,
      "num_tokens": 15255160.0,
      "step": 7355
    },
    {
      "entropy": 5.887839603424072,
      "epoch": 0.6676342525399129,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004962473753739981,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15530455261468887,
      "num_tokens": 15264641.0,
      "step": 7360
    },
    {
      "entropy": 5.847436571121216,
      "epoch": 0.6680878084179971,
      "grad_norm": 4.53125,
      "learning_rate": 0.000496241488264133,
      "loss": 5.6734,
      "mean_token_accuracy": 0.15171343982219695,
      "num_tokens": 15274945.0,
      "step": 7365
    },
    {
      "entropy": 5.727545833587646,
      "epoch": 0.6685413642960812,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004962355965789377,
      "loss": 5.5607,
      "mean_token_accuracy": 0.15616238862276077,
      "num_tokens": 15284692.0,
      "step": 7370
    },
    {
      "entropy": 5.712421989440918,
      "epoch": 0.6689949201741655,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004962297003185338,
      "loss": 5.6033,
      "mean_token_accuracy": 0.15451953560113907,
      "num_tokens": 15294199.0,
      "step": 7375
    },
    {
      "entropy": 5.770094490051269,
      "epoch": 0.6694484760522497,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004962237994830433,
      "loss": 5.5128,
      "mean_token_accuracy": 0.15364942699670792,
      "num_tokens": 15304401.0,
      "step": 7380
    },
    {
      "entropy": 5.804858446121216,
      "epoch": 0.6699020319303338,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004962178940725883,
      "loss": 5.591,
      "mean_token_accuracy": 0.16549576073884964,
      "num_tokens": 15314520.0,
      "step": 7385
    },
    {
      "entropy": 5.732750177383423,
      "epoch": 0.670355587808418,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004962119840872908,
      "loss": 5.5755,
      "mean_token_accuracy": 0.16173245906829833,
      "num_tokens": 15324071.0,
      "step": 7390
    },
    {
      "entropy": 5.73241810798645,
      "epoch": 0.6708091436865021,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004962060695272733,
      "loss": 5.5429,
      "mean_token_accuracy": 0.1608823835849762,
      "num_tokens": 15334245.0,
      "step": 7395
    },
    {
      "entropy": 5.7757049083709715,
      "epoch": 0.6712626995645864,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004962001503926577,
      "loss": 5.6373,
      "mean_token_accuracy": 0.15362300872802734,
      "num_tokens": 15344425.0,
      "step": 7400
    },
    {
      "entropy": 5.763125705718994,
      "epoch": 0.6717162554426706,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004961942266835667,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1443048521876335,
      "num_tokens": 15355237.0,
      "step": 7405
    },
    {
      "entropy": 5.750442314147949,
      "epoch": 0.6721698113207547,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004961882984001227,
      "loss": 5.5978,
      "mean_token_accuracy": 0.16058864295482636,
      "num_tokens": 15365323.0,
      "step": 7410
    },
    {
      "entropy": 5.7392988204956055,
      "epoch": 0.6726233671988389,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004961823655424484,
      "loss": 5.5328,
      "mean_token_accuracy": 0.16431051939725877,
      "num_tokens": 15374652.0,
      "step": 7415
    },
    {
      "entropy": 5.758931493759155,
      "epoch": 0.6730769230769231,
      "grad_norm": 2.125,
      "learning_rate": 0.0004961764281106664,
      "loss": 5.562,
      "mean_token_accuracy": 0.16057585924863815,
      "num_tokens": 15384582.0,
      "step": 7420
    },
    {
      "entropy": 5.7668619632720945,
      "epoch": 0.6735304789550073,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004961704861048995,
      "loss": 5.6169,
      "mean_token_accuracy": 0.15404267162084578,
      "num_tokens": 15394292.0,
      "step": 7425
    },
    {
      "entropy": 5.796910285949707,
      "epoch": 0.6739840348330914,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004961645395252705,
      "loss": 5.669,
      "mean_token_accuracy": 0.14948946088552476,
      "num_tokens": 15404036.0,
      "step": 7430
    },
    {
      "entropy": 5.787111520767212,
      "epoch": 0.6744375907111756,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004961585883719025,
      "loss": 5.5849,
      "mean_token_accuracy": 0.1580713450908661,
      "num_tokens": 15414289.0,
      "step": 7435
    },
    {
      "entropy": 5.8226477146148685,
      "epoch": 0.6748911465892597,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004961526326449186,
      "loss": 5.6941,
      "mean_token_accuracy": 0.15350158140063286,
      "num_tokens": 15425404.0,
      "step": 7440
    },
    {
      "entropy": 5.799039840698242,
      "epoch": 0.675344702467344,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004961466723444419,
      "loss": 5.6957,
      "mean_token_accuracy": 0.14718335643410682,
      "num_tokens": 15435437.0,
      "step": 7445
    },
    {
      "entropy": 5.734977293014526,
      "epoch": 0.6757982583454282,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004961407074705955,
      "loss": 5.6348,
      "mean_token_accuracy": 0.1534969449043274,
      "num_tokens": 15446434.0,
      "step": 7450
    },
    {
      "entropy": 5.886393976211548,
      "epoch": 0.6762518142235123,
      "grad_norm": 2.375,
      "learning_rate": 0.0004961347380235031,
      "loss": 5.6691,
      "mean_token_accuracy": 0.14869626760482788,
      "num_tokens": 15456346.0,
      "step": 7455
    },
    {
      "entropy": 5.814081239700317,
      "epoch": 0.6767053701015965,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004961287640032879,
      "loss": 5.6438,
      "mean_token_accuracy": 0.15773403644561768,
      "num_tokens": 15466214.0,
      "step": 7460
    },
    {
      "entropy": 5.6498833179473875,
      "epoch": 0.6771589259796807,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004961227854100734,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1564796581864357,
      "num_tokens": 15476229.0,
      "step": 7465
    },
    {
      "entropy": 5.7101973533630375,
      "epoch": 0.6776124818577649,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004961168022439835,
      "loss": 5.5777,
      "mean_token_accuracy": 0.15887040942907332,
      "num_tokens": 15486735.0,
      "step": 7470
    },
    {
      "entropy": 5.866124248504638,
      "epoch": 0.6780660377358491,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004961108145051417,
      "loss": 5.6967,
      "mean_token_accuracy": 0.1500288352370262,
      "num_tokens": 15497201.0,
      "step": 7475
    },
    {
      "entropy": 5.75830602645874,
      "epoch": 0.6785195936139332,
      "grad_norm": 3.03125,
      "learning_rate": 0.000496104822193672,
      "loss": 5.5525,
      "mean_token_accuracy": 0.1587577998638153,
      "num_tokens": 15507053.0,
      "step": 7480
    },
    {
      "entropy": 5.746447610855102,
      "epoch": 0.6789731494920174,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004960988253096982,
      "loss": 5.6633,
      "mean_token_accuracy": 0.1482136368751526,
      "num_tokens": 15518032.0,
      "step": 7485
    },
    {
      "entropy": 5.688390159606934,
      "epoch": 0.6794267053701016,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004960928238533443,
      "loss": 5.6136,
      "mean_token_accuracy": 0.1566487208008766,
      "num_tokens": 15529034.0,
      "step": 7490
    },
    {
      "entropy": 5.769197702407837,
      "epoch": 0.6798802612481858,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004960868178247344,
      "loss": 5.5861,
      "mean_token_accuracy": 0.15801844000816345,
      "num_tokens": 15539153.0,
      "step": 7495
    },
    {
      "entropy": 5.762501573562622,
      "epoch": 0.68033381712627,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004960808072239929,
      "loss": 5.5257,
      "mean_token_accuracy": 0.16039573326706885,
      "num_tokens": 15549427.0,
      "step": 7500
    },
    {
      "entropy": 5.753922939300537,
      "epoch": 0.6807873730043541,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004960747920512439,
      "loss": 5.6457,
      "mean_token_accuracy": 0.15405466556549072,
      "num_tokens": 15558834.0,
      "step": 7505
    },
    {
      "entropy": 5.762908124923706,
      "epoch": 0.6812409288824384,
      "grad_norm": 2.75,
      "learning_rate": 0.000496068772306612,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15160716623067855,
      "num_tokens": 15569209.0,
      "step": 7510
    },
    {
      "entropy": 5.750273180007935,
      "epoch": 0.6816944847605225,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004960627479902214,
      "loss": 5.5825,
      "mean_token_accuracy": 0.15494249761104584,
      "num_tokens": 15579239.0,
      "step": 7515
    },
    {
      "entropy": 5.739710664749145,
      "epoch": 0.6821480406386067,
      "grad_norm": 3.609375,
      "learning_rate": 0.000496056719102197,
      "loss": 5.5608,
      "mean_token_accuracy": 0.1585078254342079,
      "num_tokens": 15590036.0,
      "step": 7520
    },
    {
      "entropy": 5.736978149414062,
      "epoch": 0.6826015965166908,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004960506856426631,
      "loss": 5.599,
      "mean_token_accuracy": 0.16314262002706528,
      "num_tokens": 15600375.0,
      "step": 7525
    },
    {
      "entropy": 5.713376426696778,
      "epoch": 0.683055152394775,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004960446476117448,
      "loss": 5.6464,
      "mean_token_accuracy": 0.15013087242841722,
      "num_tokens": 15610859.0,
      "step": 7530
    },
    {
      "entropy": 5.815583229064941,
      "epoch": 0.6835087082728593,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004960386050095668,
      "loss": 5.6439,
      "mean_token_accuracy": 0.1477643832564354,
      "num_tokens": 15621914.0,
      "step": 7535
    },
    {
      "entropy": 5.8543415546417235,
      "epoch": 0.6839622641509434,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004960325578362543,
      "loss": 5.7415,
      "mean_token_accuracy": 0.1533266767859459,
      "num_tokens": 15632379.0,
      "step": 7540
    },
    {
      "entropy": 5.728507375717163,
      "epoch": 0.6844158200290276,
      "grad_norm": 2.890625,
      "learning_rate": 0.000496026506091932,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1556667357683182,
      "num_tokens": 15643253.0,
      "step": 7545
    },
    {
      "entropy": 5.796116685867309,
      "epoch": 0.6848693759071117,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004960204497767253,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1523918628692627,
      "num_tokens": 15653557.0,
      "step": 7550
    },
    {
      "entropy": 5.780343818664551,
      "epoch": 0.685322931785196,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004960143888907594,
      "loss": 5.5956,
      "mean_token_accuracy": 0.1578650578856468,
      "num_tokens": 15664034.0,
      "step": 7555
    },
    {
      "entropy": 5.793070936203003,
      "epoch": 0.6857764876632801,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004960083234341595,
      "loss": 5.653,
      "mean_token_accuracy": 0.15654636919498444,
      "num_tokens": 15674951.0,
      "step": 7560
    },
    {
      "entropy": 5.752449369430542,
      "epoch": 0.6862300435413643,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004960022534070513,
      "loss": 5.5547,
      "mean_token_accuracy": 0.16121672391891478,
      "num_tokens": 15684446.0,
      "step": 7565
    },
    {
      "entropy": 5.732784175872803,
      "epoch": 0.6866835994194485,
      "grad_norm": 2.203125,
      "learning_rate": 0.00049599617880956,
      "loss": 5.5322,
      "mean_token_accuracy": 0.1626044824719429,
      "num_tokens": 15694672.0,
      "step": 7570
    },
    {
      "entropy": 5.7013763904571535,
      "epoch": 0.6871371552975326,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004959900996418115,
      "loss": 5.6033,
      "mean_token_accuracy": 0.15530735701322557,
      "num_tokens": 15704637.0,
      "step": 7575
    },
    {
      "entropy": 5.715182638168335,
      "epoch": 0.6875907111756169,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004959840159039315,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1568412646651268,
      "num_tokens": 15713865.0,
      "step": 7580
    },
    {
      "entropy": 5.808747720718384,
      "epoch": 0.688044267053701,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004959779275960456,
      "loss": 5.6708,
      "mean_token_accuracy": 0.15349928438663482,
      "num_tokens": 15724258.0,
      "step": 7585
    },
    {
      "entropy": 5.770686054229737,
      "epoch": 0.6884978229317852,
      "grad_norm": 2.25,
      "learning_rate": 0.00049597183471828,
      "loss": 5.5887,
      "mean_token_accuracy": 0.15678472965955734,
      "num_tokens": 15735180.0,
      "step": 7590
    },
    {
      "entropy": 5.754590940475464,
      "epoch": 0.6889513788098693,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004959657372707604,
      "loss": 5.6069,
      "mean_token_accuracy": 0.1512817457318306,
      "num_tokens": 15745209.0,
      "step": 7595
    },
    {
      "entropy": 5.839451551437378,
      "epoch": 0.6894049346879536,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004959596352536132,
      "loss": 5.7081,
      "mean_token_accuracy": 0.1502077430486679,
      "num_tokens": 15756976.0,
      "step": 7600
    },
    {
      "entropy": 5.798298025131226,
      "epoch": 0.6898584905660378,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004959535286669643,
      "loss": 5.6943,
      "mean_token_accuracy": 0.1565860003232956,
      "num_tokens": 15767362.0,
      "step": 7605
    },
    {
      "entropy": 5.7720561027526855,
      "epoch": 0.6903120464441219,
      "grad_norm": 2.125,
      "learning_rate": 0.0004959474175109402,
      "loss": 5.669,
      "mean_token_accuracy": 0.1568397358059883,
      "num_tokens": 15778260.0,
      "step": 7610
    },
    {
      "entropy": 5.784616327285766,
      "epoch": 0.6907656023222061,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004959413017856672,
      "loss": 5.6426,
      "mean_token_accuracy": 0.15804268270730973,
      "num_tokens": 15789025.0,
      "step": 7615
    },
    {
      "entropy": 5.771534585952759,
      "epoch": 0.6912191582002902,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004959351814912716,
      "loss": 5.5859,
      "mean_token_accuracy": 0.1604921668767929,
      "num_tokens": 15798706.0,
      "step": 7620
    },
    {
      "entropy": 5.746160173416138,
      "epoch": 0.6916727140783745,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004959290566278803,
      "loss": 5.6062,
      "mean_token_accuracy": 0.16115242391824722,
      "num_tokens": 15808071.0,
      "step": 7625
    },
    {
      "entropy": 5.828521633148194,
      "epoch": 0.6921262699564587,
      "grad_norm": 3.125,
      "learning_rate": 0.0004959229271956199,
      "loss": 5.6257,
      "mean_token_accuracy": 0.15412043929100036,
      "num_tokens": 15819317.0,
      "step": 7630
    },
    {
      "entropy": 5.742588758468628,
      "epoch": 0.6925798258345428,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004959167931946169,
      "loss": 5.6581,
      "mean_token_accuracy": 0.15941284745931625,
      "num_tokens": 15830147.0,
      "step": 7635
    },
    {
      "entropy": 5.749380874633789,
      "epoch": 0.693033381712627,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004959106546249984,
      "loss": 5.5866,
      "mean_token_accuracy": 0.17110419124364853,
      "num_tokens": 15840107.0,
      "step": 7640
    },
    {
      "entropy": 5.819102001190186,
      "epoch": 0.6934869375907112,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004959045114868913,
      "loss": 5.676,
      "mean_token_accuracy": 0.15552862137556075,
      "num_tokens": 15849987.0,
      "step": 7645
    },
    {
      "entropy": 5.849531269073486,
      "epoch": 0.6939404934687954,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004958983637804225,
      "loss": 5.6429,
      "mean_token_accuracy": 0.15638294517993928,
      "num_tokens": 15859837.0,
      "step": 7650
    },
    {
      "entropy": 5.783810472488403,
      "epoch": 0.6943940493468795,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004958922115057194,
      "loss": 5.7022,
      "mean_token_accuracy": 0.1505706086754799,
      "num_tokens": 15871504.0,
      "step": 7655
    },
    {
      "entropy": 5.751116704940796,
      "epoch": 0.6948476052249637,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004958860546629089,
      "loss": 5.5217,
      "mean_token_accuracy": 0.16299662739038467,
      "num_tokens": 15881745.0,
      "step": 7660
    },
    {
      "entropy": 5.777749395370483,
      "epoch": 0.6953011611030478,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004958798932521186,
      "loss": 5.6762,
      "mean_token_accuracy": 0.15549626499414443,
      "num_tokens": 15892320.0,
      "step": 7665
    },
    {
      "entropy": 5.7372252464294435,
      "epoch": 0.6957547169811321,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004958737272734759,
      "loss": 5.6126,
      "mean_token_accuracy": 0.15911678075790406,
      "num_tokens": 15902754.0,
      "step": 7670
    },
    {
      "entropy": 5.759205341339111,
      "epoch": 0.6962082728592163,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004958675567271081,
      "loss": 5.6246,
      "mean_token_accuracy": 0.16049185246229172,
      "num_tokens": 15913247.0,
      "step": 7675
    },
    {
      "entropy": 5.815482807159424,
      "epoch": 0.6966618287373004,
      "grad_norm": 3.046875,
      "learning_rate": 0.000495861381613143,
      "loss": 5.6395,
      "mean_token_accuracy": 0.15289407670497895,
      "num_tokens": 15922828.0,
      "step": 7680
    },
    {
      "entropy": 5.720240736007691,
      "epoch": 0.6971153846153846,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004958552019317082,
      "loss": 5.5945,
      "mean_token_accuracy": 0.15427517890930176,
      "num_tokens": 15932921.0,
      "step": 7685
    },
    {
      "entropy": 5.761741065979004,
      "epoch": 0.6975689404934688,
      "grad_norm": 3.375,
      "learning_rate": 0.0004958490176829316,
      "loss": 5.5826,
      "mean_token_accuracy": 0.1560848221182823,
      "num_tokens": 15942457.0,
      "step": 7690
    },
    {
      "entropy": 5.747310972213745,
      "epoch": 0.698022496371553,
      "grad_norm": 6.875,
      "learning_rate": 0.000495842828866941,
      "loss": 5.6064,
      "mean_token_accuracy": 0.15387596637010575,
      "num_tokens": 15952275.0,
      "step": 7695
    },
    {
      "entropy": 5.649684906005859,
      "epoch": 0.6984760522496372,
      "grad_norm": 4.5,
      "learning_rate": 0.0004958366354838645,
      "loss": 5.4991,
      "mean_token_accuracy": 0.16529976576566696,
      "num_tokens": 15962110.0,
      "step": 7700
    },
    {
      "entropy": 5.704395055770874,
      "epoch": 0.6989296081277213,
      "grad_norm": 2.515625,
      "learning_rate": 0.00049583043753383,
      "loss": 5.5698,
      "mean_token_accuracy": 0.15852983072400093,
      "num_tokens": 15972634.0,
      "step": 7705
    },
    {
      "entropy": 5.791731977462769,
      "epoch": 0.6993831640058055,
      "grad_norm": 3.25,
      "learning_rate": 0.0004958242350169658,
      "loss": 5.6513,
      "mean_token_accuracy": 0.15203508138656616,
      "num_tokens": 15983132.0,
      "step": 7710
    },
    {
      "entropy": 5.738736534118653,
      "epoch": 0.6998367198838897,
      "grad_norm": 2.75,
      "learning_rate": 0.0004958180279334001,
      "loss": 5.5749,
      "mean_token_accuracy": 0.154992463439703,
      "num_tokens": 15994149.0,
      "step": 7715
    },
    {
      "entropy": 5.822169208526612,
      "epoch": 0.7002902757619739,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004958118162832614,
      "loss": 5.6743,
      "mean_token_accuracy": 0.15574501156806947,
      "num_tokens": 16005104.0,
      "step": 7720
    },
    {
      "entropy": 5.780132913589478,
      "epoch": 0.700743831640058,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004958056000666781,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15719846412539482,
      "num_tokens": 16015629.0,
      "step": 7725
    },
    {
      "entropy": 5.852429294586182,
      "epoch": 0.7011973875181422,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004957993792837786,
      "loss": 5.7046,
      "mean_token_accuracy": 0.1519222602248192,
      "num_tokens": 16025960.0,
      "step": 7730
    },
    {
      "entropy": 5.691937446594238,
      "epoch": 0.7016509433962265,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004957931539346917,
      "loss": 5.571,
      "mean_token_accuracy": 0.16283216029405595,
      "num_tokens": 16036083.0,
      "step": 7735
    },
    {
      "entropy": 5.707127094268799,
      "epoch": 0.7021044992743106,
      "grad_norm": 2.25,
      "learning_rate": 0.0004957869240195462,
      "loss": 5.5767,
      "mean_token_accuracy": 0.16615657061338424,
      "num_tokens": 16046448.0,
      "step": 7740
    },
    {
      "entropy": 5.784661531448364,
      "epoch": 0.7025580551523948,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004957806895384707,
      "loss": 5.6397,
      "mean_token_accuracy": 0.15175288319587707,
      "num_tokens": 16056807.0,
      "step": 7745
    },
    {
      "entropy": 5.795204210281372,
      "epoch": 0.7030116110304789,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004957744504915944,
      "loss": 5.6161,
      "mean_token_accuracy": 0.15674785524606705,
      "num_tokens": 16067129.0,
      "step": 7750
    },
    {
      "entropy": 5.7949577331542965,
      "epoch": 0.7034651669085631,
      "grad_norm": 2.375,
      "learning_rate": 0.0004957682068790463,
      "loss": 5.6636,
      "mean_token_accuracy": 0.15278994292020798,
      "num_tokens": 16078565.0,
      "step": 7755
    },
    {
      "entropy": 5.699900054931641,
      "epoch": 0.7039187227866474,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004957619587009553,
      "loss": 5.5113,
      "mean_token_accuracy": 0.15987293422222137,
      "num_tokens": 16089068.0,
      "step": 7760
    },
    {
      "entropy": 5.74468960762024,
      "epoch": 0.7043722786647315,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004957557059574507,
      "loss": 5.7092,
      "mean_token_accuracy": 0.14614630565047265,
      "num_tokens": 16100008.0,
      "step": 7765
    },
    {
      "entropy": 5.792629146575928,
      "epoch": 0.7048258345428157,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004957494486486618,
      "loss": 5.6674,
      "mean_token_accuracy": 0.14775043427944184,
      "num_tokens": 16111444.0,
      "step": 7770
    },
    {
      "entropy": 5.866617870330811,
      "epoch": 0.7052793904208998,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004957431867747182,
      "loss": 5.641,
      "mean_token_accuracy": 0.1542788103222847,
      "num_tokens": 16122020.0,
      "step": 7775
    },
    {
      "entropy": 5.778904294967651,
      "epoch": 0.7057329462989841,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004957369203357492,
      "loss": 5.5682,
      "mean_token_accuracy": 0.16568806916475295,
      "num_tokens": 16131281.0,
      "step": 7780
    },
    {
      "entropy": 5.7513092994689945,
      "epoch": 0.7061865021770682,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004957306493318844,
      "loss": 5.631,
      "mean_token_accuracy": 0.15769600570201875,
      "num_tokens": 16142087.0,
      "step": 7785
    },
    {
      "entropy": 5.760253143310547,
      "epoch": 0.7066400580551524,
      "grad_norm": 4.125,
      "learning_rate": 0.0004957243737632535,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16361806243658067,
      "num_tokens": 16152120.0,
      "step": 7790
    },
    {
      "entropy": 5.787371969223022,
      "epoch": 0.7070936139332366,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004957180936299863,
      "loss": 5.584,
      "mean_token_accuracy": 0.1611701652407646,
      "num_tokens": 16163432.0,
      "step": 7795
    },
    {
      "entropy": 5.750684547424316,
      "epoch": 0.7075471698113207,
      "grad_norm": 2.75,
      "learning_rate": 0.0004957118089322127,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1521909862756729,
      "num_tokens": 16174070.0,
      "step": 7800
    },
    {
      "entropy": 5.7357985496521,
      "epoch": 0.708000725689405,
      "grad_norm": 2.875,
      "learning_rate": 0.0004957055196700628,
      "loss": 5.5134,
      "mean_token_accuracy": 0.1614537939429283,
      "num_tokens": 16183727.0,
      "step": 7805
    },
    {
      "entropy": 5.744387292861939,
      "epoch": 0.7084542815674891,
      "grad_norm": 4.0,
      "learning_rate": 0.0004956992258436663,
      "loss": 5.5607,
      "mean_token_accuracy": 0.16315926909446715,
      "num_tokens": 16194056.0,
      "step": 7810
    },
    {
      "entropy": 5.729212713241577,
      "epoch": 0.7089078374455733,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004956929274531536,
      "loss": 5.5252,
      "mean_token_accuracy": 0.15964651107788086,
      "num_tokens": 16204039.0,
      "step": 7815
    },
    {
      "entropy": 5.736894512176514,
      "epoch": 0.7093613933236574,
      "grad_norm": 2.515625,
      "learning_rate": 0.000495686624498655,
      "loss": 5.5608,
      "mean_token_accuracy": 0.15911562889814376,
      "num_tokens": 16214329.0,
      "step": 7820
    },
    {
      "entropy": 5.819330215454102,
      "epoch": 0.7098149492017417,
      "grad_norm": 2.75,
      "learning_rate": 0.0004956803169803007,
      "loss": 5.729,
      "mean_token_accuracy": 0.14540325105190277,
      "num_tokens": 16226396.0,
      "step": 7825
    },
    {
      "entropy": 5.778207397460937,
      "epoch": 0.7102685050798259,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004956740048982212,
      "loss": 5.6165,
      "mean_token_accuracy": 0.16062480062246323,
      "num_tokens": 16236753.0,
      "step": 7830
    },
    {
      "entropy": 5.836630821228027,
      "epoch": 0.71072206095791,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004956676882525472,
      "loss": 5.6971,
      "mean_token_accuracy": 0.14720023572444915,
      "num_tokens": 16246270.0,
      "step": 7835
    },
    {
      "entropy": 5.851115322113037,
      "epoch": 0.7111756168359942,
      "grad_norm": 2.21875,
      "learning_rate": 0.000495661367043409,
      "loss": 5.6441,
      "mean_token_accuracy": 0.16027983278036118,
      "num_tokens": 16256071.0,
      "step": 7840
    },
    {
      "entropy": 5.766817235946656,
      "epoch": 0.7116291727140783,
      "grad_norm": 2.25,
      "learning_rate": 0.0004956550412709376,
      "loss": 5.6256,
      "mean_token_accuracy": 0.15271955505013465,
      "num_tokens": 16266995.0,
      "step": 7845
    },
    {
      "entropy": 5.844926357269287,
      "epoch": 0.7120827285921626,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004956487109352637,
      "loss": 5.7578,
      "mean_token_accuracy": 0.15099674612283706,
      "num_tokens": 16279787.0,
      "step": 7850
    },
    {
      "entropy": 5.8208479404449465,
      "epoch": 0.7125362844702468,
      "grad_norm": 3.0,
      "learning_rate": 0.0004956423760365181,
      "loss": 5.6524,
      "mean_token_accuracy": 0.1538202553987503,
      "num_tokens": 16290826.0,
      "step": 7855
    },
    {
      "entropy": 5.783289957046509,
      "epoch": 0.7129898403483309,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004956360365748321,
      "loss": 5.5786,
      "mean_token_accuracy": 0.16150108724832535,
      "num_tokens": 16300599.0,
      "step": 7860
    },
    {
      "entropy": 5.755193996429443,
      "epoch": 0.7134433962264151,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004956296925503367,
      "loss": 5.6058,
      "mean_token_accuracy": 0.15755392462015153,
      "num_tokens": 16310240.0,
      "step": 7865
    },
    {
      "entropy": 5.810258483886718,
      "epoch": 0.7138969521044993,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004956233439631628,
      "loss": 5.598,
      "mean_token_accuracy": 0.1576920837163925,
      "num_tokens": 16320580.0,
      "step": 7870
    },
    {
      "entropy": 5.671935606002807,
      "epoch": 0.7143505079825835,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004956169908134421,
      "loss": 5.6137,
      "mean_token_accuracy": 0.15596351623535157,
      "num_tokens": 16330246.0,
      "step": 7875
    },
    {
      "entropy": 5.774977493286133,
      "epoch": 0.7148040638606676,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004956106331013057,
      "loss": 5.6166,
      "mean_token_accuracy": 0.16086281836032867,
      "num_tokens": 16339769.0,
      "step": 7880
    },
    {
      "entropy": 5.781713151931763,
      "epoch": 0.7152576197387518,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004956042708268853,
      "loss": 5.6062,
      "mean_token_accuracy": 0.15815801322460174,
      "num_tokens": 16349196.0,
      "step": 7885
    },
    {
      "entropy": 5.7052558898925785,
      "epoch": 0.715711175616836,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004955979039903123,
      "loss": 5.6141,
      "mean_token_accuracy": 0.15750451236963273,
      "num_tokens": 16360072.0,
      "step": 7890
    },
    {
      "entropy": 5.69798698425293,
      "epoch": 0.7161647314949202,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004955915325917183,
      "loss": 5.5756,
      "mean_token_accuracy": 0.15827955752611161,
      "num_tokens": 16370469.0,
      "step": 7895
    },
    {
      "entropy": 5.796889019012451,
      "epoch": 0.7166182873730044,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004955851566312354,
      "loss": 5.6185,
      "mean_token_accuracy": 0.1537104293704033,
      "num_tokens": 16381252.0,
      "step": 7900
    },
    {
      "entropy": 5.848946905136108,
      "epoch": 0.7170718432510885,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004955787761089952,
      "loss": 5.5936,
      "mean_token_accuracy": 0.15734365135431289,
      "num_tokens": 16391785.0,
      "step": 7905
    },
    {
      "entropy": 5.804285717010498,
      "epoch": 0.7175253991291727,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004955723910251296,
      "loss": 5.5986,
      "mean_token_accuracy": 0.15833539739251137,
      "num_tokens": 16401937.0,
      "step": 7910
    },
    {
      "entropy": 5.739145565032959,
      "epoch": 0.717978955007257,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004955660013797709,
      "loss": 5.5496,
      "mean_token_accuracy": 0.1670544058084488,
      "num_tokens": 16411816.0,
      "step": 7915
    },
    {
      "entropy": 5.644951677322387,
      "epoch": 0.7184325108853411,
      "grad_norm": 3.140625,
      "learning_rate": 0.000495559607173051,
      "loss": 5.6196,
      "mean_token_accuracy": 0.1513867750763893,
      "num_tokens": 16423010.0,
      "step": 7920
    },
    {
      "entropy": 5.748967218399048,
      "epoch": 0.7188860667634253,
      "grad_norm": 2.25,
      "learning_rate": 0.0004955532084051022,
      "loss": 5.5803,
      "mean_token_accuracy": 0.15736083835363388,
      "num_tokens": 16433345.0,
      "step": 7925
    },
    {
      "entropy": 5.803564500808716,
      "epoch": 0.7193396226415094,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004955468050760568,
      "loss": 5.5904,
      "mean_token_accuracy": 0.15658929497003554,
      "num_tokens": 16444508.0,
      "step": 7930
    },
    {
      "entropy": 5.7014247417449955,
      "epoch": 0.7197931785195936,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004955403971860474,
      "loss": 5.5047,
      "mean_token_accuracy": 0.16188954263925553,
      "num_tokens": 16454248.0,
      "step": 7935
    },
    {
      "entropy": 5.683907747268677,
      "epoch": 0.7202467343976778,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004955339847352063,
      "loss": 5.5327,
      "mean_token_accuracy": 0.1645732879638672,
      "num_tokens": 16463762.0,
      "step": 7940
    },
    {
      "entropy": 5.744384622573852,
      "epoch": 0.720700290275762,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004955275677236662,
      "loss": 5.6089,
      "mean_token_accuracy": 0.1514907330274582,
      "num_tokens": 16474059.0,
      "step": 7945
    },
    {
      "entropy": 5.758716773986817,
      "epoch": 0.7211538461538461,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004955211461515599,
      "loss": 5.5488,
      "mean_token_accuracy": 0.17130780220031738,
      "num_tokens": 16483352.0,
      "step": 7950
    },
    {
      "entropy": 5.820909595489502,
      "epoch": 0.7216074020319303,
      "grad_norm": 2.90625,
      "learning_rate": 0.00049551472001902,
      "loss": 5.678,
      "mean_token_accuracy": 0.15455454140901564,
      "num_tokens": 16494714.0,
      "step": 7955
    },
    {
      "entropy": 5.76343355178833,
      "epoch": 0.7220609579100146,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004955082893261796,
      "loss": 5.5869,
      "mean_token_accuracy": 0.15796139240264892,
      "num_tokens": 16505610.0,
      "step": 7960
    },
    {
      "entropy": 5.7450456619262695,
      "epoch": 0.7225145137880987,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004955018540731714,
      "loss": 5.5919,
      "mean_token_accuracy": 0.15026113539934158,
      "num_tokens": 16515515.0,
      "step": 7965
    },
    {
      "entropy": 5.799071168899536,
      "epoch": 0.7229680696661829,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004954954142601288,
      "loss": 5.6342,
      "mean_token_accuracy": 0.14915079474449158,
      "num_tokens": 16525359.0,
      "step": 7970
    },
    {
      "entropy": 5.8566121578216555,
      "epoch": 0.723421625544267,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004954889698871849,
      "loss": 5.6136,
      "mean_token_accuracy": 0.15901191681623458,
      "num_tokens": 16535113.0,
      "step": 7975
    },
    {
      "entropy": 5.744398498535157,
      "epoch": 0.7238751814223512,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004954825209544728,
      "loss": 5.5453,
      "mean_token_accuracy": 0.16029145419597626,
      "num_tokens": 16545638.0,
      "step": 7980
    },
    {
      "entropy": 5.814827013015747,
      "epoch": 0.7243287373004355,
      "grad_norm": 4.34375,
      "learning_rate": 0.000495476067462126,
      "loss": 5.6753,
      "mean_token_accuracy": 0.1573385030031204,
      "num_tokens": 16555807.0,
      "step": 7985
    },
    {
      "entropy": 5.8214493751525875,
      "epoch": 0.7247822931785196,
      "grad_norm": 2.59375,
      "learning_rate": 0.000495469609410278,
      "loss": 5.6385,
      "mean_token_accuracy": 0.15442904829978943,
      "num_tokens": 16566143.0,
      "step": 7990
    },
    {
      "entropy": 5.75347580909729,
      "epoch": 0.7252358490566038,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004954631467990622,
      "loss": 5.6463,
      "mean_token_accuracy": 0.15699395090341567,
      "num_tokens": 16576192.0,
      "step": 7995
    },
    {
      "entropy": 5.796687459945678,
      "epoch": 0.7256894049346879,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004954566796286124,
      "loss": 5.683,
      "mean_token_accuracy": 0.14983606934547425,
      "num_tokens": 16587841.0,
      "step": 8000
    },
    {
      "entropy": 5.856884336471557,
      "epoch": 0.7261429608127722,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004954502078990623,
      "loss": 5.6618,
      "mean_token_accuracy": 0.1537812352180481,
      "num_tokens": 16598700.0,
      "step": 8005
    },
    {
      "entropy": 5.831228494644165,
      "epoch": 0.7265965166908563,
      "grad_norm": 2.25,
      "learning_rate": 0.0004954437316105457,
      "loss": 5.6951,
      "mean_token_accuracy": 0.14877633303403853,
      "num_tokens": 16609916.0,
      "step": 8010
    },
    {
      "entropy": 5.7455119609832765,
      "epoch": 0.7270500725689405,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004954372507631965,
      "loss": 5.5545,
      "mean_token_accuracy": 0.15734955966472625,
      "num_tokens": 16620508.0,
      "step": 8015
    },
    {
      "entropy": 5.800483751296997,
      "epoch": 0.7275036284470247,
      "grad_norm": 3.25,
      "learning_rate": 0.0004954307653571489,
      "loss": 5.6854,
      "mean_token_accuracy": 0.15541286170482635,
      "num_tokens": 16632128.0,
      "step": 8020
    },
    {
      "entropy": 5.829589605331421,
      "epoch": 0.7279571843251088,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004954242753925367,
      "loss": 5.677,
      "mean_token_accuracy": 0.1550593614578247,
      "num_tokens": 16643447.0,
      "step": 8025
    },
    {
      "entropy": 5.80330080986023,
      "epoch": 0.7284107402031931,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004954177808694946,
      "loss": 5.6144,
      "mean_token_accuracy": 0.16055452525615693,
      "num_tokens": 16653237.0,
      "step": 8030
    },
    {
      "entropy": 5.792874145507812,
      "epoch": 0.7288642960812772,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004954112817881565,
      "loss": 5.713,
      "mean_token_accuracy": 0.15150015503168107,
      "num_tokens": 16663532.0,
      "step": 8035
    },
    {
      "entropy": 5.802073526382446,
      "epoch": 0.7293178519593614,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004954047781486568,
      "loss": 5.5477,
      "mean_token_accuracy": 0.16309236139059066,
      "num_tokens": 16673236.0,
      "step": 8040
    },
    {
      "entropy": 5.775296831130982,
      "epoch": 0.7297714078374455,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004953982699511303,
      "loss": 5.6399,
      "mean_token_accuracy": 0.15158257484436036,
      "num_tokens": 16684731.0,
      "step": 8045
    },
    {
      "entropy": 5.714089298248291,
      "epoch": 0.7302249637155298,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004953917571957114,
      "loss": 5.5682,
      "mean_token_accuracy": 0.15534164160490035,
      "num_tokens": 16694498.0,
      "step": 8050
    },
    {
      "entropy": 5.83448486328125,
      "epoch": 0.730678519593614,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004953852398825346,
      "loss": 5.6853,
      "mean_token_accuracy": 0.15397260934114457,
      "num_tokens": 16705514.0,
      "step": 8055
    },
    {
      "entropy": 5.70955262184143,
      "epoch": 0.7311320754716981,
      "grad_norm": 2.421875,
      "learning_rate": 0.000495378718011735,
      "loss": 5.5292,
      "mean_token_accuracy": 0.15810254365205764,
      "num_tokens": 16714926.0,
      "step": 8060
    },
    {
      "entropy": 5.682147645950318,
      "epoch": 0.7315856313497823,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004953721915834473,
      "loss": 5.5382,
      "mean_token_accuracy": 0.15791282206773757,
      "num_tokens": 16725894.0,
      "step": 8065
    },
    {
      "entropy": 5.730485343933106,
      "epoch": 0.7320391872278664,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004953656605978067,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1612203910946846,
      "num_tokens": 16735551.0,
      "step": 8070
    },
    {
      "entropy": 5.785016202926636,
      "epoch": 0.7324927431059507,
      "grad_norm": 2.546875,
      "learning_rate": 0.000495359125054948,
      "loss": 5.5677,
      "mean_token_accuracy": 0.15639185756444932,
      "num_tokens": 16746840.0,
      "step": 8075
    },
    {
      "entropy": 5.784290552139282,
      "epoch": 0.7329462989840348,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004953525849550063,
      "loss": 5.569,
      "mean_token_accuracy": 0.1611862748861313,
      "num_tokens": 16756653.0,
      "step": 8080
    },
    {
      "entropy": 5.719831895828247,
      "epoch": 0.733399854862119,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004953460402981172,
      "loss": 5.706,
      "mean_token_accuracy": 0.15862376540899276,
      "num_tokens": 16767767.0,
      "step": 8085
    },
    {
      "entropy": 5.76441707611084,
      "epoch": 0.7338534107402032,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004953394910844157,
      "loss": 5.6634,
      "mean_token_accuracy": 0.1488182619214058,
      "num_tokens": 16778345.0,
      "step": 8090
    },
    {
      "entropy": 5.759904861450195,
      "epoch": 0.7343069666182874,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004953329373140374,
      "loss": 5.5696,
      "mean_token_accuracy": 0.15454471111297607,
      "num_tokens": 16788739.0,
      "step": 8095
    },
    {
      "entropy": 5.811275768280029,
      "epoch": 0.7347605224963716,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004953263789871179,
      "loss": 5.6876,
      "mean_token_accuracy": 0.15244855135679244,
      "num_tokens": 16799439.0,
      "step": 8100
    },
    {
      "entropy": 5.757402181625366,
      "epoch": 0.7352140783744557,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004953198161037925,
      "loss": 5.55,
      "mean_token_accuracy": 0.16849865913391113,
      "num_tokens": 16808756.0,
      "step": 8105
    },
    {
      "entropy": 5.7503485679626465,
      "epoch": 0.7356676342525399,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004953132486641975,
      "loss": 5.6065,
      "mean_token_accuracy": 0.15480702742934227,
      "num_tokens": 16818973.0,
      "step": 8110
    },
    {
      "entropy": 5.709172677993775,
      "epoch": 0.736121190130624,
      "grad_norm": 1.65625,
      "learning_rate": 0.000495306676668468,
      "loss": 5.509,
      "mean_token_accuracy": 0.15501116514205932,
      "num_tokens": 16828635.0,
      "step": 8115
    },
    {
      "entropy": 5.714795160293579,
      "epoch": 0.7365747460087083,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004953001001167406,
      "loss": 5.6449,
      "mean_token_accuracy": 0.1607098326086998,
      "num_tokens": 16838762.0,
      "step": 8120
    },
    {
      "entropy": 5.690447664260864,
      "epoch": 0.7370283018867925,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004952935190091508,
      "loss": 5.5113,
      "mean_token_accuracy": 0.15998412072658538,
      "num_tokens": 16849623.0,
      "step": 8125
    },
    {
      "entropy": 5.686318635940552,
      "epoch": 0.7374818577648766,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004952869333458349,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1523797407746315,
      "num_tokens": 16860214.0,
      "step": 8130
    },
    {
      "entropy": 5.70399751663208,
      "epoch": 0.7379354136429608,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004952803431269292,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1519754208624363,
      "num_tokens": 16871723.0,
      "step": 8135
    },
    {
      "entropy": 5.808839225769043,
      "epoch": 0.738388969521045,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004952737483525699,
      "loss": 5.702,
      "mean_token_accuracy": 0.1518370732665062,
      "num_tokens": 16882500.0,
      "step": 8140
    },
    {
      "entropy": 5.797318315505981,
      "epoch": 0.7388425253991292,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004952671490228932,
      "loss": 5.6748,
      "mean_token_accuracy": 0.15521494448184966,
      "num_tokens": 16893386.0,
      "step": 8145
    },
    {
      "entropy": 5.798702096939087,
      "epoch": 0.7392960812772134,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004952605451380357,
      "loss": 5.704,
      "mean_token_accuracy": 0.1514693468809128,
      "num_tokens": 16904770.0,
      "step": 8150
    },
    {
      "entropy": 5.919784784317017,
      "epoch": 0.7397496371552975,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004952539366981342,
      "loss": 5.7549,
      "mean_token_accuracy": 0.15494561344385147,
      "num_tokens": 16915858.0,
      "step": 8155
    },
    {
      "entropy": 5.744434642791748,
      "epoch": 0.7402031930333817,
      "grad_norm": 2.125,
      "learning_rate": 0.000495247323703325,
      "loss": 5.6026,
      "mean_token_accuracy": 0.15603308826684953,
      "num_tokens": 16926731.0,
      "step": 8160
    },
    {
      "entropy": 5.759461069107056,
      "epoch": 0.7406567489114659,
      "grad_norm": 3.765625,
      "learning_rate": 0.000495240706153745,
      "loss": 5.6387,
      "mean_token_accuracy": 0.15482465475797652,
      "num_tokens": 16938108.0,
      "step": 8165
    },
    {
      "entropy": 5.690951871871948,
      "epoch": 0.7411103047895501,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004952340840495312,
      "loss": 5.485,
      "mean_token_accuracy": 0.1668120801448822,
      "num_tokens": 16948516.0,
      "step": 8170
    },
    {
      "entropy": 5.74252872467041,
      "epoch": 0.7415638606676342,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004952274573908203,
      "loss": 5.5839,
      "mean_token_accuracy": 0.15991057008504866,
      "num_tokens": 16958642.0,
      "step": 8175
    },
    {
      "entropy": 5.717390298843384,
      "epoch": 0.7420174165457184,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004952208261777496,
      "loss": 5.5796,
      "mean_token_accuracy": 0.15802238434553145,
      "num_tokens": 16968429.0,
      "step": 8180
    },
    {
      "entropy": 5.745699214935303,
      "epoch": 0.7424709724238027,
      "grad_norm": 2.234375,
      "learning_rate": 0.000495214190410456,
      "loss": 5.6007,
      "mean_token_accuracy": 0.16200145781040193,
      "num_tokens": 16978846.0,
      "step": 8185
    },
    {
      "entropy": 5.645253658294678,
      "epoch": 0.7429245283018868,
      "grad_norm": 5.25,
      "learning_rate": 0.0004952075500890768,
      "loss": 5.512,
      "mean_token_accuracy": 0.1618196815252304,
      "num_tokens": 16988353.0,
      "step": 8190
    },
    {
      "entropy": 5.665030670166016,
      "epoch": 0.743378084179971,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004952009052137493,
      "loss": 5.5325,
      "mean_token_accuracy": 0.1606283813714981,
      "num_tokens": 16998774.0,
      "step": 8195
    },
    {
      "entropy": 5.729429912567139,
      "epoch": 0.7438316400580551,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004951942557846112,
      "loss": 5.6005,
      "mean_token_accuracy": 0.15656651556491852,
      "num_tokens": 17009286.0,
      "step": 8200
    },
    {
      "entropy": 5.716187524795532,
      "epoch": 0.7442851959361393,
      "grad_norm": 2.0,
      "learning_rate": 0.0004951876018017997,
      "loss": 5.5942,
      "mean_token_accuracy": 0.15400404408574103,
      "num_tokens": 17020067.0,
      "step": 8205
    },
    {
      "entropy": 5.793141412734985,
      "epoch": 0.7447387518142236,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004951809432654524,
      "loss": 5.6861,
      "mean_token_accuracy": 0.1583374872803688,
      "num_tokens": 17031512.0,
      "step": 8210
    },
    {
      "entropy": 5.776523923873901,
      "epoch": 0.7451923076923077,
      "grad_norm": 2.40625,
      "learning_rate": 0.000495174280175707,
      "loss": 5.57,
      "mean_token_accuracy": 0.15665111243724822,
      "num_tokens": 17042928.0,
      "step": 8215
    },
    {
      "entropy": 5.641525316238403,
      "epoch": 0.7456458635703919,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004951676125327015,
      "loss": 5.6478,
      "mean_token_accuracy": 0.14974198192358018,
      "num_tokens": 17053327.0,
      "step": 8220
    },
    {
      "entropy": 5.80580325126648,
      "epoch": 0.746099419448476,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004951609403365737,
      "loss": 5.6529,
      "mean_token_accuracy": 0.15747711509466172,
      "num_tokens": 17063100.0,
      "step": 8225
    },
    {
      "entropy": 5.8813153266906735,
      "epoch": 0.7465529753265603,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004951542635874615,
      "loss": 5.6347,
      "mean_token_accuracy": 0.15747971534729005,
      "num_tokens": 17072715.0,
      "step": 8230
    },
    {
      "entropy": 5.8548980236053465,
      "epoch": 0.7470065312046444,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004951475822855029,
      "loss": 5.7601,
      "mean_token_accuracy": 0.15097569525241852,
      "num_tokens": 17082889.0,
      "step": 8235
    },
    {
      "entropy": 5.713144540786743,
      "epoch": 0.7474600870827286,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004951408964308364,
      "loss": 5.5784,
      "mean_token_accuracy": 0.15977087765932083,
      "num_tokens": 17093143.0,
      "step": 8240
    },
    {
      "entropy": 5.691396188735962,
      "epoch": 0.7479136429608128,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004951342060236,
      "loss": 5.5846,
      "mean_token_accuracy": 0.15049274563789367,
      "num_tokens": 17103818.0,
      "step": 8245
    },
    {
      "entropy": 5.7963649272918705,
      "epoch": 0.7483671988388969,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004951275110639321,
      "loss": 5.6328,
      "mean_token_accuracy": 0.15465376079082488,
      "num_tokens": 17114299.0,
      "step": 8250
    },
    {
      "entropy": 5.75806770324707,
      "epoch": 0.7488207547169812,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004951208115519712,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1583488866686821,
      "num_tokens": 17124309.0,
      "step": 8255
    },
    {
      "entropy": 5.780774116516113,
      "epoch": 0.7492743105950653,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004951141074878558,
      "loss": 5.6307,
      "mean_token_accuracy": 0.1523465171456337,
      "num_tokens": 17135493.0,
      "step": 8260
    },
    {
      "entropy": 5.785015106201172,
      "epoch": 0.7497278664731495,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004951073988717246,
      "loss": 5.6135,
      "mean_token_accuracy": 0.15537115484476088,
      "num_tokens": 17145054.0,
      "step": 8265
    },
    {
      "entropy": 5.764077663421631,
      "epoch": 0.7501814223512336,
      "grad_norm": 2.25,
      "learning_rate": 0.0004951006857037163,
      "loss": 5.6486,
      "mean_token_accuracy": 0.15082504153251647,
      "num_tokens": 17155401.0,
      "step": 8270
    },
    {
      "entropy": 5.824543714523315,
      "epoch": 0.7506349782293179,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004950939679839696,
      "loss": 5.6661,
      "mean_token_accuracy": 0.15418027192354203,
      "num_tokens": 17165989.0,
      "step": 8275
    },
    {
      "entropy": 5.841431903839111,
      "epoch": 0.7510885341074021,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004950872457126237,
      "loss": 5.6754,
      "mean_token_accuracy": 0.15596235692501068,
      "num_tokens": 17176072.0,
      "step": 8280
    },
    {
      "entropy": 5.784214973449707,
      "epoch": 0.7515420899854862,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004950805188898175,
      "loss": 5.6662,
      "mean_token_accuracy": 0.15422232523560525,
      "num_tokens": 17186261.0,
      "step": 8285
    },
    {
      "entropy": 5.748292112350464,
      "epoch": 0.7519956458635704,
      "grad_norm": 2.375,
      "learning_rate": 0.00049507378751569,
      "loss": 5.6683,
      "mean_token_accuracy": 0.15558717995882035,
      "num_tokens": 17197738.0,
      "step": 8290
    },
    {
      "entropy": 5.874657726287841,
      "epoch": 0.7524492017416545,
      "grad_norm": 2.375,
      "learning_rate": 0.0004950670515903805,
      "loss": 5.6666,
      "mean_token_accuracy": 0.15427183508872985,
      "num_tokens": 17208038.0,
      "step": 8295
    },
    {
      "entropy": 5.794052314758301,
      "epoch": 0.7529027576197388,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004950603111140283,
      "loss": 5.6122,
      "mean_token_accuracy": 0.16207809001207352,
      "num_tokens": 17218145.0,
      "step": 8300
    },
    {
      "entropy": 5.809648418426514,
      "epoch": 0.753356313497823,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004950535660867728,
      "loss": 5.7441,
      "mean_token_accuracy": 0.14609585851430892,
      "num_tokens": 17227281.0,
      "step": 8305
    },
    {
      "entropy": 5.744253730773925,
      "epoch": 0.7538098693759071,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004950468165087535,
      "loss": 5.501,
      "mean_token_accuracy": 0.1575057849287987,
      "num_tokens": 17237571.0,
      "step": 8310
    },
    {
      "entropy": 5.737435293197632,
      "epoch": 0.7542634252539913,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004950400623801098,
      "loss": 5.6568,
      "mean_token_accuracy": 0.1578634724020958,
      "num_tokens": 17247088.0,
      "step": 8315
    },
    {
      "entropy": 5.797852087020874,
      "epoch": 0.7547169811320755,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004950333037009818,
      "loss": 5.5828,
      "mean_token_accuracy": 0.1587523862719536,
      "num_tokens": 17257638.0,
      "step": 8320
    },
    {
      "entropy": 5.803178405761718,
      "epoch": 0.7551705370101597,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004950265404715089,
      "loss": 5.596,
      "mean_token_accuracy": 0.1560966521501541,
      "num_tokens": 17268421.0,
      "step": 8325
    },
    {
      "entropy": 5.787026023864746,
      "epoch": 0.7556240928882438,
      "grad_norm": 1.78125,
      "learning_rate": 0.000495019772691831,
      "loss": 5.6711,
      "mean_token_accuracy": 0.14926135241985322,
      "num_tokens": 17279264.0,
      "step": 8330
    },
    {
      "entropy": 5.759759044647216,
      "epoch": 0.756077648766328,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004950130003620882,
      "loss": 5.6537,
      "mean_token_accuracy": 0.15707915723323823,
      "num_tokens": 17290913.0,
      "step": 8335
    },
    {
      "entropy": 5.737857341766357,
      "epoch": 0.7565312046444121,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004950062234824204,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1518915608525276,
      "num_tokens": 17302366.0,
      "step": 8340
    },
    {
      "entropy": 5.773913192749023,
      "epoch": 0.7569847605224964,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004949994420529679,
      "loss": 5.6031,
      "mean_token_accuracy": 0.16011033803224564,
      "num_tokens": 17312855.0,
      "step": 8345
    },
    {
      "entropy": 5.766496276855468,
      "epoch": 0.7574383164005806,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004949926560738709,
      "loss": 5.6185,
      "mean_token_accuracy": 0.15467959642410278,
      "num_tokens": 17324644.0,
      "step": 8350
    },
    {
      "entropy": 5.727313995361328,
      "epoch": 0.7578918722786647,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004949858655452697,
      "loss": 5.6207,
      "mean_token_accuracy": 0.149393293261528,
      "num_tokens": 17334862.0,
      "step": 8355
    },
    {
      "entropy": 5.771681880950927,
      "epoch": 0.7583454281567489,
      "grad_norm": 2.375,
      "learning_rate": 0.0004949790704673048,
      "loss": 5.6407,
      "mean_token_accuracy": 0.15446821749210357,
      "num_tokens": 17345369.0,
      "step": 8360
    },
    {
      "entropy": 5.800436162948609,
      "epoch": 0.7587989840348331,
      "grad_norm": 2.125,
      "learning_rate": 0.0004949722708401166,
      "loss": 5.6299,
      "mean_token_accuracy": 0.15178182274103164,
      "num_tokens": 17356521.0,
      "step": 8365
    },
    {
      "entropy": 5.7372547626495365,
      "epoch": 0.7592525399129173,
      "grad_norm": 2.375,
      "learning_rate": 0.0004949654666638457,
      "loss": 5.5404,
      "mean_token_accuracy": 0.15822072923183442,
      "num_tokens": 17367919.0,
      "step": 8370
    },
    {
      "entropy": 5.7388671875,
      "epoch": 0.7597060957910015,
      "grad_norm": 2.0625,
      "learning_rate": 0.000494958657938633,
      "loss": 5.5983,
      "mean_token_accuracy": 0.15761556178331376,
      "num_tokens": 17378377.0,
      "step": 8375
    },
    {
      "entropy": 5.727004051208496,
      "epoch": 0.7601596516690856,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004949518446646192,
      "loss": 5.6018,
      "mean_token_accuracy": 0.1600989006459713,
      "num_tokens": 17388657.0,
      "step": 8380
    },
    {
      "entropy": 5.759643650054931,
      "epoch": 0.7606132075471698,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004949450268419451,
      "loss": 5.6012,
      "mean_token_accuracy": 0.1576221317052841,
      "num_tokens": 17399168.0,
      "step": 8385
    },
    {
      "entropy": 5.613120746612549,
      "epoch": 0.761066763425254,
      "grad_norm": 2.375,
      "learning_rate": 0.000494938204470752,
      "loss": 5.428,
      "mean_token_accuracy": 0.16362162977457045,
      "num_tokens": 17409426.0,
      "step": 8390
    },
    {
      "entropy": 5.706241512298584,
      "epoch": 0.7615203193033382,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004949313775511806,
      "loss": 5.5514,
      "mean_token_accuracy": 0.15745215564966203,
      "num_tokens": 17419850.0,
      "step": 8395
    },
    {
      "entropy": 5.735220050811767,
      "epoch": 0.7619738751814223,
      "grad_norm": 2.5,
      "learning_rate": 0.0004949245460833723,
      "loss": 5.643,
      "mean_token_accuracy": 0.15268349796533584,
      "num_tokens": 17431936.0,
      "step": 8400
    },
    {
      "entropy": 5.778987169265747,
      "epoch": 0.7624274310595065,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004949177100674684,
      "loss": 5.6947,
      "mean_token_accuracy": 0.1501627579331398,
      "num_tokens": 17442814.0,
      "step": 8405
    },
    {
      "entropy": 5.8641212463378904,
      "epoch": 0.7628809869375908,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004949108695036102,
      "loss": 5.6124,
      "mean_token_accuracy": 0.15486170947551728,
      "num_tokens": 17452723.0,
      "step": 8410
    },
    {
      "entropy": 5.77443037033081,
      "epoch": 0.7633345428156749,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004949040243919392,
      "loss": 5.629,
      "mean_token_accuracy": 0.15803532302379608,
      "num_tokens": 17463332.0,
      "step": 8415
    },
    {
      "entropy": 5.717780447006225,
      "epoch": 0.7637880986937591,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004948971747325968,
      "loss": 5.6607,
      "mean_token_accuracy": 0.1539319097995758,
      "num_tokens": 17474338.0,
      "step": 8420
    },
    {
      "entropy": 5.765063953399658,
      "epoch": 0.7642416545718432,
      "grad_norm": 2.640625,
      "learning_rate": 0.000494890320525725,
      "loss": 5.6424,
      "mean_token_accuracy": 0.15345966815948486,
      "num_tokens": 17485236.0,
      "step": 8425
    },
    {
      "entropy": 5.743877935409546,
      "epoch": 0.7646952104499274,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004948834617714652,
      "loss": 5.5655,
      "mean_token_accuracy": 0.15759633481502533,
      "num_tokens": 17495795.0,
      "step": 8430
    },
    {
      "entropy": 5.741341543197632,
      "epoch": 0.7651487663280117,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004948765984699595,
      "loss": 5.6819,
      "mean_token_accuracy": 0.14975285530090332,
      "num_tokens": 17506756.0,
      "step": 8435
    },
    {
      "entropy": 5.7913984775543215,
      "epoch": 0.7656023222060958,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004948697306213494,
      "loss": 5.5843,
      "mean_token_accuracy": 0.1652184545993805,
      "num_tokens": 17516198.0,
      "step": 8440
    },
    {
      "entropy": 5.8136650085449215,
      "epoch": 0.76605587808418,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004948628582257775,
      "loss": 5.5901,
      "mean_token_accuracy": 0.15337755233049394,
      "num_tokens": 17527024.0,
      "step": 8445
    },
    {
      "entropy": 5.654916906356812,
      "epoch": 0.7665094339622641,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004948559812833857,
      "loss": 5.473,
      "mean_token_accuracy": 0.16364773660898208,
      "num_tokens": 17536952.0,
      "step": 8450
    },
    {
      "entropy": 5.75989294052124,
      "epoch": 0.7669629898403484,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004948490997943162,
      "loss": 5.6805,
      "mean_token_accuracy": 0.15758652240037918,
      "num_tokens": 17548879.0,
      "step": 8455
    },
    {
      "entropy": 5.799036121368408,
      "epoch": 0.7674165457184325,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004948422137587111,
      "loss": 5.679,
      "mean_token_accuracy": 0.16287029832601546,
      "num_tokens": 17559149.0,
      "step": 8460
    },
    {
      "entropy": 5.788899993896484,
      "epoch": 0.7678701015965167,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004948353231767131,
      "loss": 5.6528,
      "mean_token_accuracy": 0.15707586258649825,
      "num_tokens": 17569995.0,
      "step": 8465
    },
    {
      "entropy": 5.841172218322754,
      "epoch": 0.7683236574746009,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004948284280484645,
      "loss": 5.6797,
      "mean_token_accuracy": 0.15439939200878144,
      "num_tokens": 17581336.0,
      "step": 8470
    },
    {
      "entropy": 5.832214212417602,
      "epoch": 0.768777213352685,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004948215283741081,
      "loss": 5.6359,
      "mean_token_accuracy": 0.1576690025627613,
      "num_tokens": 17592680.0,
      "step": 8475
    },
    {
      "entropy": 5.737134313583374,
      "epoch": 0.7692307692307693,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004948146241537866,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16077874451875687,
      "num_tokens": 17603049.0,
      "step": 8480
    },
    {
      "entropy": 5.670020675659179,
      "epoch": 0.7696843251088534,
      "grad_norm": 8.0625,
      "learning_rate": 0.0004948077153876425,
      "loss": 5.5748,
      "mean_token_accuracy": 0.1587527170777321,
      "num_tokens": 17612761.0,
      "step": 8485
    },
    {
      "entropy": 5.740362167358398,
      "epoch": 0.7701378809869376,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004948008020758188,
      "loss": 5.6174,
      "mean_token_accuracy": 0.15385191291570663,
      "num_tokens": 17622343.0,
      "step": 8490
    },
    {
      "entropy": 5.714895582199096,
      "epoch": 0.7705914368650217,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004947938842184585,
      "loss": 5.5325,
      "mean_token_accuracy": 0.158719938993454,
      "num_tokens": 17632949.0,
      "step": 8495
    },
    {
      "entropy": 5.802890920639038,
      "epoch": 0.771044992743106,
      "grad_norm": 5.0,
      "learning_rate": 0.0004947869618157048,
      "loss": 5.6555,
      "mean_token_accuracy": 0.15329065024852753,
      "num_tokens": 17643199.0,
      "step": 8500
    },
    {
      "entropy": 5.879572439193725,
      "epoch": 0.7714985486211902,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004947800348677006,
      "loss": 5.6597,
      "mean_token_accuracy": 0.14939892888069153,
      "num_tokens": 17653708.0,
      "step": 8505
    },
    {
      "entropy": 5.754754877090454,
      "epoch": 0.7719521044992743,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004947731033745893,
      "loss": 5.674,
      "mean_token_accuracy": 0.15467614233493804,
      "num_tokens": 17664360.0,
      "step": 8510
    },
    {
      "entropy": 5.729792642593384,
      "epoch": 0.7724056603773585,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004947661673365142,
      "loss": 5.5138,
      "mean_token_accuracy": 0.15840732157230378,
      "num_tokens": 17674614.0,
      "step": 8515
    },
    {
      "entropy": 5.753319120407104,
      "epoch": 0.7728592162554426,
      "grad_norm": 3.625,
      "learning_rate": 0.0004947592267536188,
      "loss": 5.5562,
      "mean_token_accuracy": 0.15683589577674867,
      "num_tokens": 17683990.0,
      "step": 8520
    },
    {
      "entropy": 5.761773586273193,
      "epoch": 0.7733127721335269,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004947522816260466,
      "loss": 5.6298,
      "mean_token_accuracy": 0.1477172538638115,
      "num_tokens": 17694188.0,
      "step": 8525
    },
    {
      "entropy": 5.736271381378174,
      "epoch": 0.773766328011611,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004947453319539411,
      "loss": 5.628,
      "mean_token_accuracy": 0.1537310317158699,
      "num_tokens": 17705466.0,
      "step": 8530
    },
    {
      "entropy": 5.7209672927856445,
      "epoch": 0.7742198838896952,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004947383777374462,
      "loss": 5.5032,
      "mean_token_accuracy": 0.16311814337968827,
      "num_tokens": 17715002.0,
      "step": 8535
    },
    {
      "entropy": 5.681128740310669,
      "epoch": 0.7746734397677794,
      "grad_norm": 2.125,
      "learning_rate": 0.0004947314189767058,
      "loss": 5.5371,
      "mean_token_accuracy": 0.15525227189064025,
      "num_tokens": 17725616.0,
      "step": 8540
    },
    {
      "entropy": 5.754605531692505,
      "epoch": 0.7751269956458636,
      "grad_norm": 3.25,
      "learning_rate": 0.0004947244556718635,
      "loss": 5.6756,
      "mean_token_accuracy": 0.1498018816113472,
      "num_tokens": 17736970.0,
      "step": 8545
    },
    {
      "entropy": 5.7690338611602785,
      "epoch": 0.7755805515239478,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004947174878230635,
      "loss": 5.571,
      "mean_token_accuracy": 0.15429942607879638,
      "num_tokens": 17747235.0,
      "step": 8550
    },
    {
      "entropy": 5.724594306945801,
      "epoch": 0.7760341074020319,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004947105154304498,
      "loss": 5.6478,
      "mean_token_accuracy": 0.1567047655582428,
      "num_tokens": 17758345.0,
      "step": 8555
    },
    {
      "entropy": 5.745009088516236,
      "epoch": 0.7764876632801161,
      "grad_norm": 2.625,
      "learning_rate": 0.0004947035384941667,
      "loss": 5.6231,
      "mean_token_accuracy": 0.15353951454162598,
      "num_tokens": 17768226.0,
      "step": 8560
    },
    {
      "entropy": 5.820920133590699,
      "epoch": 0.7769412191582002,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004946965570143584,
      "loss": 5.6703,
      "mean_token_accuracy": 0.15955789536237716,
      "num_tokens": 17777742.0,
      "step": 8565
    },
    {
      "entropy": 5.8193521976470945,
      "epoch": 0.7773947750362845,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004946895709911695,
      "loss": 5.5796,
      "mean_token_accuracy": 0.1583583325147629,
      "num_tokens": 17787529.0,
      "step": 8570
    },
    {
      "entropy": 5.719511461257935,
      "epoch": 0.7778483309143687,
      "grad_norm": 3.0,
      "learning_rate": 0.0004946825804247442,
      "loss": 5.6178,
      "mean_token_accuracy": 0.15740730315446855,
      "num_tokens": 17799499.0,
      "step": 8575
    },
    {
      "entropy": 5.65519962310791,
      "epoch": 0.7783018867924528,
      "grad_norm": 4.28125,
      "learning_rate": 0.000494675585315227,
      "loss": 5.5611,
      "mean_token_accuracy": 0.16181525439023972,
      "num_tokens": 17809622.0,
      "step": 8580
    },
    {
      "entropy": 5.734025955200195,
      "epoch": 0.778755442670537,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004946685856627628,
      "loss": 5.5496,
      "mean_token_accuracy": 0.1616109773516655,
      "num_tokens": 17820677.0,
      "step": 8585
    },
    {
      "entropy": 5.733445692062378,
      "epoch": 0.7792089985486212,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004946615814674963,
      "loss": 5.5909,
      "mean_token_accuracy": 0.15883141607046128,
      "num_tokens": 17830288.0,
      "step": 8590
    },
    {
      "entropy": 5.746526098251342,
      "epoch": 0.7796625544267054,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004946545727295724,
      "loss": 5.6091,
      "mean_token_accuracy": 0.15067204385995864,
      "num_tokens": 17840854.0,
      "step": 8595
    },
    {
      "entropy": 5.709014177322388,
      "epoch": 0.7801161103047896,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004946475594491359,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1614265039563179,
      "num_tokens": 17850860.0,
      "step": 8600
    },
    {
      "entropy": 5.734700679779053,
      "epoch": 0.7805696661828737,
      "grad_norm": 3.5,
      "learning_rate": 0.0004946405416263319,
      "loss": 5.5694,
      "mean_token_accuracy": 0.15939305275678634,
      "num_tokens": 17860254.0,
      "step": 8605
    },
    {
      "entropy": 5.708762884140015,
      "epoch": 0.7810232220609579,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004946335192613054,
      "loss": 5.5874,
      "mean_token_accuracy": 0.15273386538028716,
      "num_tokens": 17871705.0,
      "step": 8610
    },
    {
      "entropy": 5.860450887680054,
      "epoch": 0.7814767779390421,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004946264923542019,
      "loss": 5.6162,
      "mean_token_accuracy": 0.15346125960350038,
      "num_tokens": 17883184.0,
      "step": 8615
    },
    {
      "entropy": 5.795368385314942,
      "epoch": 0.7819303338171263,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004946194609051664,
      "loss": 5.6044,
      "mean_token_accuracy": 0.1526493489742279,
      "num_tokens": 17894696.0,
      "step": 8620
    },
    {
      "entropy": 5.6946038722991945,
      "epoch": 0.7823838896952104,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004946124249143447,
      "loss": 5.564,
      "mean_token_accuracy": 0.157182414829731,
      "num_tokens": 17904454.0,
      "step": 8625
    },
    {
      "entropy": 5.7621626377105715,
      "epoch": 0.7828374455732946,
      "grad_norm": 2.21875,
      "learning_rate": 0.000494605384381882,
      "loss": 5.6534,
      "mean_token_accuracy": 0.15995606780052185,
      "num_tokens": 17915097.0,
      "step": 8630
    },
    {
      "entropy": 5.823960685729981,
      "epoch": 0.7832910014513788,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004945983393079239,
      "loss": 5.622,
      "mean_token_accuracy": 0.15308469980955125,
      "num_tokens": 17926234.0,
      "step": 8635
    },
    {
      "entropy": 5.75256199836731,
      "epoch": 0.783744557329463,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004945912896926163,
      "loss": 5.5466,
      "mean_token_accuracy": 0.16132265329360962,
      "num_tokens": 17936492.0,
      "step": 8640
    },
    {
      "entropy": 5.686544179916382,
      "epoch": 0.7841981132075472,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004945842355361049,
      "loss": 5.6114,
      "mean_token_accuracy": 0.15452968776226045,
      "num_tokens": 17946541.0,
      "step": 8645
    },
    {
      "entropy": 5.759810876846314,
      "epoch": 0.7846516690856313,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004945771768385355,
      "loss": 5.5403,
      "mean_token_accuracy": 0.17018820494413375,
      "num_tokens": 17956257.0,
      "step": 8650
    },
    {
      "entropy": 5.747081756591797,
      "epoch": 0.7851052249637155,
      "grad_norm": 3.515625,
      "learning_rate": 0.000494570113600054,
      "loss": 5.5013,
      "mean_token_accuracy": 0.16424116641283035,
      "num_tokens": 17966108.0,
      "step": 8655
    },
    {
      "entropy": 5.712142896652222,
      "epoch": 0.7855587808417998,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004945630458208068,
      "loss": 5.5759,
      "mean_token_accuracy": 0.16179561316967012,
      "num_tokens": 17975487.0,
      "step": 8660
    },
    {
      "entropy": 5.725969219207764,
      "epoch": 0.7860123367198839,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004945559735009397,
      "loss": 5.6136,
      "mean_token_accuracy": 0.1575325086712837,
      "num_tokens": 17986308.0,
      "step": 8665
    },
    {
      "entropy": 5.775984191894532,
      "epoch": 0.7864658925979681,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004945488966405993,
      "loss": 5.6626,
      "mean_token_accuracy": 0.15036723241209984,
      "num_tokens": 17996844.0,
      "step": 8670
    },
    {
      "entropy": 5.75177230834961,
      "epoch": 0.7869194484760522,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004945418152399317,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15092529505491256,
      "num_tokens": 18007648.0,
      "step": 8675
    },
    {
      "entropy": 5.729442882537842,
      "epoch": 0.7873730043541364,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004945347292990833,
      "loss": 5.6208,
      "mean_token_accuracy": 0.1617960065603256,
      "num_tokens": 18017971.0,
      "step": 8680
    },
    {
      "entropy": 5.704108619689942,
      "epoch": 0.7878265602322206,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004945276388182008,
      "loss": 5.4916,
      "mean_token_accuracy": 0.16510131061077118,
      "num_tokens": 18028500.0,
      "step": 8685
    },
    {
      "entropy": 5.76072244644165,
      "epoch": 0.7882801161103048,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004945205437974309,
      "loss": 5.5338,
      "mean_token_accuracy": 0.16115738749504088,
      "num_tokens": 18038101.0,
      "step": 8690
    },
    {
      "entropy": 5.81218113899231,
      "epoch": 0.788733671988389,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004945134442369201,
      "loss": 5.5593,
      "mean_token_accuracy": 0.15752734541893004,
      "num_tokens": 18048416.0,
      "step": 8695
    },
    {
      "entropy": 5.749841165542603,
      "epoch": 0.7891872278664731,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004945063401368154,
      "loss": 5.5134,
      "mean_token_accuracy": 0.15486560091376306,
      "num_tokens": 18058711.0,
      "step": 8700
    },
    {
      "entropy": 5.753841733932495,
      "epoch": 0.7896407837445574,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004944992314972637,
      "loss": 5.561,
      "mean_token_accuracy": 0.1602311372756958,
      "num_tokens": 18068855.0,
      "step": 8705
    },
    {
      "entropy": 5.681754302978516,
      "epoch": 0.7900943396226415,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004944921183184118,
      "loss": 5.5391,
      "mean_token_accuracy": 0.16171472668647766,
      "num_tokens": 18079025.0,
      "step": 8710
    },
    {
      "entropy": 5.77334885597229,
      "epoch": 0.7905478955007257,
      "grad_norm": 2.640625,
      "learning_rate": 0.000494485000600407,
      "loss": 5.7387,
      "mean_token_accuracy": 0.14843253791332245,
      "num_tokens": 18091032.0,
      "step": 8715
    },
    {
      "entropy": 5.69298734664917,
      "epoch": 0.7910014513788098,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004944778783433965,
      "loss": 5.5925,
      "mean_token_accuracy": 0.16092147380113603,
      "num_tokens": 18101398.0,
      "step": 8720
    },
    {
      "entropy": 5.712299537658692,
      "epoch": 0.791455007256894,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004944707515475276,
      "loss": 5.5059,
      "mean_token_accuracy": 0.15996126383543013,
      "num_tokens": 18111029.0,
      "step": 8725
    },
    {
      "entropy": 5.668330574035645,
      "epoch": 0.7919085631349783,
      "grad_norm": 3.125,
      "learning_rate": 0.0004944636202129475,
      "loss": 5.4802,
      "mean_token_accuracy": 0.1612555891275406,
      "num_tokens": 18120593.0,
      "step": 8730
    },
    {
      "entropy": 5.784639167785644,
      "epoch": 0.7923621190130624,
      "grad_norm": 4.0,
      "learning_rate": 0.0004944564843398039,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1531424716114998,
      "num_tokens": 18131692.0,
      "step": 8735
    },
    {
      "entropy": 5.766321182250977,
      "epoch": 0.7928156748911466,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004944493439282443,
      "loss": 5.5567,
      "mean_token_accuracy": 0.16134166419506074,
      "num_tokens": 18141244.0,
      "step": 8740
    },
    {
      "entropy": 5.73597526550293,
      "epoch": 0.7932692307692307,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004944421989784162,
      "loss": 5.5659,
      "mean_token_accuracy": 0.15590226352214814,
      "num_tokens": 18150585.0,
      "step": 8745
    },
    {
      "entropy": 5.686997890472412,
      "epoch": 0.793722786647315,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004944350494904676,
      "loss": 5.5977,
      "mean_token_accuracy": 0.16071152836084365,
      "num_tokens": 18160981.0,
      "step": 8750
    },
    {
      "entropy": 5.771206426620483,
      "epoch": 0.7941763425253991,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004944278954645463,
      "loss": 5.5691,
      "mean_token_accuracy": 0.15511913746595382,
      "num_tokens": 18171181.0,
      "step": 8755
    },
    {
      "entropy": 5.806949472427368,
      "epoch": 0.7946298984034833,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004944207369008002,
      "loss": 5.6374,
      "mean_token_accuracy": 0.1530435189604759,
      "num_tokens": 18181095.0,
      "step": 8760
    },
    {
      "entropy": 5.734293699264526,
      "epoch": 0.7950834542815675,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004944135737993773,
      "loss": 5.5335,
      "mean_token_accuracy": 0.15290747284889222,
      "num_tokens": 18190153.0,
      "step": 8765
    },
    {
      "entropy": 5.646497964859009,
      "epoch": 0.7955370101596516,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004944064061604258,
      "loss": 5.6361,
      "mean_token_accuracy": 0.15470516681671143,
      "num_tokens": 18201413.0,
      "step": 8770
    },
    {
      "entropy": 5.656438779830933,
      "epoch": 0.7959905660377359,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004943992339840939,
      "loss": 5.5157,
      "mean_token_accuracy": 0.16347565948963166,
      "num_tokens": 18211200.0,
      "step": 8775
    },
    {
      "entropy": 5.8274603366851805,
      "epoch": 0.79644412191582,
      "grad_norm": 3.09375,
      "learning_rate": 0.00049439205727053,
      "loss": 5.7226,
      "mean_token_accuracy": 0.14984723031520844,
      "num_tokens": 18222730.0,
      "step": 8780
    },
    {
      "entropy": 5.841176557540893,
      "epoch": 0.7968976777939042,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004943848760198823,
      "loss": 5.6332,
      "mean_token_accuracy": 0.15421689599752425,
      "num_tokens": 18232905.0,
      "step": 8785
    },
    {
      "entropy": 5.892225837707519,
      "epoch": 0.7973512336719883,
      "grad_norm": 3.0,
      "learning_rate": 0.0004943776902322995,
      "loss": 5.7162,
      "mean_token_accuracy": 0.14882886409759521,
      "num_tokens": 18243010.0,
      "step": 8790
    },
    {
      "entropy": 5.775762557983398,
      "epoch": 0.7978047895500726,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004943704999079302,
      "loss": 5.5456,
      "mean_token_accuracy": 0.16070357263088225,
      "num_tokens": 18254093.0,
      "step": 8795
    },
    {
      "entropy": 5.7349199771881105,
      "epoch": 0.7982583454281568,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004943633050469231,
      "loss": 5.5307,
      "mean_token_accuracy": 0.15499844700098037,
      "num_tokens": 18264103.0,
      "step": 8800
    },
    {
      "entropy": 5.7692304134368895,
      "epoch": 0.7987119013062409,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004943561056494269,
      "loss": 5.6894,
      "mean_token_accuracy": 0.15571623742580415,
      "num_tokens": 18276051.0,
      "step": 8805
    },
    {
      "entropy": 5.779341125488282,
      "epoch": 0.7991654571843251,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004943489017155906,
      "loss": 5.591,
      "mean_token_accuracy": 0.15946148782968522,
      "num_tokens": 18286854.0,
      "step": 8810
    },
    {
      "entropy": 5.752212572097778,
      "epoch": 0.7996190130624092,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004943416932455631,
      "loss": 5.6967,
      "mean_token_accuracy": 0.15069335103034973,
      "num_tokens": 18298404.0,
      "step": 8815
    },
    {
      "entropy": 5.673287105560303,
      "epoch": 0.8000725689404935,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004943344802394935,
      "loss": 5.5652,
      "mean_token_accuracy": 0.15464189201593398,
      "num_tokens": 18309541.0,
      "step": 8820
    },
    {
      "entropy": 5.746518754959107,
      "epoch": 0.8005261248185777,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004943272626975309,
      "loss": 5.5413,
      "mean_token_accuracy": 0.1546555534005165,
      "num_tokens": 18320438.0,
      "step": 8825
    },
    {
      "entropy": 5.830396270751953,
      "epoch": 0.8009796806966618,
      "grad_norm": 2.75,
      "learning_rate": 0.0004943200406198246,
      "loss": 5.6868,
      "mean_token_accuracy": 0.15161113142967225,
      "num_tokens": 18331964.0,
      "step": 8830
    },
    {
      "entropy": 5.8303650379180905,
      "epoch": 0.801433236574746,
      "grad_norm": 2.484375,
      "learning_rate": 0.000494312814006524,
      "loss": 5.637,
      "mean_token_accuracy": 0.149752576649189,
      "num_tokens": 18342479.0,
      "step": 8835
    },
    {
      "entropy": 5.762335300445557,
      "epoch": 0.8018867924528302,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004943055828577784,
      "loss": 5.5754,
      "mean_token_accuracy": 0.15963243693113327,
      "num_tokens": 18353428.0,
      "step": 8840
    },
    {
      "entropy": 5.714380741119385,
      "epoch": 0.8023403483309144,
      "grad_norm": 3.875,
      "learning_rate": 0.0004942983471737376,
      "loss": 5.5301,
      "mean_token_accuracy": 0.16554586738348007,
      "num_tokens": 18362440.0,
      "step": 8845
    },
    {
      "entropy": 5.777411890029907,
      "epoch": 0.8027939042089985,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004942911069545511,
      "loss": 5.6455,
      "mean_token_accuracy": 0.1575615018606186,
      "num_tokens": 18372130.0,
      "step": 8850
    },
    {
      "entropy": 5.781487846374512,
      "epoch": 0.8032474600870827,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004942838622003686,
      "loss": 5.6686,
      "mean_token_accuracy": 0.15949161052703859,
      "num_tokens": 18383262.0,
      "step": 8855
    },
    {
      "entropy": 5.784262704849243,
      "epoch": 0.8037010159651669,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004942766129113399,
      "loss": 5.6117,
      "mean_token_accuracy": 0.15410858765244484,
      "num_tokens": 18394036.0,
      "step": 8860
    },
    {
      "entropy": 5.731306695938111,
      "epoch": 0.8041545718432511,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004942693590876151,
      "loss": 5.5792,
      "mean_token_accuracy": 0.16315408200025558,
      "num_tokens": 18404249.0,
      "step": 8865
    },
    {
      "entropy": 5.722614669799805,
      "epoch": 0.8046081277213353,
      "grad_norm": 3.109375,
      "learning_rate": 0.000494262100729344,
      "loss": 5.5843,
      "mean_token_accuracy": 0.15265329629182817,
      "num_tokens": 18415410.0,
      "step": 8870
    },
    {
      "entropy": 5.736243772506714,
      "epoch": 0.8050616835994194,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004942548378366768,
      "loss": 5.6165,
      "mean_token_accuracy": 0.15731293857097625,
      "num_tokens": 18425762.0,
      "step": 8875
    },
    {
      "entropy": 5.70797061920166,
      "epoch": 0.8055152394775036,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004942475704097635,
      "loss": 5.5511,
      "mean_token_accuracy": 0.15828103423118592,
      "num_tokens": 18436378.0,
      "step": 8880
    },
    {
      "entropy": 5.767156314849854,
      "epoch": 0.8059687953555879,
      "grad_norm": 2.625,
      "learning_rate": 0.0004942402984487548,
      "loss": 5.6017,
      "mean_token_accuracy": 0.15951280891895295,
      "num_tokens": 18445062.0,
      "step": 8885
    },
    {
      "entropy": 5.708318996429443,
      "epoch": 0.806422351233672,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004942330219538008,
      "loss": 5.5803,
      "mean_token_accuracy": 0.15844014585018157,
      "num_tokens": 18455555.0,
      "step": 8890
    },
    {
      "entropy": 5.820754480361939,
      "epoch": 0.8068759071117562,
      "grad_norm": 3.421875,
      "learning_rate": 0.000494225740925052,
      "loss": 5.5268,
      "mean_token_accuracy": 0.162959423661232,
      "num_tokens": 18465703.0,
      "step": 8895
    },
    {
      "entropy": 5.713505792617798,
      "epoch": 0.8073294629898403,
      "grad_norm": 6.53125,
      "learning_rate": 0.0004942184553626591,
      "loss": 5.5666,
      "mean_token_accuracy": 0.16358105093240738,
      "num_tokens": 18475640.0,
      "step": 8900
    },
    {
      "entropy": 5.80126953125,
      "epoch": 0.8077830188679245,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004942111652667726,
      "loss": 5.5944,
      "mean_token_accuracy": 0.15995924472808837,
      "num_tokens": 18485135.0,
      "step": 8905
    },
    {
      "entropy": 5.706670236587525,
      "epoch": 0.8082365747460087,
      "grad_norm": 3.0,
      "learning_rate": 0.0004942038706375433,
      "loss": 5.5074,
      "mean_token_accuracy": 0.1538008600473404,
      "num_tokens": 18496086.0,
      "step": 8910
    },
    {
      "entropy": 5.665920352935791,
      "epoch": 0.8086901306240929,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004941965714751222,
      "loss": 5.5418,
      "mean_token_accuracy": 0.15753740072250366,
      "num_tokens": 18506004.0,
      "step": 8915
    },
    {
      "entropy": 5.753987216949463,
      "epoch": 0.809143686502177,
      "grad_norm": 5.5625,
      "learning_rate": 0.0004941892677796601,
      "loss": 5.6414,
      "mean_token_accuracy": 0.15734172016382217,
      "num_tokens": 18515732.0,
      "step": 8920
    },
    {
      "entropy": 5.8059326171875,
      "epoch": 0.8095972423802612,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004941819595513081,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15974941700696946,
      "num_tokens": 18525782.0,
      "step": 8925
    },
    {
      "entropy": 5.767988824844361,
      "epoch": 0.8100507982583455,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004941746467902173,
      "loss": 5.5561,
      "mean_token_accuracy": 0.1544948175549507,
      "num_tokens": 18536049.0,
      "step": 8930
    },
    {
      "entropy": 5.749485301971435,
      "epoch": 0.8105043541364296,
      "grad_norm": 4.5,
      "learning_rate": 0.000494167329496539,
      "loss": 5.6174,
      "mean_token_accuracy": 0.16105745136737823,
      "num_tokens": 18547392.0,
      "step": 8935
    },
    {
      "entropy": 5.813601493835449,
      "epoch": 0.8109579100145138,
      "grad_norm": 2.125,
      "learning_rate": 0.0004941600076704244,
      "loss": 5.6572,
      "mean_token_accuracy": 0.15198875218629837,
      "num_tokens": 18558841.0,
      "step": 8940
    },
    {
      "entropy": 5.809838914871216,
      "epoch": 0.8114114658925979,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004941526813120251,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1533591032028198,
      "num_tokens": 18570727.0,
      "step": 8945
    },
    {
      "entropy": 5.755561828613281,
      "epoch": 0.8118650217706821,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004941453504214924,
      "loss": 5.6126,
      "mean_token_accuracy": 0.15869819819927217,
      "num_tokens": 18581168.0,
      "step": 8950
    },
    {
      "entropy": 5.794755649566651,
      "epoch": 0.8123185776487664,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004941380149989781,
      "loss": 5.6081,
      "mean_token_accuracy": 0.15217462480068206,
      "num_tokens": 18590800.0,
      "step": 8955
    },
    {
      "entropy": 5.759859037399292,
      "epoch": 0.8127721335268505,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004941306750446336,
      "loss": 5.5685,
      "mean_token_accuracy": 0.1570788413286209,
      "num_tokens": 18601847.0,
      "step": 8960
    },
    {
      "entropy": 5.813338756561279,
      "epoch": 0.8132256894049347,
      "grad_norm": 2.640625,
      "learning_rate": 0.000494123330558611,
      "loss": 5.6805,
      "mean_token_accuracy": 0.15837844759225844,
      "num_tokens": 18613105.0,
      "step": 8965
    },
    {
      "entropy": 5.797929859161377,
      "epoch": 0.8136792452830188,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004941159815410622,
      "loss": 5.5947,
      "mean_token_accuracy": 0.1564161814749241,
      "num_tokens": 18622945.0,
      "step": 8970
    },
    {
      "entropy": 5.773069620132446,
      "epoch": 0.8141328011611031,
      "grad_norm": 3.203125,
      "learning_rate": 0.000494108627992139,
      "loss": 5.559,
      "mean_token_accuracy": 0.15942140966653823,
      "num_tokens": 18632726.0,
      "step": 8975
    },
    {
      "entropy": 5.771838808059693,
      "epoch": 0.8145863570391872,
      "grad_norm": 4.0,
      "learning_rate": 0.0004941012699119936,
      "loss": 5.5819,
      "mean_token_accuracy": 0.1561790868639946,
      "num_tokens": 18642984.0,
      "step": 8980
    },
    {
      "entropy": 5.749499177932739,
      "epoch": 0.8150399129172714,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004940939073007779,
      "loss": 5.6157,
      "mean_token_accuracy": 0.15468558818101882,
      "num_tokens": 18653399.0,
      "step": 8985
    },
    {
      "entropy": 5.741241455078125,
      "epoch": 0.8154934687953556,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004940865401586446,
      "loss": 5.517,
      "mean_token_accuracy": 0.15652735829353331,
      "num_tokens": 18664641.0,
      "step": 8990
    },
    {
      "entropy": 5.765671825408935,
      "epoch": 0.8159470246734397,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004940791684857458,
      "loss": 5.5623,
      "mean_token_accuracy": 0.15982917249202727,
      "num_tokens": 18674890.0,
      "step": 8995
    },
    {
      "entropy": 5.710493898391723,
      "epoch": 0.816400580551524,
      "grad_norm": 6.9375,
      "learning_rate": 0.0004940717922822339,
      "loss": 5.6301,
      "mean_token_accuracy": 0.15173522233963013,
      "num_tokens": 18686455.0,
      "step": 9000
    },
    {
      "epoch": 0.816400580551524,
      "eval_entropy": 5.556482378929449,
      "eval_loss": 5.628951549530029,
      "eval_mean_token_accuracy": 0.16274379363275088,
      "eval_num_tokens": 18686455.0,
      "eval_runtime": 26.5865,
      "eval_samples_per_second": 1329.999,
      "eval_steps_per_second": 166.25,
      "step": 9000
    },
    {
      "entropy": 5.776840400695801,
      "epoch": 0.8168541364296081,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004940644115482615,
      "loss": 5.7014,
      "mean_token_accuracy": 0.15635610818862916,
      "num_tokens": 18696785.0,
      "step": 9005
    },
    {
      "entropy": 5.853718233108521,
      "epoch": 0.8173076923076923,
      "grad_norm": 2.75,
      "learning_rate": 0.0004940570262839812,
      "loss": 5.5313,
      "mean_token_accuracy": 0.162844155728817,
      "num_tokens": 18708052.0,
      "step": 9010
    },
    {
      "entropy": 5.820675611495972,
      "epoch": 0.8177612481857764,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004940496364895457,
      "loss": 5.6566,
      "mean_token_accuracy": 0.15677399486303328,
      "num_tokens": 18718266.0,
      "step": 9015
    },
    {
      "entropy": 5.723703670501709,
      "epoch": 0.8182148040638607,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004940422421651082,
      "loss": 5.6018,
      "mean_token_accuracy": 0.15171107798814773,
      "num_tokens": 18729802.0,
      "step": 9020
    },
    {
      "entropy": 5.786997842788696,
      "epoch": 0.8186683599419449,
      "grad_norm": 2.5,
      "learning_rate": 0.0004940348433108211,
      "loss": 5.5415,
      "mean_token_accuracy": 0.16219059526920318,
      "num_tokens": 18740628.0,
      "step": 9025
    },
    {
      "entropy": 5.8281457901000975,
      "epoch": 0.819121915820029,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004940274399268377,
      "loss": 5.6195,
      "mean_token_accuracy": 0.16045619994401933,
      "num_tokens": 18749934.0,
      "step": 9030
    },
    {
      "entropy": 5.837336826324463,
      "epoch": 0.8195754716981132,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004940200320133109,
      "loss": 5.7029,
      "mean_token_accuracy": 0.15211922377347947,
      "num_tokens": 18760696.0,
      "step": 9035
    },
    {
      "entropy": 5.779109716415405,
      "epoch": 0.8200290275761973,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004940126195703941,
      "loss": 5.7071,
      "mean_token_accuracy": 0.14571218341588973,
      "num_tokens": 18770158.0,
      "step": 9040
    },
    {
      "entropy": 5.817840719223023,
      "epoch": 0.8204825834542816,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004940052025982405,
      "loss": 5.6353,
      "mean_token_accuracy": 0.15551646649837494,
      "num_tokens": 18780329.0,
      "step": 9045
    },
    {
      "entropy": 5.734896945953369,
      "epoch": 0.8209361393323658,
      "grad_norm": 3.828125,
      "learning_rate": 0.0004939977810970035,
      "loss": 5.5576,
      "mean_token_accuracy": 0.15384935587644577,
      "num_tokens": 18791032.0,
      "step": 9050
    },
    {
      "entropy": 5.807877874374389,
      "epoch": 0.8213896952104499,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004939903550668366,
      "loss": 5.6168,
      "mean_token_accuracy": 0.15695300102233886,
      "num_tokens": 18800674.0,
      "step": 9055
    },
    {
      "entropy": 5.803936243057251,
      "epoch": 0.8218432510885341,
      "grad_norm": 2.75,
      "learning_rate": 0.0004939829245078934,
      "loss": 5.6352,
      "mean_token_accuracy": 0.1526229090988636,
      "num_tokens": 18811234.0,
      "step": 9060
    },
    {
      "entropy": 5.793316888809204,
      "epoch": 0.8222968069666183,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004939754894203275,
      "loss": 5.5669,
      "mean_token_accuracy": 0.15826106667518616,
      "num_tokens": 18820753.0,
      "step": 9065
    },
    {
      "entropy": 5.785939168930054,
      "epoch": 0.8227503628447025,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004939680498042926,
      "loss": 5.6611,
      "mean_token_accuracy": 0.15249083191156387,
      "num_tokens": 18830994.0,
      "step": 9070
    },
    {
      "entropy": 5.846006917953491,
      "epoch": 0.8232039187227866,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004939606056599428,
      "loss": 5.677,
      "mean_token_accuracy": 0.15244084745645523,
      "num_tokens": 18842120.0,
      "step": 9075
    },
    {
      "entropy": 5.778271722793579,
      "epoch": 0.8236574746008708,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004939531569874316,
      "loss": 5.6602,
      "mean_token_accuracy": 0.1516385942697525,
      "num_tokens": 18852290.0,
      "step": 9080
    },
    {
      "entropy": 5.831738042831421,
      "epoch": 0.824111030478955,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004939457037869135,
      "loss": 5.6692,
      "mean_token_accuracy": 0.15011354088783263,
      "num_tokens": 18862890.0,
      "step": 9085
    },
    {
      "entropy": 5.780765581130981,
      "epoch": 0.8245645863570392,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004939382460585424,
      "loss": 5.6373,
      "mean_token_accuracy": 0.15401018559932708,
      "num_tokens": 18873827.0,
      "step": 9090
    },
    {
      "entropy": 5.780360746383667,
      "epoch": 0.8250181422351234,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004939307838024726,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1578843414783478,
      "num_tokens": 18884581.0,
      "step": 9095
    },
    {
      "entropy": 5.7969330787658695,
      "epoch": 0.8254716981132075,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004939233170188585,
      "loss": 5.625,
      "mean_token_accuracy": 0.15827976167201996,
      "num_tokens": 18895441.0,
      "step": 9100
    },
    {
      "entropy": 5.77012267112732,
      "epoch": 0.8259252539912917,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004939158457078543,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1628742828965187,
      "num_tokens": 18905232.0,
      "step": 9105
    },
    {
      "entropy": 5.767564058303833,
      "epoch": 0.826378809869376,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004939083698696146,
      "loss": 5.5406,
      "mean_token_accuracy": 0.16362984776496886,
      "num_tokens": 18915080.0,
      "step": 9110
    },
    {
      "entropy": 5.888275289535523,
      "epoch": 0.8268323657474601,
      "grad_norm": 4.25,
      "learning_rate": 0.000493900889504294,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1495535746216774,
      "num_tokens": 18926705.0,
      "step": 9115
    },
    {
      "entropy": 5.786359024047852,
      "epoch": 0.8272859216255443,
      "grad_norm": 4.25,
      "learning_rate": 0.0004938934046120473,
      "loss": 5.535,
      "mean_token_accuracy": 0.1591508522629738,
      "num_tokens": 18936866.0,
      "step": 9120
    },
    {
      "entropy": 5.7610677719116214,
      "epoch": 0.8277394775036284,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004938859151930293,
      "loss": 5.5204,
      "mean_token_accuracy": 0.15995817333459855,
      "num_tokens": 18946660.0,
      "step": 9125
    },
    {
      "entropy": 5.774123811721802,
      "epoch": 0.8281930333817126,
      "grad_norm": 4.90625,
      "learning_rate": 0.0004938784212473947,
      "loss": 5.6123,
      "mean_token_accuracy": 0.1515015497803688,
      "num_tokens": 18957973.0,
      "step": 9130
    },
    {
      "entropy": 5.882508897781372,
      "epoch": 0.8286465892597968,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004938709227752985,
      "loss": 5.7078,
      "mean_token_accuracy": 0.1531926952302456,
      "num_tokens": 18969882.0,
      "step": 9135
    },
    {
      "entropy": 5.814449310302734,
      "epoch": 0.829100145137881,
      "grad_norm": 3.078125,
      "learning_rate": 0.000493863419776896,
      "loss": 5.6625,
      "mean_token_accuracy": 0.15482495725154877,
      "num_tokens": 18979633.0,
      "step": 9140
    },
    {
      "entropy": 5.886104106903076,
      "epoch": 0.8295537010159652,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004938559122523421,
      "loss": 5.6607,
      "mean_token_accuracy": 0.15840226411819458,
      "num_tokens": 18989537.0,
      "step": 9145
    },
    {
      "entropy": 5.738320064544678,
      "epoch": 0.8300072568940493,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004938484002017921,
      "loss": 5.5471,
      "mean_token_accuracy": 0.15736908465623856,
      "num_tokens": 18999465.0,
      "step": 9150
    },
    {
      "entropy": 5.802149248123169,
      "epoch": 0.8304608127721336,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004938408836254015,
      "loss": 5.5701,
      "mean_token_accuracy": 0.17053684890270232,
      "num_tokens": 19008990.0,
      "step": 9155
    },
    {
      "entropy": 5.775449275970459,
      "epoch": 0.8309143686502177,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004938333625233258,
      "loss": 5.5815,
      "mean_token_accuracy": 0.15539330691099168,
      "num_tokens": 19018710.0,
      "step": 9160
    },
    {
      "entropy": 5.727145051956176,
      "epoch": 0.8313679245283019,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004938258368957202,
      "loss": 5.5912,
      "mean_token_accuracy": 0.15500498116016387,
      "num_tokens": 19028572.0,
      "step": 9165
    },
    {
      "entropy": 5.819398593902588,
      "epoch": 0.831821480406386,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004938183067427406,
      "loss": 5.6596,
      "mean_token_accuracy": 0.15367700532078743,
      "num_tokens": 19039169.0,
      "step": 9170
    },
    {
      "entropy": 5.900253486633301,
      "epoch": 0.8322750362844702,
      "grad_norm": 2.125,
      "learning_rate": 0.0004938107720645426,
      "loss": 5.7029,
      "mean_token_accuracy": 0.15673841312527656,
      "num_tokens": 19049905.0,
      "step": 9175
    },
    {
      "entropy": 5.807830953598023,
      "epoch": 0.8327285921625545,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004938032328612821,
      "loss": 5.6255,
      "mean_token_accuracy": 0.15786124914884567,
      "num_tokens": 19059842.0,
      "step": 9180
    },
    {
      "entropy": 5.821698379516602,
      "epoch": 0.8331821480406386,
      "grad_norm": 4.0625,
      "learning_rate": 0.000493795689133115,
      "loss": 5.7331,
      "mean_token_accuracy": 0.14656377360224723,
      "num_tokens": 19070850.0,
      "step": 9185
    },
    {
      "entropy": 5.765946578979492,
      "epoch": 0.8336357039187228,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004937881408801974,
      "loss": 5.5614,
      "mean_token_accuracy": 0.15952363461256028,
      "num_tokens": 19081526.0,
      "step": 9190
    },
    {
      "entropy": 5.677028512954712,
      "epoch": 0.8340892597968069,
      "grad_norm": 4.28125,
      "learning_rate": 0.000493780588102685,
      "loss": 5.4916,
      "mean_token_accuracy": 0.15835061222314833,
      "num_tokens": 19090698.0,
      "step": 9195
    },
    {
      "entropy": 5.775712013244629,
      "epoch": 0.8345428156748912,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004937730308007346,
      "loss": 5.5994,
      "mean_token_accuracy": 0.156134045869112,
      "num_tokens": 19100862.0,
      "step": 9200
    },
    {
      "entropy": 5.851109027862549,
      "epoch": 0.8349963715529753,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004937654689745021,
      "loss": 5.5883,
      "mean_token_accuracy": 0.15505269169807434,
      "num_tokens": 19111228.0,
      "step": 9205
    },
    {
      "entropy": 5.849263095855713,
      "epoch": 0.8354499274310595,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004937579026241439,
      "loss": 5.7082,
      "mean_token_accuracy": 0.152321557700634,
      "num_tokens": 19122372.0,
      "step": 9210
    },
    {
      "entropy": 5.821360969543457,
      "epoch": 0.8359034833091437,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004937503317498165,
      "loss": 5.5872,
      "mean_token_accuracy": 0.15431858152151107,
      "num_tokens": 19132678.0,
      "step": 9215
    },
    {
      "entropy": 5.80333194732666,
      "epoch": 0.8363570391872278,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004937427563516765,
      "loss": 5.6358,
      "mean_token_accuracy": 0.16021214723587035,
      "num_tokens": 19142644.0,
      "step": 9220
    },
    {
      "entropy": 5.866559982299805,
      "epoch": 0.8368105950653121,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004937351764298805,
      "loss": 5.704,
      "mean_token_accuracy": 0.1565028116106987,
      "num_tokens": 19152755.0,
      "step": 9225
    },
    {
      "entropy": 5.898500871658325,
      "epoch": 0.8372641509433962,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004937275919845855,
      "loss": 5.8207,
      "mean_token_accuracy": 0.14555472284555435,
      "num_tokens": 19164320.0,
      "step": 9230
    },
    {
      "entropy": 5.7658318996429445,
      "epoch": 0.8377177068214804,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004937200030159481,
      "loss": 5.5323,
      "mean_token_accuracy": 0.16410059928894044,
      "num_tokens": 19175954.0,
      "step": 9235
    },
    {
      "entropy": 5.690430736541748,
      "epoch": 0.8381712626995645,
      "grad_norm": 4.0,
      "learning_rate": 0.0004937124095241252,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16230606585741042,
      "num_tokens": 19185762.0,
      "step": 9240
    },
    {
      "entropy": 5.745036363601685,
      "epoch": 0.8386248185776488,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004937048115092739,
      "loss": 5.6076,
      "mean_token_accuracy": 0.1604189783334732,
      "num_tokens": 19195826.0,
      "step": 9245
    },
    {
      "entropy": 5.7351212978363035,
      "epoch": 0.839078374455733,
      "grad_norm": 3.40625,
      "learning_rate": 0.0004936972089715515,
      "loss": 5.5433,
      "mean_token_accuracy": 0.1597079038619995,
      "num_tokens": 19206127.0,
      "step": 9250
    },
    {
      "entropy": 5.850226879119873,
      "epoch": 0.8395319303338171,
      "grad_norm": 3.109375,
      "learning_rate": 0.000493689601911115,
      "loss": 5.5892,
      "mean_token_accuracy": 0.1550763726234436,
      "num_tokens": 19216642.0,
      "step": 9255
    },
    {
      "entropy": 5.724189805984497,
      "epoch": 0.8399854862119013,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004936819903281219,
      "loss": 5.5651,
      "mean_token_accuracy": 0.16378083527088166,
      "num_tokens": 19226721.0,
      "step": 9260
    },
    {
      "entropy": 5.6341856002807615,
      "epoch": 0.8404390420899854,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004936743742227295,
      "loss": 5.5622,
      "mean_token_accuracy": 0.1577123984694481,
      "num_tokens": 19237490.0,
      "step": 9265
    },
    {
      "entropy": 5.773586225509644,
      "epoch": 0.8408925979680697,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004936667535950952,
      "loss": 5.6782,
      "mean_token_accuracy": 0.15248237550258636,
      "num_tokens": 19248093.0,
      "step": 9270
    },
    {
      "entropy": 5.871043682098389,
      "epoch": 0.8413461538461539,
      "grad_norm": 4.25,
      "learning_rate": 0.0004936591284453768,
      "loss": 5.5921,
      "mean_token_accuracy": 0.16444561034440994,
      "num_tokens": 19258892.0,
      "step": 9275
    },
    {
      "entropy": 5.840593242645264,
      "epoch": 0.841799709724238,
      "grad_norm": 3.0,
      "learning_rate": 0.0004936514987737319,
      "loss": 5.5768,
      "mean_token_accuracy": 0.16426993161439896,
      "num_tokens": 19268666.0,
      "step": 9280
    },
    {
      "entropy": 5.6901709079742435,
      "epoch": 0.8422532656023222,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004936438645803182,
      "loss": 5.6096,
      "mean_token_accuracy": 0.15355295538902283,
      "num_tokens": 19279111.0,
      "step": 9285
    },
    {
      "entropy": 5.839419603347778,
      "epoch": 0.8427068214804064,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004936362258652938,
      "loss": 5.7105,
      "mean_token_accuracy": 0.15006187558174133,
      "num_tokens": 19290660.0,
      "step": 9290
    },
    {
      "entropy": 5.7745153427124025,
      "epoch": 0.8431603773584906,
      "grad_norm": 5.875,
      "learning_rate": 0.0004936285826288165,
      "loss": 5.5657,
      "mean_token_accuracy": 0.1510035626590252,
      "num_tokens": 19300988.0,
      "step": 9295
    },
    {
      "entropy": 5.777370119094849,
      "epoch": 0.8436139332365747,
      "grad_norm": 3.71875,
      "learning_rate": 0.0004936209348710444,
      "loss": 5.6131,
      "mean_token_accuracy": 0.15827206075191497,
      "num_tokens": 19312392.0,
      "step": 9300
    },
    {
      "entropy": 5.794031047821045,
      "epoch": 0.8440674891146589,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004936132825921355,
      "loss": 5.6242,
      "mean_token_accuracy": 0.1563329927623272,
      "num_tokens": 19322311.0,
      "step": 9305
    },
    {
      "entropy": 5.797012424468994,
      "epoch": 0.844521044992743,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004936056257922484,
      "loss": 5.6559,
      "mean_token_accuracy": 0.15734238773584366,
      "num_tokens": 19332289.0,
      "step": 9310
    },
    {
      "entropy": 5.810130405426025,
      "epoch": 0.8449746008708273,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004935979644715412,
      "loss": 5.6737,
      "mean_token_accuracy": 0.15293284207582475,
      "num_tokens": 19342950.0,
      "step": 9315
    },
    {
      "entropy": 5.800215196609497,
      "epoch": 0.8454281567489115,
      "grad_norm": 3.0,
      "learning_rate": 0.0004935902986301724,
      "loss": 5.5848,
      "mean_token_accuracy": 0.15654435902833938,
      "num_tokens": 19353288.0,
      "step": 9320
    },
    {
      "entropy": 5.771015691757202,
      "epoch": 0.8458817126269956,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004935826282683005,
      "loss": 5.589,
      "mean_token_accuracy": 0.15742939412593843,
      "num_tokens": 19362934.0,
      "step": 9325
    },
    {
      "entropy": 5.758797359466553,
      "epoch": 0.8463352685050798,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004935749533860842,
      "loss": 5.5659,
      "mean_token_accuracy": 0.15728280022740365,
      "num_tokens": 19373149.0,
      "step": 9330
    },
    {
      "entropy": 5.759192991256714,
      "epoch": 0.846788824383164,
      "grad_norm": 2.375,
      "learning_rate": 0.000493567273983682,
      "loss": 5.5133,
      "mean_token_accuracy": 0.1573615312576294,
      "num_tokens": 19383424.0,
      "step": 9335
    },
    {
      "entropy": 5.78565936088562,
      "epoch": 0.8472423802612482,
      "grad_norm": 2.140625,
      "learning_rate": 0.000493559590061253,
      "loss": 5.611,
      "mean_token_accuracy": 0.15459365248680115,
      "num_tokens": 19393400.0,
      "step": 9340
    },
    {
      "entropy": 5.784577941894531,
      "epoch": 0.8476959361393324,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004935519016189559,
      "loss": 5.5871,
      "mean_token_accuracy": 0.1565425544977188,
      "num_tokens": 19403876.0,
      "step": 9345
    },
    {
      "entropy": 5.697234487533569,
      "epoch": 0.8481494920174165,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004935442086569498,
      "loss": 5.5098,
      "mean_token_accuracy": 0.15983743071556092,
      "num_tokens": 19413291.0,
      "step": 9350
    },
    {
      "entropy": 5.8318124294281,
      "epoch": 0.8486030478955007,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004935365111753939,
      "loss": 5.6087,
      "mean_token_accuracy": 0.15553602278232576,
      "num_tokens": 19422997.0,
      "step": 9355
    },
    {
      "entropy": 5.764914274215698,
      "epoch": 0.8490566037735849,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004935288091744472,
      "loss": 5.6413,
      "mean_token_accuracy": 0.15595665872097014,
      "num_tokens": 19433476.0,
      "step": 9360
    },
    {
      "entropy": 5.782774972915649,
      "epoch": 0.8495101596516691,
      "grad_norm": 2.5625,
      "learning_rate": 0.000493521102654269,
      "loss": 5.5964,
      "mean_token_accuracy": 0.16230583041906357,
      "num_tokens": 19443314.0,
      "step": 9365
    },
    {
      "entropy": 5.867934370040894,
      "epoch": 0.8499637155297532,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004935133916150187,
      "loss": 5.695,
      "mean_token_accuracy": 0.15295173227787018,
      "num_tokens": 19454990.0,
      "step": 9370
    },
    {
      "entropy": 5.8137526512146,
      "epoch": 0.8504172714078374,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004935056760568558,
      "loss": 5.6449,
      "mean_token_accuracy": 0.15060910433530808,
      "num_tokens": 19465575.0,
      "step": 9375
    },
    {
      "entropy": 5.797601222991943,
      "epoch": 0.8508708272859217,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004934979559799399,
      "loss": 5.6522,
      "mean_token_accuracy": 0.15404964089393616,
      "num_tokens": 19475841.0,
      "step": 9380
    },
    {
      "entropy": 5.909706163406372,
      "epoch": 0.8513243831640058,
      "grad_norm": 3.8125,
      "learning_rate": 0.0004934902313844306,
      "loss": 5.6509,
      "mean_token_accuracy": 0.14857155978679656,
      "num_tokens": 19485725.0,
      "step": 9385
    },
    {
      "entropy": 5.798264360427856,
      "epoch": 0.85177793904209,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004934825022704875,
      "loss": 5.6811,
      "mean_token_accuracy": 0.15481986552476884,
      "num_tokens": 19496102.0,
      "step": 9390
    },
    {
      "entropy": 5.7493983745574955,
      "epoch": 0.8522314949201741,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004934747686382707,
      "loss": 5.623,
      "mean_token_accuracy": 0.15276634693145752,
      "num_tokens": 19506437.0,
      "step": 9395
    },
    {
      "entropy": 5.769255828857422,
      "epoch": 0.8526850507982583,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004934670304879398,
      "loss": 5.6218,
      "mean_token_accuracy": 0.15732096135616302,
      "num_tokens": 19517942.0,
      "step": 9400
    },
    {
      "entropy": 5.811493396759033,
      "epoch": 0.8531386066763426,
      "grad_norm": 2.125,
      "learning_rate": 0.0004934592878196553,
      "loss": 5.5183,
      "mean_token_accuracy": 0.15871499180793763,
      "num_tokens": 19526468.0,
      "step": 9405
    },
    {
      "entropy": 5.786588859558106,
      "epoch": 0.8535921625544267,
      "grad_norm": 3.109375,
      "learning_rate": 0.000493451540633577,
      "loss": 5.5343,
      "mean_token_accuracy": 0.160228730738163,
      "num_tokens": 19535977.0,
      "step": 9410
    },
    {
      "entropy": 5.721152448654175,
      "epoch": 0.8540457184325109,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004934437889298651,
      "loss": 5.5079,
      "mean_token_accuracy": 0.15876430720090867,
      "num_tokens": 19545617.0,
      "step": 9415
    },
    {
      "entropy": 5.724825286865235,
      "epoch": 0.854499274310595,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004934360327086799,
      "loss": 5.653,
      "mean_token_accuracy": 0.15359345376491546,
      "num_tokens": 19554976.0,
      "step": 9420
    },
    {
      "entropy": 5.815794324874878,
      "epoch": 0.8549528301886793,
      "grad_norm": 2.390625,
      "learning_rate": 0.000493428271970182,
      "loss": 5.6908,
      "mean_token_accuracy": 0.15186690390110016,
      "num_tokens": 19565431.0,
      "step": 9425
    },
    {
      "entropy": 5.908529567718506,
      "epoch": 0.8554063860667634,
      "grad_norm": 2.375,
      "learning_rate": 0.0004934205067145318,
      "loss": 5.6843,
      "mean_token_accuracy": 0.15194858014583587,
      "num_tokens": 19576280.0,
      "step": 9430
    },
    {
      "entropy": 5.827503490447998,
      "epoch": 0.8558599419448476,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004934127369418898,
      "loss": 5.587,
      "mean_token_accuracy": 0.1574860021471977,
      "num_tokens": 19587058.0,
      "step": 9435
    },
    {
      "entropy": 5.712215805053711,
      "epoch": 0.8563134978229318,
      "grad_norm": 2.25,
      "learning_rate": 0.0004934049626524168,
      "loss": 5.5281,
      "mean_token_accuracy": 0.15933785885572432,
      "num_tokens": 19596606.0,
      "step": 9440
    },
    {
      "entropy": 5.731781482696533,
      "epoch": 0.8567670537010159,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004933971838462734,
      "loss": 5.616,
      "mean_token_accuracy": 0.15793745815753937,
      "num_tokens": 19607060.0,
      "step": 9445
    },
    {
      "entropy": 5.7539434909820555,
      "epoch": 0.8572206095791002,
      "grad_norm": 2.125,
      "learning_rate": 0.0004933894005236207,
      "loss": 5.587,
      "mean_token_accuracy": 0.15431178510189056,
      "num_tokens": 19617378.0,
      "step": 9450
    },
    {
      "entropy": 5.874515628814697,
      "epoch": 0.8576741654571843,
      "grad_norm": 6.6875,
      "learning_rate": 0.0004933816126846195,
      "loss": 5.6567,
      "mean_token_accuracy": 0.1632300779223442,
      "num_tokens": 19627071.0,
      "step": 9455
    },
    {
      "entropy": 5.7735847473144535,
      "epoch": 0.8581277213352685,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004933738203294309,
      "loss": 5.5182,
      "mean_token_accuracy": 0.15498260110616685,
      "num_tokens": 19637491.0,
      "step": 9460
    },
    {
      "entropy": 5.76154260635376,
      "epoch": 0.8585812772133526,
      "grad_norm": 3.546875,
      "learning_rate": 0.000493366023458216,
      "loss": 5.5862,
      "mean_token_accuracy": 0.16241435706615448,
      "num_tokens": 19647062.0,
      "step": 9465
    },
    {
      "entropy": 5.8701849460601805,
      "epoch": 0.8590348330914369,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004933582220711362,
      "loss": 5.7338,
      "mean_token_accuracy": 0.15406497269868852,
      "num_tokens": 19657753.0,
      "step": 9470
    },
    {
      "entropy": 5.758343648910523,
      "epoch": 0.8594883889695211,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004933504161683526,
      "loss": 5.5168,
      "mean_token_accuracy": 0.16479631960392,
      "num_tokens": 19667472.0,
      "step": 9475
    },
    {
      "entropy": 5.823338270187378,
      "epoch": 0.8599419448476052,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004933426057500269,
      "loss": 5.6487,
      "mean_token_accuracy": 0.15531134903430938,
      "num_tokens": 19677847.0,
      "step": 9480
    },
    {
      "entropy": 5.757506895065307,
      "epoch": 0.8603955007256894,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004933347908163203,
      "loss": 5.5647,
      "mean_token_accuracy": 0.15997239351272582,
      "num_tokens": 19687795.0,
      "step": 9485
    },
    {
      "entropy": 5.74517970085144,
      "epoch": 0.8608490566037735,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004933269713673947,
      "loss": 5.5521,
      "mean_token_accuracy": 0.1630155086517334,
      "num_tokens": 19698174.0,
      "step": 9490
    },
    {
      "entropy": 5.7923808097839355,
      "epoch": 0.8613026124818578,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004933191474034116,
      "loss": 5.6266,
      "mean_token_accuracy": 0.15424173325300217,
      "num_tokens": 19708658.0,
      "step": 9495
    },
    {
      "entropy": 5.824654769897461,
      "epoch": 0.861756168359942,
      "grad_norm": 1.875,
      "learning_rate": 0.000493311318924533,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1592453107237816,
      "num_tokens": 19719068.0,
      "step": 9500
    },
    {
      "entropy": 5.773280429840088,
      "epoch": 0.8622097242380261,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004933034859309208,
      "loss": 5.536,
      "mean_token_accuracy": 0.16038210988044738,
      "num_tokens": 19729611.0,
      "step": 9505
    },
    {
      "entropy": 5.717344141006469,
      "epoch": 0.8626632801161103,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004932956484227367,
      "loss": 5.5654,
      "mean_token_accuracy": 0.15528836846351624,
      "num_tokens": 19739772.0,
      "step": 9510
    },
    {
      "entropy": 5.761060428619385,
      "epoch": 0.8631168359941945,
      "grad_norm": 2.609375,
      "learning_rate": 0.000493287806400143,
      "loss": 5.5981,
      "mean_token_accuracy": 0.16070275753736496,
      "num_tokens": 19749360.0,
      "step": 9515
    },
    {
      "entropy": 5.776277732849121,
      "epoch": 0.8635703918722787,
      "grad_norm": 2.125,
      "learning_rate": 0.0004932799598633018,
      "loss": 5.5583,
      "mean_token_accuracy": 0.1544738754630089,
      "num_tokens": 19759803.0,
      "step": 9520
    },
    {
      "entropy": 5.845948219299316,
      "epoch": 0.8640239477503628,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004932721088123753,
      "loss": 5.6368,
      "mean_token_accuracy": 0.1543733984231949,
      "num_tokens": 19771357.0,
      "step": 9525
    },
    {
      "entropy": 5.771435499191284,
      "epoch": 0.864477503628447,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004932642532475262,
      "loss": 5.5486,
      "mean_token_accuracy": 0.15931447744369506,
      "num_tokens": 19781654.0,
      "step": 9530
    },
    {
      "entropy": 5.724986839294433,
      "epoch": 0.8649310595065312,
      "grad_norm": 2.5,
      "learning_rate": 0.0004932563931689168,
      "loss": 5.6259,
      "mean_token_accuracy": 0.15499583929777144,
      "num_tokens": 19792188.0,
      "step": 9535
    },
    {
      "entropy": 5.733976268768311,
      "epoch": 0.8653846153846154,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004932485285767093,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16302815675735474,
      "num_tokens": 19802442.0,
      "step": 9540
    },
    {
      "entropy": 5.718762636184692,
      "epoch": 0.8658381712626996,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004932406594710667,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1554040014743805,
      "num_tokens": 19813214.0,
      "step": 9545
    },
    {
      "entropy": 5.6916275978088375,
      "epoch": 0.8662917271407837,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004932327858521517,
      "loss": 5.5866,
      "mean_token_accuracy": 0.1576170027256012,
      "num_tokens": 19824771.0,
      "step": 9550
    },
    {
      "entropy": 5.7513017654418945,
      "epoch": 0.8667452830188679,
      "grad_norm": 2.0625,
      "learning_rate": 0.000493224907720127,
      "loss": 5.5695,
      "mean_token_accuracy": 0.162931127846241,
      "num_tokens": 19834047.0,
      "step": 9555
    },
    {
      "entropy": 5.827654314041138,
      "epoch": 0.8671988388969522,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004932170250751556,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1605009138584137,
      "num_tokens": 19843868.0,
      "step": 9560
    },
    {
      "entropy": 5.783278512954712,
      "epoch": 0.8676523947750363,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004932091379174006,
      "loss": 5.7224,
      "mean_token_accuracy": 0.14755918383598327,
      "num_tokens": 19854764.0,
      "step": 9565
    },
    {
      "entropy": 5.796078443527222,
      "epoch": 0.8681059506531205,
      "grad_norm": 3.21875,
      "learning_rate": 0.000493201246247025,
      "loss": 5.6008,
      "mean_token_accuracy": 0.1576782613992691,
      "num_tokens": 19866189.0,
      "step": 9570
    },
    {
      "entropy": 5.901652526855469,
      "epoch": 0.8685595065312046,
      "grad_norm": 2.09375,
      "learning_rate": 0.000493193350064192,
      "loss": 5.6786,
      "mean_token_accuracy": 0.15492264330387115,
      "num_tokens": 19876800.0,
      "step": 9575
    },
    {
      "entropy": 5.7523252964019775,
      "epoch": 0.8690130624092888,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004931854493690649,
      "loss": 5.5915,
      "mean_token_accuracy": 0.1607486441731453,
      "num_tokens": 19887330.0,
      "step": 9580
    },
    {
      "entropy": 5.815252017974854,
      "epoch": 0.869466618287373,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004931775441618071,
      "loss": 5.6539,
      "mean_token_accuracy": 0.15631003677845,
      "num_tokens": 19897464.0,
      "step": 9585
    },
    {
      "entropy": 5.716238689422608,
      "epoch": 0.8699201741654572,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004931696344425821,
      "loss": 5.5648,
      "mean_token_accuracy": 0.15558705255389213,
      "num_tokens": 19908325.0,
      "step": 9590
    },
    {
      "entropy": 5.784364986419678,
      "epoch": 0.8703737300435413,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004931617202115535,
      "loss": 5.6063,
      "mean_token_accuracy": 0.15765802562236786,
      "num_tokens": 19918847.0,
      "step": 9595
    },
    {
      "entropy": 5.728229761123657,
      "epoch": 0.8708272859216255,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004931538014688849,
      "loss": 5.548,
      "mean_token_accuracy": 0.16058621555566788,
      "num_tokens": 19929177.0,
      "step": 9600
    },
    {
      "entropy": 5.700157022476196,
      "epoch": 0.8712808417997098,
      "grad_norm": 3.25,
      "learning_rate": 0.0004931458782147401,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1719153866171837,
      "num_tokens": 19940878.0,
      "step": 9605
    },
    {
      "entropy": 5.760498952865601,
      "epoch": 0.8717343976777939,
      "grad_norm": 4.09375,
      "learning_rate": 0.000493137950449283,
      "loss": 5.6104,
      "mean_token_accuracy": 0.1593600779771805,
      "num_tokens": 19950314.0,
      "step": 9610
    },
    {
      "entropy": 5.7314128398895265,
      "epoch": 0.8721879535558781,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004931300181726775,
      "loss": 5.642,
      "mean_token_accuracy": 0.1519266977906227,
      "num_tokens": 19960405.0,
      "step": 9615
    },
    {
      "entropy": 5.804286861419678,
      "epoch": 0.8726415094339622,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004931220813850875,
      "loss": 5.6097,
      "mean_token_accuracy": 0.16163116693496704,
      "num_tokens": 19970667.0,
      "step": 9620
    },
    {
      "entropy": 5.721092748641968,
      "epoch": 0.8730950653120464,
      "grad_norm": 3.25,
      "learning_rate": 0.0004931141400866773,
      "loss": 5.5032,
      "mean_token_accuracy": 0.1610335499048233,
      "num_tokens": 19980109.0,
      "step": 9625
    },
    {
      "entropy": 5.761779069900513,
      "epoch": 0.8735486211901307,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004931061942776112,
      "loss": 5.6322,
      "mean_token_accuracy": 0.1590443417429924,
      "num_tokens": 19989678.0,
      "step": 9630
    },
    {
      "entropy": 5.7973034381866455,
      "epoch": 0.8740021770682148,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004930982439580534,
      "loss": 5.6777,
      "mean_token_accuracy": 0.15426260232925415,
      "num_tokens": 19999889.0,
      "step": 9635
    },
    {
      "entropy": 5.805342197418213,
      "epoch": 0.874455732946299,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004930902891281683,
      "loss": 5.5839,
      "mean_token_accuracy": 0.16272704303264618,
      "num_tokens": 20009483.0,
      "step": 9640
    },
    {
      "entropy": 5.818941307067871,
      "epoch": 0.8749092888243831,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004930823297881205,
      "loss": 5.5941,
      "mean_token_accuracy": 0.16335407346487046,
      "num_tokens": 20019240.0,
      "step": 9645
    },
    {
      "entropy": 5.793977880477906,
      "epoch": 0.8753628447024674,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004930743659380745,
      "loss": 5.5909,
      "mean_token_accuracy": 0.15915590524673462,
      "num_tokens": 20030437.0,
      "step": 9650
    },
    {
      "entropy": 5.777416133880616,
      "epoch": 0.8758164005805515,
      "grad_norm": 2.765625,
      "learning_rate": 0.000493066397578195,
      "loss": 5.5893,
      "mean_token_accuracy": 0.16071107983589172,
      "num_tokens": 20041706.0,
      "step": 9655
    },
    {
      "entropy": 5.808022689819336,
      "epoch": 0.8762699564586357,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004930584247086468,
      "loss": 5.6608,
      "mean_token_accuracy": 0.15009503960609435,
      "num_tokens": 20051780.0,
      "step": 9660
    },
    {
      "entropy": 5.731687116622925,
      "epoch": 0.8767235123367199,
      "grad_norm": 2.46875,
      "learning_rate": 0.000493050447329595,
      "loss": 5.5504,
      "mean_token_accuracy": 0.16191339939832688,
      "num_tokens": 20061912.0,
      "step": 9665
    },
    {
      "entropy": 5.76660647392273,
      "epoch": 0.877177068214804,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004930424654412043,
      "loss": 5.6892,
      "mean_token_accuracy": 0.1501455396413803,
      "num_tokens": 20072691.0,
      "step": 9670
    },
    {
      "entropy": 5.813216924667358,
      "epoch": 0.8776306240928883,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004930344790436398,
      "loss": 5.5371,
      "mean_token_accuracy": 0.16547645032405853,
      "num_tokens": 20082165.0,
      "step": 9675
    },
    {
      "entropy": 5.814589977264404,
      "epoch": 0.8780841799709724,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004930264881370666,
      "loss": 5.5713,
      "mean_token_accuracy": 0.16154917776584626,
      "num_tokens": 20092840.0,
      "step": 9680
    },
    {
      "entropy": 5.805503416061401,
      "epoch": 0.8785377358490566,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004930184927216504,
      "loss": 5.6238,
      "mean_token_accuracy": 0.1623796045780182,
      "num_tokens": 20102338.0,
      "step": 9685
    },
    {
      "entropy": 5.770312309265137,
      "epoch": 0.8789912917271407,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004930104927975559,
      "loss": 5.5845,
      "mean_token_accuracy": 0.15509213283658027,
      "num_tokens": 20113944.0,
      "step": 9690
    },
    {
      "entropy": 5.803833150863648,
      "epoch": 0.879444847605225,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004930024883649489,
      "loss": 5.651,
      "mean_token_accuracy": 0.1510723277926445,
      "num_tokens": 20125548.0,
      "step": 9695
    },
    {
      "entropy": 5.768340015411377,
      "epoch": 0.8798984034833092,
      "grad_norm": 3.90625,
      "learning_rate": 0.000492994479423995,
      "loss": 5.5779,
      "mean_token_accuracy": 0.15510473996400834,
      "num_tokens": 20136437.0,
      "step": 9700
    },
    {
      "entropy": 5.7656957626342775,
      "epoch": 0.8803519593613933,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004929864659748597,
      "loss": 5.6108,
      "mean_token_accuracy": 0.1533298522233963,
      "num_tokens": 20146585.0,
      "step": 9705
    },
    {
      "entropy": 5.813033246994019,
      "epoch": 0.8808055152394775,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004929784480177087,
      "loss": 5.57,
      "mean_token_accuracy": 0.15888408571481705,
      "num_tokens": 20156882.0,
      "step": 9710
    },
    {
      "entropy": 5.763446950912476,
      "epoch": 0.8812590711175616,
      "grad_norm": 2.125,
      "learning_rate": 0.0004929704255527078,
      "loss": 5.6782,
      "mean_token_accuracy": 0.1537492588162422,
      "num_tokens": 20167885.0,
      "step": 9715
    },
    {
      "entropy": 5.762321615219117,
      "epoch": 0.8817126269956459,
      "grad_norm": 2.984375,
      "learning_rate": 0.000492962398580023,
      "loss": 5.6116,
      "mean_token_accuracy": 0.16395186334848405,
      "num_tokens": 20177860.0,
      "step": 9720
    },
    {
      "entropy": 5.716923189163208,
      "epoch": 0.88216618287373,
      "grad_norm": 2.5,
      "learning_rate": 0.0004929543670998203,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1595002979040146,
      "num_tokens": 20188192.0,
      "step": 9725
    },
    {
      "entropy": 5.767555141448975,
      "epoch": 0.8826197387518142,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004929463311122657,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1648905321955681,
      "num_tokens": 20198164.0,
      "step": 9730
    },
    {
      "entropy": 5.78915376663208,
      "epoch": 0.8830732946298984,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004929382906175253,
      "loss": 5.6113,
      "mean_token_accuracy": 0.15980602204799652,
      "num_tokens": 20208799.0,
      "step": 9735
    },
    {
      "entropy": 5.778161287307739,
      "epoch": 0.8835268505079826,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004929302456157657,
      "loss": 5.6503,
      "mean_token_accuracy": 0.1607613205909729,
      "num_tokens": 20219666.0,
      "step": 9740
    },
    {
      "entropy": 5.7221434116363525,
      "epoch": 0.8839804063860668,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004929221961071531,
      "loss": 5.5635,
      "mean_token_accuracy": 0.15667997747659684,
      "num_tokens": 20229118.0,
      "step": 9745
    },
    {
      "entropy": 5.880081701278686,
      "epoch": 0.8844339622641509,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004929141420918538,
      "loss": 5.6855,
      "mean_token_accuracy": 0.15291736274957657,
      "num_tokens": 20240925.0,
      "step": 9750
    },
    {
      "entropy": 5.78455958366394,
      "epoch": 0.8848875181422351,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004929060835700347,
      "loss": 5.5856,
      "mean_token_accuracy": 0.1504445791244507,
      "num_tokens": 20250244.0,
      "step": 9755
    },
    {
      "entropy": 5.693233871459961,
      "epoch": 0.8853410740203193,
      "grad_norm": 2.671875,
      "learning_rate": 0.000492898020541862,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1591622620820999,
      "num_tokens": 20260947.0,
      "step": 9760
    },
    {
      "entropy": 5.6994953632354735,
      "epoch": 0.8857946298984035,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004928899530075029,
      "loss": 5.498,
      "mean_token_accuracy": 0.17445793598890305,
      "num_tokens": 20270666.0,
      "step": 9765
    },
    {
      "entropy": 5.679705047607422,
      "epoch": 0.8862481857764877,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004928818809671241,
      "loss": 5.5605,
      "mean_token_accuracy": 0.15611713379621506,
      "num_tokens": 20282100.0,
      "step": 9770
    },
    {
      "entropy": 5.657387399673462,
      "epoch": 0.8867017416545718,
      "grad_norm": 9.3125,
      "learning_rate": 0.0004928738044208923,
      "loss": 5.4528,
      "mean_token_accuracy": 0.1579565018415451,
      "num_tokens": 20293114.0,
      "step": 9775
    },
    {
      "entropy": 5.708540773391723,
      "epoch": 0.887155297532656,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004928657233689748,
      "loss": 5.5515,
      "mean_token_accuracy": 0.16061361730098725,
      "num_tokens": 20303325.0,
      "step": 9780
    },
    {
      "entropy": 5.797102165222168,
      "epoch": 0.8876088534107403,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004928576378115387,
      "loss": 5.6114,
      "mean_token_accuracy": 0.15374773889780044,
      "num_tokens": 20313959.0,
      "step": 9785
    },
    {
      "entropy": 5.750910425186158,
      "epoch": 0.8880624092888244,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004928495477487509,
      "loss": 5.4853,
      "mean_token_accuracy": 0.16549808382987977,
      "num_tokens": 20323253.0,
      "step": 9790
    },
    {
      "entropy": 5.734001302719117,
      "epoch": 0.8885159651669086,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004928414531807792,
      "loss": 5.6341,
      "mean_token_accuracy": 0.15663368105888367,
      "num_tokens": 20333798.0,
      "step": 9795
    },
    {
      "entropy": 5.731454753875733,
      "epoch": 0.8889695210449927,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004928333541077906,
      "loss": 5.5311,
      "mean_token_accuracy": 0.15974932461977004,
      "num_tokens": 20343735.0,
      "step": 9800
    },
    {
      "entropy": 5.777715635299683,
      "epoch": 0.8894230769230769,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004928252505299527,
      "loss": 5.545,
      "mean_token_accuracy": 0.15425665825605392,
      "num_tokens": 20353876.0,
      "step": 9805
    },
    {
      "entropy": 5.753269290924072,
      "epoch": 0.8898766328011611,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004928171424474332,
      "loss": 5.5575,
      "mean_token_accuracy": 0.16936624497175218,
      "num_tokens": 20363590.0,
      "step": 9810
    },
    {
      "entropy": 5.729820632934571,
      "epoch": 0.8903301886792453,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004928090298603996,
      "loss": 5.5634,
      "mean_token_accuracy": 0.1631142571568489,
      "num_tokens": 20374621.0,
      "step": 9815
    },
    {
      "entropy": 5.742217540740967,
      "epoch": 0.8907837445573294,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004928009127690199,
      "loss": 5.5922,
      "mean_token_accuracy": 0.15664929151535034,
      "num_tokens": 20385473.0,
      "step": 9820
    },
    {
      "entropy": 5.760408592224121,
      "epoch": 0.8912373004354136,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004927927911734617,
      "loss": 5.5394,
      "mean_token_accuracy": 0.16518695801496505,
      "num_tokens": 20395581.0,
      "step": 9825
    },
    {
      "entropy": 5.691333436965943,
      "epoch": 0.8916908563134979,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004927846650738931,
      "loss": 5.5448,
      "mean_token_accuracy": 0.15757792294025422,
      "num_tokens": 20405584.0,
      "step": 9830
    },
    {
      "entropy": 5.735745763778686,
      "epoch": 0.892144412191582,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004927765344704821,
      "loss": 5.6676,
      "mean_token_accuracy": 0.1564056694507599,
      "num_tokens": 20416564.0,
      "step": 9835
    },
    {
      "entropy": 5.705387163162231,
      "epoch": 0.8925979680696662,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004927683993633969,
      "loss": 5.5763,
      "mean_token_accuracy": 0.1650102511048317,
      "num_tokens": 20426819.0,
      "step": 9840
    },
    {
      "entropy": 5.733826971054077,
      "epoch": 0.8930515239477503,
      "grad_norm": 6.0625,
      "learning_rate": 0.0004927602597528057,
      "loss": 5.5938,
      "mean_token_accuracy": 0.1572804979979992,
      "num_tokens": 20438232.0,
      "step": 9845
    },
    {
      "entropy": 5.772125101089477,
      "epoch": 0.8935050798258345,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004927521156388769,
      "loss": 5.6274,
      "mean_token_accuracy": 0.1541820600628853,
      "num_tokens": 20448821.0,
      "step": 9850
    },
    {
      "entropy": 5.793784284591675,
      "epoch": 0.8939586357039188,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004927439670217788,
      "loss": 5.5961,
      "mean_token_accuracy": 0.1655206337571144,
      "num_tokens": 20458414.0,
      "step": 9855
    },
    {
      "entropy": 5.816038131713867,
      "epoch": 0.8944121915820029,
      "grad_norm": 2.65625,
      "learning_rate": 0.00049273581390168,
      "loss": 5.5755,
      "mean_token_accuracy": 0.15802491158246995,
      "num_tokens": 20468722.0,
      "step": 9860
    },
    {
      "entropy": 5.699888753890991,
      "epoch": 0.8948657474600871,
      "grad_norm": 2.4375,
      "learning_rate": 0.000492727656278749,
      "loss": 5.5023,
      "mean_token_accuracy": 0.15789415538311005,
      "num_tokens": 20479917.0,
      "step": 9865
    },
    {
      "entropy": 5.764579629898071,
      "epoch": 0.8953193033381712,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004927194941531547,
      "loss": 5.6212,
      "mean_token_accuracy": 0.15586561113595962,
      "num_tokens": 20489537.0,
      "step": 9870
    },
    {
      "entropy": 5.741041851043701,
      "epoch": 0.8957728592162555,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004927113275250657,
      "loss": 5.5907,
      "mean_token_accuracy": 0.15124815702438354,
      "num_tokens": 20500855.0,
      "step": 9875
    },
    {
      "entropy": 5.736967849731445,
      "epoch": 0.8962264150943396,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004927031563946511,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1567249208688736,
      "num_tokens": 20511349.0,
      "step": 9880
    },
    {
      "entropy": 5.7952371597290036,
      "epoch": 0.8966799709724238,
      "grad_norm": 3.0,
      "learning_rate": 0.0004926949807620796,
      "loss": 5.6202,
      "mean_token_accuracy": 0.15793016850948333,
      "num_tokens": 20521886.0,
      "step": 9885
    },
    {
      "entropy": 5.801976299285888,
      "epoch": 0.897133526850508,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004926868006275204,
      "loss": 5.5804,
      "mean_token_accuracy": 0.1503107041120529,
      "num_tokens": 20531207.0,
      "step": 9890
    },
    {
      "entropy": 5.732528829574585,
      "epoch": 0.8975870827285921,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004926786159911427,
      "loss": 5.6059,
      "mean_token_accuracy": 0.15419580042362213,
      "num_tokens": 20540854.0,
      "step": 9895
    },
    {
      "entropy": 5.723742437362671,
      "epoch": 0.8980406386066764,
      "grad_norm": 2.46875,
      "learning_rate": 0.000492670426853116,
      "loss": 5.5523,
      "mean_token_accuracy": 0.15889148712158202,
      "num_tokens": 20550404.0,
      "step": 9900
    },
    {
      "entropy": 5.715986490249634,
      "epoch": 0.8984941944847605,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004926622332136092,
      "loss": 5.6393,
      "mean_token_accuracy": 0.15383820980787277,
      "num_tokens": 20560478.0,
      "step": 9905
    },
    {
      "entropy": 5.836056280136108,
      "epoch": 0.8989477503628447,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004926540350727918,
      "loss": 5.7131,
      "mean_token_accuracy": 0.1518503852188587,
      "num_tokens": 20571944.0,
      "step": 9910
    },
    {
      "entropy": 5.8559455394744875,
      "epoch": 0.8994013062409288,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004926458324308337,
      "loss": 5.6337,
      "mean_token_accuracy": 0.1575888216495514,
      "num_tokens": 20582035.0,
      "step": 9915
    },
    {
      "entropy": 5.725993013381958,
      "epoch": 0.8998548621190131,
      "grad_norm": 4.0,
      "learning_rate": 0.0004926376252879043,
      "loss": 5.5562,
      "mean_token_accuracy": 0.16676303446292878,
      "num_tokens": 20592772.0,
      "step": 9920
    },
    {
      "entropy": 5.7645509243011475,
      "epoch": 0.9003084179970973,
      "grad_norm": 2.125,
      "learning_rate": 0.0004926294136441733,
      "loss": 5.6714,
      "mean_token_accuracy": 0.15678180009126663,
      "num_tokens": 20603905.0,
      "step": 9925
    },
    {
      "entropy": 5.8177484512329105,
      "epoch": 0.9007619738751814,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004926211974998106,
      "loss": 5.6132,
      "mean_token_accuracy": 0.15980715155601502,
      "num_tokens": 20614030.0,
      "step": 9930
    },
    {
      "entropy": 5.827275085449219,
      "epoch": 0.9012155297532656,
      "grad_norm": 2.21875,
      "learning_rate": 0.000492612976854986,
      "loss": 5.707,
      "mean_token_accuracy": 0.15005202889442443,
      "num_tokens": 20625065.0,
      "step": 9935
    },
    {
      "entropy": 5.738382196426391,
      "epoch": 0.9016690856313497,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004926047517098697,
      "loss": 5.5821,
      "mean_token_accuracy": 0.16158285439014436,
      "num_tokens": 20635109.0,
      "step": 9940
    },
    {
      "entropy": 5.671261310577393,
      "epoch": 0.902122641509434,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004925965220646315,
      "loss": 5.5381,
      "mean_token_accuracy": 0.16070961207151413,
      "num_tokens": 20645591.0,
      "step": 9945
    },
    {
      "entropy": 5.719666433334351,
      "epoch": 0.9025761973875182,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004925882879194419,
      "loss": 5.4918,
      "mean_token_accuracy": 0.16208715736865997,
      "num_tokens": 20655363.0,
      "step": 9950
    },
    {
      "entropy": 5.804986095428466,
      "epoch": 0.9030297532656023,
      "grad_norm": 3.0,
      "learning_rate": 0.0004925800492744712,
      "loss": 5.5852,
      "mean_token_accuracy": 0.15884936302900315,
      "num_tokens": 20666612.0,
      "step": 9955
    },
    {
      "entropy": 5.772980499267578,
      "epoch": 0.9034833091436865,
      "grad_norm": 3.125,
      "learning_rate": 0.0004925718061298895,
      "loss": 5.671,
      "mean_token_accuracy": 0.15561782270669938,
      "num_tokens": 20677752.0,
      "step": 9960
    },
    {
      "entropy": 5.703854322433472,
      "epoch": 0.9039368650217707,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004925635584858675,
      "loss": 5.5748,
      "mean_token_accuracy": 0.15889802277088166,
      "num_tokens": 20687964.0,
      "step": 9965
    },
    {
      "entropy": 5.761281108856201,
      "epoch": 0.9043904208998549,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004925553063425755,
      "loss": 5.5619,
      "mean_token_accuracy": 0.16269438713788986,
      "num_tokens": 20698404.0,
      "step": 9970
    },
    {
      "entropy": 5.697084140777588,
      "epoch": 0.904843976777939,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004925470497001844,
      "loss": 5.4898,
      "mean_token_accuracy": 0.16590061336755751,
      "num_tokens": 20708789.0,
      "step": 9975
    },
    {
      "entropy": 5.719874048233033,
      "epoch": 0.9052975326560232,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004925387885588648,
      "loss": 5.5746,
      "mean_token_accuracy": 0.1578325554728508,
      "num_tokens": 20719207.0,
      "step": 9980
    },
    {
      "entropy": 5.8075329780578615,
      "epoch": 0.9057510885341074,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004925305229187877,
      "loss": 5.6434,
      "mean_token_accuracy": 0.15681922882795335,
      "num_tokens": 20730171.0,
      "step": 9985
    },
    {
      "entropy": 5.88227391242981,
      "epoch": 0.9062046444121916,
      "grad_norm": 3.5,
      "learning_rate": 0.000492522252780124,
      "loss": 5.7143,
      "mean_token_accuracy": 0.1521185390651226,
      "num_tokens": 20741212.0,
      "step": 9990
    },
    {
      "entropy": 5.824160528182984,
      "epoch": 0.9066582002902758,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004925139781430446,
      "loss": 5.6392,
      "mean_token_accuracy": 0.15082903504371642,
      "num_tokens": 20752424.0,
      "step": 9995
    },
    {
      "entropy": 5.785498142242432,
      "epoch": 0.9071117561683599,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004925056990077207,
      "loss": 5.6224,
      "mean_token_accuracy": 0.15814245343208314,
      "num_tokens": 20763198.0,
      "step": 10000
    },
    {
      "entropy": 5.741479301452637,
      "epoch": 0.9075653120464441,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004924974153743235,
      "loss": 5.5897,
      "mean_token_accuracy": 0.15915692299604417,
      "num_tokens": 20773957.0,
      "step": 10005
    },
    {
      "entropy": 5.788909578323365,
      "epoch": 0.9080188679245284,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004924891272430244,
      "loss": 5.5763,
      "mean_token_accuracy": 0.15208770185709,
      "num_tokens": 20784643.0,
      "step": 10010
    },
    {
      "entropy": 5.68543496131897,
      "epoch": 0.9084724238026125,
      "grad_norm": 2.5,
      "learning_rate": 0.0004924808346139948,
      "loss": 5.4338,
      "mean_token_accuracy": 0.16410207599401475,
      "num_tokens": 20794805.0,
      "step": 10015
    },
    {
      "entropy": 5.744233226776123,
      "epoch": 0.9089259796806967,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004924725374874059,
      "loss": 5.5307,
      "mean_token_accuracy": 0.16546066105365753,
      "num_tokens": 20807132.0,
      "step": 10020
    },
    {
      "entropy": 5.734099721908569,
      "epoch": 0.9093795355587808,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004924642358634297,
      "loss": 5.532,
      "mean_token_accuracy": 0.16355610489845276,
      "num_tokens": 20816683.0,
      "step": 10025
    },
    {
      "entropy": 5.702745008468628,
      "epoch": 0.909833091436865,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004924559297422376,
      "loss": 5.5731,
      "mean_token_accuracy": 0.1543893814086914,
      "num_tokens": 20827390.0,
      "step": 10030
    },
    {
      "entropy": 5.754718494415283,
      "epoch": 0.9102866473149492,
      "grad_norm": 4.0,
      "learning_rate": 0.0004924476191240014,
      "loss": 5.5452,
      "mean_token_accuracy": 0.1629335328936577,
      "num_tokens": 20837204.0,
      "step": 10035
    },
    {
      "entropy": 5.728320837020874,
      "epoch": 0.9107402031930334,
      "grad_norm": 2.625,
      "learning_rate": 0.000492439304008893,
      "loss": 5.5349,
      "mean_token_accuracy": 0.16521793901920317,
      "num_tokens": 20846250.0,
      "step": 10040
    },
    {
      "entropy": 5.712874174118042,
      "epoch": 0.9111937590711175,
      "grad_norm": 3.75,
      "learning_rate": 0.0004924309843970847,
      "loss": 5.6773,
      "mean_token_accuracy": 0.14820607528090476,
      "num_tokens": 20856351.0,
      "step": 10045
    },
    {
      "entropy": 5.873688220977783,
      "epoch": 0.9116473149492017,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004924226602887479,
      "loss": 5.7037,
      "mean_token_accuracy": 0.14887718558311464,
      "num_tokens": 20865889.0,
      "step": 10050
    },
    {
      "entropy": 5.851448631286621,
      "epoch": 0.912100870827286,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004924143316840552,
      "loss": 5.6439,
      "mean_token_accuracy": 0.1539697453379631,
      "num_tokens": 20876977.0,
      "step": 10055
    },
    {
      "entropy": 5.807049989700317,
      "epoch": 0.9125544267053701,
      "grad_norm": 4.03125,
      "learning_rate": 0.0004924059985831787,
      "loss": 5.6156,
      "mean_token_accuracy": 0.15294692516326905,
      "num_tokens": 20888302.0,
      "step": 10060
    },
    {
      "entropy": 5.701662063598633,
      "epoch": 0.9130079825834543,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004923976609862908,
      "loss": 5.5576,
      "mean_token_accuracy": 0.15857545137405396,
      "num_tokens": 20898881.0,
      "step": 10065
    },
    {
      "entropy": 5.742940425872803,
      "epoch": 0.9134615384615384,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004923893188935638,
      "loss": 5.6034,
      "mean_token_accuracy": 0.1555649533867836,
      "num_tokens": 20910818.0,
      "step": 10070
    },
    {
      "entropy": 5.79479341506958,
      "epoch": 0.9139150943396226,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004923809723051702,
      "loss": 5.5446,
      "mean_token_accuracy": 0.15759580582380295,
      "num_tokens": 20919930.0,
      "step": 10075
    },
    {
      "entropy": 5.719135856628418,
      "epoch": 0.9143686502177069,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004923726212212829,
      "loss": 5.5202,
      "mean_token_accuracy": 0.1566787302494049,
      "num_tokens": 20930524.0,
      "step": 10080
    },
    {
      "entropy": 5.7719869136810305,
      "epoch": 0.914822206095791,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004923642656420742,
      "loss": 5.5749,
      "mean_token_accuracy": 0.15472060441970825,
      "num_tokens": 20939617.0,
      "step": 10085
    },
    {
      "entropy": 5.811985015869141,
      "epoch": 0.9152757619738752,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004923559055677173,
      "loss": 5.5612,
      "mean_token_accuracy": 0.15483544915914535,
      "num_tokens": 20949624.0,
      "step": 10090
    },
    {
      "entropy": 5.826816177368164,
      "epoch": 0.9157293178519593,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004923475409983848,
      "loss": 5.6449,
      "mean_token_accuracy": 0.14818937033414842,
      "num_tokens": 20961496.0,
      "step": 10095
    },
    {
      "entropy": 5.798296356201172,
      "epoch": 0.9161828737300436,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004923391719342497,
      "loss": 5.6068,
      "mean_token_accuracy": 0.15662551075220107,
      "num_tokens": 20970842.0,
      "step": 10100
    },
    {
      "entropy": 5.793449449539184,
      "epoch": 0.9166364296081277,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004923307983754852,
      "loss": 5.6034,
      "mean_token_accuracy": 0.16058154106140138,
      "num_tokens": 20981442.0,
      "step": 10105
    },
    {
      "entropy": 5.712341213226319,
      "epoch": 0.9170899854862119,
      "grad_norm": 3.375,
      "learning_rate": 0.0004923224203222645,
      "loss": 5.6194,
      "mean_token_accuracy": 0.15990293473005296,
      "num_tokens": 20991652.0,
      "step": 10110
    },
    {
      "entropy": 5.8263936042785645,
      "epoch": 0.9175435413642961,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004923140377747606,
      "loss": 5.6374,
      "mean_token_accuracy": 0.15728904753923417,
      "num_tokens": 21003147.0,
      "step": 10115
    },
    {
      "entropy": 5.766634225845337,
      "epoch": 0.9179970972423802,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004923056507331471,
      "loss": 5.5746,
      "mean_token_accuracy": 0.15852663964033126,
      "num_tokens": 21013123.0,
      "step": 10120
    },
    {
      "entropy": 5.7576963901519775,
      "epoch": 0.9184506531204645,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004922972591975974,
      "loss": 5.5503,
      "mean_token_accuracy": 0.1584336444735527,
      "num_tokens": 21023203.0,
      "step": 10125
    },
    {
      "entropy": 5.7687952518463135,
      "epoch": 0.9189042089985486,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004922888631682849,
      "loss": 5.5691,
      "mean_token_accuracy": 0.15644412040710448,
      "num_tokens": 21034068.0,
      "step": 10130
    },
    {
      "entropy": 5.7613122940063475,
      "epoch": 0.9193577648766328,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004922804626453834,
      "loss": 5.6061,
      "mean_token_accuracy": 0.15606467425823212,
      "num_tokens": 21044092.0,
      "step": 10135
    },
    {
      "entropy": 5.8270978927612305,
      "epoch": 0.9198113207547169,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004922720576290665,
      "loss": 5.714,
      "mean_token_accuracy": 0.14804730117321013,
      "num_tokens": 21056014.0,
      "step": 10140
    },
    {
      "entropy": 5.796032333374024,
      "epoch": 0.9202648766328012,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004922636481195082,
      "loss": 5.5853,
      "mean_token_accuracy": 0.15590243190526962,
      "num_tokens": 21066052.0,
      "step": 10145
    },
    {
      "entropy": 5.742114925384522,
      "epoch": 0.9207184325108854,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004922552341168823,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1572227418422699,
      "num_tokens": 21075321.0,
      "step": 10150
    },
    {
      "entropy": 5.762363767623901,
      "epoch": 0.9211719883889695,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004922468156213627,
      "loss": 5.6602,
      "mean_token_accuracy": 0.15420769900083542,
      "num_tokens": 21086130.0,
      "step": 10155
    },
    {
      "entropy": 5.792463064193726,
      "epoch": 0.9216255442670537,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004922383926331236,
      "loss": 5.6249,
      "mean_token_accuracy": 0.1511351302266121,
      "num_tokens": 21095994.0,
      "step": 10160
    },
    {
      "entropy": 5.775228452682495,
      "epoch": 0.9220791001451378,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004922299651523392,
      "loss": 5.6356,
      "mean_token_accuracy": 0.15297728925943374,
      "num_tokens": 21106554.0,
      "step": 10165
    },
    {
      "entropy": 5.831748723983765,
      "epoch": 0.9225326560232221,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004922215331791837,
      "loss": 5.6789,
      "mean_token_accuracy": 0.1488865226507187,
      "num_tokens": 21118047.0,
      "step": 10170
    },
    {
      "entropy": 5.831108951568604,
      "epoch": 0.9229862119013063,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004922130967138316,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1671900048851967,
      "num_tokens": 21127864.0,
      "step": 10175
    },
    {
      "entropy": 5.710483074188232,
      "epoch": 0.9234397677793904,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004922046557564574,
      "loss": 5.5351,
      "mean_token_accuracy": 0.15462752878665925,
      "num_tokens": 21137817.0,
      "step": 10180
    },
    {
      "entropy": 5.736862659454346,
      "epoch": 0.9238933236574746,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004921962103072355,
      "loss": 5.6293,
      "mean_token_accuracy": 0.15744008719921113,
      "num_tokens": 21147545.0,
      "step": 10185
    },
    {
      "entropy": 5.707754707336425,
      "epoch": 0.9243468795355588,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004921877603663406,
      "loss": 5.575,
      "mean_token_accuracy": 0.1553434267640114,
      "num_tokens": 21156458.0,
      "step": 10190
    },
    {
      "entropy": 5.796969556808472,
      "epoch": 0.924800435413643,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004921793059339474,
      "loss": 5.6146,
      "mean_token_accuracy": 0.15412977188825608,
      "num_tokens": 21165699.0,
      "step": 10195
    },
    {
      "entropy": 5.80735559463501,
      "epoch": 0.9252539912917271,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004921708470102309,
      "loss": 5.5835,
      "mean_token_accuracy": 0.16180741041898727,
      "num_tokens": 21175922.0,
      "step": 10200
    },
    {
      "entropy": 5.795233964920044,
      "epoch": 0.9257075471698113,
      "grad_norm": 2.53125,
      "learning_rate": 0.000492162383595366,
      "loss": 5.6801,
      "mean_token_accuracy": 0.15241990536451339,
      "num_tokens": 21187123.0,
      "step": 10205
    },
    {
      "entropy": 5.815777683258057,
      "epoch": 0.9261611030478955,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004921539156895275,
      "loss": 5.606,
      "mean_token_accuracy": 0.15472038686275483,
      "num_tokens": 21197298.0,
      "step": 10210
    },
    {
      "entropy": 5.759130859375,
      "epoch": 0.9266146589259797,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004921454432928908,
      "loss": 5.4791,
      "mean_token_accuracy": 0.16374792605638505,
      "num_tokens": 21206437.0,
      "step": 10215
    },
    {
      "entropy": 5.861420822143555,
      "epoch": 0.9270682148040639,
      "grad_norm": 2.5,
      "learning_rate": 0.000492136966405631,
      "loss": 5.7374,
      "mean_token_accuracy": 0.15485017597675324,
      "num_tokens": 21217106.0,
      "step": 10220
    },
    {
      "entropy": 5.706059837341309,
      "epoch": 0.927521770682148,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004921284850279233,
      "loss": 5.5796,
      "mean_token_accuracy": 0.16197270005941392,
      "num_tokens": 21228397.0,
      "step": 10225
    },
    {
      "entropy": 5.769831991195678,
      "epoch": 0.9279753265602322,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004921199991599433,
      "loss": 5.6177,
      "mean_token_accuracy": 0.15558795779943466,
      "num_tokens": 21239846.0,
      "step": 10230
    },
    {
      "entropy": 5.827559280395508,
      "epoch": 0.9284288824383164,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004921115088018665,
      "loss": 5.6064,
      "mean_token_accuracy": 0.1562991499900818,
      "num_tokens": 21250335.0,
      "step": 10235
    },
    {
      "entropy": 5.816254186630249,
      "epoch": 0.9288824383164006,
      "grad_norm": 3.3125,
      "learning_rate": 0.000492103013953868,
      "loss": 5.6339,
      "mean_token_accuracy": 0.1633976072072983,
      "num_tokens": 21260075.0,
      "step": 10240
    },
    {
      "entropy": 5.787323236465454,
      "epoch": 0.9293359941944848,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004920945146161242,
      "loss": 5.65,
      "mean_token_accuracy": 0.15700064599514008,
      "num_tokens": 21271144.0,
      "step": 10245
    },
    {
      "entropy": 5.813453245162964,
      "epoch": 0.9297895500725689,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004920860107888105,
      "loss": 5.6505,
      "mean_token_accuracy": 0.15433294624090194,
      "num_tokens": 21283353.0,
      "step": 10250
    },
    {
      "entropy": 5.755099153518676,
      "epoch": 0.9302431059506531,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004920775024721026,
      "loss": 5.5848,
      "mean_token_accuracy": 0.16015762835741043,
      "num_tokens": 21293949.0,
      "step": 10255
    },
    {
      "entropy": 5.713023853302002,
      "epoch": 0.9306966618287373,
      "grad_norm": 3.0,
      "learning_rate": 0.0004920689896661767,
      "loss": 5.5015,
      "mean_token_accuracy": 0.1701981395483017,
      "num_tokens": 21303725.0,
      "step": 10260
    },
    {
      "entropy": 5.735786437988281,
      "epoch": 0.9311502177068215,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004920604723712088,
      "loss": 5.5812,
      "mean_token_accuracy": 0.16076552867889404,
      "num_tokens": 21314320.0,
      "step": 10265
    },
    {
      "entropy": 5.713997173309326,
      "epoch": 0.9316037735849056,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004920519505873751,
      "loss": 5.5496,
      "mean_token_accuracy": 0.1650150254368782,
      "num_tokens": 21324593.0,
      "step": 10270
    },
    {
      "entropy": 5.8037193775177,
      "epoch": 0.9320573294629898,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004920434243148515,
      "loss": 5.6227,
      "mean_token_accuracy": 0.1541777528822422,
      "num_tokens": 21335548.0,
      "step": 10275
    },
    {
      "entropy": 5.785840082168579,
      "epoch": 0.9325108853410741,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004920348935538147,
      "loss": 5.6059,
      "mean_token_accuracy": 0.1548403188586235,
      "num_tokens": 21345430.0,
      "step": 10280
    },
    {
      "entropy": 5.792271614074707,
      "epoch": 0.9329644412191582,
      "grad_norm": 2.09375,
      "learning_rate": 0.000492026358304441,
      "loss": 5.5746,
      "mean_token_accuracy": 0.16436884850263594,
      "num_tokens": 21355240.0,
      "step": 10285
    },
    {
      "entropy": 5.786996459960937,
      "epoch": 0.9334179970972424,
      "grad_norm": 1.859375,
      "learning_rate": 0.000492017818566907,
      "loss": 5.5827,
      "mean_token_accuracy": 0.15481290817260743,
      "num_tokens": 21365152.0,
      "step": 10290
    },
    {
      "entropy": 5.72499270439148,
      "epoch": 0.9338715529753265,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004920092743413892,
      "loss": 5.6222,
      "mean_token_accuracy": 0.16012369394302367,
      "num_tokens": 21375857.0,
      "step": 10295
    },
    {
      "entropy": 5.754889488220215,
      "epoch": 0.9343251088534107,
      "grad_norm": 2.625,
      "learning_rate": 0.0004920007256280642,
      "loss": 5.5619,
      "mean_token_accuracy": 0.16167580634355544,
      "num_tokens": 21385669.0,
      "step": 10300
    },
    {
      "entropy": 5.766891765594482,
      "epoch": 0.934778664731495,
      "grad_norm": 4.03125,
      "learning_rate": 0.000491992172427109,
      "loss": 5.5203,
      "mean_token_accuracy": 0.16033307909965516,
      "num_tokens": 21395691.0,
      "step": 10305
    },
    {
      "entropy": 5.785992431640625,
      "epoch": 0.9352322206095791,
      "grad_norm": 2.875,
      "learning_rate": 0.0004919836147387002,
      "loss": 5.5498,
      "mean_token_accuracy": 0.1563458576798439,
      "num_tokens": 21406103.0,
      "step": 10310
    },
    {
      "entropy": 5.759999513626099,
      "epoch": 0.9356857764876633,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004919750525630151,
      "loss": 5.599,
      "mean_token_accuracy": 0.16221485137939454,
      "num_tokens": 21416778.0,
      "step": 10315
    },
    {
      "entropy": 5.77952127456665,
      "epoch": 0.9361393323657474,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004919664859002306,
      "loss": 5.587,
      "mean_token_accuracy": 0.16231129914522172,
      "num_tokens": 21426090.0,
      "step": 10320
    },
    {
      "entropy": 5.835115242004394,
      "epoch": 0.9365928882438317,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004919579147505239,
      "loss": 5.5766,
      "mean_token_accuracy": 0.15895685851573943,
      "num_tokens": 21435506.0,
      "step": 10325
    },
    {
      "entropy": 5.767234992980957,
      "epoch": 0.9370464441219158,
      "grad_norm": 2.25,
      "learning_rate": 0.0004919493391140721,
      "loss": 5.6474,
      "mean_token_accuracy": 0.15202571153640748,
      "num_tokens": 21446155.0,
      "step": 10330
    },
    {
      "entropy": 5.748478555679322,
      "epoch": 0.9375,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004919407589910529,
      "loss": 5.5911,
      "mean_token_accuracy": 0.1629505142569542,
      "num_tokens": 21456746.0,
      "step": 10335
    },
    {
      "entropy": 5.824591398239136,
      "epoch": 0.9379535558780842,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004919321743816434,
      "loss": 5.6103,
      "mean_token_accuracy": 0.15974183082580568,
      "num_tokens": 21467347.0,
      "step": 10340
    },
    {
      "entropy": 5.721471118927002,
      "epoch": 0.9384071117561683,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004919235852860213,
      "loss": 5.5102,
      "mean_token_accuracy": 0.16396656036376953,
      "num_tokens": 21477005.0,
      "step": 10345
    },
    {
      "entropy": 5.767013454437256,
      "epoch": 0.9388606676342526,
      "grad_norm": 2.25,
      "learning_rate": 0.0004919149917043641,
      "loss": 5.5806,
      "mean_token_accuracy": 0.1651973158121109,
      "num_tokens": 21486922.0,
      "step": 10350
    },
    {
      "entropy": 5.837851095199585,
      "epoch": 0.9393142235123367,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004919063936368496,
      "loss": 5.6697,
      "mean_token_accuracy": 0.15222235172986984,
      "num_tokens": 21498681.0,
      "step": 10355
    },
    {
      "entropy": 5.730506801605225,
      "epoch": 0.9397677793904209,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004918977910836557,
      "loss": 5.56,
      "mean_token_accuracy": 0.16230792850255965,
      "num_tokens": 21509445.0,
      "step": 10360
    },
    {
      "entropy": 5.757676792144776,
      "epoch": 0.940221335268505,
      "grad_norm": 4.5,
      "learning_rate": 0.0004918891840449601,
      "loss": 5.6865,
      "mean_token_accuracy": 0.15417071878910066,
      "num_tokens": 21520141.0,
      "step": 10365
    },
    {
      "entropy": 5.719749307632446,
      "epoch": 0.9406748911465893,
      "grad_norm": 3.71875,
      "learning_rate": 0.000491880572520941,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16636838912963867,
      "num_tokens": 21529747.0,
      "step": 10370
    },
    {
      "entropy": 5.857556772232056,
      "epoch": 0.9411284470246735,
      "grad_norm": 4.25,
      "learning_rate": 0.0004918719565117763,
      "loss": 5.625,
      "mean_token_accuracy": 0.15749348253011702,
      "num_tokens": 21540813.0,
      "step": 10375
    },
    {
      "entropy": 5.789521265029907,
      "epoch": 0.9415820029027576,
      "grad_norm": 3.625,
      "learning_rate": 0.0004918633360176444,
      "loss": 5.5358,
      "mean_token_accuracy": 0.16570295989513398,
      "num_tokens": 21550447.0,
      "step": 10380
    },
    {
      "entropy": 5.689071893692017,
      "epoch": 0.9420355587808418,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004918547110387234,
      "loss": 5.5487,
      "mean_token_accuracy": 0.16338303089141845,
      "num_tokens": 21560741.0,
      "step": 10385
    },
    {
      "entropy": 5.791003274917602,
      "epoch": 0.9424891146589259,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004918460815751916,
      "loss": 5.611,
      "mean_token_accuracy": 0.15666065067052842,
      "num_tokens": 21571742.0,
      "step": 10390
    },
    {
      "entropy": 5.7272461414337155,
      "epoch": 0.9429426705370102,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004918374476272277,
      "loss": 5.5708,
      "mean_token_accuracy": 0.15909540504217148,
      "num_tokens": 21582712.0,
      "step": 10395
    },
    {
      "entropy": 5.72408275604248,
      "epoch": 0.9433962264150944,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004918288091950101,
      "loss": 5.5559,
      "mean_token_accuracy": 0.15907078236341476,
      "num_tokens": 21592350.0,
      "step": 10400
    },
    {
      "entropy": 5.747805404663086,
      "epoch": 0.9438497822931785,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004918201662787174,
      "loss": 5.5725,
      "mean_token_accuracy": 0.16181823015213012,
      "num_tokens": 21603098.0,
      "step": 10405
    },
    {
      "entropy": 5.715806007385254,
      "epoch": 0.9443033381712627,
      "grad_norm": 3.59375,
      "learning_rate": 0.0004918115188785285,
      "loss": 5.4944,
      "mean_token_accuracy": 0.16417356431484223,
      "num_tokens": 21613084.0,
      "step": 10410
    },
    {
      "entropy": 5.796920728683472,
      "epoch": 0.9447568940493469,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004918028669946221,
      "loss": 5.6835,
      "mean_token_accuracy": 0.14824492186307908,
      "num_tokens": 21624895.0,
      "step": 10415
    },
    {
      "entropy": 5.849762630462647,
      "epoch": 0.9452104499274311,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004917942106271771,
      "loss": 5.7,
      "mean_token_accuracy": 0.15705374330282212,
      "num_tokens": 21636061.0,
      "step": 10420
    },
    {
      "entropy": 5.770350408554077,
      "epoch": 0.9456640058055152,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004917855497763727,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15630857199430465,
      "num_tokens": 21646579.0,
      "step": 10425
    },
    {
      "entropy": 5.7390271663665775,
      "epoch": 0.9461175616835994,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004917768844423878,
      "loss": 5.5737,
      "mean_token_accuracy": 0.16065034717321397,
      "num_tokens": 21657750.0,
      "step": 10430
    },
    {
      "entropy": 5.743339967727661,
      "epoch": 0.9465711175616836,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004917682146254017,
      "loss": 5.5508,
      "mean_token_accuracy": 0.15963664948940276,
      "num_tokens": 21668030.0,
      "step": 10435
    },
    {
      "entropy": 5.691131162643432,
      "epoch": 0.9470246734397678,
      "grad_norm": 2.75,
      "learning_rate": 0.0004917595403255938,
      "loss": 5.473,
      "mean_token_accuracy": 0.1650802195072174,
      "num_tokens": 21678377.0,
      "step": 10440
    },
    {
      "entropy": 5.699818277359009,
      "epoch": 0.947478229317852,
      "grad_norm": 2.875,
      "learning_rate": 0.0004917508615431431,
      "loss": 5.6944,
      "mean_token_accuracy": 0.1529602140188217,
      "num_tokens": 21689258.0,
      "step": 10445
    },
    {
      "entropy": 5.667859506607056,
      "epoch": 0.9479317851959361,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004917421782782296,
      "loss": 5.4919,
      "mean_token_accuracy": 0.1657070502638817,
      "num_tokens": 21699202.0,
      "step": 10450
    },
    {
      "entropy": 5.770370101928711,
      "epoch": 0.9483853410740203,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004917334905310325,
      "loss": 5.53,
      "mean_token_accuracy": 0.16051478683948517,
      "num_tokens": 21709165.0,
      "step": 10455
    },
    {
      "entropy": 5.710021829605102,
      "epoch": 0.9488388969521045,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004917247983017317,
      "loss": 5.4932,
      "mean_token_accuracy": 0.15949437022209167,
      "num_tokens": 21718291.0,
      "step": 10460
    },
    {
      "entropy": 5.745303773880005,
      "epoch": 0.9492924528301887,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004917161015905067,
      "loss": 5.665,
      "mean_token_accuracy": 0.15907982587814332,
      "num_tokens": 21728942.0,
      "step": 10465
    },
    {
      "entropy": 5.809605932235717,
      "epoch": 0.9497460087082729,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004917074003975376,
      "loss": 5.5494,
      "mean_token_accuracy": 0.16044687032699584,
      "num_tokens": 21739017.0,
      "step": 10470
    },
    {
      "entropy": 5.653786849975586,
      "epoch": 0.950199564586357,
      "grad_norm": 3.125,
      "learning_rate": 0.0004916986947230042,
      "loss": 5.4219,
      "mean_token_accuracy": 0.16775716841220856,
      "num_tokens": 21748923.0,
      "step": 10475
    },
    {
      "entropy": 5.770017862319946,
      "epoch": 0.9506531204644412,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004916899845670866,
      "loss": 5.623,
      "mean_token_accuracy": 0.155046346783638,
      "num_tokens": 21759744.0,
      "step": 10480
    },
    {
      "entropy": 5.782181167602539,
      "epoch": 0.9511066763425254,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004916812699299649,
      "loss": 5.5613,
      "mean_token_accuracy": 0.16836415380239486,
      "num_tokens": 21769020.0,
      "step": 10485
    },
    {
      "entropy": 5.768579149246216,
      "epoch": 0.9515602322206096,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004916725508118193,
      "loss": 5.5013,
      "mean_token_accuracy": 0.16281820237636566,
      "num_tokens": 21779841.0,
      "step": 10490
    },
    {
      "entropy": 5.7496484279632565,
      "epoch": 0.9520137880986937,
      "grad_norm": 2.75,
      "learning_rate": 0.0004916638272128301,
      "loss": 5.5269,
      "mean_token_accuracy": 0.16326934546232225,
      "num_tokens": 21788481.0,
      "step": 10495
    },
    {
      "entropy": 5.658914518356323,
      "epoch": 0.9524673439767779,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004916550991331777,
      "loss": 5.4921,
      "mean_token_accuracy": 0.16592521518468856,
      "num_tokens": 21798158.0,
      "step": 10500
    },
    {
      "entropy": 5.765426206588745,
      "epoch": 0.9529208998548622,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004916463665730427,
      "loss": 5.6626,
      "mean_token_accuracy": 0.15711542814970017,
      "num_tokens": 21808375.0,
      "step": 10505
    },
    {
      "entropy": 5.819419193267822,
      "epoch": 0.9533744557329463,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004916376295326057,
      "loss": 5.5823,
      "mean_token_accuracy": 0.16107871234416962,
      "num_tokens": 21819450.0,
      "step": 10510
    },
    {
      "entropy": 5.7885795593261715,
      "epoch": 0.9538280116110305,
      "grad_norm": 6.28125,
      "learning_rate": 0.0004916288880120472,
      "loss": 5.5793,
      "mean_token_accuracy": 0.1666354462504387,
      "num_tokens": 21829526.0,
      "step": 10515
    },
    {
      "entropy": 5.785041809082031,
      "epoch": 0.9542815674891146,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004916201420115482,
      "loss": 5.4924,
      "mean_token_accuracy": 0.16208628416061402,
      "num_tokens": 21839595.0,
      "step": 10520
    },
    {
      "entropy": 5.711234903335571,
      "epoch": 0.9547351233671988,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004916113915312895,
      "loss": 5.5352,
      "mean_token_accuracy": 0.16675786823034286,
      "num_tokens": 21850063.0,
      "step": 10525
    },
    {
      "entropy": 5.736117029190064,
      "epoch": 0.9551886792452831,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004916026365714519,
      "loss": 5.5229,
      "mean_token_accuracy": 0.16148243993520736,
      "num_tokens": 21860605.0,
      "step": 10530
    },
    {
      "entropy": 5.703631210327148,
      "epoch": 0.9556422351233672,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004915938771322165,
      "loss": 5.5474,
      "mean_token_accuracy": 0.17460711002349855,
      "num_tokens": 21870283.0,
      "step": 10535
    },
    {
      "entropy": 5.769282722473145,
      "epoch": 0.9560957910014514,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004915851132137646,
      "loss": 5.6335,
      "mean_token_accuracy": 0.15640568286180495,
      "num_tokens": 21880840.0,
      "step": 10540
    },
    {
      "entropy": 5.778675127029419,
      "epoch": 0.9565493468795355,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004915763448162774,
      "loss": 5.578,
      "mean_token_accuracy": 0.16251935064792633,
      "num_tokens": 21890829.0,
      "step": 10545
    },
    {
      "entropy": 5.763799858093262,
      "epoch": 0.9570029027576198,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004915675719399362,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16117385774850845,
      "num_tokens": 21900365.0,
      "step": 10550
    },
    {
      "entropy": 5.79035439491272,
      "epoch": 0.9574564586357039,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004915587945849225,
      "loss": 5.6357,
      "mean_token_accuracy": 0.1530054971575737,
      "num_tokens": 21911807.0,
      "step": 10555
    },
    {
      "entropy": 5.730452966690064,
      "epoch": 0.9579100145137881,
      "grad_norm": 3.671875,
      "learning_rate": 0.0004915500127514176,
      "loss": 5.5841,
      "mean_token_accuracy": 0.16075941026210785,
      "num_tokens": 21921609.0,
      "step": 10560
    },
    {
      "entropy": 5.85571608543396,
      "epoch": 0.9583635703918723,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004915412264396034,
      "loss": 5.5806,
      "mean_token_accuracy": 0.16014135479927064,
      "num_tokens": 21932104.0,
      "step": 10565
    },
    {
      "entropy": 5.771084833145141,
      "epoch": 0.9588171262699564,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004915324356496613,
      "loss": 5.5422,
      "mean_token_accuracy": 0.16742627024650575,
      "num_tokens": 21941620.0,
      "step": 10570
    },
    {
      "entropy": 5.757388067245484,
      "epoch": 0.9592706821480407,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004915236403817733,
      "loss": 5.6446,
      "mean_token_accuracy": 0.15428871661424637,
      "num_tokens": 21951654.0,
      "step": 10575
    },
    {
      "entropy": 5.75521502494812,
      "epoch": 0.9597242380261248,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004915148406361213,
      "loss": 5.4659,
      "mean_token_accuracy": 0.16495732814073563,
      "num_tokens": 21962004.0,
      "step": 10580
    },
    {
      "entropy": 5.776280927658081,
      "epoch": 0.960177793904209,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004915060364128872,
      "loss": 5.5561,
      "mean_token_accuracy": 0.15665401369333268,
      "num_tokens": 21972042.0,
      "step": 10585
    },
    {
      "entropy": 5.733834362030029,
      "epoch": 0.9606313497822931,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004914972277122532,
      "loss": 5.6099,
      "mean_token_accuracy": 0.14938019216060638,
      "num_tokens": 21983299.0,
      "step": 10590
    },
    {
      "entropy": 5.75510687828064,
      "epoch": 0.9610849056603774,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004914884145344012,
      "loss": 5.6093,
      "mean_token_accuracy": 0.16269951909780503,
      "num_tokens": 21994085.0,
      "step": 10595
    },
    {
      "entropy": 5.790181970596313,
      "epoch": 0.9615384615384616,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004914795968795137,
      "loss": 5.6473,
      "mean_token_accuracy": 0.1541765734553337,
      "num_tokens": 22004389.0,
      "step": 10600
    },
    {
      "entropy": 5.763106012344361,
      "epoch": 0.9619920174165457,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004914707747477728,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1612747848033905,
      "num_tokens": 22014198.0,
      "step": 10605
    },
    {
      "entropy": 5.724530124664307,
      "epoch": 0.9624455732946299,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004914619481393613,
      "loss": 5.4855,
      "mean_token_accuracy": 0.16010512858629228,
      "num_tokens": 22024994.0,
      "step": 10610
    },
    {
      "entropy": 5.741182470321656,
      "epoch": 0.962899129172714,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004914531170544615,
      "loss": 5.5643,
      "mean_token_accuracy": 0.15734463334083557,
      "num_tokens": 22035757.0,
      "step": 10615
    },
    {
      "entropy": 5.813630533218384,
      "epoch": 0.9633526850507983,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004914442814932561,
      "loss": 5.643,
      "mean_token_accuracy": 0.1584210902452469,
      "num_tokens": 22046097.0,
      "step": 10620
    },
    {
      "entropy": 5.7486159324646,
      "epoch": 0.9638062409288825,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004914354414559278,
      "loss": 5.6015,
      "mean_token_accuracy": 0.1594898521900177,
      "num_tokens": 22055457.0,
      "step": 10625
    },
    {
      "entropy": 5.793947839736939,
      "epoch": 0.9642597968069666,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004914265969426594,
      "loss": 5.5833,
      "mean_token_accuracy": 0.16348688751459123,
      "num_tokens": 22065687.0,
      "step": 10630
    },
    {
      "entropy": 5.834330081939697,
      "epoch": 0.9647133526850508,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004914177479536338,
      "loss": 5.5697,
      "mean_token_accuracy": 0.164402736723423,
      "num_tokens": 22076580.0,
      "step": 10635
    },
    {
      "entropy": 5.75211124420166,
      "epoch": 0.965166908563135,
      "grad_norm": 2.640625,
      "learning_rate": 0.000491408894489034,
      "loss": 5.5599,
      "mean_token_accuracy": 0.15735766291618347,
      "num_tokens": 22086493.0,
      "step": 10640
    },
    {
      "entropy": 5.764957618713379,
      "epoch": 0.9656204644412192,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004914000365490432,
      "loss": 5.5823,
      "mean_token_accuracy": 0.15760962963104247,
      "num_tokens": 22098018.0,
      "step": 10645
    },
    {
      "entropy": 5.76650595664978,
      "epoch": 0.9660740203193033,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004913911741338445,
      "loss": 5.5725,
      "mean_token_accuracy": 0.15959393829107285,
      "num_tokens": 22108548.0,
      "step": 10650
    },
    {
      "entropy": 5.726461505889892,
      "epoch": 0.9665275761973875,
      "grad_norm": 3.0,
      "learning_rate": 0.0004913823072436211,
      "loss": 5.5361,
      "mean_token_accuracy": 0.15646940916776658,
      "num_tokens": 22118580.0,
      "step": 10655
    },
    {
      "entropy": 5.737323522567749,
      "epoch": 0.9669811320754716,
      "grad_norm": 7.125,
      "learning_rate": 0.0004913734358785565,
      "loss": 5.5489,
      "mean_token_accuracy": 0.16140337139368058,
      "num_tokens": 22128912.0,
      "step": 10660
    },
    {
      "entropy": 5.739957046508789,
      "epoch": 0.9674346879535559,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004913645600388341,
      "loss": 5.6296,
      "mean_token_accuracy": 0.15566144436597823,
      "num_tokens": 22139627.0,
      "step": 10665
    },
    {
      "entropy": 5.761223649978637,
      "epoch": 0.9678882438316401,
      "grad_norm": 2.875,
      "learning_rate": 0.0004913556797246374,
      "loss": 5.5996,
      "mean_token_accuracy": 0.16123740524053573,
      "num_tokens": 22150136.0,
      "step": 10670
    },
    {
      "entropy": 5.779002857208252,
      "epoch": 0.9683417997097242,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004913467949361503,
      "loss": 5.6004,
      "mean_token_accuracy": 0.15770047903060913,
      "num_tokens": 22161738.0,
      "step": 10675
    },
    {
      "entropy": 5.73794527053833,
      "epoch": 0.9687953555878084,
      "grad_norm": 4.8125,
      "learning_rate": 0.0004913379056735561,
      "loss": 5.5502,
      "mean_token_accuracy": 0.16065671890974045,
      "num_tokens": 22172734.0,
      "step": 10680
    },
    {
      "entropy": 5.764235162734986,
      "epoch": 0.9692489114658926,
      "grad_norm": 5.1875,
      "learning_rate": 0.000491329011937039,
      "loss": 5.5383,
      "mean_token_accuracy": 0.16276451647281648,
      "num_tokens": 22183833.0,
      "step": 10685
    },
    {
      "entropy": 5.756505298614502,
      "epoch": 0.9697024673439768,
      "grad_norm": 3.5,
      "learning_rate": 0.0004913201137267828,
      "loss": 5.5449,
      "mean_token_accuracy": 0.15776949524879455,
      "num_tokens": 22194769.0,
      "step": 10690
    },
    {
      "entropy": 5.7486083984375,
      "epoch": 0.970156023222061,
      "grad_norm": 3.5,
      "learning_rate": 0.0004913112110429715,
      "loss": 5.5049,
      "mean_token_accuracy": 0.16457379311323167,
      "num_tokens": 22204824.0,
      "step": 10695
    },
    {
      "entropy": 5.76269850730896,
      "epoch": 0.9706095791001451,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004913023038857892,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1676845893263817,
      "num_tokens": 22215840.0,
      "step": 10700
    },
    {
      "entropy": 5.760590267181397,
      "epoch": 0.9710631349782293,
      "grad_norm": 7.125,
      "learning_rate": 0.0004912933922554203,
      "loss": 5.5863,
      "mean_token_accuracy": 0.15847639739513397,
      "num_tokens": 22226792.0,
      "step": 10705
    },
    {
      "entropy": 5.687443590164184,
      "epoch": 0.9715166908563135,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004912844761520487,
      "loss": 5.5824,
      "mean_token_accuracy": 0.16106030493974685,
      "num_tokens": 22237059.0,
      "step": 10710
    },
    {
      "entropy": 5.762354946136474,
      "epoch": 0.9719702467343977,
      "grad_norm": 3.484375,
      "learning_rate": 0.0004912755555758592,
      "loss": 5.5099,
      "mean_token_accuracy": 0.16265950351953506,
      "num_tokens": 22246570.0,
      "step": 10715
    },
    {
      "entropy": 5.80545072555542,
      "epoch": 0.9724238026124818,
      "grad_norm": 3.640625,
      "learning_rate": 0.000491266630527036,
      "loss": 5.6176,
      "mean_token_accuracy": 0.1568292886018753,
      "num_tokens": 22257076.0,
      "step": 10720
    },
    {
      "entropy": 5.754783725738525,
      "epoch": 0.972877358490566,
      "grad_norm": 5.875,
      "learning_rate": 0.0004912577010057638,
      "loss": 5.5771,
      "mean_token_accuracy": 0.16309200078248978,
      "num_tokens": 22269061.0,
      "step": 10725
    },
    {
      "entropy": 5.728498458862305,
      "epoch": 0.9733309143686503,
      "grad_norm": 4.25,
      "learning_rate": 0.0004912487670122273,
      "loss": 5.5139,
      "mean_token_accuracy": 0.16305593997240067,
      "num_tokens": 22279764.0,
      "step": 10730
    },
    {
      "entropy": 5.748498868942261,
      "epoch": 0.9737844702467344,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004912398285466111,
      "loss": 5.5019,
      "mean_token_accuracy": 0.16102107018232345,
      "num_tokens": 22289725.0,
      "step": 10735
    },
    {
      "entropy": 5.726200675964355,
      "epoch": 0.9742380261248186,
      "grad_norm": 3.546875,
      "learning_rate": 0.0004912308856091002,
      "loss": 5.5607,
      "mean_token_accuracy": 0.1602289706468582,
      "num_tokens": 22299263.0,
      "step": 10740
    },
    {
      "entropy": 5.725038957595825,
      "epoch": 0.9746915820029027,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004912219381998795,
      "loss": 5.5657,
      "mean_token_accuracy": 0.16212027221918107,
      "num_tokens": 22308138.0,
      "step": 10745
    },
    {
      "entropy": 5.744976282119751,
      "epoch": 0.9751451378809869,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004912129863191339,
      "loss": 5.6019,
      "mean_token_accuracy": 0.15792572423815726,
      "num_tokens": 22318144.0,
      "step": 10750
    },
    {
      "entropy": 5.839464616775513,
      "epoch": 0.9755986937590712,
      "grad_norm": 5.96875,
      "learning_rate": 0.0004912040299670487,
      "loss": 5.4867,
      "mean_token_accuracy": 0.1658570110797882,
      "num_tokens": 22328040.0,
      "step": 10755
    },
    {
      "entropy": 5.74774284362793,
      "epoch": 0.9760522496371553,
      "grad_norm": 4.96875,
      "learning_rate": 0.0004911950691438091,
      "loss": 5.5915,
      "mean_token_accuracy": 0.15731143206357956,
      "num_tokens": 22338217.0,
      "step": 10760
    },
    {
      "entropy": 5.777115201950073,
      "epoch": 0.9765058055152395,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004911861038496004,
      "loss": 5.5744,
      "mean_token_accuracy": 0.15651740729808808,
      "num_tokens": 22348545.0,
      "step": 10765
    },
    {
      "entropy": 5.8585512161254885,
      "epoch": 0.9769593613933236,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004911771340846079,
      "loss": 5.5995,
      "mean_token_accuracy": 0.15708330720663072,
      "num_tokens": 22358457.0,
      "step": 10770
    },
    {
      "entropy": 5.85707368850708,
      "epoch": 0.9774129172714079,
      "grad_norm": 5.625,
      "learning_rate": 0.0004911681598490172,
      "loss": 5.5627,
      "mean_token_accuracy": 0.16272783577442168,
      "num_tokens": 22368035.0,
      "step": 10775
    },
    {
      "entropy": 5.766559314727783,
      "epoch": 0.977866473149492,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004911591811430139,
      "loss": 5.5536,
      "mean_token_accuracy": 0.16507384777069092,
      "num_tokens": 22377654.0,
      "step": 10780
    },
    {
      "entropy": 5.711172866821289,
      "epoch": 0.9783200290275762,
      "grad_norm": 3.875,
      "learning_rate": 0.0004911501979667836,
      "loss": 5.6606,
      "mean_token_accuracy": 0.15210606753826142,
      "num_tokens": 22388691.0,
      "step": 10785
    },
    {
      "entropy": 5.827831363677978,
      "epoch": 0.9787735849056604,
      "grad_norm": 2.625,
      "learning_rate": 0.0004911412103205122,
      "loss": 5.6667,
      "mean_token_accuracy": 0.1560517132282257,
      "num_tokens": 22399579.0,
      "step": 10790
    },
    {
      "entropy": 5.853053903579712,
      "epoch": 0.9792271407837445,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004911322182043855,
      "loss": 5.5741,
      "mean_token_accuracy": 0.16505400687456132,
      "num_tokens": 22409387.0,
      "step": 10795
    },
    {
      "entropy": 5.755277395248413,
      "epoch": 0.9796806966618288,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004911232216185893,
      "loss": 5.546,
      "mean_token_accuracy": 0.16382538825273513,
      "num_tokens": 22419263.0,
      "step": 10800
    },
    {
      "entropy": 5.742936325073242,
      "epoch": 0.9801342525399129,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004911142205633099,
      "loss": 5.5552,
      "mean_token_accuracy": 0.1672627180814743,
      "num_tokens": 22429717.0,
      "step": 10805
    },
    {
      "entropy": 5.781493234634399,
      "epoch": 0.9805878084179971,
      "grad_norm": 4.875,
      "learning_rate": 0.0004911052150387334,
      "loss": 5.5944,
      "mean_token_accuracy": 0.16006461083889006,
      "num_tokens": 22439713.0,
      "step": 10810
    },
    {
      "entropy": 5.69221134185791,
      "epoch": 0.9810413642960812,
      "grad_norm": 4.125,
      "learning_rate": 0.0004910962050450458,
      "loss": 5.4911,
      "mean_token_accuracy": 0.16025393307209015,
      "num_tokens": 22450808.0,
      "step": 10815
    },
    {
      "entropy": 5.71334490776062,
      "epoch": 0.9814949201741655,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004910871905824338,
      "loss": 5.5605,
      "mean_token_accuracy": 0.15803758502006532,
      "num_tokens": 22460723.0,
      "step": 10820
    },
    {
      "entropy": 5.787650632858276,
      "epoch": 0.9819484760522497,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004910781716510837,
      "loss": 5.5369,
      "mean_token_accuracy": 0.1604889377951622,
      "num_tokens": 22470342.0,
      "step": 10825
    },
    {
      "entropy": 5.807727479934693,
      "epoch": 0.9824020319303338,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004910691482511817,
      "loss": 5.6041,
      "mean_token_accuracy": 0.15597862154245376,
      "num_tokens": 22481387.0,
      "step": 10830
    },
    {
      "entropy": 5.683573007583618,
      "epoch": 0.982855587808418,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004910601203829148,
      "loss": 5.5619,
      "mean_token_accuracy": 0.16236660182476043,
      "num_tokens": 22491342.0,
      "step": 10835
    },
    {
      "entropy": 5.796797037124634,
      "epoch": 0.9833091436865021,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004910510880464696,
      "loss": 5.5938,
      "mean_token_accuracy": 0.16152505874633788,
      "num_tokens": 22501951.0,
      "step": 10840
    },
    {
      "entropy": 5.839929294586182,
      "epoch": 0.9837626995645864,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004910420512420327,
      "loss": 5.6229,
      "mean_token_accuracy": 0.15495607405900955,
      "num_tokens": 22511598.0,
      "step": 10845
    },
    {
      "entropy": 5.788203430175781,
      "epoch": 0.9842162554426706,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004910330099697913,
      "loss": 5.5451,
      "mean_token_accuracy": 0.15964120328426362,
      "num_tokens": 22522719.0,
      "step": 10850
    },
    {
      "entropy": 5.775867033004761,
      "epoch": 0.9846698113207547,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004910239642299321,
      "loss": 5.5608,
      "mean_token_accuracy": 0.16430606842041015,
      "num_tokens": 22533073.0,
      "step": 10855
    },
    {
      "entropy": 5.78825945854187,
      "epoch": 0.9851233671988389,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004910149140226422,
      "loss": 5.5785,
      "mean_token_accuracy": 0.16027194261550903,
      "num_tokens": 22542819.0,
      "step": 10860
    },
    {
      "entropy": 5.791331720352173,
      "epoch": 0.9855769230769231,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004910058593481089,
      "loss": 5.6116,
      "mean_token_accuracy": 0.15468202531337738,
      "num_tokens": 22552920.0,
      "step": 10865
    },
    {
      "entropy": 5.8213968753814695,
      "epoch": 0.9860304789550073,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004909968002065195,
      "loss": 5.5946,
      "mean_token_accuracy": 0.1593124508857727,
      "num_tokens": 22563491.0,
      "step": 10870
    },
    {
      "entropy": 5.797160339355469,
      "epoch": 0.9864840348330914,
      "grad_norm": 2.203125,
      "learning_rate": 0.000490987736598061,
      "loss": 5.6419,
      "mean_token_accuracy": 0.15619842261075972,
      "num_tokens": 22573337.0,
      "step": 10875
    },
    {
      "entropy": 5.774786424636841,
      "epoch": 0.9869375907111756,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004909786685229211,
      "loss": 5.6271,
      "mean_token_accuracy": 0.15950653403997422,
      "num_tokens": 22583317.0,
      "step": 10880
    },
    {
      "entropy": 5.7209861278533936,
      "epoch": 0.9873911465892597,
      "grad_norm": 2.625,
      "learning_rate": 0.0004909695959812875,
      "loss": 5.5313,
      "mean_token_accuracy": 0.16710339933633805,
      "num_tokens": 22593133.0,
      "step": 10885
    },
    {
      "entropy": 5.774353790283203,
      "epoch": 0.987844702467344,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004909605189733474,
      "loss": 5.5885,
      "mean_token_accuracy": 0.16063532531261443,
      "num_tokens": 22603117.0,
      "step": 10890
    },
    {
      "entropy": 5.73704047203064,
      "epoch": 0.9882982583454282,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004909514374992889,
      "loss": 5.4692,
      "mean_token_accuracy": 0.16348331570625305,
      "num_tokens": 22612980.0,
      "step": 10895
    },
    {
      "entropy": 5.750940370559692,
      "epoch": 0.9887518142235123,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004909423515592996,
      "loss": 5.55,
      "mean_token_accuracy": 0.1583065539598465,
      "num_tokens": 22622610.0,
      "step": 10900
    },
    {
      "entropy": 5.737071990966797,
      "epoch": 0.9892053701015965,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004909332611535675,
      "loss": 5.562,
      "mean_token_accuracy": 0.16179215013980866,
      "num_tokens": 22632065.0,
      "step": 10905
    },
    {
      "entropy": 5.7813292980194095,
      "epoch": 0.9896589259796807,
      "grad_norm": 5.21875,
      "learning_rate": 0.0004909241662822804,
      "loss": 5.6146,
      "mean_token_accuracy": 0.16099244952201844,
      "num_tokens": 22643162.0,
      "step": 10910
    },
    {
      "entropy": 5.793786287307739,
      "epoch": 0.9901124818577649,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004909150669456265,
      "loss": 5.5602,
      "mean_token_accuracy": 0.1629994332790375,
      "num_tokens": 22653388.0,
      "step": 10915
    },
    {
      "entropy": 5.7237261772155765,
      "epoch": 0.9905660377358491,
      "grad_norm": 5.09375,
      "learning_rate": 0.000490905963143794,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16971467882394792,
      "num_tokens": 22662714.0,
      "step": 10920
    },
    {
      "entropy": 5.715773057937622,
      "epoch": 0.9910195936139332,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004908968548769713,
      "loss": 5.561,
      "mean_token_accuracy": 0.15682685673236846,
      "num_tokens": 22672184.0,
      "step": 10925
    },
    {
      "entropy": 5.7049195766448975,
      "epoch": 0.9914731494920174,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004908877421453466,
      "loss": 5.5117,
      "mean_token_accuracy": 0.1609465077519417,
      "num_tokens": 22682590.0,
      "step": 10930
    },
    {
      "entropy": 5.81724853515625,
      "epoch": 0.9919267053701016,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004908786249491082,
      "loss": 5.6579,
      "mean_token_accuracy": 0.16107190251350403,
      "num_tokens": 22692986.0,
      "step": 10935
    },
    {
      "entropy": 5.773460054397583,
      "epoch": 0.9923802612481858,
      "grad_norm": 2.59375,
      "learning_rate": 0.000490869503288445,
      "loss": 5.4626,
      "mean_token_accuracy": 0.16461695581674576,
      "num_tokens": 22702680.0,
      "step": 10940
    },
    {
      "entropy": 5.753363990783692,
      "epoch": 0.99283381712627,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004908603771635454,
      "loss": 5.5829,
      "mean_token_accuracy": 0.1634833410382271,
      "num_tokens": 22712712.0,
      "step": 10945
    },
    {
      "entropy": 5.825459384918213,
      "epoch": 0.9932873730043541,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004908512465745982,
      "loss": 5.6463,
      "mean_token_accuracy": 0.15797701254487037,
      "num_tokens": 22723231.0,
      "step": 10950
    },
    {
      "entropy": 5.815076637268066,
      "epoch": 0.9937409288824384,
      "grad_norm": 5.78125,
      "learning_rate": 0.0004908421115217922,
      "loss": 5.6154,
      "mean_token_accuracy": 0.1607817977666855,
      "num_tokens": 22733909.0,
      "step": 10955
    },
    {
      "entropy": 5.711560440063477,
      "epoch": 0.9941944847605225,
      "grad_norm": 3.0,
      "learning_rate": 0.0004908329720053164,
      "loss": 5.5506,
      "mean_token_accuracy": 0.15961403399705887,
      "num_tokens": 22744558.0,
      "step": 10960
    },
    {
      "entropy": 5.75459852218628,
      "epoch": 0.9946480406386067,
      "grad_norm": 3.25,
      "learning_rate": 0.0004908238280253597,
      "loss": 5.5382,
      "mean_token_accuracy": 0.16251804083585739,
      "num_tokens": 22755018.0,
      "step": 10965
    },
    {
      "entropy": 5.789547157287598,
      "epoch": 0.9951015965166908,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004908146795821112,
      "loss": 5.5888,
      "mean_token_accuracy": 0.16327035203576087,
      "num_tokens": 22764936.0,
      "step": 10970
    },
    {
      "entropy": 5.791564130783081,
      "epoch": 0.995555152394775,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004908055266757602,
      "loss": 5.6661,
      "mean_token_accuracy": 0.1596682995557785,
      "num_tokens": 22775216.0,
      "step": 10975
    },
    {
      "entropy": 5.726565504074097,
      "epoch": 0.9960087082728593,
      "grad_norm": 4.65625,
      "learning_rate": 0.0004907963693064958,
      "loss": 5.544,
      "mean_token_accuracy": 0.16100151985883712,
      "num_tokens": 22784668.0,
      "step": 10980
    },
    {
      "entropy": 5.787533235549927,
      "epoch": 0.9964622641509434,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004907872074745075,
      "loss": 5.5543,
      "mean_token_accuracy": 0.15683875605463982,
      "num_tokens": 22795027.0,
      "step": 10985
    },
    {
      "entropy": 5.765175819396973,
      "epoch": 0.9969158200290276,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004907780411799849,
      "loss": 5.6061,
      "mean_token_accuracy": 0.15528410971164702,
      "num_tokens": 22806370.0,
      "step": 10990
    },
    {
      "entropy": 5.699238300323486,
      "epoch": 0.9973693759071117,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004907688704231173,
      "loss": 5.5173,
      "mean_token_accuracy": 0.163358948379755,
      "num_tokens": 22816212.0,
      "step": 10995
    },
    {
      "entropy": 5.719211769104004,
      "epoch": 0.997822931785196,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004907596952040945,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1681550547480583,
      "num_tokens": 22825813.0,
      "step": 11000
    },
    {
      "entropy": 5.716399335861206,
      "epoch": 0.9982764876632801,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004907505155231063,
      "loss": 5.5526,
      "mean_token_accuracy": 0.16009844988584518,
      "num_tokens": 22836457.0,
      "step": 11005
    },
    {
      "entropy": 5.756836032867431,
      "epoch": 0.9987300435413643,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004907413313803424,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1575374871492386,
      "num_tokens": 22847859.0,
      "step": 11010
    },
    {
      "entropy": 5.800490951538086,
      "epoch": 0.9991835994194485,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004907321427759928,
      "loss": 5.6086,
      "mean_token_accuracy": 0.15785778760910035,
      "num_tokens": 22857956.0,
      "step": 11015
    },
    {
      "entropy": 5.788890838623047,
      "epoch": 0.9996371552975326,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004907229497102475,
      "loss": 5.6357,
      "mean_token_accuracy": 0.15690969675779343,
      "num_tokens": 22868901.0,
      "step": 11020
    },
    {
      "entropy": 5.772069644927979,
      "epoch": 1.0000907111756168,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004907137521832966,
      "loss": 5.575,
      "mean_token_accuracy": 0.15224703401327133,
      "num_tokens": 22879116.0,
      "step": 11025
    },
    {
      "entropy": 5.67910590171814,
      "epoch": 1.000544267053701,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004907045501953304,
      "loss": 5.4774,
      "mean_token_accuracy": 0.16393017172813415,
      "num_tokens": 22889919.0,
      "step": 11030
    },
    {
      "entropy": 5.786879205703736,
      "epoch": 1.0009978229317853,
      "grad_norm": 2.671875,
      "learning_rate": 0.000490695343746539,
      "loss": 5.5289,
      "mean_token_accuracy": 0.1584495037794113,
      "num_tokens": 22899434.0,
      "step": 11035
    },
    {
      "entropy": 5.757241725921631,
      "epoch": 1.0014513788098693,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004906861328371131,
      "loss": 5.524,
      "mean_token_accuracy": 0.16017119586467743,
      "num_tokens": 22909829.0,
      "step": 11040
    },
    {
      "entropy": 5.683193874359131,
      "epoch": 1.0019049346879536,
      "grad_norm": 4.5,
      "learning_rate": 0.0004906769174672429,
      "loss": 5.5337,
      "mean_token_accuracy": 0.15968889743089676,
      "num_tokens": 22921961.0,
      "step": 11045
    },
    {
      "entropy": 5.726322793960572,
      "epoch": 1.0023584905660377,
      "grad_norm": 4.875,
      "learning_rate": 0.000490667697637119,
      "loss": 5.5021,
      "mean_token_accuracy": 0.158238422870636,
      "num_tokens": 22932718.0,
      "step": 11050
    },
    {
      "entropy": 5.732589626312256,
      "epoch": 1.002812046444122,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004906584733469324,
      "loss": 5.4611,
      "mean_token_accuracy": 0.16289833188056946,
      "num_tokens": 22942844.0,
      "step": 11055
    },
    {
      "entropy": 5.744976043701172,
      "epoch": 1.0032656023222062,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004906492445968735,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1638798162341118,
      "num_tokens": 22952370.0,
      "step": 11060
    },
    {
      "entropy": 5.751995182037353,
      "epoch": 1.0037191582002902,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004906400113871332,
      "loss": 5.4562,
      "mean_token_accuracy": 0.15912625789642335,
      "num_tokens": 22962449.0,
      "step": 11065
    },
    {
      "entropy": 5.719634914398194,
      "epoch": 1.0041727140783745,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004906307737179026,
      "loss": 5.454,
      "mean_token_accuracy": 0.16155023872852325,
      "num_tokens": 22973952.0,
      "step": 11070
    },
    {
      "entropy": 5.664208793640137,
      "epoch": 1.0046262699564585,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004906215315893726,
      "loss": 5.488,
      "mean_token_accuracy": 0.16348590701818466,
      "num_tokens": 22983798.0,
      "step": 11075
    },
    {
      "entropy": 5.738824558258057,
      "epoch": 1.0050798258345428,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004906122850017344,
      "loss": 5.4982,
      "mean_token_accuracy": 0.16925758570432664,
      "num_tokens": 22993803.0,
      "step": 11080
    },
    {
      "entropy": 5.745176935195923,
      "epoch": 1.005533381712627,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004906030339551792,
      "loss": 5.4949,
      "mean_token_accuracy": 0.15942434668540956,
      "num_tokens": 23003727.0,
      "step": 11085
    },
    {
      "entropy": 5.750190925598145,
      "epoch": 1.0059869375907111,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004905937784498983,
      "loss": 5.5479,
      "mean_token_accuracy": 0.1587470144033432,
      "num_tokens": 23014344.0,
      "step": 11090
    },
    {
      "entropy": 5.746530437469483,
      "epoch": 1.0064404934687954,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004905845184860831,
      "loss": 5.5124,
      "mean_token_accuracy": 0.15678538531064987,
      "num_tokens": 23025082.0,
      "step": 11095
    },
    {
      "entropy": 5.767651557922363,
      "epoch": 1.0068940493468794,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004905752540639252,
      "loss": 5.5547,
      "mean_token_accuracy": 0.15660654306411742,
      "num_tokens": 23035513.0,
      "step": 11100
    },
    {
      "entropy": 5.799265193939209,
      "epoch": 1.0073476052249637,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004905659851836161,
      "loss": 5.5421,
      "mean_token_accuracy": 0.1640932857990265,
      "num_tokens": 23045778.0,
      "step": 11105
    },
    {
      "entropy": 5.741215562820434,
      "epoch": 1.007801161103048,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004905567118453475,
      "loss": 5.5088,
      "mean_token_accuracy": 0.1637937992811203,
      "num_tokens": 23056244.0,
      "step": 11110
    },
    {
      "entropy": 5.701434564590454,
      "epoch": 1.008254716981132,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004905474340493111,
      "loss": 5.4695,
      "mean_token_accuracy": 0.16456196308135987,
      "num_tokens": 23066225.0,
      "step": 11115
    },
    {
      "entropy": 5.617484903335571,
      "epoch": 1.0087082728592163,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004905381517956988,
      "loss": 5.4018,
      "mean_token_accuracy": 0.1635944664478302,
      "num_tokens": 23076075.0,
      "step": 11120
    },
    {
      "entropy": 5.736221885681152,
      "epoch": 1.0091618287373005,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004905288650847027,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1566113531589508,
      "num_tokens": 23086437.0,
      "step": 11125
    },
    {
      "entropy": 5.849229145050049,
      "epoch": 1.0096153846153846,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004905195739165146,
      "loss": 5.5684,
      "mean_token_accuracy": 0.1568540006875992,
      "num_tokens": 23096994.0,
      "step": 11130
    },
    {
      "entropy": 5.786506128311157,
      "epoch": 1.0100689404934688,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000490510278291327,
      "loss": 5.4766,
      "mean_token_accuracy": 0.16240027248859407,
      "num_tokens": 23106918.0,
      "step": 11135
    },
    {
      "entropy": 5.702534294128418,
      "epoch": 1.010522496371553,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004905009782093318,
      "loss": 5.4397,
      "mean_token_accuracy": 0.16481769382953643,
      "num_tokens": 23115981.0,
      "step": 11140
    },
    {
      "entropy": 5.7717200756073,
      "epoch": 1.0109760522496372,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004904916736707214,
      "loss": 5.5469,
      "mean_token_accuracy": 0.17090407609939576,
      "num_tokens": 23124689.0,
      "step": 11145
    },
    {
      "entropy": 5.774488735198974,
      "epoch": 1.0114296081277214,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004904823646756885,
      "loss": 5.4875,
      "mean_token_accuracy": 0.16207441836595535,
      "num_tokens": 23134985.0,
      "step": 11150
    },
    {
      "entropy": 5.645842599868774,
      "epoch": 1.0118831640058055,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004904730512244252,
      "loss": 5.4925,
      "mean_token_accuracy": 0.1594291403889656,
      "num_tokens": 23145738.0,
      "step": 11155
    },
    {
      "entropy": 5.732175397872925,
      "epoch": 1.0123367198838897,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004904637333171243,
      "loss": 5.5633,
      "mean_token_accuracy": 0.15769633799791336,
      "num_tokens": 23156206.0,
      "step": 11160
    },
    {
      "entropy": 5.69897689819336,
      "epoch": 1.0127902757619738,
      "grad_norm": 6.875,
      "learning_rate": 0.0004904544109539785,
      "loss": 5.3355,
      "mean_token_accuracy": 0.17022090703248977,
      "num_tokens": 23166556.0,
      "step": 11165
    },
    {
      "entropy": 5.7195000648498535,
      "epoch": 1.013243831640058,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004904450841351805,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16423276215791702,
      "num_tokens": 23176998.0,
      "step": 11170
    },
    {
      "entropy": 5.722396802902222,
      "epoch": 1.0136973875181423,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004904357528609236,
      "loss": 5.4624,
      "mean_token_accuracy": 0.16163630038499832,
      "num_tokens": 23187203.0,
      "step": 11175
    },
    {
      "entropy": 5.6736852645874025,
      "epoch": 1.0141509433962264,
      "grad_norm": 2.75,
      "learning_rate": 0.0004904264171314,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1669812649488449,
      "num_tokens": 23197123.0,
      "step": 11180
    },
    {
      "entropy": 5.704225254058838,
      "epoch": 1.0146044992743106,
      "grad_norm": 3.65625,
      "learning_rate": 0.0004904170769468035,
      "loss": 5.562,
      "mean_token_accuracy": 0.16195669025182724,
      "num_tokens": 23207915.0,
      "step": 11185
    },
    {
      "entropy": 5.704606676101685,
      "epoch": 1.0150580551523947,
      "grad_norm": 2.125,
      "learning_rate": 0.0004904077323073268,
      "loss": 5.519,
      "mean_token_accuracy": 0.16205753535032272,
      "num_tokens": 23218173.0,
      "step": 11190
    },
    {
      "entropy": 5.80987319946289,
      "epoch": 1.015511611030479,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004903983832131633,
      "loss": 5.6491,
      "mean_token_accuracy": 0.15827425122261046,
      "num_tokens": 23229667.0,
      "step": 11195
    },
    {
      "entropy": 5.819153451919556,
      "epoch": 1.0159651669085632,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004903890296645063,
      "loss": 5.5613,
      "mean_token_accuracy": 0.15614357143640517,
      "num_tokens": 23240190.0,
      "step": 11200
    },
    {
      "entropy": 5.801133632659912,
      "epoch": 1.0164187227866472,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004903796716615495,
      "loss": 5.5918,
      "mean_token_accuracy": 0.16640986800193786,
      "num_tokens": 23250145.0,
      "step": 11205
    },
    {
      "entropy": 5.786750411987304,
      "epoch": 1.0168722786647315,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004903703092044861,
      "loss": 5.5102,
      "mean_token_accuracy": 0.16329634487628936,
      "num_tokens": 23259600.0,
      "step": 11210
    },
    {
      "entropy": 5.67227954864502,
      "epoch": 1.0173258345428158,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004903609422935097,
      "loss": 5.4887,
      "mean_token_accuracy": 0.16372834891080856,
      "num_tokens": 23269906.0,
      "step": 11215
    },
    {
      "entropy": 5.684049034118653,
      "epoch": 1.0177793904208998,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004903515709288143,
      "loss": 5.4681,
      "mean_token_accuracy": 0.16762581169605256,
      "num_tokens": 23280273.0,
      "step": 11220
    },
    {
      "entropy": 5.772693252563476,
      "epoch": 1.018232946298984,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004903421951105933,
      "loss": 5.6052,
      "mean_token_accuracy": 0.1527007684111595,
      "num_tokens": 23290889.0,
      "step": 11225
    },
    {
      "entropy": 5.781169986724853,
      "epoch": 1.0186865021770681,
      "grad_norm": 3.25,
      "learning_rate": 0.0004903328148390411,
      "loss": 5.4906,
      "mean_token_accuracy": 0.15907640308141707,
      "num_tokens": 23300971.0,
      "step": 11230
    },
    {
      "entropy": 5.7909337997436525,
      "epoch": 1.0191400580551524,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004903234301143513,
      "loss": 5.4903,
      "mean_token_accuracy": 0.16083768904209136,
      "num_tokens": 23312082.0,
      "step": 11235
    },
    {
      "entropy": 5.807578611373901,
      "epoch": 1.0195936139332367,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004903140409367181,
      "loss": 5.6234,
      "mean_token_accuracy": 0.15780077129602432,
      "num_tokens": 23321755.0,
      "step": 11240
    },
    {
      "entropy": 5.69726710319519,
      "epoch": 1.0200471698113207,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004903046473063357,
      "loss": 5.4972,
      "mean_token_accuracy": 0.1563010722398758,
      "num_tokens": 23332064.0,
      "step": 11245
    },
    {
      "entropy": 5.806121349334717,
      "epoch": 1.020500725689405,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004902952492233983,
      "loss": 5.6475,
      "mean_token_accuracy": 0.1551139011979103,
      "num_tokens": 23342282.0,
      "step": 11250
    },
    {
      "entropy": 5.865414810180664,
      "epoch": 1.020954281567489,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004902858466881002,
      "loss": 5.5232,
      "mean_token_accuracy": 0.15560048967599868,
      "num_tokens": 23352821.0,
      "step": 11255
    },
    {
      "entropy": 5.774457550048828,
      "epoch": 1.0214078374455733,
      "grad_norm": 2.828125,
      "learning_rate": 0.000490276439700636,
      "loss": 5.5442,
      "mean_token_accuracy": 0.16058065891265869,
      "num_tokens": 23363530.0,
      "step": 11260
    },
    {
      "entropy": 5.708051300048828,
      "epoch": 1.0218613933236576,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004902670282612002,
      "loss": 5.4959,
      "mean_token_accuracy": 0.16534019708633424,
      "num_tokens": 23373209.0,
      "step": 11265
    },
    {
      "entropy": 5.7487485885620115,
      "epoch": 1.0223149492017416,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004902576123699873,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1639833018183708,
      "num_tokens": 23383677.0,
      "step": 11270
    },
    {
      "entropy": 5.677666568756104,
      "epoch": 1.0227685050798259,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004902481920271922,
      "loss": 5.4614,
      "mean_token_accuracy": 0.16406575441360474,
      "num_tokens": 23394447.0,
      "step": 11275
    },
    {
      "entropy": 5.765851402282715,
      "epoch": 1.02322206095791,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004902387672330097,
      "loss": 5.5203,
      "mean_token_accuracy": 0.15985085368156432,
      "num_tokens": 23404783.0,
      "step": 11280
    },
    {
      "entropy": 5.689364624023438,
      "epoch": 1.0236756168359942,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004902293379876345,
      "loss": 5.4045,
      "mean_token_accuracy": 0.16567852199077607,
      "num_tokens": 23413271.0,
      "step": 11285
    },
    {
      "entropy": 5.6645372867584225,
      "epoch": 1.0241291727140784,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004902199042912618,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1613709047436714,
      "num_tokens": 23423192.0,
      "step": 11290
    },
    {
      "entropy": 5.638602209091187,
      "epoch": 1.0245827285921625,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004902104661440868,
      "loss": 5.4604,
      "mean_token_accuracy": 0.161133249104023,
      "num_tokens": 23433767.0,
      "step": 11295
    },
    {
      "entropy": 5.741264009475708,
      "epoch": 1.0250362844702468,
      "grad_norm": 3.421875,
      "learning_rate": 0.0004902010235463042,
      "loss": 5.6019,
      "mean_token_accuracy": 0.1633840635418892,
      "num_tokens": 23444371.0,
      "step": 11300
    },
    {
      "entropy": 5.835170602798462,
      "epoch": 1.025489840348331,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004901915764981098,
      "loss": 5.4799,
      "mean_token_accuracy": 0.1616680547595024,
      "num_tokens": 23454702.0,
      "step": 11305
    },
    {
      "entropy": 5.742986583709717,
      "epoch": 1.025943396226415,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004901821249996988,
      "loss": 5.514,
      "mean_token_accuracy": 0.1604331076145172,
      "num_tokens": 23464381.0,
      "step": 11310
    },
    {
      "entropy": 5.724117469787598,
      "epoch": 1.0263969521044993,
      "grad_norm": 2.25,
      "learning_rate": 0.0004901726690512665,
      "loss": 5.5326,
      "mean_token_accuracy": 0.1670032724738121,
      "num_tokens": 23474018.0,
      "step": 11315
    },
    {
      "entropy": 5.822562122344971,
      "epoch": 1.0268505079825834,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004901632086530087,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1553361251950264,
      "num_tokens": 23486047.0,
      "step": 11320
    },
    {
      "entropy": 5.827811527252197,
      "epoch": 1.0273040638606676,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004901537438051207,
      "loss": 5.5424,
      "mean_token_accuracy": 0.15312551259994506,
      "num_tokens": 23496356.0,
      "step": 11325
    },
    {
      "entropy": 5.742318344116211,
      "epoch": 1.027757619738752,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004901442745077986,
      "loss": 5.5069,
      "mean_token_accuracy": 0.16515136361122132,
      "num_tokens": 23506345.0,
      "step": 11330
    },
    {
      "entropy": 5.705273389816284,
      "epoch": 1.028211175616836,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004901348007612381,
      "loss": 5.5488,
      "mean_token_accuracy": 0.1617298886179924,
      "num_tokens": 23516464.0,
      "step": 11335
    },
    {
      "entropy": 5.683308219909668,
      "epoch": 1.0286647314949202,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004901253225656351,
      "loss": 5.4264,
      "mean_token_accuracy": 0.1645350381731987,
      "num_tokens": 23526067.0,
      "step": 11340
    },
    {
      "entropy": 5.684415197372436,
      "epoch": 1.0291182873730043,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004901158399211855,
      "loss": 5.4328,
      "mean_token_accuracy": 0.16813672930002213,
      "num_tokens": 23536208.0,
      "step": 11345
    },
    {
      "entropy": 5.719861078262329,
      "epoch": 1.0295718432510885,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004901063528280856,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1711297169327736,
      "num_tokens": 23546393.0,
      "step": 11350
    },
    {
      "entropy": 5.730941581726074,
      "epoch": 1.0300253991291728,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004900968612865314,
      "loss": 5.4696,
      "mean_token_accuracy": 0.1664518266916275,
      "num_tokens": 23556869.0,
      "step": 11355
    },
    {
      "entropy": 5.722088813781738,
      "epoch": 1.0304789550072568,
      "grad_norm": 3.78125,
      "learning_rate": 0.0004900873652967195,
      "loss": 5.5667,
      "mean_token_accuracy": 0.15868754237890242,
      "num_tokens": 23568358.0,
      "step": 11360
    },
    {
      "entropy": 5.672901201248169,
      "epoch": 1.030932510885341,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004900778648588459,
      "loss": 5.4596,
      "mean_token_accuracy": 0.16742920875549316,
      "num_tokens": 23578748.0,
      "step": 11365
    },
    {
      "entropy": 5.839376068115234,
      "epoch": 1.0313860667634251,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004900683599731073,
      "loss": 5.5862,
      "mean_token_accuracy": 0.15849231481552123,
      "num_tokens": 23589390.0,
      "step": 11370
    },
    {
      "entropy": 5.787117052078247,
      "epoch": 1.0318396226415094,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004900588506397002,
      "loss": 5.5585,
      "mean_token_accuracy": 0.16275690943002702,
      "num_tokens": 23600858.0,
      "step": 11375
    },
    {
      "entropy": 5.764139890670776,
      "epoch": 1.0322931785195937,
      "grad_norm": 2.5,
      "learning_rate": 0.0004900493368588213,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16228657215833664,
      "num_tokens": 23611296.0,
      "step": 11380
    },
    {
      "entropy": 5.679147863388062,
      "epoch": 1.0327467343976777,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004900398186306672,
      "loss": 5.4926,
      "mean_token_accuracy": 0.16124074310064315,
      "num_tokens": 23622584.0,
      "step": 11385
    },
    {
      "entropy": 5.717840909957886,
      "epoch": 1.033200290275762,
      "grad_norm": 3.484375,
      "learning_rate": 0.000490030295955435,
      "loss": 5.5206,
      "mean_token_accuracy": 0.16097282767295837,
      "num_tokens": 23632831.0,
      "step": 11390
    },
    {
      "entropy": 5.793515682220459,
      "epoch": 1.0336538461538463,
      "grad_norm": 3.5,
      "learning_rate": 0.0004900207688333215,
      "loss": 5.5317,
      "mean_token_accuracy": 0.15800091475248337,
      "num_tokens": 23643025.0,
      "step": 11395
    },
    {
      "entropy": 5.826427888870239,
      "epoch": 1.0341074020319303,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004900112372645236,
      "loss": 5.6093,
      "mean_token_accuracy": 0.1544092506170273,
      "num_tokens": 23653367.0,
      "step": 11400
    },
    {
      "entropy": 5.8057561874389645,
      "epoch": 1.0345609579100146,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004900017012492386,
      "loss": 5.6086,
      "mean_token_accuracy": 0.15091422721743583,
      "num_tokens": 23663423.0,
      "step": 11405
    },
    {
      "entropy": 5.723331689834595,
      "epoch": 1.0350145137880986,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004899921607876634,
      "loss": 5.5817,
      "mean_token_accuracy": 0.15385486781597138,
      "num_tokens": 23674350.0,
      "step": 11410
    },
    {
      "entropy": 5.721215438842774,
      "epoch": 1.0354680696661829,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004899826158799958,
      "loss": 5.3994,
      "mean_token_accuracy": 0.17262798696756362,
      "num_tokens": 23683771.0,
      "step": 11415
    },
    {
      "entropy": 5.744855260848999,
      "epoch": 1.0359216255442671,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004899730665264328,
      "loss": 5.5606,
      "mean_token_accuracy": 0.15634527653455735,
      "num_tokens": 23694310.0,
      "step": 11420
    },
    {
      "entropy": 5.777623414993286,
      "epoch": 1.0363751814223512,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004899635127271721,
      "loss": 5.6153,
      "mean_token_accuracy": 0.1526750758290291,
      "num_tokens": 23704957.0,
      "step": 11425
    },
    {
      "entropy": 5.794167947769165,
      "epoch": 1.0368287373004355,
      "grad_norm": 3.328125,
      "learning_rate": 0.0004899539544824111,
      "loss": 5.495,
      "mean_token_accuracy": 0.16210370808839797,
      "num_tokens": 23714847.0,
      "step": 11430
    },
    {
      "entropy": 5.761078119277954,
      "epoch": 1.0372822931785195,
      "grad_norm": 2.25,
      "learning_rate": 0.0004899443917923476,
      "loss": 5.434,
      "mean_token_accuracy": 0.16689908802509307,
      "num_tokens": 23725035.0,
      "step": 11435
    },
    {
      "entropy": 5.665680885314941,
      "epoch": 1.0377358490566038,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004899348246571792,
      "loss": 5.461,
      "mean_token_accuracy": 0.1694757267832756,
      "num_tokens": 23735543.0,
      "step": 11440
    },
    {
      "entropy": 5.667950630187988,
      "epoch": 1.038189404934688,
      "grad_norm": 4.3125,
      "learning_rate": 0.000489925253077104,
      "loss": 5.46,
      "mean_token_accuracy": 0.16655135452747344,
      "num_tokens": 23746799.0,
      "step": 11445
    },
    {
      "entropy": 5.680819177627564,
      "epoch": 1.038642960812772,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004899156770523196,
      "loss": 5.5024,
      "mean_token_accuracy": 0.17014334797859193,
      "num_tokens": 23756068.0,
      "step": 11450
    },
    {
      "entropy": 5.695856380462646,
      "epoch": 1.0390965166908563,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004899060965830244,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16628849804401397,
      "num_tokens": 23766175.0,
      "step": 11455
    },
    {
      "entropy": 5.692382431030273,
      "epoch": 1.0395500725689404,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004898965116694162,
      "loss": 5.4652,
      "mean_token_accuracy": 0.16252631396055223,
      "num_tokens": 23776490.0,
      "step": 11460
    },
    {
      "entropy": 5.721606874465943,
      "epoch": 1.0400036284470247,
      "grad_norm": 2.5,
      "learning_rate": 0.0004898869223116935,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16547683775424957,
      "num_tokens": 23786925.0,
      "step": 11465
    },
    {
      "entropy": 5.7865630149841305,
      "epoch": 1.040457184325109,
      "grad_norm": 2.875,
      "learning_rate": 0.0004898773285100546,
      "loss": 5.5556,
      "mean_token_accuracy": 0.16131402999162675,
      "num_tokens": 23798918.0,
      "step": 11470
    },
    {
      "entropy": 5.801124954223633,
      "epoch": 1.040910740203193,
      "grad_norm": 3.375,
      "learning_rate": 0.0004898677302646975,
      "loss": 5.5235,
      "mean_token_accuracy": 0.16115800738334657,
      "num_tokens": 23809473.0,
      "step": 11475
    },
    {
      "entropy": 5.7377008438110355,
      "epoch": 1.0413642960812772,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004898581275758211,
      "loss": 5.5349,
      "mean_token_accuracy": 0.16235569268465042,
      "num_tokens": 23819256.0,
      "step": 11480
    },
    {
      "entropy": 5.682327938079834,
      "epoch": 1.0418178519593615,
      "grad_norm": 3.625,
      "learning_rate": 0.0004898485204436239,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16149462908506393,
      "num_tokens": 23828591.0,
      "step": 11485
    },
    {
      "entropy": 5.720239925384521,
      "epoch": 1.0422714078374455,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004898389088683047,
      "loss": 5.4991,
      "mean_token_accuracy": 0.1606788828969002,
      "num_tokens": 23839704.0,
      "step": 11490
    },
    {
      "entropy": 5.721277570724487,
      "epoch": 1.0427249637155298,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004898292928500619,
      "loss": 5.4974,
      "mean_token_accuracy": 0.16450997292995453,
      "num_tokens": 23851117.0,
      "step": 11495
    },
    {
      "entropy": 5.714069938659668,
      "epoch": 1.0431785195936139,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004898196723890947,
      "loss": 5.5017,
      "mean_token_accuracy": 0.16115083247423173,
      "num_tokens": 23861347.0,
      "step": 11500
    },
    {
      "entropy": 5.760830593109131,
      "epoch": 1.0436320754716981,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004898100474856019,
      "loss": 5.5037,
      "mean_token_accuracy": 0.16236412078142165,
      "num_tokens": 23870910.0,
      "step": 11505
    },
    {
      "entropy": 5.782321548461914,
      "epoch": 1.0440856313497824,
      "grad_norm": 3.375,
      "learning_rate": 0.0004898004181397824,
      "loss": 5.5123,
      "mean_token_accuracy": 0.16165544986724853,
      "num_tokens": 23881194.0,
      "step": 11510
    },
    {
      "entropy": 5.719546318054199,
      "epoch": 1.0445391872278664,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004897907843518358,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16147395223379135,
      "num_tokens": 23891386.0,
      "step": 11515
    },
    {
      "entropy": 5.741972255706787,
      "epoch": 1.0449927431059507,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004897811461219609,
      "loss": 5.5938,
      "mean_token_accuracy": 0.1579977683722973,
      "num_tokens": 23902751.0,
      "step": 11520
    },
    {
      "entropy": 5.6795188903808596,
      "epoch": 1.0454462989840347,
      "grad_norm": 2.75,
      "learning_rate": 0.0004897715034503571,
      "loss": 5.4684,
      "mean_token_accuracy": 0.16300767809152603,
      "num_tokens": 23912499.0,
      "step": 11525
    },
    {
      "entropy": 5.700448322296142,
      "epoch": 1.045899854862119,
      "grad_norm": 2.640625,
      "learning_rate": 0.000489761856337224,
      "loss": 5.4842,
      "mean_token_accuracy": 0.16051235496997834,
      "num_tokens": 23923224.0,
      "step": 11530
    },
    {
      "entropy": 5.782007455825806,
      "epoch": 1.0463534107402033,
      "grad_norm": 3.125,
      "learning_rate": 0.0004897522047827609,
      "loss": 5.5414,
      "mean_token_accuracy": 0.1592988833785057,
      "num_tokens": 23933543.0,
      "step": 11535
    },
    {
      "entropy": 5.787376403808594,
      "epoch": 1.0468069666182873,
      "grad_norm": 5.0625,
      "learning_rate": 0.0004897425487871676,
      "loss": 5.5802,
      "mean_token_accuracy": 0.15683628916740416,
      "num_tokens": 23943894.0,
      "step": 11540
    },
    {
      "entropy": 5.6655004024505615,
      "epoch": 1.0472605224963716,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004897328883506436,
      "loss": 5.4971,
      "mean_token_accuracy": 0.16208865642547607,
      "num_tokens": 23953833.0,
      "step": 11545
    },
    {
      "entropy": 5.720190811157226,
      "epoch": 1.0477140783744556,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004897232234733887,
      "loss": 5.547,
      "mean_token_accuracy": 0.15657385289669037,
      "num_tokens": 23963730.0,
      "step": 11550
    },
    {
      "entropy": 5.841046476364136,
      "epoch": 1.04816763425254,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004897135541556028,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16438598334789276,
      "num_tokens": 23974640.0,
      "step": 11555
    },
    {
      "entropy": 5.743403720855713,
      "epoch": 1.0486211901306242,
      "grad_norm": 2.53125,
      "learning_rate": 0.000489703880397486,
      "loss": 5.446,
      "mean_token_accuracy": 0.1636401429772377,
      "num_tokens": 23984840.0,
      "step": 11560
    },
    {
      "entropy": 5.723632574081421,
      "epoch": 1.0490747460087082,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004896942021992382,
      "loss": 5.5092,
      "mean_token_accuracy": 0.16144366711378097,
      "num_tokens": 23995781.0,
      "step": 11565
    },
    {
      "entropy": 5.7233422756195065,
      "epoch": 1.0495283018867925,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004896845195610595,
      "loss": 5.4605,
      "mean_token_accuracy": 0.16311514675617217,
      "num_tokens": 24005420.0,
      "step": 11570
    },
    {
      "entropy": 5.7589069366455075,
      "epoch": 1.0499818577648767,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004896748324831504,
      "loss": 5.3821,
      "mean_token_accuracy": 0.17135454714298248,
      "num_tokens": 24015892.0,
      "step": 11575
    },
    {
      "entropy": 5.729975605010987,
      "epoch": 1.0504354136429608,
      "grad_norm": 3.5,
      "learning_rate": 0.0004896651409657109,
      "loss": 5.4501,
      "mean_token_accuracy": 0.16824133694171906,
      "num_tokens": 24027206.0,
      "step": 11580
    },
    {
      "entropy": 5.635092782974243,
      "epoch": 1.050888969521045,
      "grad_norm": 5.03125,
      "learning_rate": 0.0004896554450089417,
      "loss": 5.5267,
      "mean_token_accuracy": 0.15955890119075775,
      "num_tokens": 24037518.0,
      "step": 11585
    },
    {
      "entropy": 5.7148387908935545,
      "epoch": 1.051342525399129,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004896457446130431,
      "loss": 5.4886,
      "mean_token_accuracy": 0.1600768819451332,
      "num_tokens": 24048372.0,
      "step": 11590
    },
    {
      "entropy": 5.747332191467285,
      "epoch": 1.0517960812772134,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004896360397782159,
      "loss": 5.4699,
      "mean_token_accuracy": 0.16056328266859055,
      "num_tokens": 24058182.0,
      "step": 11595
    },
    {
      "entropy": 5.678472518920898,
      "epoch": 1.0522496371552976,
      "grad_norm": 2.25,
      "learning_rate": 0.0004896263305046607,
      "loss": 5.4473,
      "mean_token_accuracy": 0.16772022247314453,
      "num_tokens": 24068342.0,
      "step": 11600
    },
    {
      "entropy": 5.714088106155396,
      "epoch": 1.0527031930333817,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004896166167925782,
      "loss": 5.4445,
      "mean_token_accuracy": 0.16501030176877976,
      "num_tokens": 24078560.0,
      "step": 11605
    },
    {
      "entropy": 5.758160066604614,
      "epoch": 1.053156748911466,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004896068986421693,
      "loss": 5.6274,
      "mean_token_accuracy": 0.15621067434549332,
      "num_tokens": 24090616.0,
      "step": 11610
    },
    {
      "entropy": 5.797816801071167,
      "epoch": 1.05361030478955,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004895971760536351,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1658847376704216,
      "num_tokens": 24100301.0,
      "step": 11615
    },
    {
      "entropy": 5.727489185333252,
      "epoch": 1.0540638606676342,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004895874490271768,
      "loss": 5.4692,
      "mean_token_accuracy": 0.15674588680267335,
      "num_tokens": 24110306.0,
      "step": 11620
    },
    {
      "entropy": 5.647307968139648,
      "epoch": 1.0545174165457185,
      "grad_norm": 4.6875,
      "learning_rate": 0.0004895777175629952,
      "loss": 5.4755,
      "mean_token_accuracy": 0.16304078400135041,
      "num_tokens": 24120300.0,
      "step": 11625
    },
    {
      "entropy": 5.726163721084594,
      "epoch": 1.0549709724238026,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004895679816612917,
      "loss": 5.4802,
      "mean_token_accuracy": 0.16591937243938445,
      "num_tokens": 24130710.0,
      "step": 11630
    },
    {
      "entropy": 5.764380979537964,
      "epoch": 1.0554245283018868,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004895582413222678,
      "loss": 5.5505,
      "mean_token_accuracy": 0.16078477948904038,
      "num_tokens": 24141100.0,
      "step": 11635
    },
    {
      "entropy": 5.766975688934326,
      "epoch": 1.0558780841799709,
      "grad_norm": 2.375,
      "learning_rate": 0.0004895484965461247,
      "loss": 5.4967,
      "mean_token_accuracy": 0.15915921628475188,
      "num_tokens": 24150471.0,
      "step": 11640
    },
    {
      "entropy": 5.735536289215088,
      "epoch": 1.0563316400580551,
      "grad_norm": 2.515625,
      "learning_rate": 0.000489538747333064,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1476710170507431,
      "num_tokens": 24161122.0,
      "step": 11645
    },
    {
      "entropy": 5.7045050144195555,
      "epoch": 1.0567851959361394,
      "grad_norm": 3.53125,
      "learning_rate": 0.0004895289936832874,
      "loss": 5.5554,
      "mean_token_accuracy": 0.15462986752390862,
      "num_tokens": 24171257.0,
      "step": 11650
    },
    {
      "entropy": 5.6969311237335205,
      "epoch": 1.0572387518142234,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004895192355969966,
      "loss": 5.4499,
      "mean_token_accuracy": 0.16754541993141175,
      "num_tokens": 24181522.0,
      "step": 11655
    },
    {
      "entropy": 5.715392923355102,
      "epoch": 1.0576923076923077,
      "grad_norm": 2.5,
      "learning_rate": 0.0004895094730743932,
      "loss": 5.4643,
      "mean_token_accuracy": 0.16701870262622834,
      "num_tokens": 24192060.0,
      "step": 11660
    },
    {
      "entropy": 5.78725118637085,
      "epoch": 1.058145863570392,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004894997061156793,
      "loss": 5.6361,
      "mean_token_accuracy": 0.15588540732860565,
      "num_tokens": 24201987.0,
      "step": 11665
    },
    {
      "entropy": 5.804919004440308,
      "epoch": 1.058599419448476,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004894899347210567,
      "loss": 5.5157,
      "mean_token_accuracy": 0.16373774260282517,
      "num_tokens": 24211845.0,
      "step": 11670
    },
    {
      "entropy": 5.813258075714112,
      "epoch": 1.0590529753265603,
      "grad_norm": 2.375,
      "learning_rate": 0.0004894801588907278,
      "loss": 5.551,
      "mean_token_accuracy": 0.1600181505084038,
      "num_tokens": 24222959.0,
      "step": 11675
    },
    {
      "entropy": 5.745786905288696,
      "epoch": 1.0595065312046443,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004894703786248944,
      "loss": 5.5109,
      "mean_token_accuracy": 0.1627909779548645,
      "num_tokens": 24232815.0,
      "step": 11680
    },
    {
      "entropy": 5.625432538986206,
      "epoch": 1.0599600870827286,
      "grad_norm": 2.21875,
      "learning_rate": 0.000489460593923759,
      "loss": 5.5574,
      "mean_token_accuracy": 0.15919324979186059,
      "num_tokens": 24243606.0,
      "step": 11685
    },
    {
      "entropy": 5.674048614501953,
      "epoch": 1.0604136429608129,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004894508047875237,
      "loss": 5.4504,
      "mean_token_accuracy": 0.15797059014439582,
      "num_tokens": 24254133.0,
      "step": 11690
    },
    {
      "entropy": 5.775349187850952,
      "epoch": 1.060867198838897,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004894410112163913,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16682333052158355,
      "num_tokens": 24263442.0,
      "step": 11695
    },
    {
      "entropy": 5.7040611743927006,
      "epoch": 1.0613207547169812,
      "grad_norm": 5.875,
      "learning_rate": 0.0004894312132105641,
      "loss": 5.4525,
      "mean_token_accuracy": 0.16754695475101472,
      "num_tokens": 24273548.0,
      "step": 11700
    },
    {
      "entropy": 5.799657154083252,
      "epoch": 1.0617743105950652,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004894214107702447,
      "loss": 5.4974,
      "mean_token_accuracy": 0.16441601514816284,
      "num_tokens": 24284690.0,
      "step": 11705
    },
    {
      "entropy": 5.736912488937378,
      "epoch": 1.0622278664731495,
      "grad_norm": 2.515625,
      "learning_rate": 0.000489411603895636,
      "loss": 5.597,
      "mean_token_accuracy": 0.15134185403585435,
      "num_tokens": 24295362.0,
      "step": 11710
    },
    {
      "entropy": 5.659896516799927,
      "epoch": 1.0626814223512338,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004894017925869406,
      "loss": 5.5111,
      "mean_token_accuracy": 0.16038587540388108,
      "num_tokens": 24305032.0,
      "step": 11715
    },
    {
      "entropy": 5.738727760314942,
      "epoch": 1.0631349782293178,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004893919768443615,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1570585139095783,
      "num_tokens": 24315894.0,
      "step": 11720
    },
    {
      "entropy": 5.85515341758728,
      "epoch": 1.063588534107402,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004893821566681017,
      "loss": 5.5456,
      "mean_token_accuracy": 0.16425900757312775,
      "num_tokens": 24325445.0,
      "step": 11725
    },
    {
      "entropy": 5.696656036376953,
      "epoch": 1.0640420899854863,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004893723320583644,
      "loss": 5.4766,
      "mean_token_accuracy": 0.16507300436496736,
      "num_tokens": 24335618.0,
      "step": 11730
    },
    {
      "entropy": 5.713216543197632,
      "epoch": 1.0644956458635704,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004893625030153525,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1638691782951355,
      "num_tokens": 24344981.0,
      "step": 11735
    },
    {
      "entropy": 5.668301916122436,
      "epoch": 1.0649492017416546,
      "grad_norm": 3.125,
      "learning_rate": 0.0004893526695392694,
      "loss": 5.4327,
      "mean_token_accuracy": 0.16402191817760467,
      "num_tokens": 24354957.0,
      "step": 11740
    },
    {
      "entropy": 5.686540651321411,
      "epoch": 1.0654027576197387,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004893428316303185,
      "loss": 5.4656,
      "mean_token_accuracy": 0.15925028026103974,
      "num_tokens": 24364904.0,
      "step": 11745
    },
    {
      "entropy": 5.733685302734375,
      "epoch": 1.065856313497823,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004893329892887031,
      "loss": 5.5545,
      "mean_token_accuracy": 0.16397877037525177,
      "num_tokens": 24374904.0,
      "step": 11750
    },
    {
      "entropy": 5.788058805465698,
      "epoch": 1.0663098693759072,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004893231425146271,
      "loss": 5.5086,
      "mean_token_accuracy": 0.16399105042219161,
      "num_tokens": 24384820.0,
      "step": 11755
    },
    {
      "entropy": 5.702828598022461,
      "epoch": 1.0667634252539913,
      "grad_norm": 2.0,
      "learning_rate": 0.0004893132913082937,
      "loss": 5.5219,
      "mean_token_accuracy": 0.16408717930316924,
      "num_tokens": 24394465.0,
      "step": 11760
    },
    {
      "entropy": 5.73837456703186,
      "epoch": 1.0672169811320755,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004893034356699069,
      "loss": 5.5228,
      "mean_token_accuracy": 0.16932104676961898,
      "num_tokens": 24404361.0,
      "step": 11765
    },
    {
      "entropy": 5.730602407455445,
      "epoch": 1.0676705370101596,
      "grad_norm": 2.875,
      "learning_rate": 0.0004892935755996705,
      "loss": 5.5275,
      "mean_token_accuracy": 0.15826882123947145,
      "num_tokens": 24414841.0,
      "step": 11770
    },
    {
      "entropy": 5.682967567443848,
      "epoch": 1.0681240928882438,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004892837110977883,
      "loss": 5.4204,
      "mean_token_accuracy": 0.17234706580638887,
      "num_tokens": 24424292.0,
      "step": 11775
    },
    {
      "entropy": 5.715915298461914,
      "epoch": 1.068577648766328,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004892738421644644,
      "loss": 5.5783,
      "mean_token_accuracy": 0.15604971200227738,
      "num_tokens": 24435069.0,
      "step": 11780
    },
    {
      "entropy": 5.687619018554687,
      "epoch": 1.0690312046444121,
      "grad_norm": 13.375,
      "learning_rate": 0.0004892639687999027,
      "loss": 5.3819,
      "mean_token_accuracy": 0.16114187389612197,
      "num_tokens": 24445379.0,
      "step": 11785
    },
    {
      "entropy": 5.767459774017334,
      "epoch": 1.0694847605224964,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004892540910043076,
      "loss": 5.5067,
      "mean_token_accuracy": 0.165464349091053,
      "num_tokens": 24455094.0,
      "step": 11790
    },
    {
      "entropy": 5.6730055809021,
      "epoch": 1.0699383164005805,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004892442087778833,
      "loss": 5.5212,
      "mean_token_accuracy": 0.16512048244476318,
      "num_tokens": 24464889.0,
      "step": 11795
    },
    {
      "entropy": 5.7394458770751955,
      "epoch": 1.0703918722786647,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004892343221208342,
      "loss": 5.4888,
      "mean_token_accuracy": 0.16280749291181565,
      "num_tokens": 24474676.0,
      "step": 11800
    },
    {
      "entropy": 5.786469411849976,
      "epoch": 1.070845428156749,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004892244310333646,
      "loss": 5.514,
      "mean_token_accuracy": 0.16460341066122056,
      "num_tokens": 24483924.0,
      "step": 11805
    },
    {
      "entropy": 5.702552604675293,
      "epoch": 1.071298984034833,
      "grad_norm": 2.5,
      "learning_rate": 0.0004892145355156793,
      "loss": 5.511,
      "mean_token_accuracy": 0.1561882495880127,
      "num_tokens": 24494331.0,
      "step": 11810
    },
    {
      "entropy": 5.741305017471314,
      "epoch": 1.0717525399129173,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004892046355679829,
      "loss": 5.4959,
      "mean_token_accuracy": 0.16068216115236283,
      "num_tokens": 24504871.0,
      "step": 11815
    },
    {
      "entropy": 5.7645971298217775,
      "epoch": 1.0722060957910013,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004891947311904799,
      "loss": 5.556,
      "mean_token_accuracy": 0.15722292065620422,
      "num_tokens": 24515765.0,
      "step": 11820
    },
    {
      "entropy": 5.661913776397705,
      "epoch": 1.0726596516690856,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004891848223833753,
      "loss": 5.5071,
      "mean_token_accuracy": 0.15962351411581038,
      "num_tokens": 24528467.0,
      "step": 11825
    },
    {
      "entropy": 5.751865434646606,
      "epoch": 1.0731132075471699,
      "grad_norm": 2.03125,
      "learning_rate": 0.000489174909146874,
      "loss": 5.4926,
      "mean_token_accuracy": 0.16506169289350509,
      "num_tokens": 24538777.0,
      "step": 11830
    },
    {
      "entropy": 5.684018659591675,
      "epoch": 1.073566763425254,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004891649914811811,
      "loss": 5.4128,
      "mean_token_accuracy": 0.16000169068574904,
      "num_tokens": 24550031.0,
      "step": 11835
    },
    {
      "entropy": 5.690298461914063,
      "epoch": 1.0740203193033382,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004891550693865014,
      "loss": 5.4381,
      "mean_token_accuracy": 0.17177549600601197,
      "num_tokens": 24559577.0,
      "step": 11840
    },
    {
      "entropy": 5.705758571624756,
      "epoch": 1.0744738751814222,
      "grad_norm": 3.375,
      "learning_rate": 0.0004891451428630405,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16973724514245986,
      "num_tokens": 24569771.0,
      "step": 11845
    },
    {
      "entropy": 5.758140325546265,
      "epoch": 1.0749274310595065,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004891352119110033,
      "loss": 5.4166,
      "mean_token_accuracy": 0.1702731281518936,
      "num_tokens": 24579167.0,
      "step": 11850
    },
    {
      "entropy": 5.669926166534424,
      "epoch": 1.0753809869375908,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004891252765305954,
      "loss": 5.4338,
      "mean_token_accuracy": 0.17733323127031325,
      "num_tokens": 24589275.0,
      "step": 11855
    },
    {
      "entropy": 5.685131788253784,
      "epoch": 1.0758345428156748,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004891153367220222,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1604471132159233,
      "num_tokens": 24601118.0,
      "step": 11860
    },
    {
      "entropy": 5.748859548568726,
      "epoch": 1.076288098693759,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004891053924854894,
      "loss": 5.5698,
      "mean_token_accuracy": 0.1654861569404602,
      "num_tokens": 24611384.0,
      "step": 11865
    },
    {
      "entropy": 5.745014476776123,
      "epoch": 1.0767416545718433,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004890954438212024,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16209276914596557,
      "num_tokens": 24622050.0,
      "step": 11870
    },
    {
      "entropy": 5.673732662200928,
      "epoch": 1.0771952104499274,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004890854907293671,
      "loss": 5.4669,
      "mean_token_accuracy": 0.1674303501844406,
      "num_tokens": 24631242.0,
      "step": 11875
    },
    {
      "entropy": 5.776395988464356,
      "epoch": 1.0776487663280117,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004890755332101893,
      "loss": 5.5426,
      "mean_token_accuracy": 0.1598892480134964,
      "num_tokens": 24641716.0,
      "step": 11880
    },
    {
      "entropy": 5.7612744808197025,
      "epoch": 1.0781023222060957,
      "grad_norm": 2.75,
      "learning_rate": 0.0004890655712638749,
      "loss": 5.4916,
      "mean_token_accuracy": 0.16443771123886108,
      "num_tokens": 24652344.0,
      "step": 11885
    },
    {
      "entropy": 5.676549005508423,
      "epoch": 1.07855587808418,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048905560489063,
      "loss": 5.4418,
      "mean_token_accuracy": 0.16579073667526245,
      "num_tokens": 24662832.0,
      "step": 11890
    },
    {
      "entropy": 5.7199602127075195,
      "epoch": 1.0790094339622642,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004890456340906605,
      "loss": 5.5272,
      "mean_token_accuracy": 0.15533642172813417,
      "num_tokens": 24673137.0,
      "step": 11895
    },
    {
      "entropy": 5.772547769546509,
      "epoch": 1.0794629898403483,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004890356588641729,
      "loss": 5.4499,
      "mean_token_accuracy": 0.1572963759303093,
      "num_tokens": 24683375.0,
      "step": 11900
    },
    {
      "entropy": 5.766538906097412,
      "epoch": 1.0799165457184325,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004890256792113733,
      "loss": 5.5397,
      "mean_token_accuracy": 0.15947365313768386,
      "num_tokens": 24693679.0,
      "step": 11905
    },
    {
      "entropy": 5.651413488388061,
      "epoch": 1.0803701015965168,
      "grad_norm": 4.125,
      "learning_rate": 0.0004890156951324681,
      "loss": 5.419,
      "mean_token_accuracy": 0.16723806113004686,
      "num_tokens": 24704428.0,
      "step": 11910
    },
    {
      "entropy": 5.701001501083374,
      "epoch": 1.0808236574746009,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004890057066276636,
      "loss": 5.4255,
      "mean_token_accuracy": 0.16004096567630768,
      "num_tokens": 24714552.0,
      "step": 11915
    },
    {
      "entropy": 5.687047624588013,
      "epoch": 1.0812772133526851,
      "grad_norm": 2.625,
      "learning_rate": 0.0004889957136971668,
      "loss": 5.5179,
      "mean_token_accuracy": 0.1643939346075058,
      "num_tokens": 24724989.0,
      "step": 11920
    },
    {
      "entropy": 5.776320266723633,
      "epoch": 1.0817307692307692,
      "grad_norm": 2.15625,
      "learning_rate": 0.000488985716341184,
      "loss": 5.5158,
      "mean_token_accuracy": 0.1646418571472168,
      "num_tokens": 24736473.0,
      "step": 11925
    },
    {
      "entropy": 5.684682893753052,
      "epoch": 1.0821843251088534,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004889757145599222,
      "loss": 5.4707,
      "mean_token_accuracy": 0.1629565417766571,
      "num_tokens": 24746332.0,
      "step": 11930
    },
    {
      "entropy": 5.6986058235168455,
      "epoch": 1.0826378809869377,
      "grad_norm": 4.28125,
      "learning_rate": 0.000488965708353588,
      "loss": 5.4624,
      "mean_token_accuracy": 0.16526600122451782,
      "num_tokens": 24756164.0,
      "step": 11935
    },
    {
      "entropy": 5.769347143173218,
      "epoch": 1.0830914368650217,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004889556977223883,
      "loss": 5.5066,
      "mean_token_accuracy": 0.17026737928390503,
      "num_tokens": 24766778.0,
      "step": 11940
    },
    {
      "entropy": 5.697160387039185,
      "epoch": 1.083544992743106,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004889456826665304,
      "loss": 5.4211,
      "mean_token_accuracy": 0.1636136531829834,
      "num_tokens": 24776252.0,
      "step": 11945
    },
    {
      "entropy": 5.655699968338013,
      "epoch": 1.08399854862119,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004889356631862211,
      "loss": 5.4529,
      "mean_token_accuracy": 0.163382588326931,
      "num_tokens": 24786439.0,
      "step": 11950
    },
    {
      "entropy": 5.6982697486877445,
      "epoch": 1.0844521044992743,
      "grad_norm": 2.9375,
      "learning_rate": 0.000488925639281668,
      "loss": 5.46,
      "mean_token_accuracy": 0.16267504096031188,
      "num_tokens": 24796212.0,
      "step": 11955
    },
    {
      "entropy": 5.67959303855896,
      "epoch": 1.0849056603773586,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004889156109530781,
      "loss": 5.4703,
      "mean_token_accuracy": 0.16297507137060166,
      "num_tokens": 24807279.0,
      "step": 11960
    },
    {
      "entropy": 5.654568767547607,
      "epoch": 1.0853592162554426,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004889055782006588,
      "loss": 5.5053,
      "mean_token_accuracy": 0.16001011729240416,
      "num_tokens": 24818808.0,
      "step": 11965
    },
    {
      "entropy": 5.768672943115234,
      "epoch": 1.085812772133527,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004888955410246176,
      "loss": 5.4959,
      "mean_token_accuracy": 0.16443366855382918,
      "num_tokens": 24828938.0,
      "step": 11970
    },
    {
      "entropy": 5.745870351791382,
      "epoch": 1.086266328011611,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004888854994251621,
      "loss": 5.4936,
      "mean_token_accuracy": 0.16899036467075348,
      "num_tokens": 24838203.0,
      "step": 11975
    },
    {
      "entropy": 5.660204601287842,
      "epoch": 1.0867198838896952,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004888754534025,
      "loss": 5.5396,
      "mean_token_accuracy": 0.15934212356805802,
      "num_tokens": 24848897.0,
      "step": 11980
    },
    {
      "entropy": 5.611825609207154,
      "epoch": 1.0871734397677795,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004888654029568391,
      "loss": 5.3876,
      "mean_token_accuracy": 0.17074303776025773,
      "num_tokens": 24858883.0,
      "step": 11985
    },
    {
      "entropy": 5.719109344482422,
      "epoch": 1.0876269956458635,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004888553480883871,
      "loss": 5.58,
      "mean_token_accuracy": 0.15014299899339675,
      "num_tokens": 24870180.0,
      "step": 11990
    },
    {
      "entropy": 5.778216361999512,
      "epoch": 1.0880805515239478,
      "grad_norm": 2.46875,
      "learning_rate": 0.000488845288797352,
      "loss": 5.4928,
      "mean_token_accuracy": 0.16264058649539948,
      "num_tokens": 24881263.0,
      "step": 11995
    },
    {
      "entropy": 5.738774967193604,
      "epoch": 1.0885341074020318,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004888352250839418,
      "loss": 5.5215,
      "mean_token_accuracy": 0.16275340020656587,
      "num_tokens": 24892231.0,
      "step": 12000
    },
    {
      "epoch": 1.0885341074020318,
      "eval_entropy": 5.525525393097649,
      "eval_loss": 5.5782694816589355,
      "eval_mean_token_accuracy": 0.1667893212254905,
      "eval_num_tokens": 24892231.0,
      "eval_runtime": 26.5676,
      "eval_samples_per_second": 1330.942,
      "eval_steps_per_second": 166.368,
      "step": 12000
    },
    {
      "entropy": 5.594419288635254,
      "epoch": 1.088987663280116,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004888251569483647,
      "loss": 5.3945,
      "mean_token_accuracy": 0.16971904784440994,
      "num_tokens": 24902720.0,
      "step": 12005
    },
    {
      "entropy": 5.6569194316864015,
      "epoch": 1.0894412191582004,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004888150843908288,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1724456861615181,
      "num_tokens": 24913346.0,
      "step": 12010
    },
    {
      "entropy": 5.7564393997192385,
      "epoch": 1.0898947750362844,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004888050074115426,
      "loss": 5.5314,
      "mean_token_accuracy": 0.16175781190395355,
      "num_tokens": 24924409.0,
      "step": 12015
    },
    {
      "entropy": 5.722735404968262,
      "epoch": 1.0903483309143687,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004887949260107143,
      "loss": 5.4862,
      "mean_token_accuracy": 0.1640880823135376,
      "num_tokens": 24934711.0,
      "step": 12020
    },
    {
      "entropy": 5.687616348266602,
      "epoch": 1.0908018867924527,
      "grad_norm": 2.25,
      "learning_rate": 0.0004887848401885524,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15834415405988694,
      "num_tokens": 24945989.0,
      "step": 12025
    },
    {
      "entropy": 5.799186134338379,
      "epoch": 1.091255442670537,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004887747499452656,
      "loss": 5.5127,
      "mean_token_accuracy": 0.16659198254346846,
      "num_tokens": 24955354.0,
      "step": 12030
    },
    {
      "entropy": 5.645429420471191,
      "epoch": 1.0917089985486212,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004887646552810624,
      "loss": 5.4189,
      "mean_token_accuracy": 0.16547064781188964,
      "num_tokens": 24965906.0,
      "step": 12035
    },
    {
      "entropy": 5.751064205169678,
      "epoch": 1.0921625544267053,
      "grad_norm": 2.125,
      "learning_rate": 0.0004887545561961517,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1594035655260086,
      "num_tokens": 24975826.0,
      "step": 12040
    },
    {
      "entropy": 5.720312309265137,
      "epoch": 1.0926161103047896,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004887444526907424,
      "loss": 5.466,
      "mean_token_accuracy": 0.16367220431566237,
      "num_tokens": 24986409.0,
      "step": 12045
    },
    {
      "entropy": 5.739640808105468,
      "epoch": 1.0930696661828738,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004887343447650432,
      "loss": 5.4887,
      "mean_token_accuracy": 0.1637328378856182,
      "num_tokens": 24996455.0,
      "step": 12050
    },
    {
      "entropy": 5.693829917907715,
      "epoch": 1.0935232220609579,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004887242324192634,
      "loss": 5.4804,
      "mean_token_accuracy": 0.16175874471664428,
      "num_tokens": 25007626.0,
      "step": 12055
    },
    {
      "entropy": 5.7655041217803955,
      "epoch": 1.0939767779390421,
      "grad_norm": 4.75,
      "learning_rate": 0.0004887141156536119,
      "loss": 5.54,
      "mean_token_accuracy": 0.15850070714950562,
      "num_tokens": 25018357.0,
      "step": 12060
    },
    {
      "entropy": 5.664424991607666,
      "epoch": 1.0944303338171262,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004887039944682981,
      "loss": 5.4968,
      "mean_token_accuracy": 0.15514399111270905,
      "num_tokens": 25029356.0,
      "step": 12065
    },
    {
      "entropy": 5.724434947967529,
      "epoch": 1.0948838896952104,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004886938688635313,
      "loss": 5.5243,
      "mean_token_accuracy": 0.15394580215215684,
      "num_tokens": 25040043.0,
      "step": 12070
    },
    {
      "entropy": 5.748786497116089,
      "epoch": 1.0953374455732947,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004886837388395207,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16013906821608542,
      "num_tokens": 25050441.0,
      "step": 12075
    },
    {
      "entropy": 5.710179662704467,
      "epoch": 1.0957910014513788,
      "grad_norm": 1.71875,
      "learning_rate": 0.000488673604396476,
      "loss": 5.5373,
      "mean_token_accuracy": 0.15877657234668732,
      "num_tokens": 25060627.0,
      "step": 12080
    },
    {
      "entropy": 5.6502485275268555,
      "epoch": 1.096244557329463,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004886634655346067,
      "loss": 5.441,
      "mean_token_accuracy": 0.1655240163207054,
      "num_tokens": 25070860.0,
      "step": 12085
    },
    {
      "entropy": 5.792421865463257,
      "epoch": 1.0966981132075473,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004886533222541225,
      "loss": 5.5389,
      "mean_token_accuracy": 0.16736784130334853,
      "num_tokens": 25081525.0,
      "step": 12090
    },
    {
      "entropy": 5.756735038757324,
      "epoch": 1.0971516690856313,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004886431745552331,
      "loss": 5.4941,
      "mean_token_accuracy": 0.16310690641403197,
      "num_tokens": 25091285.0,
      "step": 12095
    },
    {
      "entropy": 5.639165449142456,
      "epoch": 1.0976052249637156,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004886330224381485,
      "loss": 5.4668,
      "mean_token_accuracy": 0.16173416823148729,
      "num_tokens": 25100798.0,
      "step": 12100
    },
    {
      "entropy": 5.7160196781158445,
      "epoch": 1.0980587808417996,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004886228659030785,
      "loss": 5.6085,
      "mean_token_accuracy": 0.15853865146636964,
      "num_tokens": 25111244.0,
      "step": 12105
    },
    {
      "entropy": 5.7493102073669435,
      "epoch": 1.098512336719884,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004886127049502331,
      "loss": 5.4585,
      "mean_token_accuracy": 0.1666278675198555,
      "num_tokens": 25120928.0,
      "step": 12110
    },
    {
      "entropy": 5.718634033203125,
      "epoch": 1.0989658925979682,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004886025395798228,
      "loss": 5.4481,
      "mean_token_accuracy": 0.17016846537590027,
      "num_tokens": 25130431.0,
      "step": 12115
    },
    {
      "entropy": 5.682046937942505,
      "epoch": 1.0994194484760522,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004885923697920573,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1700902536511421,
      "num_tokens": 25140142.0,
      "step": 12120
    },
    {
      "entropy": 5.749057817459106,
      "epoch": 1.0998730043541365,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004885821955871472,
      "loss": 5.5012,
      "mean_token_accuracy": 0.15432314872741698,
      "num_tokens": 25150976.0,
      "step": 12125
    },
    {
      "entropy": 5.697264099121094,
      "epoch": 1.1003265602322205,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004885720169653028,
      "loss": 5.4406,
      "mean_token_accuracy": 0.16945091634988785,
      "num_tokens": 25160904.0,
      "step": 12130
    },
    {
      "entropy": 5.677787446975708,
      "epoch": 1.1007801161103048,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004885618339267348,
      "loss": 5.4992,
      "mean_token_accuracy": 0.16291617900133132,
      "num_tokens": 25170164.0,
      "step": 12135
    },
    {
      "entropy": 5.71990327835083,
      "epoch": 1.101233671988389,
      "grad_norm": 2.0,
      "learning_rate": 0.0004885516464716537,
      "loss": 5.4536,
      "mean_token_accuracy": 0.15918266922235488,
      "num_tokens": 25180290.0,
      "step": 12140
    },
    {
      "entropy": 5.695782041549682,
      "epoch": 1.101687227866473,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048854145460027,
      "loss": 5.4922,
      "mean_token_accuracy": 0.15823221057653428,
      "num_tokens": 25190369.0,
      "step": 12145
    },
    {
      "entropy": 5.689862537384033,
      "epoch": 1.1021407837445574,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004885312583127946,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1673481598496437,
      "num_tokens": 25200195.0,
      "step": 12150
    },
    {
      "entropy": 5.718158006668091,
      "epoch": 1.1025943396226414,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004885210576094385,
      "loss": 5.4681,
      "mean_token_accuracy": 0.1603190615773201,
      "num_tokens": 25210915.0,
      "step": 12155
    },
    {
      "entropy": 5.778222751617432,
      "epoch": 1.1030478955007257,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004885108524904124,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1632695123553276,
      "num_tokens": 25220683.0,
      "step": 12160
    },
    {
      "entropy": 5.719636297225952,
      "epoch": 1.10350145137881,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004885006429559275,
      "loss": 5.4834,
      "mean_token_accuracy": 0.16410521566867828,
      "num_tokens": 25230662.0,
      "step": 12165
    },
    {
      "entropy": 5.716928386688233,
      "epoch": 1.103955007256894,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004884904290061949,
      "loss": 5.5265,
      "mean_token_accuracy": 0.16244978457689285,
      "num_tokens": 25240932.0,
      "step": 12170
    },
    {
      "entropy": 5.675851106643677,
      "epoch": 1.1044085631349783,
      "grad_norm": 2.25,
      "learning_rate": 0.0004884802106414258,
      "loss": 5.4477,
      "mean_token_accuracy": 0.1672980859875679,
      "num_tokens": 25250945.0,
      "step": 12175
    },
    {
      "entropy": 5.731085729598999,
      "epoch": 1.1048621190130623,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004884699878618316,
      "loss": 5.5513,
      "mean_token_accuracy": 0.1670376718044281,
      "num_tokens": 25260747.0,
      "step": 12180
    },
    {
      "entropy": 5.741350984573364,
      "epoch": 1.1053156748911466,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004884597606676236,
      "loss": 5.5265,
      "mean_token_accuracy": 0.1588592991232872,
      "num_tokens": 25271035.0,
      "step": 12185
    },
    {
      "entropy": 5.8133117198944095,
      "epoch": 1.1057692307692308,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004884495290590134,
      "loss": 5.5472,
      "mean_token_accuracy": 0.1642577216029167,
      "num_tokens": 25281669.0,
      "step": 12190
    },
    {
      "entropy": 5.638579177856445,
      "epoch": 1.1062227866473149,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004884392930362124,
      "loss": 5.4581,
      "mean_token_accuracy": 0.17585345208644867,
      "num_tokens": 25291809.0,
      "step": 12195
    },
    {
      "entropy": 5.705216646194458,
      "epoch": 1.1066763425253991,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004884290525994326,
      "loss": 5.5321,
      "mean_token_accuracy": 0.1690521389245987,
      "num_tokens": 25302277.0,
      "step": 12200
    },
    {
      "entropy": 5.704319667816162,
      "epoch": 1.1071298984034832,
      "grad_norm": 2.5,
      "learning_rate": 0.0004884188077488855,
      "loss": 5.476,
      "mean_token_accuracy": 0.15593263655900955,
      "num_tokens": 25313578.0,
      "step": 12205
    },
    {
      "entropy": 5.78058614730835,
      "epoch": 1.1075834542815675,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004884085584847831,
      "loss": 5.6233,
      "mean_token_accuracy": 0.15150113850831987,
      "num_tokens": 25323895.0,
      "step": 12210
    },
    {
      "entropy": 5.666037654876709,
      "epoch": 1.1080370101596517,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004883983048073375,
      "loss": 5.399,
      "mean_token_accuracy": 0.16374248415231704,
      "num_tokens": 25332998.0,
      "step": 12215
    },
    {
      "entropy": 5.7342503547668455,
      "epoch": 1.1084905660377358,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004883880467167603,
      "loss": 5.5132,
      "mean_token_accuracy": 0.16622536331415178,
      "num_tokens": 25343035.0,
      "step": 12220
    },
    {
      "entropy": 5.706403303146362,
      "epoch": 1.10894412191582,
      "grad_norm": 1.890625,
      "learning_rate": 0.000488377784213264,
      "loss": 5.4729,
      "mean_token_accuracy": 0.16081324517726897,
      "num_tokens": 25353539.0,
      "step": 12225
    },
    {
      "entropy": 5.786635971069336,
      "epoch": 1.1093976777939043,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004883675172970608,
      "loss": 5.5874,
      "mean_token_accuracy": 0.1633201152086258,
      "num_tokens": 25364151.0,
      "step": 12230
    },
    {
      "entropy": 5.749664163589477,
      "epoch": 1.1098512336719883,
      "grad_norm": 2.078125,
      "learning_rate": 0.000488357245968363,
      "loss": 5.4719,
      "mean_token_accuracy": 0.16135362833738326,
      "num_tokens": 25375024.0,
      "step": 12235
    },
    {
      "entropy": 5.7102744579315186,
      "epoch": 1.1103047895500726,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004883469702273829,
      "loss": 5.4557,
      "mean_token_accuracy": 0.16266125738620757,
      "num_tokens": 25385190.0,
      "step": 12240
    },
    {
      "entropy": 5.687498140335083,
      "epoch": 1.1107583454281567,
      "grad_norm": 3.125,
      "learning_rate": 0.0004883366900743331,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1691732183098793,
      "num_tokens": 25396423.0,
      "step": 12245
    },
    {
      "entropy": 5.747631454467774,
      "epoch": 1.111211901306241,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004883264055094262,
      "loss": 5.5656,
      "mean_token_accuracy": 0.16264670491218566,
      "num_tokens": 25406806.0,
      "step": 12250
    },
    {
      "entropy": 5.703668308258057,
      "epoch": 1.1116654571843252,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004883161165328749,
      "loss": 5.5483,
      "mean_token_accuracy": 0.16399476826190948,
      "num_tokens": 25417871.0,
      "step": 12255
    },
    {
      "entropy": 5.729721975326538,
      "epoch": 1.1121190130624092,
      "grad_norm": 1.734375,
      "learning_rate": 0.000488305823144892,
      "loss": 5.5693,
      "mean_token_accuracy": 0.16129586845636368,
      "num_tokens": 25428648.0,
      "step": 12260
    },
    {
      "entropy": 5.777127742767334,
      "epoch": 1.1125725689404935,
      "grad_norm": 2.625,
      "learning_rate": 0.0004882955253456902,
      "loss": 5.4778,
      "mean_token_accuracy": 0.1627258613705635,
      "num_tokens": 25439131.0,
      "step": 12265
    },
    {
      "entropy": 5.744696044921875,
      "epoch": 1.1130261248185778,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004882852231354826,
      "loss": 5.5082,
      "mean_token_accuracy": 0.16040317118167877,
      "num_tokens": 25448852.0,
      "step": 12270
    },
    {
      "entropy": 5.758439064025879,
      "epoch": 1.1134796806966618,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004882749165144822,
      "loss": 5.6123,
      "mean_token_accuracy": 0.15663883537054063,
      "num_tokens": 25459518.0,
      "step": 12275
    },
    {
      "entropy": 5.654857206344604,
      "epoch": 1.113933236574746,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004882646054829023,
      "loss": 5.4349,
      "mean_token_accuracy": 0.16036461740732194,
      "num_tokens": 25468960.0,
      "step": 12280
    },
    {
      "entropy": 5.713877868652344,
      "epoch": 1.1143867924528301,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048825429004095597,
      "loss": 5.4787,
      "mean_token_accuracy": 0.16251427829265594,
      "num_tokens": 25479163.0,
      "step": 12285
    },
    {
      "entropy": 5.777317762374878,
      "epoch": 1.1148403483309144,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004882439701888566,
      "loss": 5.4823,
      "mean_token_accuracy": 0.16781481057405473,
      "num_tokens": 25489321.0,
      "step": 12290
    },
    {
      "entropy": 5.723558712005615,
      "epoch": 1.1152939042089987,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048823364592681754,
      "loss": 5.4894,
      "mean_token_accuracy": 0.16353384405374527,
      "num_tokens": 25500056.0,
      "step": 12295
    },
    {
      "entropy": 5.697708082199097,
      "epoch": 1.1157474600870827,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048822331725505234,
      "loss": 5.4531,
      "mean_token_accuracy": 0.16378927528858184,
      "num_tokens": 25511810.0,
      "step": 12300
    },
    {
      "entropy": 5.717595338821411,
      "epoch": 1.116201015965167,
      "grad_norm": 3.8125,
      "learning_rate": 0.00048821298417377456,
      "loss": 5.567,
      "mean_token_accuracy": 0.1560910478234291,
      "num_tokens": 25523195.0,
      "step": 12305
    },
    {
      "entropy": 5.818385934829712,
      "epoch": 1.116654571843251,
      "grad_norm": 3.453125,
      "learning_rate": 0.000488202646683198,
      "loss": 5.6282,
      "mean_token_accuracy": 0.159672150015831,
      "num_tokens": 25534197.0,
      "step": 12310
    },
    {
      "entropy": 5.731664991378784,
      "epoch": 1.1171081277213353,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004881923047835364,
      "loss": 5.4684,
      "mean_token_accuracy": 0.16348556131124498,
      "num_tokens": 25545442.0,
      "step": 12315
    },
    {
      "entropy": 5.779276371002197,
      "epoch": 1.1175616835994195,
      "grad_norm": 3.859375,
      "learning_rate": 0.0004881819584750036,
      "loss": 5.5434,
      "mean_token_accuracy": 0.15985412150621414,
      "num_tokens": 25556132.0,
      "step": 12320
    },
    {
      "entropy": 5.785367727279663,
      "epoch": 1.1180152394775036,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048817160775781356,
      "loss": 5.608,
      "mean_token_accuracy": 0.16114221066236495,
      "num_tokens": 25568186.0,
      "step": 12325
    },
    {
      "entropy": 5.716791725158691,
      "epoch": 1.1184687953555879,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048816125263218023,
      "loss": 5.429,
      "mean_token_accuracy": 0.17571020126342773,
      "num_tokens": 25578316.0,
      "step": 12330
    },
    {
      "entropy": 5.658790540695191,
      "epoch": 1.118922351233672,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048815089309831803,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1708226755261421,
      "num_tokens": 25587556.0,
      "step": 12335
    },
    {
      "entropy": 5.689793920516967,
      "epoch": 1.1193759071117562,
      "grad_norm": 2.359375,
      "learning_rate": 0.000488140529156441,
      "loss": 5.476,
      "mean_token_accuracy": 0.1608987644314766,
      "num_tokens": 25597967.0,
      "step": 12340
    },
    {
      "entropy": 5.686386966705323,
      "epoch": 1.1198294629898404,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004881301608067635,
      "loss": 5.4682,
      "mean_token_accuracy": 0.16304361671209336,
      "num_tokens": 25607952.0,
      "step": 12345
    },
    {
      "entropy": 5.756064128875733,
      "epoch": 1.1202830188679245,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048811978804949997,
      "loss": 5.4685,
      "mean_token_accuracy": 0.16708625555038453,
      "num_tokens": 25617923.0,
      "step": 12350
    },
    {
      "entropy": 5.6989928722381595,
      "epoch": 1.1207365747460087,
      "grad_norm": 2.1875,
      "learning_rate": 0.000488109410884865,
      "loss": 5.4215,
      "mean_token_accuracy": 0.16821280717849732,
      "num_tokens": 25628345.0,
      "step": 12355
    },
    {
      "entropy": 5.728607225418091,
      "epoch": 1.1211901306240928,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048809902931307294,
      "loss": 5.6026,
      "mean_token_accuracy": 0.15578905194997789,
      "num_tokens": 25640059.0,
      "step": 12360
    },
    {
      "entropy": 5.719809198379517,
      "epoch": 1.121643686502177,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004880886433343388,
      "loss": 5.4262,
      "mean_token_accuracy": 0.16505700051784516,
      "num_tokens": 25650126.0,
      "step": 12365
    },
    {
      "entropy": 5.785664749145508,
      "epoch": 1.1220972423802613,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004880782529488771,
      "loss": 5.5108,
      "mean_token_accuracy": 0.1603873699903488,
      "num_tokens": 25661735.0,
      "step": 12370
    },
    {
      "entropy": 5.711270713806153,
      "epoch": 1.1225507982583454,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004880678581569029,
      "loss": 5.5509,
      "mean_token_accuracy": 0.1579383835196495,
      "num_tokens": 25673491.0,
      "step": 12375
    },
    {
      "entropy": 5.767218065261841,
      "epoch": 1.1230043541364296,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048805745895863105,
      "loss": 5.4992,
      "mean_token_accuracy": 0.1596676751971245,
      "num_tokens": 25684120.0,
      "step": 12380
    },
    {
      "entropy": 5.7552892684936525,
      "epoch": 1.1234579100145137,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004880470553542767,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16444639265537261,
      "num_tokens": 25694494.0,
      "step": 12385
    },
    {
      "entropy": 5.781294918060302,
      "epoch": 1.123911465892598,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048803664734405495,
      "loss": 5.5676,
      "mean_token_accuracy": 0.15853183716535568,
      "num_tokens": 25704925.0,
      "step": 12390
    },
    {
      "entropy": 5.608174896240234,
      "epoch": 1.1243650217706822,
      "grad_norm": 2.15625,
      "learning_rate": 0.000488026234928181,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1662816137075424,
      "num_tokens": 25714597.0,
      "step": 12395
    },
    {
      "entropy": 5.726742887496949,
      "epoch": 1.1248185776487662,
      "grad_norm": 2.375,
      "learning_rate": 0.0004880158181068703,
      "loss": 5.4576,
      "mean_token_accuracy": 0.15936375707387923,
      "num_tokens": 25725000.0,
      "step": 12400
    },
    {
      "entropy": 5.67924280166626,
      "epoch": 1.1252721335268505,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048800539688033807,
      "loss": 5.474,
      "mean_token_accuracy": 0.16664165407419204,
      "num_tokens": 25735176.0,
      "step": 12405
    },
    {
      "entropy": 5.764802408218384,
      "epoch": 1.1257256894049348,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004879949712488001,
      "loss": 5.4799,
      "mean_token_accuracy": 0.15936209708452226,
      "num_tokens": 25745613.0,
      "step": 12410
    },
    {
      "entropy": 5.726183843612671,
      "epoch": 1.1261792452830188,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048798454121247177,
      "loss": 5.4116,
      "mean_token_accuracy": 0.1699094831943512,
      "num_tokens": 25755123.0,
      "step": 12415
    },
    {
      "entropy": 5.657440900802612,
      "epoch": 1.126632801161103,
      "grad_norm": 1.625,
      "learning_rate": 0.0004879741067715688,
      "loss": 5.4576,
      "mean_token_accuracy": 0.16801634132862092,
      "num_tokens": 25765126.0,
      "step": 12420
    },
    {
      "entropy": 5.627175903320312,
      "epoch": 1.1270863570391871,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048796366792630715,
      "loss": 5.4602,
      "mean_token_accuracy": 0.1608129248023033,
      "num_tokens": 25775144.0,
      "step": 12425
    },
    {
      "entropy": 5.66564154624939,
      "epoch": 1.1275399129172714,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004879532246769025,
      "loss": 5.4195,
      "mean_token_accuracy": 0.16845256686210633,
      "num_tokens": 25784982.0,
      "step": 12430
    },
    {
      "entropy": 5.6108301162719725,
      "epoch": 1.1279934687953557,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004879427770235709,
      "loss": 5.3435,
      "mean_token_accuracy": 0.17013841420412062,
      "num_tokens": 25794786.0,
      "step": 12435
    },
    {
      "entropy": 5.724650764465332,
      "epoch": 1.1284470246734397,
      "grad_norm": 2.75,
      "learning_rate": 0.00048793232496652846,
      "loss": 5.5135,
      "mean_token_accuracy": 0.15705978721380234,
      "num_tokens": 25806226.0,
      "step": 12440
    },
    {
      "entropy": 5.705124807357788,
      "epoch": 1.128900580551524,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048792186850599124,
      "loss": 5.4582,
      "mean_token_accuracy": 0.16622921079397202,
      "num_tokens": 25816428.0,
      "step": 12445
    },
    {
      "entropy": 5.648011112213135,
      "epoch": 1.1293541364296082,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004879114076421755,
      "loss": 5.4214,
      "mean_token_accuracy": 0.1661675125360489,
      "num_tokens": 25826622.0,
      "step": 12450
    },
    {
      "entropy": 5.7611319065094,
      "epoch": 1.1298076923076923,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048790094237529763,
      "loss": 5.4986,
      "mean_token_accuracy": 0.1627287596464157,
      "num_tokens": 25836628.0,
      "step": 12455
    },
    {
      "entropy": 5.75986328125,
      "epoch": 1.1302612481857766,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000487890472705574,
      "loss": 5.5638,
      "mean_token_accuracy": 0.16533479541540147,
      "num_tokens": 25848033.0,
      "step": 12460
    },
    {
      "entropy": 5.730114698410034,
      "epoch": 1.1307148040638606,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00048787999863322113,
      "loss": 5.4863,
      "mean_token_accuracy": 0.1632877469062805,
      "num_tokens": 25859098.0,
      "step": 12465
    },
    {
      "entropy": 5.8046328067779545,
      "epoch": 1.1311683599419449,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004878695201584557,
      "loss": 5.6176,
      "mean_token_accuracy": 0.15689141005277635,
      "num_tokens": 25870029.0,
      "step": 12470
    },
    {
      "entropy": 5.754722976684571,
      "epoch": 1.1316219158200291,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004878590372814943,
      "loss": 5.4888,
      "mean_token_accuracy": 0.16249582767486573,
      "num_tokens": 25880543.0,
      "step": 12475
    },
    {
      "entropy": 5.634375858306885,
      "epoch": 1.1320754716981132,
      "grad_norm": 1.875,
      "learning_rate": 0.00048784855000255374,
      "loss": 5.4521,
      "mean_token_accuracy": 0.16282664686441423,
      "num_tokens": 25889964.0,
      "step": 12480
    },
    {
      "entropy": 5.673883485794067,
      "epoch": 1.1325290275761974,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048783805832185097,
      "loss": 5.4645,
      "mean_token_accuracy": 0.15819445624947548,
      "num_tokens": 25900673.0,
      "step": 12485
    },
    {
      "entropy": 5.802273988723755,
      "epoch": 1.1329825834542815,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004878275622396029,
      "loss": 5.565,
      "mean_token_accuracy": 0.16726450026035308,
      "num_tokens": 25911442.0,
      "step": 12490
    },
    {
      "entropy": 5.6783466815948485,
      "epoch": 1.1334361393323658,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048781706175602665,
      "loss": 5.483,
      "mean_token_accuracy": 0.1660732090473175,
      "num_tokens": 25921717.0,
      "step": 12495
    },
    {
      "entropy": 5.616969156265259,
      "epoch": 1.13388969521045,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048780655687133927,
      "loss": 5.46,
      "mean_token_accuracy": 0.16282618492841722,
      "num_tokens": 25930345.0,
      "step": 12500
    },
    {
      "entropy": 5.785351896286011,
      "epoch": 1.134343251088534,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048779604758575806,
      "loss": 5.476,
      "mean_token_accuracy": 0.1663526251912117,
      "num_tokens": 25940837.0,
      "step": 12505
    },
    {
      "entropy": 5.693494749069214,
      "epoch": 1.1347968069666183,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004877855338995004,
      "loss": 5.4813,
      "mean_token_accuracy": 0.16492131352424622,
      "num_tokens": 25951792.0,
      "step": 12510
    },
    {
      "entropy": 5.605434417724609,
      "epoch": 1.1352503628447024,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048777501581278367,
      "loss": 5.4685,
      "mean_token_accuracy": 0.15807050764560698,
      "num_tokens": 25960810.0,
      "step": 12515
    },
    {
      "entropy": 5.68025770187378,
      "epoch": 1.1357039187227866,
      "grad_norm": 2.25,
      "learning_rate": 0.0004877644933258254,
      "loss": 5.3988,
      "mean_token_accuracy": 0.17114539444446564,
      "num_tokens": 25972428.0,
      "step": 12520
    },
    {
      "entropy": 5.726837682723999,
      "epoch": 1.136157474600871,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004877539664388432,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16639023125171662,
      "num_tokens": 25983726.0,
      "step": 12525
    },
    {
      "entropy": 5.669631433486939,
      "epoch": 1.136611030478955,
      "grad_norm": 2.375,
      "learning_rate": 0.00048774343515205464,
      "loss": 5.4179,
      "mean_token_accuracy": 0.16351398825645447,
      "num_tokens": 25993781.0,
      "step": 12530
    },
    {
      "entropy": 5.720082187652588,
      "epoch": 1.1370645863570392,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004877328994656778,
      "loss": 5.5609,
      "mean_token_accuracy": 0.16483699083328246,
      "num_tokens": 26003672.0,
      "step": 12535
    },
    {
      "entropy": 5.736012125015259,
      "epoch": 1.1375181422351233,
      "grad_norm": 9.0625,
      "learning_rate": 0.0004877223593799303,
      "loss": 5.4864,
      "mean_token_accuracy": 0.16362462788820267,
      "num_tokens": 26013485.0,
      "step": 12540
    },
    {
      "entropy": 5.692254829406738,
      "epoch": 1.1379716981132075,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004877118148950302,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1586073651909828,
      "num_tokens": 26024212.0,
      "step": 12545
    },
    {
      "entropy": 5.769058895111084,
      "epoch": 1.1384252539912918,
      "grad_norm": 2.25,
      "learning_rate": 0.00048770126601119555,
      "loss": 5.5296,
      "mean_token_accuracy": 0.15578835904598237,
      "num_tokens": 26034937.0,
      "step": 12550
    },
    {
      "entropy": 5.685423088073731,
      "epoch": 1.1388788098693758,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048769071272864454,
      "loss": 5.4026,
      "mean_token_accuracy": 0.1663497045636177,
      "num_tokens": 26045182.0,
      "step": 12555
    },
    {
      "entropy": 5.693830394744873,
      "epoch": 1.13933236574746,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048768015504759536,
      "loss": 5.525,
      "mean_token_accuracy": 0.16389835774898528,
      "num_tokens": 26055263.0,
      "step": 12560
    },
    {
      "entropy": 5.677569198608398,
      "epoch": 1.1397859216255442,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004876695929682665,
      "loss": 5.4514,
      "mean_token_accuracy": 0.16440078616142273,
      "num_tokens": 26065926.0,
      "step": 12565
    },
    {
      "entropy": 5.7594481945037845,
      "epoch": 1.1402394775036284,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048765902649087605,
      "loss": 5.4892,
      "mean_token_accuracy": 0.16910947859287262,
      "num_tokens": 26075895.0,
      "step": 12570
    },
    {
      "entropy": 5.719408130645752,
      "epoch": 1.1406930333817127,
      "grad_norm": 6.25,
      "learning_rate": 0.00048764845561564285,
      "loss": 5.4899,
      "mean_token_accuracy": 0.16287877708673476,
      "num_tokens": 26085936.0,
      "step": 12575
    },
    {
      "entropy": 5.716920852661133,
      "epoch": 1.1411465892597967,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048763788034278545,
      "loss": 5.5253,
      "mean_token_accuracy": 0.1659355044364929,
      "num_tokens": 26096070.0,
      "step": 12580
    },
    {
      "entropy": 5.796630573272705,
      "epoch": 1.141600145137881,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004876273006725225,
      "loss": 5.6809,
      "mean_token_accuracy": 0.15550892055034637,
      "num_tokens": 26107681.0,
      "step": 12585
    },
    {
      "entropy": 5.7304768562316895,
      "epoch": 1.1420537010159653,
      "grad_norm": 2.875,
      "learning_rate": 0.00048761671660507266,
      "loss": 5.438,
      "mean_token_accuracy": 0.1698172867298126,
      "num_tokens": 26117610.0,
      "step": 12590
    },
    {
      "entropy": 5.682204723358154,
      "epoch": 1.1425072568940493,
      "grad_norm": 2.375,
      "learning_rate": 0.000487606128140655,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17094091325998306,
      "num_tokens": 26128394.0,
      "step": 12595
    },
    {
      "entropy": 5.652076482772827,
      "epoch": 1.1429608127721336,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004875955352794885,
      "loss": 5.5731,
      "mean_token_accuracy": 0.15528872534632682,
      "num_tokens": 26139707.0,
      "step": 12600
    },
    {
      "entropy": 5.73680157661438,
      "epoch": 1.1434143686502176,
      "grad_norm": 4.84375,
      "learning_rate": 0.0004875849380217921,
      "loss": 5.5278,
      "mean_token_accuracy": 0.1512846678495407,
      "num_tokens": 26150337.0,
      "step": 12605
    },
    {
      "entropy": 5.8371350288391115,
      "epoch": 1.1438679245283019,
      "grad_norm": 3.765625,
      "learning_rate": 0.000487574336367785,
      "loss": 5.514,
      "mean_token_accuracy": 0.16107017695903778,
      "num_tokens": 26159534.0,
      "step": 12610
    },
    {
      "entropy": 5.69606647491455,
      "epoch": 1.1443214804063861,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004875637303176865,
      "loss": 5.4042,
      "mean_token_accuracy": 0.1717361330986023,
      "num_tokens": 26168692.0,
      "step": 12615
    },
    {
      "entropy": 5.67038402557373,
      "epoch": 1.1447750362844702,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00048755311987171585,
      "loss": 5.5516,
      "mean_token_accuracy": 0.15879932343959807,
      "num_tokens": 26180127.0,
      "step": 12620
    },
    {
      "entropy": 5.723080492019653,
      "epoch": 1.1452285921625545,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048754250503009244,
      "loss": 5.4498,
      "mean_token_accuracy": 0.1681497260928154,
      "num_tokens": 26190410.0,
      "step": 12625
    },
    {
      "entropy": 5.718997764587402,
      "epoch": 1.1456821480406387,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048753188579303596,
      "loss": 5.5064,
      "mean_token_accuracy": 0.16461855918169022,
      "num_tokens": 26201030.0,
      "step": 12630
    },
    {
      "entropy": 5.554056215286255,
      "epoch": 1.1461357039187228,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004875212621607659,
      "loss": 5.3979,
      "mean_token_accuracy": 0.16387778371572495,
      "num_tokens": 26210984.0,
      "step": 12635
    },
    {
      "entropy": 5.6511646747589115,
      "epoch": 1.146589259796807,
      "grad_norm": 5.5,
      "learning_rate": 0.00048751063413350203,
      "loss": 5.4555,
      "mean_token_accuracy": 0.16519658118486405,
      "num_tokens": 26221779.0,
      "step": 12640
    },
    {
      "entropy": 5.686412000656128,
      "epoch": 1.147042815674891,
      "grad_norm": 2.6875,
      "learning_rate": 0.000487500001711464,
      "loss": 5.4561,
      "mean_token_accuracy": 0.1638703539967537,
      "num_tokens": 26232175.0,
      "step": 12645
    },
    {
      "entropy": 5.751825618743896,
      "epoch": 1.1474963715529753,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004874893648948719,
      "loss": 5.567,
      "mean_token_accuracy": 0.15980178266763687,
      "num_tokens": 26242417.0,
      "step": 12650
    },
    {
      "entropy": 5.740047311782837,
      "epoch": 1.1479499274310596,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048747872368394546,
      "loss": 5.488,
      "mean_token_accuracy": 0.16299936771392823,
      "num_tokens": 26252748.0,
      "step": 12655
    },
    {
      "entropy": 5.713072013854981,
      "epoch": 1.1484034833091437,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048746807807890496,
      "loss": 5.4811,
      "mean_token_accuracy": 0.16244303584098815,
      "num_tokens": 26263870.0,
      "step": 12660
    },
    {
      "entropy": 5.68633451461792,
      "epoch": 1.148857039187228,
      "grad_norm": 4.5625,
      "learning_rate": 0.00048745742807997046,
      "loss": 5.4531,
      "mean_token_accuracy": 0.16799683421850203,
      "num_tokens": 26273611.0,
      "step": 12665
    },
    {
      "entropy": 5.649423217773437,
      "epoch": 1.149310595065312,
      "grad_norm": 3.78125,
      "learning_rate": 0.00048744677368736216,
      "loss": 5.4584,
      "mean_token_accuracy": 0.16430903375148773,
      "num_tokens": 26284374.0,
      "step": 12670
    },
    {
      "entropy": 5.74077262878418,
      "epoch": 1.1497641509433962,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048743611490130044,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16219829767942429,
      "num_tokens": 26295907.0,
      "step": 12675
    },
    {
      "entropy": 5.786208248138427,
      "epoch": 1.1502177068214805,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048742545172200576,
      "loss": 5.5246,
      "mean_token_accuracy": 0.15778998881578446,
      "num_tokens": 26306294.0,
      "step": 12680
    },
    {
      "entropy": 5.671911525726318,
      "epoch": 1.1506712626995645,
      "grad_norm": 6.0,
      "learning_rate": 0.0004874147841496986,
      "loss": 5.5162,
      "mean_token_accuracy": 0.16119887083768844,
      "num_tokens": 26316152.0,
      "step": 12685
    },
    {
      "entropy": 5.78262095451355,
      "epoch": 1.1511248185776488,
      "grad_norm": 3.765625,
      "learning_rate": 0.0004874041121845996,
      "loss": 5.5999,
      "mean_token_accuracy": 0.15688785463571547,
      "num_tokens": 26327664.0,
      "step": 12690
    },
    {
      "entropy": 5.802752065658569,
      "epoch": 1.1515783744557329,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004873934358269294,
      "loss": 5.4894,
      "mean_token_accuracy": 0.15907246321439744,
      "num_tokens": 26339060.0,
      "step": 12695
    },
    {
      "entropy": 5.7160083770751955,
      "epoch": 1.1520319303338171,
      "grad_norm": 3.453125,
      "learning_rate": 0.0004873827550769088,
      "loss": 5.4847,
      "mean_token_accuracy": 0.16440905556082724,
      "num_tokens": 26349219.0,
      "step": 12700
    },
    {
      "entropy": 5.673494911193847,
      "epoch": 1.1524854862119014,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004873720699347587,
      "loss": 5.5158,
      "mean_token_accuracy": 0.16712939292192458,
      "num_tokens": 26359620.0,
      "step": 12705
    },
    {
      "entropy": 5.763228034973144,
      "epoch": 1.1529390420899854,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004873613804007001,
      "loss": 5.4576,
      "mean_token_accuracy": 0.16672644764184952,
      "num_tokens": 26370065.0,
      "step": 12710
    },
    {
      "entropy": 5.691225481033325,
      "epoch": 1.1533925979680697,
      "grad_norm": 4.3125,
      "learning_rate": 0.0004873506864749541,
      "loss": 5.4772,
      "mean_token_accuracy": 0.16544202119112014,
      "num_tokens": 26379878.0,
      "step": 12715
    },
    {
      "entropy": 5.6572277545928955,
      "epoch": 1.1538461538461537,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004873399881577417,
      "loss": 5.448,
      "mean_token_accuracy": 0.16319901943206788,
      "num_tokens": 26389859.0,
      "step": 12720
    },
    {
      "entropy": 5.668529033660889,
      "epoch": 1.154299709724238,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004873292854492842,
      "loss": 5.4558,
      "mean_token_accuracy": 0.16755273938179016,
      "num_tokens": 26399952.0,
      "step": 12725
    },
    {
      "entropy": 5.692190837860108,
      "epoch": 1.1547532656023223,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004873185783498031,
      "loss": 5.4495,
      "mean_token_accuracy": 0.16885923445224763,
      "num_tokens": 26410238.0,
      "step": 12730
    },
    {
      "entropy": 5.714127254486084,
      "epoch": 1.1552068214804063,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004873078668595197,
      "loss": 5.5286,
      "mean_token_accuracy": 0.1645926669239998,
      "num_tokens": 26421242.0,
      "step": 12735
    },
    {
      "entropy": 5.774478340148926,
      "epoch": 1.1556603773584906,
      "grad_norm": 5.6875,
      "learning_rate": 0.00048729715097865534,
      "loss": 5.5419,
      "mean_token_accuracy": 0.15534109622240067,
      "num_tokens": 26431890.0,
      "step": 12740
    },
    {
      "entropy": 5.756366395950318,
      "epoch": 1.1561139332365746,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004872864307074319,
      "loss": 5.5539,
      "mean_token_accuracy": 0.1584905505180359,
      "num_tokens": 26442341.0,
      "step": 12745
    },
    {
      "entropy": 5.815241813659668,
      "epoch": 1.156567489114659,
      "grad_norm": 3.625,
      "learning_rate": 0.00048727570604607096,
      "loss": 5.5554,
      "mean_token_accuracy": 0.15368589460849763,
      "num_tokens": 26454082.0,
      "step": 12750
    },
    {
      "entropy": 5.70571551322937,
      "epoch": 1.1570210449927432,
      "grad_norm": 2.578125,
      "learning_rate": 0.00048726497699479437,
      "loss": 5.4183,
      "mean_token_accuracy": 0.17059891372919084,
      "num_tokens": 26463152.0,
      "step": 12755
    },
    {
      "entropy": 5.7736592292785645,
      "epoch": 1.1574746008708272,
      "grad_norm": 12.75,
      "learning_rate": 0.000487254243553824,
      "loss": 5.6749,
      "mean_token_accuracy": 0.15674822330474852,
      "num_tokens": 26474315.0,
      "step": 12760
    },
    {
      "entropy": 5.734152221679688,
      "epoch": 1.1579281567489115,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004872435057233817,
      "loss": 5.4622,
      "mean_token_accuracy": 0.16900216192007064,
      "num_tokens": 26484029.0,
      "step": 12765
    },
    {
      "entropy": 5.733778715133667,
      "epoch": 1.1583817126269957,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048723276350368957,
      "loss": 5.54,
      "mean_token_accuracy": 0.160042804479599,
      "num_tokens": 26494315.0,
      "step": 12770
    },
    {
      "entropy": 5.714306831359863,
      "epoch": 1.1588352685050798,
      "grad_norm": 7.0625,
      "learning_rate": 0.00048722201689496984,
      "loss": 5.5482,
      "mean_token_accuracy": 0.16254013478755952,
      "num_tokens": 26504923.0,
      "step": 12775
    },
    {
      "entropy": 5.674842643737793,
      "epoch": 1.159288824383164,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004872112658974447,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1743605002760887,
      "num_tokens": 26515276.0,
      "step": 12780
    },
    {
      "entropy": 5.722029209136963,
      "epoch": 1.159742380261248,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048720051051133657,
      "loss": 5.4847,
      "mean_token_accuracy": 0.16597037762403488,
      "num_tokens": 26525075.0,
      "step": 12785
    },
    {
      "entropy": 5.6921288013458256,
      "epoch": 1.1601959361393324,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004871897507368678,
      "loss": 5.4326,
      "mean_token_accuracy": 0.16896616369485856,
      "num_tokens": 26537203.0,
      "step": 12790
    },
    {
      "entropy": 5.645166254043579,
      "epoch": 1.1606494920174166,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048717898657426093,
      "loss": 5.4903,
      "mean_token_accuracy": 0.1670250788331032,
      "num_tokens": 26545928.0,
      "step": 12795
    },
    {
      "entropy": 5.621448040008545,
      "epoch": 1.1611030478955007,
      "grad_norm": 3.1875,
      "learning_rate": 0.00048716821802373856,
      "loss": 5.4017,
      "mean_token_accuracy": 0.1775278240442276,
      "num_tokens": 26555923.0,
      "step": 12800
    },
    {
      "entropy": 5.674316120147705,
      "epoch": 1.161556603773585,
      "grad_norm": 2.625,
      "learning_rate": 0.00048715744508552334,
      "loss": 5.376,
      "mean_token_accuracy": 0.17524190992116928,
      "num_tokens": 26567334.0,
      "step": 12805
    },
    {
      "entropy": 5.710293483734131,
      "epoch": 1.1620101596516692,
      "grad_norm": 3.75,
      "learning_rate": 0.0004871466677598381,
      "loss": 5.4497,
      "mean_token_accuracy": 0.1603351578116417,
      "num_tokens": 26576534.0,
      "step": 12810
    },
    {
      "entropy": 5.755929851531983,
      "epoch": 1.1624637155297532,
      "grad_norm": 3.984375,
      "learning_rate": 0.0004871358860469057,
      "loss": 5.5363,
      "mean_token_accuracy": 0.16189496964216232,
      "num_tokens": 26586091.0,
      "step": 12815
    },
    {
      "entropy": 5.735464763641358,
      "epoch": 1.1629172714078375,
      "grad_norm": 6.0,
      "learning_rate": 0.00048712509994694903,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1690073385834694,
      "num_tokens": 26596723.0,
      "step": 12820
    },
    {
      "entropy": 5.705807065963745,
      "epoch": 1.1633708272859216,
      "grad_norm": 6.65625,
      "learning_rate": 0.0004871143094601914,
      "loss": 5.4825,
      "mean_token_accuracy": 0.1651734232902527,
      "num_tokens": 26606351.0,
      "step": 12825
    },
    {
      "entropy": 5.629496669769287,
      "epoch": 1.1638243831640058,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004871035145868557,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16704117953777314,
      "num_tokens": 26617026.0,
      "step": 12830
    },
    {
      "entropy": 5.695806264877319,
      "epoch": 1.16427793904209,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048709271532716524,
      "loss": 5.5021,
      "mean_token_accuracy": 0.15982649475336075,
      "num_tokens": 26627529.0,
      "step": 12835
    },
    {
      "entropy": 5.702539348602295,
      "epoch": 1.1647314949201741,
      "grad_norm": 5.5,
      "learning_rate": 0.0004870819116813434,
      "loss": 5.4008,
      "mean_token_accuracy": 0.16716398447751998,
      "num_tokens": 26637189.0,
      "step": 12840
    },
    {
      "entropy": 5.745016241073609,
      "epoch": 1.1651850507982584,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048707110364961363,
      "loss": 5.508,
      "mean_token_accuracy": 0.16399665027856827,
      "num_tokens": 26647457.0,
      "step": 12845
    },
    {
      "entropy": 5.706327819824219,
      "epoch": 1.1656386066763424,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048706029123219937,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1583716318011284,
      "num_tokens": 26657542.0,
      "step": 12850
    },
    {
      "entropy": 5.656693935394287,
      "epoch": 1.1660921625544267,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048704947442932426,
      "loss": 5.3955,
      "mean_token_accuracy": 0.1608876347541809,
      "num_tokens": 26667943.0,
      "step": 12855
    },
    {
      "entropy": 5.668429803848267,
      "epoch": 1.166545718432511,
      "grad_norm": 5.25,
      "learning_rate": 0.00048703865324121196,
      "loss": 5.4261,
      "mean_token_accuracy": 0.17094897031784057,
      "num_tokens": 26677882.0,
      "step": 12860
    },
    {
      "entropy": 5.705692338943481,
      "epoch": 1.166999274310595,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048702782766808637,
      "loss": 5.3845,
      "mean_token_accuracy": 0.17209497839212418,
      "num_tokens": 26687088.0,
      "step": 12865
    },
    {
      "entropy": 5.673368883132935,
      "epoch": 1.1674528301886793,
      "grad_norm": 5.53125,
      "learning_rate": 0.0004870169977101711,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1692939430475235,
      "num_tokens": 26696508.0,
      "step": 12870
    },
    {
      "entropy": 5.801535034179688,
      "epoch": 1.1679063860667633,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048700616336769037,
      "loss": 5.6098,
      "mean_token_accuracy": 0.15874230712652207,
      "num_tokens": 26707777.0,
      "step": 12875
    },
    {
      "entropy": 5.721925210952759,
      "epoch": 1.1683599419448476,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048699532464086817,
      "loss": 5.5236,
      "mean_token_accuracy": 0.158287650346756,
      "num_tokens": 26718266.0,
      "step": 12880
    },
    {
      "entropy": 5.680122375488281,
      "epoch": 1.1688134978229319,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048698448152992865,
      "loss": 5.4866,
      "mean_token_accuracy": 0.16226572990417482,
      "num_tokens": 26728802.0,
      "step": 12885
    },
    {
      "entropy": 5.799480867385864,
      "epoch": 1.169267053701016,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048697363403509605,
      "loss": 5.5028,
      "mean_token_accuracy": 0.1617770493030548,
      "num_tokens": 26739254.0,
      "step": 12890
    },
    {
      "entropy": 5.707293272018433,
      "epoch": 1.1697206095791002,
      "grad_norm": 3.140625,
      "learning_rate": 0.0004869627821565946,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1667792022228241,
      "num_tokens": 26748966.0,
      "step": 12895
    },
    {
      "entropy": 5.642256069183349,
      "epoch": 1.1701741654571842,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004869519258946489,
      "loss": 5.3788,
      "mean_token_accuracy": 0.17043628990650178,
      "num_tokens": 26758171.0,
      "step": 12900
    },
    {
      "entropy": 5.598988771438599,
      "epoch": 1.1706277213352685,
      "grad_norm": 3.96875,
      "learning_rate": 0.00048694106524948324,
      "loss": 5.3711,
      "mean_token_accuracy": 0.17249623835086822,
      "num_tokens": 26768149.0,
      "step": 12905
    },
    {
      "entropy": 5.763545751571655,
      "epoch": 1.1710812772133528,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048693020022132243,
      "loss": 5.567,
      "mean_token_accuracy": 0.16127921938896178,
      "num_tokens": 26778022.0,
      "step": 12910
    },
    {
      "entropy": 5.789980220794678,
      "epoch": 1.1715348330914368,
      "grad_norm": 2.40625,
      "learning_rate": 0.000486919330810391,
      "loss": 5.5661,
      "mean_token_accuracy": 0.15790176689624785,
      "num_tokens": 26788473.0,
      "step": 12915
    },
    {
      "entropy": 5.7036701202392575,
      "epoch": 1.171988388969521,
      "grad_norm": 3.5,
      "learning_rate": 0.0004869084570169139,
      "loss": 5.4767,
      "mean_token_accuracy": 0.16398370116949082,
      "num_tokens": 26798587.0,
      "step": 12920
    },
    {
      "entropy": 5.7239015102386475,
      "epoch": 1.172441944847605,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048689757884111585,
      "loss": 5.4697,
      "mean_token_accuracy": 0.17049850523471832,
      "num_tokens": 26808399.0,
      "step": 12925
    },
    {
      "entropy": 5.717213296890259,
      "epoch": 1.1728955007256894,
      "grad_norm": 2.875,
      "learning_rate": 0.00048688669628322185,
      "loss": 5.5652,
      "mean_token_accuracy": 0.15862731039524078,
      "num_tokens": 26818544.0,
      "step": 12930
    },
    {
      "entropy": 5.662959241867066,
      "epoch": 1.1733490566037736,
      "grad_norm": 2.375,
      "learning_rate": 0.000486875809343457,
      "loss": 5.4196,
      "mean_token_accuracy": 0.16286466717720033,
      "num_tokens": 26828805.0,
      "step": 12935
    },
    {
      "entropy": 5.725504064559937,
      "epoch": 1.1738026124818577,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048686491802204645,
      "loss": 5.4557,
      "mean_token_accuracy": 0.16511040925979614,
      "num_tokens": 26839362.0,
      "step": 12940
    },
    {
      "entropy": 5.720784521102905,
      "epoch": 1.174256168359942,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004868540223192154,
      "loss": 5.4423,
      "mean_token_accuracy": 0.16742505878210068,
      "num_tokens": 26849045.0,
      "step": 12945
    },
    {
      "entropy": 5.652719116210937,
      "epoch": 1.1747097242380262,
      "grad_norm": 3.90625,
      "learning_rate": 0.0004868431222351892,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16602034121751785,
      "num_tokens": 26859465.0,
      "step": 12950
    },
    {
      "entropy": 5.741164779663086,
      "epoch": 1.1751632801161103,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048683221777019313,
      "loss": 5.5291,
      "mean_token_accuracy": 0.16014062613248825,
      "num_tokens": 26870957.0,
      "step": 12955
    },
    {
      "entropy": 5.657290840148926,
      "epoch": 1.1756168359941945,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048682130892445293,
      "loss": 5.4877,
      "mean_token_accuracy": 0.16701532006263733,
      "num_tokens": 26881750.0,
      "step": 12960
    },
    {
      "entropy": 5.739746475219727,
      "epoch": 1.1760703918722786,
      "grad_norm": 3.296875,
      "learning_rate": 0.000486810395698194,
      "loss": 5.4852,
      "mean_token_accuracy": 0.16689557135105132,
      "num_tokens": 26891873.0,
      "step": 12965
    },
    {
      "entropy": 5.76758394241333,
      "epoch": 1.1765239477503628,
      "grad_norm": 2.125,
      "learning_rate": 0.00048679947809164223,
      "loss": 5.5152,
      "mean_token_accuracy": 0.16020269691944122,
      "num_tokens": 26902462.0,
      "step": 12970
    },
    {
      "entropy": 5.7580560684204105,
      "epoch": 1.176977503628447,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004867885561050232,
      "loss": 5.4382,
      "mean_token_accuracy": 0.16875692158937455,
      "num_tokens": 26912310.0,
      "step": 12975
    },
    {
      "entropy": 5.7133745670318605,
      "epoch": 1.1774310595065312,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004867776297385629,
      "loss": 5.4888,
      "mean_token_accuracy": 0.16140227168798446,
      "num_tokens": 26923081.0,
      "step": 12980
    },
    {
      "entropy": 5.720228815078736,
      "epoch": 1.1778846153846154,
      "grad_norm": 2.125,
      "learning_rate": 0.00048676669899248723,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16098527759313583,
      "num_tokens": 26933347.0,
      "step": 12985
    },
    {
      "entropy": 5.791284084320068,
      "epoch": 1.1783381712626997,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004867557638670222,
      "loss": 5.5083,
      "mean_token_accuracy": 0.16175081580877304,
      "num_tokens": 26943303.0,
      "step": 12990
    },
    {
      "entropy": 5.73546085357666,
      "epoch": 1.1787917271407837,
      "grad_norm": 2.125,
      "learning_rate": 0.0004867448243623941,
      "loss": 5.5208,
      "mean_token_accuracy": 0.16093730181455612,
      "num_tokens": 26953870.0,
      "step": 12995
    },
    {
      "entropy": 5.702433681488037,
      "epoch": 1.179245283018868,
      "grad_norm": 2.15625,
      "learning_rate": 0.000486733880478829,
      "loss": 5.4114,
      "mean_token_accuracy": 0.16227124482393265,
      "num_tokens": 26965054.0,
      "step": 13000
    },
    {
      "entropy": 5.7584411144256595,
      "epoch": 1.179698838896952,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048672293221655333,
      "loss": 5.5177,
      "mean_token_accuracy": 0.15639309734106063,
      "num_tokens": 26975400.0,
      "step": 13005
    },
    {
      "entropy": 5.79458327293396,
      "epoch": 1.1801523947750363,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004867119795757935,
      "loss": 5.607,
      "mean_token_accuracy": 0.15995785892009734,
      "num_tokens": 26986813.0,
      "step": 13010
    },
    {
      "entropy": 5.729414701461792,
      "epoch": 1.1806059506531206,
      "grad_norm": 3.796875,
      "learning_rate": 0.000486701022556776,
      "loss": 5.4498,
      "mean_token_accuracy": 0.16568358540534972,
      "num_tokens": 26998112.0,
      "step": 13015
    },
    {
      "entropy": 5.72943525314331,
      "epoch": 1.1810595065312046,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004866900611597273,
      "loss": 5.5015,
      "mean_token_accuracy": 0.1601800099015236,
      "num_tokens": 27008791.0,
      "step": 13020
    },
    {
      "entropy": 5.666735029220581,
      "epoch": 1.1815130624092889,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048667909538487427,
      "loss": 5.4398,
      "mean_token_accuracy": 0.16841307580471038,
      "num_tokens": 27018523.0,
      "step": 13025
    },
    {
      "entropy": 5.750645017623901,
      "epoch": 1.181966618287373,
      "grad_norm": 1.75,
      "learning_rate": 0.0004866681252324436,
      "loss": 5.5233,
      "mean_token_accuracy": 0.15650758594274522,
      "num_tokens": 27028667.0,
      "step": 13030
    },
    {
      "entropy": 5.726427841186523,
      "epoch": 1.1824201741654572,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048665715070266214,
      "loss": 5.4555,
      "mean_token_accuracy": 0.1640412002801895,
      "num_tokens": 27038476.0,
      "step": 13035
    },
    {
      "entropy": 5.651886320114135,
      "epoch": 1.1828737300435415,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048664617179575685,
      "loss": 5.4382,
      "mean_token_accuracy": 0.1654398672282696,
      "num_tokens": 27048599.0,
      "step": 13040
    },
    {
      "entropy": 5.7015196800231935,
      "epoch": 1.1833272859216255,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004866351885119548,
      "loss": 5.4807,
      "mean_token_accuracy": 0.16667191982269286,
      "num_tokens": 27059402.0,
      "step": 13045
    },
    {
      "entropy": 5.68692536354065,
      "epoch": 1.1837808417997098,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004866242008514832,
      "loss": 5.4283,
      "mean_token_accuracy": 0.16390670388936995,
      "num_tokens": 27069538.0,
      "step": 13050
    },
    {
      "entropy": 5.77054500579834,
      "epoch": 1.1842343976777938,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048661320881456904,
      "loss": 5.602,
      "mean_token_accuracy": 0.1612290099263191,
      "num_tokens": 27080007.0,
      "step": 13055
    },
    {
      "entropy": 5.76998872756958,
      "epoch": 1.184687953555878,
      "grad_norm": 3.296875,
      "learning_rate": 0.0004866022124014399,
      "loss": 5.5184,
      "mean_token_accuracy": 0.16498316675424576,
      "num_tokens": 27091314.0,
      "step": 13060
    },
    {
      "entropy": 5.723768949508667,
      "epoch": 1.1851415094339623,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048659121161232313,
      "loss": 5.4829,
      "mean_token_accuracy": 0.16116041243076323,
      "num_tokens": 27102478.0,
      "step": 13065
    },
    {
      "entropy": 5.728901433944702,
      "epoch": 1.1855950653120464,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004865802064474461,
      "loss": 5.5959,
      "mean_token_accuracy": 0.15973734408617019,
      "num_tokens": 27112995.0,
      "step": 13070
    },
    {
      "entropy": 5.746443414688111,
      "epoch": 1.1860486211901307,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048656919690703653,
      "loss": 5.6273,
      "mean_token_accuracy": 0.16198693215847015,
      "num_tokens": 27123671.0,
      "step": 13075
    },
    {
      "entropy": 5.771101617813111,
      "epoch": 1.1865021770682147,
      "grad_norm": 1.8828125,
      "learning_rate": 0.000486558182991322,
      "loss": 5.4731,
      "mean_token_accuracy": 0.16478919759392738,
      "num_tokens": 27133736.0,
      "step": 13080
    },
    {
      "entropy": 5.712762546539307,
      "epoch": 1.186955732946299,
      "grad_norm": 3.25,
      "learning_rate": 0.0004865471647005304,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1642451748251915,
      "num_tokens": 27142946.0,
      "step": 13085
    },
    {
      "entropy": 5.7169756412506105,
      "epoch": 1.1874092888243832,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048653614203488947,
      "loss": 5.4284,
      "mean_token_accuracy": 0.16038626283407212,
      "num_tokens": 27153506.0,
      "step": 13090
    },
    {
      "entropy": 5.729301643371582,
      "epoch": 1.1878628447024673,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004865251149946273,
      "loss": 5.4864,
      "mean_token_accuracy": 0.1559924900531769,
      "num_tokens": 27163562.0,
      "step": 13095
    },
    {
      "entropy": 5.791506052017212,
      "epoch": 1.1883164005805515,
      "grad_norm": 2.5,
      "learning_rate": 0.00048651408357997173,
      "loss": 5.6034,
      "mean_token_accuracy": 0.15750048607587813,
      "num_tokens": 27174739.0,
      "step": 13100
    },
    {
      "entropy": 5.799505233764648,
      "epoch": 1.1887699564586356,
      "grad_norm": 11.8125,
      "learning_rate": 0.00048650304779115114,
      "loss": 5.4975,
      "mean_token_accuracy": 0.1631581112742424,
      "num_tokens": 27185729.0,
      "step": 13105
    },
    {
      "entropy": 5.81752028465271,
      "epoch": 1.1892235123367199,
      "grad_norm": 2.5,
      "learning_rate": 0.0004864920076283935,
      "loss": 5.5285,
      "mean_token_accuracy": 0.1686689466238022,
      "num_tokens": 27196034.0,
      "step": 13110
    },
    {
      "entropy": 5.67460298538208,
      "epoch": 1.1896770682148041,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004864809630919273,
      "loss": 5.4576,
      "mean_token_accuracy": 0.16569779813289642,
      "num_tokens": 27206102.0,
      "step": 13115
    },
    {
      "entropy": 5.671186828613282,
      "epoch": 1.1901306240928882,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004864699141819809,
      "loss": 5.4308,
      "mean_token_accuracy": 0.1750039964914322,
      "num_tokens": 27215430.0,
      "step": 13120
    },
    {
      "entropy": 5.741038274765015,
      "epoch": 1.1905841799709724,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048645886089878276,
      "loss": 5.4708,
      "mean_token_accuracy": 0.1684890702366829,
      "num_tokens": 27225289.0,
      "step": 13125
    },
    {
      "entropy": 5.632559108734131,
      "epoch": 1.1910377358490567,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048644780324256153,
      "loss": 5.3083,
      "mean_token_accuracy": 0.17981231659650804,
      "num_tokens": 27234979.0,
      "step": 13130
    },
    {
      "entropy": 5.68755464553833,
      "epoch": 1.1914912917271407,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004864367412135458,
      "loss": 5.4718,
      "mean_token_accuracy": 0.16708918958902358,
      "num_tokens": 27244885.0,
      "step": 13135
    },
    {
      "entropy": 5.684140157699585,
      "epoch": 1.191944847605225,
      "grad_norm": 3.265625,
      "learning_rate": 0.00048642567481196446,
      "loss": 5.4114,
      "mean_token_accuracy": 0.16903109848499298,
      "num_tokens": 27255383.0,
      "step": 13140
    },
    {
      "entropy": 5.6808459758758545,
      "epoch": 1.192398403483309,
      "grad_norm": 2.75,
      "learning_rate": 0.0004864146040380462,
      "loss": 5.4845,
      "mean_token_accuracy": 0.1659761667251587,
      "num_tokens": 27265966.0,
      "step": 13145
    },
    {
      "entropy": 5.7317462921142575,
      "epoch": 1.1928519593613933,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004864035288920202,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1639991819858551,
      "num_tokens": 27275832.0,
      "step": 13150
    },
    {
      "entropy": 5.67484941482544,
      "epoch": 1.1933055152394776,
      "grad_norm": 2.375,
      "learning_rate": 0.00048639244937411516,
      "loss": 5.4656,
      "mean_token_accuracy": 0.16640899032354356,
      "num_tokens": 27285854.0,
      "step": 13155
    },
    {
      "entropy": 5.678304624557495,
      "epoch": 1.1937590711175616,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048638136548456045,
      "loss": 5.4715,
      "mean_token_accuracy": 0.16945944130420684,
      "num_tokens": 27296066.0,
      "step": 13160
    },
    {
      "entropy": 5.685620307922363,
      "epoch": 1.194212626995646,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048637027722358536,
      "loss": 5.414,
      "mean_token_accuracy": 0.17039725482463836,
      "num_tokens": 27305562.0,
      "step": 13165
    },
    {
      "entropy": 5.669494867324829,
      "epoch": 1.1946661828737302,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048635918459141893,
      "loss": 5.4135,
      "mean_token_accuracy": 0.17342533618211747,
      "num_tokens": 27315221.0,
      "step": 13170
    },
    {
      "entropy": 5.728151798248291,
      "epoch": 1.1951197387518142,
      "grad_norm": 2.0,
      "learning_rate": 0.00048634808758829073,
      "loss": 5.4753,
      "mean_token_accuracy": 0.16145778745412825,
      "num_tokens": 27326711.0,
      "step": 13175
    },
    {
      "entropy": 5.72562804222107,
      "epoch": 1.1955732946298985,
      "grad_norm": 2.625,
      "learning_rate": 0.00048633698621443024,
      "loss": 5.5684,
      "mean_token_accuracy": 0.14819405823946,
      "num_tokens": 27338079.0,
      "step": 13180
    },
    {
      "entropy": 5.715652942657471,
      "epoch": 1.1960268505079825,
      "grad_norm": 3.34375,
      "learning_rate": 0.000486325880470067,
      "loss": 5.4579,
      "mean_token_accuracy": 0.16527933925390242,
      "num_tokens": 27349660.0,
      "step": 13185
    },
    {
      "entropy": 5.744312715530396,
      "epoch": 1.1964804063860668,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004863147703554307,
      "loss": 5.465,
      "mean_token_accuracy": 0.16554103344678878,
      "num_tokens": 27360072.0,
      "step": 13190
    },
    {
      "entropy": 5.733823633193969,
      "epoch": 1.196933962264151,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048630365587075115,
      "loss": 5.4984,
      "mean_token_accuracy": 0.16040853559970855,
      "num_tokens": 27370536.0,
      "step": 13195
    },
    {
      "entropy": 5.796777963638306,
      "epoch": 1.197387518142235,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048629253701625816,
      "loss": 5.5431,
      "mean_token_accuracy": 0.15520152151584626,
      "num_tokens": 27381311.0,
      "step": 13200
    },
    {
      "entropy": 5.690798664093018,
      "epoch": 1.1978410740203194,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004862814137921816,
      "loss": 5.4764,
      "mean_token_accuracy": 0.16387361586093901,
      "num_tokens": 27391540.0,
      "step": 13205
    },
    {
      "entropy": 5.7300519943237305,
      "epoch": 1.1982946298984034,
      "grad_norm": 3.953125,
      "learning_rate": 0.00048627028619875157,
      "loss": 5.5674,
      "mean_token_accuracy": 0.16046540588140487,
      "num_tokens": 27401726.0,
      "step": 13210
    },
    {
      "entropy": 5.76314640045166,
      "epoch": 1.1987481857764877,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004862591542361982,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1727989763021469,
      "num_tokens": 27411802.0,
      "step": 13215
    },
    {
      "entropy": 5.736448478698731,
      "epoch": 1.199201741654572,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004862480179047516,
      "loss": 5.4479,
      "mean_token_accuracy": 0.17110642194747924,
      "num_tokens": 27421471.0,
      "step": 13220
    },
    {
      "entropy": 5.736895990371704,
      "epoch": 1.199655297532656,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004862368772046423,
      "loss": 5.536,
      "mean_token_accuracy": 0.16687171757221222,
      "num_tokens": 27430397.0,
      "step": 13225
    },
    {
      "entropy": 5.733551931381226,
      "epoch": 1.2001088534107403,
      "grad_norm": 3.71875,
      "learning_rate": 0.00048622573213610057,
      "loss": 5.4704,
      "mean_token_accuracy": 0.16447337418794633,
      "num_tokens": 27441104.0,
      "step": 13230
    },
    {
      "entropy": 5.7476215839385985,
      "epoch": 1.2005624092888243,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048621458269935676,
      "loss": 5.5473,
      "mean_token_accuracy": 0.16173536628484725,
      "num_tokens": 27451398.0,
      "step": 13235
    },
    {
      "entropy": 5.843199014663696,
      "epoch": 1.2010159651669086,
      "grad_norm": 2.796875,
      "learning_rate": 0.00048620342889464153,
      "loss": 5.568,
      "mean_token_accuracy": 0.15448569506406784,
      "num_tokens": 27461332.0,
      "step": 13240
    },
    {
      "entropy": 5.747753763198853,
      "epoch": 1.2014695210449928,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048619227072218565,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16710079908370973,
      "num_tokens": 27470898.0,
      "step": 13245
    },
    {
      "entropy": 5.70690221786499,
      "epoch": 1.2019230769230769,
      "grad_norm": 6.9375,
      "learning_rate": 0.0004861811081822198,
      "loss": 5.5332,
      "mean_token_accuracy": 0.16594915688037873,
      "num_tokens": 27481817.0,
      "step": 13250
    },
    {
      "entropy": 5.70338191986084,
      "epoch": 1.2023766328011611,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004861699412749748,
      "loss": 5.5256,
      "mean_token_accuracy": 0.16427023261785506,
      "num_tokens": 27492237.0,
      "step": 13255
    },
    {
      "entropy": 5.684726715087891,
      "epoch": 1.2028301886792452,
      "grad_norm": 1.875,
      "learning_rate": 0.0004861587700006816,
      "loss": 5.5437,
      "mean_token_accuracy": 0.1636929288506508,
      "num_tokens": 27502326.0,
      "step": 13260
    },
    {
      "entropy": 5.839702081680298,
      "epoch": 1.2032837445573294,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048614759435957114,
      "loss": 5.5175,
      "mean_token_accuracy": 0.16045489013195038,
      "num_tokens": 27512644.0,
      "step": 13265
    },
    {
      "entropy": 5.677344512939453,
      "epoch": 1.2037373004354137,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004861364143518747,
      "loss": 5.3831,
      "mean_token_accuracy": 0.1730010762810707,
      "num_tokens": 27521821.0,
      "step": 13270
    },
    {
      "entropy": 5.7301225662231445,
      "epoch": 1.2041908563134978,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004861252299778234,
      "loss": 5.5125,
      "mean_token_accuracy": 0.16092113256454468,
      "num_tokens": 27531445.0,
      "step": 13275
    },
    {
      "entropy": 5.761343336105346,
      "epoch": 1.204644412191582,
      "grad_norm": 2.875,
      "learning_rate": 0.00048611404123764855,
      "loss": 5.4875,
      "mean_token_accuracy": 0.1645788073539734,
      "num_tokens": 27541844.0,
      "step": 13280
    },
    {
      "entropy": 5.763279867172241,
      "epoch": 1.205097968069666,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048610284813158154,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1666392371058464,
      "num_tokens": 27552995.0,
      "step": 13285
    },
    {
      "entropy": 5.652666187286377,
      "epoch": 1.2055515239477503,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004860916506598538,
      "loss": 5.5456,
      "mean_token_accuracy": 0.15889296531677247,
      "num_tokens": 27563664.0,
      "step": 13290
    },
    {
      "entropy": 5.762326955795288,
      "epoch": 1.2060050798258346,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048608044882269695,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1588563546538353,
      "num_tokens": 27575245.0,
      "step": 13295
    },
    {
      "entropy": 5.77012996673584,
      "epoch": 1.2064586357039186,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004860692426203427,
      "loss": 5.4653,
      "mean_token_accuracy": 0.1646143913269043,
      "num_tokens": 27585416.0,
      "step": 13300
    },
    {
      "entropy": 5.679472303390503,
      "epoch": 1.206912191582003,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004860580320530226,
      "loss": 5.5095,
      "mean_token_accuracy": 0.15809310972690582,
      "num_tokens": 27595901.0,
      "step": 13305
    },
    {
      "entropy": 5.724077749252319,
      "epoch": 1.2073657474600872,
      "grad_norm": 4.5,
      "learning_rate": 0.0004860468171209687,
      "loss": 5.5086,
      "mean_token_accuracy": 0.16687510013580323,
      "num_tokens": 27605936.0,
      "step": 13310
    },
    {
      "entropy": 5.697344160079956,
      "epoch": 1.2078193033381712,
      "grad_norm": 3.890625,
      "learning_rate": 0.0004860355978244128,
      "loss": 5.4722,
      "mean_token_accuracy": 0.16212105005979538,
      "num_tokens": 27615469.0,
      "step": 13315
    },
    {
      "entropy": 5.7590583801269535,
      "epoch": 1.2082728592162555,
      "grad_norm": 5.15625,
      "learning_rate": 0.000486024374163587,
      "loss": 5.4743,
      "mean_token_accuracy": 0.16246298104524612,
      "num_tokens": 27626864.0,
      "step": 13320
    },
    {
      "entropy": 5.675850534439087,
      "epoch": 1.2087264150943395,
      "grad_norm": 3.640625,
      "learning_rate": 0.00048601314613872334,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1633027032017708,
      "num_tokens": 27636556.0,
      "step": 13325
    },
    {
      "entropy": 5.678484153747559,
      "epoch": 1.2091799709724238,
      "grad_norm": 3.15625,
      "learning_rate": 0.000486001913750054,
      "loss": 5.4918,
      "mean_token_accuracy": 0.16721775233745576,
      "num_tokens": 27647703.0,
      "step": 13330
    },
    {
      "entropy": 5.8305683612823485,
      "epoch": 1.209633526850508,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004859906769978113,
      "loss": 5.5189,
      "mean_token_accuracy": 0.16830294579267502,
      "num_tokens": 27657845.0,
      "step": 13335
    },
    {
      "entropy": 5.753674840927124,
      "epoch": 1.2100870827285921,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048597943588222776,
      "loss": 5.4492,
      "mean_token_accuracy": 0.16290076225996017,
      "num_tokens": 27667080.0,
      "step": 13340
    },
    {
      "entropy": 5.743187189102173,
      "epoch": 1.2105406386066764,
      "grad_norm": 1.765625,
      "learning_rate": 0.00048596819040353564,
      "loss": 5.5165,
      "mean_token_accuracy": 0.16817213743925094,
      "num_tokens": 27676805.0,
      "step": 13345
    },
    {
      "entropy": 5.703033304214477,
      "epoch": 1.2109941944847606,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004859569405619676,
      "loss": 5.4696,
      "mean_token_accuracy": 0.16901935040950775,
      "num_tokens": 27686640.0,
      "step": 13350
    },
    {
      "entropy": 5.691295337677002,
      "epoch": 1.2114477503628447,
      "grad_norm": 2.875,
      "learning_rate": 0.0004859456863577562,
      "loss": 5.4992,
      "mean_token_accuracy": 0.16624235212802888,
      "num_tokens": 27696960.0,
      "step": 13355
    },
    {
      "entropy": 5.7300293922424315,
      "epoch": 1.211901306240929,
      "grad_norm": 5.5625,
      "learning_rate": 0.00048593442779113437,
      "loss": 5.5133,
      "mean_token_accuracy": 0.16317834556102753,
      "num_tokens": 27706562.0,
      "step": 13360
    },
    {
      "entropy": 5.6981652736663815,
      "epoch": 1.212354862119013,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004859231648623348,
      "loss": 5.4943,
      "mean_token_accuracy": 0.1671316757798195,
      "num_tokens": 27716693.0,
      "step": 13365
    },
    {
      "entropy": 5.786464405059815,
      "epoch": 1.2128084179970973,
      "grad_norm": 2.5,
      "learning_rate": 0.00048591189757159035,
      "loss": 5.5329,
      "mean_token_accuracy": 0.1609986752271652,
      "num_tokens": 27727163.0,
      "step": 13370
    },
    {
      "entropy": 5.784613084793091,
      "epoch": 1.2132619738751815,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004859006259191342,
      "loss": 5.4463,
      "mean_token_accuracy": 0.16525145322084428,
      "num_tokens": 27737905.0,
      "step": 13375
    },
    {
      "entropy": 5.71513729095459,
      "epoch": 1.2137155297532656,
      "grad_norm": 3.75,
      "learning_rate": 0.0004858893499051993,
      "loss": 5.5315,
      "mean_token_accuracy": 0.16561403274536132,
      "num_tokens": 27748974.0,
      "step": 13380
    },
    {
      "entropy": 5.689414310455322,
      "epoch": 1.2141690856313498,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004858780695300189,
      "loss": 5.5089,
      "mean_token_accuracy": 0.16324905455112457,
      "num_tokens": 27759626.0,
      "step": 13385
    },
    {
      "entropy": 5.703880405426025,
      "epoch": 1.2146226415094339,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004858667847938263,
      "loss": 5.4986,
      "mean_token_accuracy": 0.16525344252586366,
      "num_tokens": 27769769.0,
      "step": 13390
    },
    {
      "entropy": 5.742326259613037,
      "epoch": 1.2150761973875182,
      "grad_norm": 5.40625,
      "learning_rate": 0.00048585549569685486,
      "loss": 5.4722,
      "mean_token_accuracy": 0.16570081263780595,
      "num_tokens": 27779369.0,
      "step": 13395
    },
    {
      "entropy": 5.725001049041748,
      "epoch": 1.2155297532656024,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048584420223933805,
      "loss": 5.5159,
      "mean_token_accuracy": 0.1556485190987587,
      "num_tokens": 27790194.0,
      "step": 13400
    },
    {
      "entropy": 5.698969888687134,
      "epoch": 1.2159833091436865,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004858329044215094,
      "loss": 5.3931,
      "mean_token_accuracy": 0.17435964643955232,
      "num_tokens": 27800354.0,
      "step": 13405
    },
    {
      "entropy": 5.691322898864746,
      "epoch": 1.2164368650217707,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048582160224360253,
      "loss": 5.4883,
      "mean_token_accuracy": 0.16315832287073134,
      "num_tokens": 27811771.0,
      "step": 13410
    },
    {
      "entropy": 5.710209703445434,
      "epoch": 1.2168904208998548,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004858102957058513,
      "loss": 5.4943,
      "mean_token_accuracy": 0.16843938529491426,
      "num_tokens": 27821516.0,
      "step": 13415
    },
    {
      "entropy": 5.780866622924805,
      "epoch": 1.217343976777939,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048579898480848933,
      "loss": 5.5191,
      "mean_token_accuracy": 0.16266807913780212,
      "num_tokens": 27832000.0,
      "step": 13420
    },
    {
      "entropy": 5.748239469528198,
      "epoch": 1.2177975326560233,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004857876695517507,
      "loss": 5.5164,
      "mean_token_accuracy": 0.16004256904125214,
      "num_tokens": 27843436.0,
      "step": 13425
    },
    {
      "entropy": 5.662473726272583,
      "epoch": 1.2182510885341074,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048577634993586923,
      "loss": 5.4653,
      "mean_token_accuracy": 0.15929367691278457,
      "num_tokens": 27853418.0,
      "step": 13430
    },
    {
      "entropy": 5.767095851898193,
      "epoch": 1.2187046444121916,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048576502596107924,
      "loss": 5.4801,
      "mean_token_accuracy": 0.16274332702159883,
      "num_tokens": 27863253.0,
      "step": 13435
    },
    {
      "entropy": 5.841007566452026,
      "epoch": 1.2191582002902757,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004857536976276148,
      "loss": 5.5385,
      "mean_token_accuracy": 0.15988081991672515,
      "num_tokens": 27873560.0,
      "step": 13440
    },
    {
      "entropy": 5.742250108718872,
      "epoch": 1.21961175616836,
      "grad_norm": 4.125,
      "learning_rate": 0.0004857423649357102,
      "loss": 5.4669,
      "mean_token_accuracy": 0.16766862124204635,
      "num_tokens": 27884516.0,
      "step": 13445
    },
    {
      "entropy": 5.706092786788941,
      "epoch": 1.2200653120464442,
      "grad_norm": 2.375,
      "learning_rate": 0.00048573102788559974,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16122180670499803,
      "num_tokens": 27893664.0,
      "step": 13450
    },
    {
      "entropy": 5.734672737121582,
      "epoch": 1.2205188679245282,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004857196864775179,
      "loss": 5.4483,
      "mean_token_accuracy": 0.164666149020195,
      "num_tokens": 27903953.0,
      "step": 13455
    },
    {
      "entropy": 5.674837923049926,
      "epoch": 1.2209724238026125,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048570834071169926,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1688227981328964,
      "num_tokens": 27914664.0,
      "step": 13460
    },
    {
      "entropy": 5.629163312911987,
      "epoch": 1.2214259796806965,
      "grad_norm": 3.453125,
      "learning_rate": 0.00048569699058837856,
      "loss": 5.4212,
      "mean_token_accuracy": 0.17492074221372605,
      "num_tokens": 27924321.0,
      "step": 13465
    },
    {
      "entropy": 5.70867509841919,
      "epoch": 1.2218795355587808,
      "grad_norm": 2.5,
      "learning_rate": 0.0004856856361077903,
      "loss": 5.441,
      "mean_token_accuracy": 0.16939868330955504,
      "num_tokens": 27934196.0,
      "step": 13470
    },
    {
      "entropy": 5.725666093826294,
      "epoch": 1.222333091436865,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004856742772701694,
      "loss": 5.4802,
      "mean_token_accuracy": 0.1733180820941925,
      "num_tokens": 27945195.0,
      "step": 13475
    },
    {
      "entropy": 5.7089954853057865,
      "epoch": 1.2227866473149491,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048566291407575073,
      "loss": 5.4718,
      "mean_token_accuracy": 0.1729305014014244,
      "num_tokens": 27955286.0,
      "step": 13480
    },
    {
      "entropy": 5.611675262451172,
      "epoch": 1.2232402031930334,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048565154652476937,
      "loss": 5.4204,
      "mean_token_accuracy": 0.16516849398612976,
      "num_tokens": 27965422.0,
      "step": 13485
    },
    {
      "entropy": 5.680658197402954,
      "epoch": 1.2236937590711174,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004856401746174603,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17226017415523528,
      "num_tokens": 27975831.0,
      "step": 13490
    },
    {
      "entropy": 5.775923109054565,
      "epoch": 1.2241473149492017,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048562879835405874,
      "loss": 5.568,
      "mean_token_accuracy": 0.1648126944899559,
      "num_tokens": 27986922.0,
      "step": 13495
    },
    {
      "entropy": 5.732413625717163,
      "epoch": 1.224600870827286,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004856174177348,
      "loss": 5.4651,
      "mean_token_accuracy": 0.16148756593465804,
      "num_tokens": 27997926.0,
      "step": 13500
    },
    {
      "entropy": 5.741351747512818,
      "epoch": 1.22505442670537,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004856060327599194,
      "loss": 5.487,
      "mean_token_accuracy": 0.1622229352593422,
      "num_tokens": 28007709.0,
      "step": 13505
    },
    {
      "entropy": 5.686682796478271,
      "epoch": 1.2255079825834543,
      "grad_norm": 2.890625,
      "learning_rate": 0.00048559464342965233,
      "loss": 5.4067,
      "mean_token_accuracy": 0.1723533809185028,
      "num_tokens": 28017671.0,
      "step": 13510
    },
    {
      "entropy": 5.6439460754394535,
      "epoch": 1.2259615384615385,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004855832497442344,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16372669637203216,
      "num_tokens": 28027772.0,
      "step": 13515
    },
    {
      "entropy": 5.7187474250793455,
      "epoch": 1.2264150943396226,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004855718517039012,
      "loss": 5.5,
      "mean_token_accuracy": 0.15953412503004075,
      "num_tokens": 28038365.0,
      "step": 13520
    },
    {
      "entropy": 5.736917495727539,
      "epoch": 1.2268686502177069,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048556044930888844,
      "loss": 5.4079,
      "mean_token_accuracy": 0.16804706305265427,
      "num_tokens": 28048569.0,
      "step": 13525
    },
    {
      "entropy": 5.749265384674072,
      "epoch": 1.2273222060957911,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004855490425594319,
      "loss": 5.519,
      "mean_token_accuracy": 0.16731065958738328,
      "num_tokens": 28059297.0,
      "step": 13530
    },
    {
      "entropy": 5.68096890449524,
      "epoch": 1.2277757619738752,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004855376314557675,
      "loss": 5.5245,
      "mean_token_accuracy": 0.16125939339399337,
      "num_tokens": 28070684.0,
      "step": 13535
    },
    {
      "entropy": 5.696394062042236,
      "epoch": 1.2282293178519594,
      "grad_norm": 3.25,
      "learning_rate": 0.0004855262159981313,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1765495553612709,
      "num_tokens": 28080621.0,
      "step": 13540
    },
    {
      "entropy": 5.808384370803833,
      "epoch": 1.2286828737300435,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048551479618675925,
      "loss": 5.5098,
      "mean_token_accuracy": 0.16822903454303742,
      "num_tokens": 28089987.0,
      "step": 13545
    },
    {
      "entropy": 5.779904222488403,
      "epoch": 1.2291364296081277,
      "grad_norm": 2.65625,
      "learning_rate": 0.00048550337202188756,
      "loss": 5.5802,
      "mean_token_accuracy": 0.15943180099129678,
      "num_tokens": 28100063.0,
      "step": 13550
    },
    {
      "entropy": 5.73337345123291,
      "epoch": 1.229589985486212,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004854919435037525,
      "loss": 5.5801,
      "mean_token_accuracy": 0.15865944400429727,
      "num_tokens": 28111235.0,
      "step": 13555
    },
    {
      "entropy": 5.809113025665283,
      "epoch": 1.230043541364296,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004854805106325904,
      "loss": 5.6037,
      "mean_token_accuracy": 0.16124303936958312,
      "num_tokens": 28122214.0,
      "step": 13560
    },
    {
      "entropy": 5.827086353302002,
      "epoch": 1.2304970972423803,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004854690734086378,
      "loss": 5.5779,
      "mean_token_accuracy": 0.16593380123376847,
      "num_tokens": 28132919.0,
      "step": 13565
    },
    {
      "entropy": 5.69459753036499,
      "epoch": 1.2309506531204644,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004854576318321311,
      "loss": 5.4378,
      "mean_token_accuracy": 0.15904999077320098,
      "num_tokens": 28143056.0,
      "step": 13570
    },
    {
      "entropy": 5.722627782821656,
      "epoch": 1.2314042089985486,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048544618590330685,
      "loss": 5.4205,
      "mean_token_accuracy": 0.16908029466867447,
      "num_tokens": 28153470.0,
      "step": 13575
    },
    {
      "entropy": 5.768122911453247,
      "epoch": 1.231857764876633,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048543473562240194,
      "loss": 5.5181,
      "mean_token_accuracy": 0.1596188098192215,
      "num_tokens": 28163391.0,
      "step": 13580
    },
    {
      "entropy": 5.722019338607788,
      "epoch": 1.232311320754717,
      "grad_norm": 1.96875,
      "learning_rate": 0.000485423280989653,
      "loss": 5.4392,
      "mean_token_accuracy": 0.16849638968706132,
      "num_tokens": 28174460.0,
      "step": 13585
    },
    {
      "entropy": 5.684997272491455,
      "epoch": 1.2327648766328012,
      "grad_norm": 4.1875,
      "learning_rate": 0.000485411822005297,
      "loss": 5.42,
      "mean_token_accuracy": 0.1652347832918167,
      "num_tokens": 28184977.0,
      "step": 13590
    },
    {
      "entropy": 5.688517475128174,
      "epoch": 1.2332184325108853,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048540035866957097,
      "loss": 5.453,
      "mean_token_accuracy": 0.16974510103464127,
      "num_tokens": 28194914.0,
      "step": 13595
    },
    {
      "entropy": 5.719634294509888,
      "epoch": 1.2336719883889695,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004853888909827118,
      "loss": 5.4442,
      "mean_token_accuracy": 0.16463993042707442,
      "num_tokens": 28205347.0,
      "step": 13600
    },
    {
      "entropy": 5.686352252960205,
      "epoch": 1.2341255442670538,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004853774189449568,
      "loss": 5.4589,
      "mean_token_accuracy": 0.1688178539276123,
      "num_tokens": 28215710.0,
      "step": 13605
    },
    {
      "entropy": 5.731152725219727,
      "epoch": 1.2345791001451378,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048536594255654315,
      "loss": 5.5269,
      "mean_token_accuracy": 0.16374423503875732,
      "num_tokens": 28227043.0,
      "step": 13610
    },
    {
      "entropy": 5.702719068527221,
      "epoch": 1.235032656023222,
      "grad_norm": 2.375,
      "learning_rate": 0.0004853544618177083,
      "loss": 5.4859,
      "mean_token_accuracy": 0.16956159025430678,
      "num_tokens": 28237460.0,
      "step": 13615
    },
    {
      "entropy": 5.624921703338623,
      "epoch": 1.2354862119013061,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048534297672868945,
      "loss": 5.3965,
      "mean_token_accuracy": 0.17451781779527664,
      "num_tokens": 28246734.0,
      "step": 13620
    },
    {
      "entropy": 5.656033372879028,
      "epoch": 1.2359397677793904,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048533148728972434,
      "loss": 5.4815,
      "mean_token_accuracy": 0.16094603687524794,
      "num_tokens": 28257941.0,
      "step": 13625
    },
    {
      "entropy": 5.730606889724731,
      "epoch": 1.2363933236574747,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00048531999350105037,
      "loss": 5.5081,
      "mean_token_accuracy": 0.16635871976613997,
      "num_tokens": 28267839.0,
      "step": 13630
    },
    {
      "entropy": 5.702928590774536,
      "epoch": 1.2368468795355587,
      "grad_norm": 3.125,
      "learning_rate": 0.00048530849536290537,
      "loss": 5.4275,
      "mean_token_accuracy": 0.16330047696828842,
      "num_tokens": 28279581.0,
      "step": 13635
    },
    {
      "entropy": 5.698543357849121,
      "epoch": 1.237300435413643,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004852969928755271,
      "loss": 5.461,
      "mean_token_accuracy": 0.1689723551273346,
      "num_tokens": 28289240.0,
      "step": 13640
    },
    {
      "entropy": 5.645983123779297,
      "epoch": 1.237753991291727,
      "grad_norm": 3.171875,
      "learning_rate": 0.00048528548603915344,
      "loss": 5.487,
      "mean_token_accuracy": 0.16145799458026885,
      "num_tokens": 28300652.0,
      "step": 13645
    },
    {
      "entropy": 5.6258398532867435,
      "epoch": 1.2382075471698113,
      "grad_norm": 1.875,
      "learning_rate": 0.00048527397485402236,
      "loss": 5.394,
      "mean_token_accuracy": 0.16563512831926347,
      "num_tokens": 28312065.0,
      "step": 13650
    },
    {
      "entropy": 5.717063093185425,
      "epoch": 1.2386611030478956,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048526245932037186,
      "loss": 5.4322,
      "mean_token_accuracy": 0.16594297140836717,
      "num_tokens": 28322675.0,
      "step": 13655
    },
    {
      "entropy": 5.634007596969605,
      "epoch": 1.2391146589259796,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048525093943844013,
      "loss": 5.4319,
      "mean_token_accuracy": 0.16979459673166275,
      "num_tokens": 28332692.0,
      "step": 13660
    },
    {
      "entropy": 5.667316055297851,
      "epoch": 1.2395682148040639,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004852394152084654,
      "loss": 5.4748,
      "mean_token_accuracy": 0.16718326061964034,
      "num_tokens": 28342643.0,
      "step": 13665
    },
    {
      "entropy": 5.637993335723877,
      "epoch": 1.240021770682148,
      "grad_norm": 1.9609375,
      "learning_rate": 0.000485227886630686,
      "loss": 5.3552,
      "mean_token_accuracy": 0.16825830638408662,
      "num_tokens": 28352217.0,
      "step": 13670
    },
    {
      "entropy": 5.6174647331237795,
      "epoch": 1.2404753265602322,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004852163537053403,
      "loss": 5.4457,
      "mean_token_accuracy": 0.17103493064641953,
      "num_tokens": 28362254.0,
      "step": 13675
    },
    {
      "entropy": 5.649400758743286,
      "epoch": 1.2409288824383164,
      "grad_norm": 2.5,
      "learning_rate": 0.0004852048164326669,
      "loss": 5.4219,
      "mean_token_accuracy": 0.16227339208126068,
      "num_tokens": 28372174.0,
      "step": 13680
    },
    {
      "entropy": 5.751662826538086,
      "epoch": 1.2413824383164005,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004851932748129043,
      "loss": 5.483,
      "mean_token_accuracy": 0.16633015424013137,
      "num_tokens": 28382302.0,
      "step": 13685
    },
    {
      "entropy": 5.752591514587403,
      "epoch": 1.2418359941944848,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004851817288462912,
      "loss": 5.4527,
      "mean_token_accuracy": 0.1669698566198349,
      "num_tokens": 28392581.0,
      "step": 13690
    },
    {
      "entropy": 5.700544166564941,
      "epoch": 1.242289550072569,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004851701785330663,
      "loss": 5.5437,
      "mean_token_accuracy": 0.1639311894774437,
      "num_tokens": 28403089.0,
      "step": 13695
    },
    {
      "entropy": 5.752901840209961,
      "epoch": 1.242743105950653,
      "grad_norm": 2.875,
      "learning_rate": 0.0004851586238734687,
      "loss": 5.5828,
      "mean_token_accuracy": 0.15601109266281127,
      "num_tokens": 28415249.0,
      "step": 13700
    },
    {
      "entropy": 5.744243717193603,
      "epoch": 1.2431966618287373,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004851470648677371,
      "loss": 5.5409,
      "mean_token_accuracy": 0.16179286539554597,
      "num_tokens": 28426580.0,
      "step": 13705
    },
    {
      "entropy": 5.737151765823365,
      "epoch": 1.2436502177068216,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004851355015161107,
      "loss": 5.4252,
      "mean_token_accuracy": 0.16839856058359146,
      "num_tokens": 28436501.0,
      "step": 13710
    },
    {
      "entropy": 5.742711925506592,
      "epoch": 1.2441037735849056,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004851239338188286,
      "loss": 5.5359,
      "mean_token_accuracy": 0.16015122532844545,
      "num_tokens": 28447749.0,
      "step": 13715
    },
    {
      "entropy": 5.682984161376953,
      "epoch": 1.24455732946299,
      "grad_norm": 3.03125,
      "learning_rate": 0.00048511236177612994,
      "loss": 5.4629,
      "mean_token_accuracy": 0.16792826801538469,
      "num_tokens": 28458028.0,
      "step": 13720
    },
    {
      "entropy": 5.748624849319458,
      "epoch": 1.245010885341074,
      "grad_norm": 4.15625,
      "learning_rate": 0.00048510078538825413,
      "loss": 5.4938,
      "mean_token_accuracy": 0.1586667537689209,
      "num_tokens": 28468460.0,
      "step": 13725
    },
    {
      "entropy": 5.632420539855957,
      "epoch": 1.2454644412191582,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048508920465544056,
      "loss": 5.3975,
      "mean_token_accuracy": 0.16563227474689485,
      "num_tokens": 28478634.0,
      "step": 13730
    },
    {
      "entropy": 5.630134010314942,
      "epoch": 1.2459179970972425,
      "grad_norm": 2.5,
      "learning_rate": 0.0004850776195779286,
      "loss": 5.375,
      "mean_token_accuracy": 0.16670473366975785,
      "num_tokens": 28488277.0,
      "step": 13735
    },
    {
      "entropy": 5.736680030822754,
      "epoch": 1.2463715529753265,
      "grad_norm": 2.390625,
      "learning_rate": 0.000485066030155958,
      "loss": 5.5073,
      "mean_token_accuracy": 0.16161807477474213,
      "num_tokens": 28499398.0,
      "step": 13740
    },
    {
      "entropy": 5.722391080856323,
      "epoch": 1.2468251088534108,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004850544363897684,
      "loss": 5.4159,
      "mean_token_accuracy": 0.1701399192214012,
      "num_tokens": 28509111.0,
      "step": 13745
    },
    {
      "entropy": 5.639610338211059,
      "epoch": 1.2472786647314948,
      "grad_norm": 4.78125,
      "learning_rate": 0.00048504283827959947,
      "loss": 5.4069,
      "mean_token_accuracy": 0.1631857842206955,
      "num_tokens": 28519499.0,
      "step": 13750
    },
    {
      "entropy": 5.751363277435303,
      "epoch": 1.2477322206095791,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004850312358256912,
      "loss": 5.5723,
      "mean_token_accuracy": 0.16463851928710938,
      "num_tokens": 28529927.0,
      "step": 13755
    },
    {
      "entropy": 5.697699117660522,
      "epoch": 1.2481857764876634,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004850196290282834,
      "loss": 5.4474,
      "mean_token_accuracy": 0.1634952798485756,
      "num_tokens": 28541635.0,
      "step": 13760
    },
    {
      "entropy": 5.633901739120484,
      "epoch": 1.2486393323657474,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048500801788761613,
      "loss": 5.4384,
      "mean_token_accuracy": 0.16753406822681427,
      "num_tokens": 28552248.0,
      "step": 13765
    },
    {
      "entropy": 5.753492546081543,
      "epoch": 1.2490928882438317,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004849964024039296,
      "loss": 5.517,
      "mean_token_accuracy": 0.16746966689825057,
      "num_tokens": 28562658.0,
      "step": 13770
    },
    {
      "entropy": 5.771497344970703,
      "epoch": 1.2495464441219157,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048498478257746387,
      "loss": 5.4941,
      "mean_token_accuracy": 0.1662286788225174,
      "num_tokens": 28574403.0,
      "step": 13775
    },
    {
      "entropy": 5.755687046051025,
      "epoch": 1.25,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048497315840845934,
      "loss": 5.5501,
      "mean_token_accuracy": 0.15899156928062438,
      "num_tokens": 28583807.0,
      "step": 13780
    },
    {
      "entropy": 5.783190536499023,
      "epoch": 1.2504535558780843,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004849615298971565,
      "loss": 5.4921,
      "mean_token_accuracy": 0.16355013698339463,
      "num_tokens": 28593641.0,
      "step": 13785
    },
    {
      "entropy": 5.750645351409912,
      "epoch": 1.2509071117561683,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048494989704379556,
      "loss": 5.5056,
      "mean_token_accuracy": 0.1673031911253929,
      "num_tokens": 28603573.0,
      "step": 13790
    },
    {
      "entropy": 5.655308532714844,
      "epoch": 1.2513606676342526,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004849382598486173,
      "loss": 5.4134,
      "mean_token_accuracy": 0.17760835438966752,
      "num_tokens": 28613368.0,
      "step": 13795
    },
    {
      "entropy": 5.715953588485718,
      "epoch": 1.2518142235123366,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00048492661831186243,
      "loss": 5.5423,
      "mean_token_accuracy": 0.16676028966903686,
      "num_tokens": 28624249.0,
      "step": 13800
    },
    {
      "entropy": 5.803564977645874,
      "epoch": 1.2522677793904209,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004849149724337715,
      "loss": 5.5312,
      "mean_token_accuracy": 0.1625569760799408,
      "num_tokens": 28634391.0,
      "step": 13805
    },
    {
      "entropy": 5.754829406738281,
      "epoch": 1.2527213352685052,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048490332221458543,
      "loss": 5.4883,
      "mean_token_accuracy": 0.16254082769155503,
      "num_tokens": 28645855.0,
      "step": 13810
    },
    {
      "entropy": 5.754119491577148,
      "epoch": 1.2531748911465892,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004848916676545452,
      "loss": 5.5677,
      "mean_token_accuracy": 0.1539563149213791,
      "num_tokens": 28656804.0,
      "step": 13815
    },
    {
      "entropy": 5.775495719909668,
      "epoch": 1.2536284470246735,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048488000875389174,
      "loss": 5.5337,
      "mean_token_accuracy": 0.15982940942049026,
      "num_tokens": 28667971.0,
      "step": 13820
    },
    {
      "entropy": 5.7915550708770756,
      "epoch": 1.2540820029027575,
      "grad_norm": 4.4375,
      "learning_rate": 0.0004848683455128663,
      "loss": 5.4951,
      "mean_token_accuracy": 0.16234183311462402,
      "num_tokens": 28679584.0,
      "step": 13825
    },
    {
      "entropy": 5.6818804264068605,
      "epoch": 1.2545355587808418,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048485667793170994,
      "loss": 5.4827,
      "mean_token_accuracy": 0.1631179854273796,
      "num_tokens": 28689056.0,
      "step": 13830
    },
    {
      "entropy": 5.684180068969726,
      "epoch": 1.254989114658926,
      "grad_norm": 3.234375,
      "learning_rate": 0.000484845006010664,
      "loss": 5.479,
      "mean_token_accuracy": 0.1726303815841675,
      "num_tokens": 28698986.0,
      "step": 13835
    },
    {
      "entropy": 5.722275876998902,
      "epoch": 1.25544267053701,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004848333297499699,
      "loss": 5.5025,
      "mean_token_accuracy": 0.16858190596103667,
      "num_tokens": 28709533.0,
      "step": 13840
    },
    {
      "entropy": 5.829799747467041,
      "epoch": 1.2558962264150944,
      "grad_norm": 3.203125,
      "learning_rate": 0.00048482164914986903,
      "loss": 5.5514,
      "mean_token_accuracy": 0.16070029139518738,
      "num_tokens": 28719836.0,
      "step": 13845
    },
    {
      "entropy": 5.778800821304321,
      "epoch": 1.2563497822931784,
      "grad_norm": 5.5,
      "learning_rate": 0.00048480996421060307,
      "loss": 5.5516,
      "mean_token_accuracy": 0.15735034942626952,
      "num_tokens": 28731481.0,
      "step": 13850
    },
    {
      "entropy": 5.756581783294678,
      "epoch": 1.2568033381712627,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048479827493241347,
      "loss": 5.4427,
      "mean_token_accuracy": 0.16463713198900223,
      "num_tokens": 28741037.0,
      "step": 13855
    },
    {
      "entropy": 5.680491542816162,
      "epoch": 1.257256894049347,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048478658131554216,
      "loss": 5.551,
      "mean_token_accuracy": 0.16687646508216858,
      "num_tokens": 28750816.0,
      "step": 13860
    },
    {
      "entropy": 5.683056735992432,
      "epoch": 1.2577104499274312,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004847748833602308,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16662055850028992,
      "num_tokens": 28760354.0,
      "step": 13865
    },
    {
      "entropy": 5.712932777404785,
      "epoch": 1.2581640058055152,
      "grad_norm": 6.21875,
      "learning_rate": 0.0004847631810667215,
      "loss": 5.4763,
      "mean_token_accuracy": 0.16387179344892502,
      "num_tokens": 28770444.0,
      "step": 13870
    },
    {
      "entropy": 5.761759185791016,
      "epoch": 1.2586175616835993,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004847514744352561,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1623402789235115,
      "num_tokens": 28781195.0,
      "step": 13875
    },
    {
      "entropy": 5.684118556976318,
      "epoch": 1.2590711175616836,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004847397634660768,
      "loss": 5.4538,
      "mean_token_accuracy": 0.16507753729820251,
      "num_tokens": 28791506.0,
      "step": 13880
    },
    {
      "entropy": 5.6993873596191404,
      "epoch": 1.2595246734397678,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048472804815942575,
      "loss": 5.4845,
      "mean_token_accuracy": 0.1625525549054146,
      "num_tokens": 28801945.0,
      "step": 13885
    },
    {
      "entropy": 5.782627725601197,
      "epoch": 1.259978229317852,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048471632851554517,
      "loss": 5.4648,
      "mean_token_accuracy": 0.16511930376291276,
      "num_tokens": 28811286.0,
      "step": 13890
    },
    {
      "entropy": 5.757046556472778,
      "epoch": 1.2604317851959361,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004847046045346775,
      "loss": 5.5927,
      "mean_token_accuracy": 0.1556539848446846,
      "num_tokens": 28821765.0,
      "step": 13895
    },
    {
      "entropy": 5.694489908218384,
      "epoch": 1.2608853410740204,
      "grad_norm": 2.171875,
      "learning_rate": 0.00048469287621706516,
      "loss": 5.4603,
      "mean_token_accuracy": 0.16464832574129104,
      "num_tokens": 28832689.0,
      "step": 13900
    },
    {
      "entropy": 5.6573761940002445,
      "epoch": 1.2613388969521044,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004846811435629507,
      "loss": 5.3944,
      "mean_token_accuracy": 0.17215001583099365,
      "num_tokens": 28842345.0,
      "step": 13905
    },
    {
      "entropy": 5.700132894515991,
      "epoch": 1.2617924528301887,
      "grad_norm": 2.25,
      "learning_rate": 0.0004846694065725768,
      "loss": 5.426,
      "mean_token_accuracy": 0.16272615343332292,
      "num_tokens": 28852720.0,
      "step": 13910
    },
    {
      "entropy": 5.656035566329956,
      "epoch": 1.262246008708273,
      "grad_norm": 3.375,
      "learning_rate": 0.0004846576652461861,
      "loss": 5.4578,
      "mean_token_accuracy": 0.16809033900499343,
      "num_tokens": 28862609.0,
      "step": 13915
    },
    {
      "entropy": 5.694993209838867,
      "epoch": 1.262699564586357,
      "grad_norm": 2.375,
      "learning_rate": 0.00048464591958402147,
      "loss": 5.5202,
      "mean_token_accuracy": 0.16439792662858962,
      "num_tokens": 28874179.0,
      "step": 13920
    },
    {
      "entropy": 5.715208196640015,
      "epoch": 1.2631531204644413,
      "grad_norm": 2.75,
      "learning_rate": 0.00048463416958632573,
      "loss": 5.3536,
      "mean_token_accuracy": 0.168010613322258,
      "num_tokens": 28884512.0,
      "step": 13925
    },
    {
      "entropy": 5.6756147861480715,
      "epoch": 1.2636066763425253,
      "grad_norm": 3.21875,
      "learning_rate": 0.00048462241525334206,
      "loss": 5.4274,
      "mean_token_accuracy": 0.16767284721136094,
      "num_tokens": 28895062.0,
      "step": 13930
    },
    {
      "entropy": 5.720499658584595,
      "epoch": 1.2640602322206096,
      "grad_norm": 5.0,
      "learning_rate": 0.0004846106565853133,
      "loss": 5.5531,
      "mean_token_accuracy": 0.15743291825056077,
      "num_tokens": 28905723.0,
      "step": 13935
    },
    {
      "entropy": 5.768528938293457,
      "epoch": 1.2645137880986939,
      "grad_norm": 3.078125,
      "learning_rate": 0.00048459889358248284,
      "loss": 5.5177,
      "mean_token_accuracy": 0.163484562933445,
      "num_tokens": 28916159.0,
      "step": 13940
    },
    {
      "entropy": 5.79841570854187,
      "epoch": 1.264967343976778,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048458712624509383,
      "loss": 5.5154,
      "mean_token_accuracy": 0.1650406002998352,
      "num_tokens": 28926421.0,
      "step": 13945
    },
    {
      "entropy": 5.657514762878418,
      "epoch": 1.2654208998548622,
      "grad_norm": 2.625,
      "learning_rate": 0.00048457535457338964,
      "loss": 5.4703,
      "mean_token_accuracy": 0.17192776501178741,
      "num_tokens": 28935805.0,
      "step": 13950
    },
    {
      "entropy": 5.758550691604614,
      "epoch": 1.2658744557329462,
      "grad_norm": 1.875,
      "learning_rate": 0.0004845635785676137,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1637763574719429,
      "num_tokens": 28946708.0,
      "step": 13955
    },
    {
      "entropy": 5.69978928565979,
      "epoch": 1.2663280116110305,
      "grad_norm": 2.5625,
      "learning_rate": 0.00048455179822800957,
      "loss": 5.3758,
      "mean_token_accuracy": 0.16850202530622482,
      "num_tokens": 28956667.0,
      "step": 13960
    },
    {
      "entropy": 5.765382575988769,
      "epoch": 1.2667815674891147,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004845400135548208,
      "loss": 5.5635,
      "mean_token_accuracy": 0.156741002202034,
      "num_tokens": 28966941.0,
      "step": 13965
    },
    {
      "entropy": 5.6996674060821535,
      "epoch": 1.2672351233671988,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004845282245482912,
      "loss": 5.4919,
      "mean_token_accuracy": 0.16281982362270356,
      "num_tokens": 28977192.0,
      "step": 13970
    },
    {
      "entropy": 5.699718189239502,
      "epoch": 1.267688679245283,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004845164312086646,
      "loss": 5.3879,
      "mean_token_accuracy": 0.17075175791978836,
      "num_tokens": 28986965.0,
      "step": 13975
    },
    {
      "entropy": 5.752448558807373,
      "epoch": 1.268142235123367,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004845046335361847,
      "loss": 5.5561,
      "mean_token_accuracy": 0.1561591550707817,
      "num_tokens": 28996648.0,
      "step": 13980
    },
    {
      "entropy": 5.737544584274292,
      "epoch": 1.2685957910014514,
      "grad_norm": 1.84375,
      "learning_rate": 0.00048449283153109566,
      "loss": 5.564,
      "mean_token_accuracy": 0.15724655091762543,
      "num_tokens": 29008221.0,
      "step": 13985
    },
    {
      "entropy": 5.632817268371582,
      "epoch": 1.2690493468795356,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004844810251936413,
      "loss": 5.4437,
      "mean_token_accuracy": 0.1635597601532936,
      "num_tokens": 29019026.0,
      "step": 13990
    },
    {
      "entropy": 5.757634878158569,
      "epoch": 1.2695029027576197,
      "grad_norm": 3.703125,
      "learning_rate": 0.0004844692145240661,
      "loss": 5.5266,
      "mean_token_accuracy": 0.1603964701294899,
      "num_tokens": 29030305.0,
      "step": 13995
    },
    {
      "entropy": 5.81465482711792,
      "epoch": 1.269956458635704,
      "grad_norm": 4.78125,
      "learning_rate": 0.0004844573995226142,
      "loss": 5.5973,
      "mean_token_accuracy": 0.15718501359224318,
      "num_tokens": 29040765.0,
      "step": 14000
    },
    {
      "entropy": 5.7442961692810055,
      "epoch": 1.270410014513788,
      "grad_norm": 3.140625,
      "learning_rate": 0.00048444558018952984,
      "loss": 5.5063,
      "mean_token_accuracy": 0.17170140743255616,
      "num_tokens": 29052028.0,
      "step": 14005
    },
    {
      "entropy": 5.761911249160766,
      "epoch": 1.2708635703918723,
      "grad_norm": 4.28125,
      "learning_rate": 0.00048443375652505754,
      "loss": 5.4738,
      "mean_token_accuracy": 0.16050581634044647,
      "num_tokens": 29062546.0,
      "step": 14010
    },
    {
      "entropy": 5.7274175643920895,
      "epoch": 1.2713171262699565,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004844219285294418,
      "loss": 5.5428,
      "mean_token_accuracy": 0.16050905734300613,
      "num_tokens": 29072694.0,
      "step": 14015
    },
    {
      "entropy": 5.692515325546265,
      "epoch": 1.2717706821480406,
      "grad_norm": 7.65625,
      "learning_rate": 0.00048441009620292714,
      "loss": 5.4554,
      "mean_token_accuracy": 0.16051364690065384,
      "num_tokens": 29082387.0,
      "step": 14020
    },
    {
      "entropy": 5.6992795944213865,
      "epoch": 1.2722242380261248,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004843982595457584,
      "loss": 5.4644,
      "mean_token_accuracy": 0.17075041234493255,
      "num_tokens": 29092798.0,
      "step": 14025
    },
    {
      "entropy": 5.625979614257813,
      "epoch": 1.2726777939042089,
      "grad_norm": 1.9375,
      "learning_rate": 0.00048438641855818026,
      "loss": 5.4498,
      "mean_token_accuracy": 0.1683127611875534,
      "num_tokens": 29103076.0,
      "step": 14030
    },
    {
      "entropy": 5.766232967376709,
      "epoch": 1.2731313497822931,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004843745732404376,
      "loss": 5.5148,
      "mean_token_accuracy": 0.16077880710363388,
      "num_tokens": 29113805.0,
      "step": 14035
    },
    {
      "entropy": 5.714926385879517,
      "epoch": 1.2735849056603774,
      "grad_norm": 5.28125,
      "learning_rate": 0.0004843627235927755,
      "loss": 5.4654,
      "mean_token_accuracy": 0.1656664490699768,
      "num_tokens": 29124851.0,
      "step": 14040
    },
    {
      "entropy": 5.6357948780059814,
      "epoch": 1.2740384615384617,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004843508696154388,
      "loss": 5.3924,
      "mean_token_accuracy": 0.16911513805389405,
      "num_tokens": 29134896.0,
      "step": 14045
    },
    {
      "entropy": 5.737551879882813,
      "epoch": 1.2744920174165457,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048433901130867283,
      "loss": 5.5087,
      "mean_token_accuracy": 0.1708175614476204,
      "num_tokens": 29145381.0,
      "step": 14050
    },
    {
      "entropy": 5.696426820755005,
      "epoch": 1.2749455732946298,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004843271486727228,
      "loss": 5.48,
      "mean_token_accuracy": 0.16629390120506288,
      "num_tokens": 29155569.0,
      "step": 14055
    },
    {
      "entropy": 5.677827787399292,
      "epoch": 1.275399129172714,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048431528170783395,
      "loss": 5.4094,
      "mean_token_accuracy": 0.171189184486866,
      "num_tokens": 29166249.0,
      "step": 14060
    },
    {
      "entropy": 5.7350945472717285,
      "epoch": 1.2758526850507983,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004843034104142517,
      "loss": 5.5567,
      "mean_token_accuracy": 0.15646651685237883,
      "num_tokens": 29176789.0,
      "step": 14065
    },
    {
      "entropy": 5.730005931854248,
      "epoch": 1.2763062409288826,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004842915347922216,
      "loss": 5.44,
      "mean_token_accuracy": 0.17230269312858582,
      "num_tokens": 29187686.0,
      "step": 14070
    },
    {
      "entropy": 5.768440914154053,
      "epoch": 1.2767597968069666,
      "grad_norm": 2.375,
      "learning_rate": 0.0004842796548419893,
      "loss": 5.5388,
      "mean_token_accuracy": 0.1594705268740654,
      "num_tokens": 29198392.0,
      "step": 14075
    },
    {
      "entropy": 5.781287670135498,
      "epoch": 1.2772133526850509,
      "grad_norm": 4.625,
      "learning_rate": 0.00048426777056380035,
      "loss": 5.5392,
      "mean_token_accuracy": 0.1647916778922081,
      "num_tokens": 29210280.0,
      "step": 14080
    },
    {
      "entropy": 5.68105149269104,
      "epoch": 1.277666908563135,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004842558819579006,
      "loss": 5.4701,
      "mean_token_accuracy": 0.16973872780799865,
      "num_tokens": 29220440.0,
      "step": 14085
    },
    {
      "entropy": 5.677794075012207,
      "epoch": 1.2781204644412192,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004842439890245358,
      "loss": 5.4787,
      "mean_token_accuracy": 0.16356311440467836,
      "num_tokens": 29231810.0,
      "step": 14090
    },
    {
      "entropy": 5.710329914093018,
      "epoch": 1.2785740203193035,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048423209176395204,
      "loss": 5.4858,
      "mean_token_accuracy": 0.15729893147945403,
      "num_tokens": 29241842.0,
      "step": 14095
    },
    {
      "entropy": 5.696225833892822,
      "epoch": 1.2790275761973875,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004842201901763954,
      "loss": 5.4308,
      "mean_token_accuracy": 0.16821990460157393,
      "num_tokens": 29251926.0,
      "step": 14100
    },
    {
      "entropy": 5.681302070617676,
      "epoch": 1.2794811320754718,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004842082842621118,
      "loss": 5.4348,
      "mean_token_accuracy": 0.1750992238521576,
      "num_tokens": 29261840.0,
      "step": 14105
    },
    {
      "entropy": 5.647224998474121,
      "epoch": 1.2799346879535558,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004841963740213476,
      "loss": 5.4437,
      "mean_token_accuracy": 0.16671071350574493,
      "num_tokens": 29271559.0,
      "step": 14110
    },
    {
      "entropy": 5.706625699996948,
      "epoch": 1.28038824383164,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004841844594543491,
      "loss": 5.4025,
      "mean_token_accuracy": 0.1679767906665802,
      "num_tokens": 29281818.0,
      "step": 14115
    },
    {
      "entropy": 5.696302366256714,
      "epoch": 1.2808417997097243,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004841725405613626,
      "loss": 5.3855,
      "mean_token_accuracy": 0.16827781945466996,
      "num_tokens": 29291157.0,
      "step": 14120
    },
    {
      "entropy": 5.682747840881348,
      "epoch": 1.2812953555878084,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004841606173426347,
      "loss": 5.6125,
      "mean_token_accuracy": 0.15545516163110734,
      "num_tokens": 29301749.0,
      "step": 14125
    },
    {
      "entropy": 5.73097448348999,
      "epoch": 1.2817489114658926,
      "grad_norm": 2.125,
      "learning_rate": 0.00048414868979841197,
      "loss": 5.4975,
      "mean_token_accuracy": 0.16907256245613098,
      "num_tokens": 29312042.0,
      "step": 14130
    },
    {
      "entropy": 5.812651538848877,
      "epoch": 1.2822024673439767,
      "grad_norm": 2.125,
      "learning_rate": 0.00048413675792894105,
      "loss": 5.5224,
      "mean_token_accuracy": 0.1609726220369339,
      "num_tokens": 29323906.0,
      "step": 14135
    },
    {
      "entropy": 5.665666294097901,
      "epoch": 1.282656023222061,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004841248217344686,
      "loss": 5.4629,
      "mean_token_accuracy": 0.16233468055725098,
      "num_tokens": 29334508.0,
      "step": 14140
    },
    {
      "entropy": 5.7009509086608885,
      "epoch": 1.2831095791001452,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004841128812152415,
      "loss": 5.5463,
      "mean_token_accuracy": 0.1583101660013199,
      "num_tokens": 29345363.0,
      "step": 14145
    },
    {
      "entropy": 5.788045263290405,
      "epoch": 1.2835631349782293,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004841009363715068,
      "loss": 5.5182,
      "mean_token_accuracy": 0.15558721572160722,
      "num_tokens": 29355493.0,
      "step": 14150
    },
    {
      "entropy": 5.668447017669678,
      "epoch": 1.2840166908563135,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004840889872035115,
      "loss": 5.4221,
      "mean_token_accuracy": 0.1688051328063011,
      "num_tokens": 29365518.0,
      "step": 14155
    },
    {
      "entropy": 5.668225526809692,
      "epoch": 1.2844702467343976,
      "grad_norm": 3.375,
      "learning_rate": 0.0004840770337115026,
      "loss": 5.4746,
      "mean_token_accuracy": 0.16167976409196855,
      "num_tokens": 29376022.0,
      "step": 14160
    },
    {
      "entropy": 5.729173564910889,
      "epoch": 1.2849238026124818,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048406507589572747,
      "loss": 5.4838,
      "mean_token_accuracy": 0.1656462863087654,
      "num_tokens": 29385696.0,
      "step": 14165
    },
    {
      "entropy": 5.743153619766235,
      "epoch": 1.2853773584905661,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004840531137564331,
      "loss": 5.4933,
      "mean_token_accuracy": 0.1646637126803398,
      "num_tokens": 29396533.0,
      "step": 14170
    },
    {
      "entropy": 5.788575744628906,
      "epoch": 1.2858309143686502,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004840411472938672,
      "loss": 5.5519,
      "mean_token_accuracy": 0.16011734306812286,
      "num_tokens": 29408204.0,
      "step": 14175
    },
    {
      "entropy": 5.655870246887207,
      "epoch": 1.2862844702467344,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004840291765082771,
      "loss": 5.4338,
      "mean_token_accuracy": 0.16536842286586761,
      "num_tokens": 29418563.0,
      "step": 14180
    },
    {
      "entropy": 5.7787682056427006,
      "epoch": 1.2867380261248185,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048401720139991036,
      "loss": 5.529,
      "mean_token_accuracy": 0.1596818298101425,
      "num_tokens": 29428874.0,
      "step": 14185
    },
    {
      "entropy": 5.771486759185791,
      "epoch": 1.2871915820029027,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004840052219690146,
      "loss": 5.5162,
      "mean_token_accuracy": 0.16028618216514587,
      "num_tokens": 29438440.0,
      "step": 14190
    },
    {
      "entropy": 5.636241102218628,
      "epoch": 1.287645137880987,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004839932382158376,
      "loss": 5.4249,
      "mean_token_accuracy": 0.1655709907412529,
      "num_tokens": 29448454.0,
      "step": 14195
    },
    {
      "entropy": 5.640407419204712,
      "epoch": 1.288098693759071,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004839812501406272,
      "loss": 5.5046,
      "mean_token_accuracy": 0.1649011805653572,
      "num_tokens": 29458552.0,
      "step": 14200
    },
    {
      "entropy": 5.730694103240967,
      "epoch": 1.2885522496371553,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004839692577436313,
      "loss": 5.4269,
      "mean_token_accuracy": 0.16814237982034683,
      "num_tokens": 29468548.0,
      "step": 14205
    },
    {
      "entropy": 5.666298151016235,
      "epoch": 1.2890058055152394,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004839572610250978,
      "loss": 5.3741,
      "mean_token_accuracy": 0.1724609389901161,
      "num_tokens": 29480058.0,
      "step": 14210
    },
    {
      "entropy": 5.741631555557251,
      "epoch": 1.2894593613933236,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048394525998527496,
      "loss": 5.5439,
      "mean_token_accuracy": 0.16496948897838593,
      "num_tokens": 29489946.0,
      "step": 14215
    },
    {
      "entropy": 5.683090353012085,
      "epoch": 1.2899129172714079,
      "grad_norm": 9.125,
      "learning_rate": 0.0004839332546244109,
      "loss": 5.4522,
      "mean_token_accuracy": 0.1704390048980713,
      "num_tokens": 29499994.0,
      "step": 14220
    },
    {
      "entropy": 5.739141225814819,
      "epoch": 1.2903664731494922,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00048392124494275385,
      "loss": 5.5007,
      "mean_token_accuracy": 0.16371714919805527,
      "num_tokens": 29509823.0,
      "step": 14225
    },
    {
      "entropy": 5.671052932739258,
      "epoch": 1.2908200290275762,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004839092309405523,
      "loss": 5.4256,
      "mean_token_accuracy": 0.16412421464920043,
      "num_tokens": 29519211.0,
      "step": 14230
    },
    {
      "entropy": 5.719418478012085,
      "epoch": 1.2912735849056602,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048389721261805457,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16387570649385452,
      "num_tokens": 29529593.0,
      "step": 14235
    },
    {
      "entropy": 5.715094423294067,
      "epoch": 1.2917271407837445,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048388518997550926,
      "loss": 5.4868,
      "mean_token_accuracy": 0.15572591125965118,
      "num_tokens": 29540377.0,
      "step": 14240
    },
    {
      "entropy": 5.656461763381958,
      "epoch": 1.2921806966618288,
      "grad_norm": 3.828125,
      "learning_rate": 0.00048387316301316493,
      "loss": 5.4701,
      "mean_token_accuracy": 0.17048799395561218,
      "num_tokens": 29550500.0,
      "step": 14245
    },
    {
      "entropy": 5.702421855926514,
      "epoch": 1.292634252539913,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004838611317312705,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17157652080059052,
      "num_tokens": 29561043.0,
      "step": 14250
    },
    {
      "entropy": 5.770909357070923,
      "epoch": 1.293087808417997,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004838490961300745,
      "loss": 5.5818,
      "mean_token_accuracy": 0.15611080527305604,
      "num_tokens": 29572010.0,
      "step": 14255
    },
    {
      "entropy": 5.764454889297485,
      "epoch": 1.2935413642960814,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048383705620982606,
      "loss": 5.5086,
      "mean_token_accuracy": 0.16535808891057968,
      "num_tokens": 29582636.0,
      "step": 14260
    },
    {
      "entropy": 5.685814428329468,
      "epoch": 1.2939949201741654,
      "grad_norm": 6.40625,
      "learning_rate": 0.00048382501197077405,
      "loss": 5.4885,
      "mean_token_accuracy": 0.16161974221467973,
      "num_tokens": 29592563.0,
      "step": 14265
    },
    {
      "entropy": 5.7629331111907955,
      "epoch": 1.2944484760522497,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004838129634131675,
      "loss": 5.5094,
      "mean_token_accuracy": 0.16433310210704805,
      "num_tokens": 29603490.0,
      "step": 14270
    },
    {
      "entropy": 5.741161775588989,
      "epoch": 1.294902031930334,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004838009105372558,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1626087784767151,
      "num_tokens": 29613700.0,
      "step": 14275
    },
    {
      "entropy": 5.695987558364868,
      "epoch": 1.295355587808418,
      "grad_norm": 2.46875,
      "learning_rate": 0.00048378885334328793,
      "loss": 5.4049,
      "mean_token_accuracy": 0.17186198085546495,
      "num_tokens": 29624040.0,
      "step": 14280
    },
    {
      "entropy": 5.677083539962768,
      "epoch": 1.2958091436865022,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004837767918315135,
      "loss": 5.4362,
      "mean_token_accuracy": 0.16934436857700347,
      "num_tokens": 29634264.0,
      "step": 14285
    },
    {
      "entropy": 5.68390121459961,
      "epoch": 1.2962626995645863,
      "grad_norm": 1.984375,
      "learning_rate": 0.00048376472600218174,
      "loss": 5.45,
      "mean_token_accuracy": 0.16321310251951218,
      "num_tokens": 29644990.0,
      "step": 14290
    },
    {
      "entropy": 5.7680949687957765,
      "epoch": 1.2967162554426706,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004837526558555422,
      "loss": 5.5987,
      "mean_token_accuracy": 0.1635652780532837,
      "num_tokens": 29656988.0,
      "step": 14295
    },
    {
      "entropy": 5.731340932846069,
      "epoch": 1.2971698113207548,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004837405813918446,
      "loss": 5.473,
      "mean_token_accuracy": 0.16106108129024505,
      "num_tokens": 29667477.0,
      "step": 14300
    },
    {
      "entropy": 5.706531190872193,
      "epoch": 1.2976233671988389,
      "grad_norm": 2.234375,
      "learning_rate": 0.00048372850261133856,
      "loss": 5.4277,
      "mean_token_accuracy": 0.16967263370752333,
      "num_tokens": 29677588.0,
      "step": 14305
    },
    {
      "entropy": 5.613648700714111,
      "epoch": 1.2980769230769231,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048371641951427393,
      "loss": 5.4076,
      "mean_token_accuracy": 0.17583250552415847,
      "num_tokens": 29687717.0,
      "step": 14310
    },
    {
      "entropy": 5.729506731033325,
      "epoch": 1.2985304789550072,
      "grad_norm": 2.140625,
      "learning_rate": 0.00048370433210090055,
      "loss": 5.527,
      "mean_token_accuracy": 0.16463506668806077,
      "num_tokens": 29698247.0,
      "step": 14315
    },
    {
      "entropy": 5.77975697517395,
      "epoch": 1.2989840348330914,
      "grad_norm": 2.5,
      "learning_rate": 0.0004836922403714683,
      "loss": 5.5005,
      "mean_token_accuracy": 0.16675624251365662,
      "num_tokens": 29709105.0,
      "step": 14320
    },
    {
      "entropy": 5.725807523727417,
      "epoch": 1.2994375907111757,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048368014432622746,
      "loss": 5.4661,
      "mean_token_accuracy": 0.16289403885602952,
      "num_tokens": 29718848.0,
      "step": 14325
    },
    {
      "entropy": 5.687685155868531,
      "epoch": 1.2998911465892597,
      "grad_norm": 1.9765625,
      "learning_rate": 0.000483668043965428,
      "loss": 5.3635,
      "mean_token_accuracy": 0.17059047371149064,
      "num_tokens": 29728653.0,
      "step": 14330
    },
    {
      "entropy": 5.77165379524231,
      "epoch": 1.300344702467344,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004836559392893203,
      "loss": 5.6057,
      "mean_token_accuracy": 0.16050591915845872,
      "num_tokens": 29739503.0,
      "step": 14335
    },
    {
      "entropy": 5.750324249267578,
      "epoch": 1.300798258345428,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004836438302981545,
      "loss": 5.4269,
      "mean_token_accuracy": 0.1676155596971512,
      "num_tokens": 29749344.0,
      "step": 14340
    },
    {
      "entropy": 5.766082096099853,
      "epoch": 1.3012518142235123,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004836317169921811,
      "loss": 5.5282,
      "mean_token_accuracy": 0.15744034349918365,
      "num_tokens": 29760042.0,
      "step": 14345
    },
    {
      "entropy": 5.696500253677368,
      "epoch": 1.3017053701015966,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048361959937165075,
      "loss": 5.4682,
      "mean_token_accuracy": 0.160247366130352,
      "num_tokens": 29771133.0,
      "step": 14350
    },
    {
      "entropy": 5.720372152328491,
      "epoch": 1.3021589259796806,
      "grad_norm": 2.875,
      "learning_rate": 0.0004836074774368138,
      "loss": 5.4891,
      "mean_token_accuracy": 0.159864205121994,
      "num_tokens": 29781320.0,
      "step": 14355
    },
    {
      "entropy": 5.780493307113647,
      "epoch": 1.302612481857765,
      "grad_norm": 3.28125,
      "learning_rate": 0.00048359535118792116,
      "loss": 5.5861,
      "mean_token_accuracy": 0.15875804871320726,
      "num_tokens": 29792661.0,
      "step": 14360
    },
    {
      "entropy": 5.6871710300445555,
      "epoch": 1.303066037735849,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004835832206252235,
      "loss": 5.3988,
      "mean_token_accuracy": 0.17436134517192842,
      "num_tokens": 29802006.0,
      "step": 14365
    },
    {
      "entropy": 5.731395101547241,
      "epoch": 1.3035195936139332,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004835710857489716,
      "loss": 5.5667,
      "mean_token_accuracy": 0.15697238892316817,
      "num_tokens": 29811941.0,
      "step": 14370
    },
    {
      "entropy": 5.807639741897583,
      "epoch": 1.3039731494920175,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004835589465594165,
      "loss": 5.5701,
      "mean_token_accuracy": 0.15942388474941255,
      "num_tokens": 29823401.0,
      "step": 14375
    },
    {
      "entropy": 5.7819427967071535,
      "epoch": 1.3044267053701015,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048354680305680926,
      "loss": 5.3894,
      "mean_token_accuracy": 0.1747710257768631,
      "num_tokens": 29833946.0,
      "step": 14380
    },
    {
      "entropy": 5.648218250274658,
      "epoch": 1.3048802612481858,
      "grad_norm": 2.9375,
      "learning_rate": 0.00048353465524140105,
      "loss": 5.395,
      "mean_token_accuracy": 0.16791066378355027,
      "num_tokens": 29844324.0,
      "step": 14385
    },
    {
      "entropy": 5.673121786117553,
      "epoch": 1.3053338171262698,
      "grad_norm": 2.125,
      "learning_rate": 0.000483522503113443,
      "loss": 5.5105,
      "mean_token_accuracy": 0.16608888655900955,
      "num_tokens": 29855949.0,
      "step": 14390
    },
    {
      "entropy": 5.7863787651062015,
      "epoch": 1.305787373004354,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048351034667318634,
      "loss": 5.4756,
      "mean_token_accuracy": 0.16756484359502793,
      "num_tokens": 29865628.0,
      "step": 14395
    },
    {
      "entropy": 5.7888243198394775,
      "epoch": 1.3062409288824384,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004834981859208827,
      "loss": 5.473,
      "mean_token_accuracy": 0.17243407815694808,
      "num_tokens": 29875329.0,
      "step": 14400
    },
    {
      "entropy": 5.553763437271118,
      "epoch": 1.3066944847605226,
      "grad_norm": 2.203125,
      "learning_rate": 0.00048348602085678346,
      "loss": 5.363,
      "mean_token_accuracy": 0.17741848677396774,
      "num_tokens": 29884871.0,
      "step": 14405
    },
    {
      "entropy": 5.648448514938354,
      "epoch": 1.3071480406386067,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048347385148114004,
      "loss": 5.4525,
      "mean_token_accuracy": 0.1690407931804657,
      "num_tokens": 29895625.0,
      "step": 14410
    },
    {
      "entropy": 5.763497686386108,
      "epoch": 1.3076015965166907,
      "grad_norm": 4.40625,
      "learning_rate": 0.00048346167779420436,
      "loss": 5.489,
      "mean_token_accuracy": 0.16140507012605668,
      "num_tokens": 29906554.0,
      "step": 14415
    },
    {
      "entropy": 5.65837459564209,
      "epoch": 1.308055152394775,
      "grad_norm": 1.78125,
      "learning_rate": 0.00048344949979622806,
      "loss": 5.4107,
      "mean_token_accuracy": 0.17000339776277543,
      "num_tokens": 29916714.0,
      "step": 14420
    },
    {
      "entropy": 5.806635093688965,
      "epoch": 1.3085087082728593,
      "grad_norm": 2.71875,
      "learning_rate": 0.00048343731748746294,
      "loss": 5.5384,
      "mean_token_accuracy": 0.1570013076066971,
      "num_tokens": 29926265.0,
      "step": 14425
    },
    {
      "entropy": 5.760943794250489,
      "epoch": 1.3089622641509435,
      "grad_norm": 3.25,
      "learning_rate": 0.000483425130868161,
      "loss": 5.4123,
      "mean_token_accuracy": 0.16816662847995759,
      "num_tokens": 29936658.0,
      "step": 14430
    },
    {
      "entropy": 5.641919374465942,
      "epoch": 1.3094158200290276,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004834129399385742,
      "loss": 5.3944,
      "mean_token_accuracy": 0.17047334760427474,
      "num_tokens": 29946593.0,
      "step": 14435
    },
    {
      "entropy": 5.655229377746582,
      "epoch": 1.3098693759071118,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004834007446989548,
      "loss": 5.488,
      "mean_token_accuracy": 0.1567129909992218,
      "num_tokens": 29955984.0,
      "step": 14440
    },
    {
      "entropy": 5.797909259796143,
      "epoch": 1.3103229317851959,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048338854514955474,
      "loss": 5.5286,
      "mean_token_accuracy": 0.15591030567884445,
      "num_tokens": 29965551.0,
      "step": 14445
    },
    {
      "entropy": 5.796445989608765,
      "epoch": 1.3107764876632801,
      "grad_norm": 2.375,
      "learning_rate": 0.00048337634129062655,
      "loss": 5.45,
      "mean_token_accuracy": 0.17060483396053314,
      "num_tokens": 29976035.0,
      "step": 14450
    },
    {
      "entropy": 5.710024499893189,
      "epoch": 1.3112300435413644,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048336413312242255,
      "loss": 5.4189,
      "mean_token_accuracy": 0.17439668327569963,
      "num_tokens": 29985420.0,
      "step": 14455
    },
    {
      "entropy": 5.644829893112183,
      "epoch": 1.3116835994194485,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048335192064519514,
      "loss": 5.4973,
      "mean_token_accuracy": 0.1659938797354698,
      "num_tokens": 29995815.0,
      "step": 14460
    },
    {
      "entropy": 5.6930211067199705,
      "epoch": 1.3121371552975327,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048333970385919695,
      "loss": 5.4798,
      "mean_token_accuracy": 0.17020788490772248,
      "num_tokens": 30004960.0,
      "step": 14465
    },
    {
      "entropy": 5.6714136600494385,
      "epoch": 1.3125907111756168,
      "grad_norm": 5.65625,
      "learning_rate": 0.00048332748276468053,
      "loss": 5.3824,
      "mean_token_accuracy": 0.1702604740858078,
      "num_tokens": 30014928.0,
      "step": 14470
    },
    {
      "entropy": 5.731176328659058,
      "epoch": 1.313044267053701,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004833152573618987,
      "loss": 5.404,
      "mean_token_accuracy": 0.16557410806417466,
      "num_tokens": 30025179.0,
      "step": 14475
    },
    {
      "entropy": 5.5935252666473385,
      "epoch": 1.3134978229317853,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048330302765110436,
      "loss": 5.3779,
      "mean_token_accuracy": 0.16939875483512878,
      "num_tokens": 30035285.0,
      "step": 14480
    },
    {
      "entropy": 5.6008336544036865,
      "epoch": 1.3139513788098693,
      "grad_norm": 3.046875,
      "learning_rate": 0.00048329079363255024,
      "loss": 5.4222,
      "mean_token_accuracy": 0.16378167867660523,
      "num_tokens": 30045297.0,
      "step": 14485
    },
    {
      "entropy": 5.762645483016968,
      "epoch": 1.3144049346879536,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048327855530648934,
      "loss": 5.5474,
      "mean_token_accuracy": 0.15336880534887315,
      "num_tokens": 30056511.0,
      "step": 14490
    },
    {
      "entropy": 5.845439195632935,
      "epoch": 1.3148584905660377,
      "grad_norm": 2.484375,
      "learning_rate": 0.000483266312673175,
      "loss": 5.5762,
      "mean_token_accuracy": 0.16314998865127564,
      "num_tokens": 30066996.0,
      "step": 14495
    },
    {
      "entropy": 5.786246824264526,
      "epoch": 1.315312046444122,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004832540657328602,
      "loss": 5.5687,
      "mean_token_accuracy": 0.16032708436250687,
      "num_tokens": 30078067.0,
      "step": 14500
    },
    {
      "entropy": 5.702739334106445,
      "epoch": 1.3157656023222062,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004832418144857982,
      "loss": 5.4358,
      "mean_token_accuracy": 0.16805187463760377,
      "num_tokens": 30088319.0,
      "step": 14505
    },
    {
      "entropy": 5.682297229766846,
      "epoch": 1.3162191582002902,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048322955893224245,
      "loss": 5.5091,
      "mean_token_accuracy": 0.16495788097381592,
      "num_tokens": 30099376.0,
      "step": 14510
    },
    {
      "entropy": 5.709005737304688,
      "epoch": 1.3166727140783745,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048321729907244626,
      "loss": 5.5147,
      "mean_token_accuracy": 0.1579695910215378,
      "num_tokens": 30109649.0,
      "step": 14515
    },
    {
      "entropy": 5.792888355255127,
      "epoch": 1.3171262699564585,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004832050349066634,
      "loss": 5.4848,
      "mean_token_accuracy": 0.16476464569568633,
      "num_tokens": 30120180.0,
      "step": 14520
    },
    {
      "entropy": 5.733684492111206,
      "epoch": 1.3175798258345428,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004831927664351473,
      "loss": 5.452,
      "mean_token_accuracy": 0.1688125640153885,
      "num_tokens": 30130127.0,
      "step": 14525
    },
    {
      "entropy": 5.668943262100219,
      "epoch": 1.318033381712627,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004831804936581518,
      "loss": 5.4154,
      "mean_token_accuracy": 0.17015041261911393,
      "num_tokens": 30140221.0,
      "step": 14530
    },
    {
      "entropy": 5.651942920684815,
      "epoch": 1.3184869375907111,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004831682165759305,
      "loss": 5.382,
      "mean_token_accuracy": 0.17183762639760972,
      "num_tokens": 30149555.0,
      "step": 14535
    },
    {
      "entropy": 5.657701301574707,
      "epoch": 1.3189404934687954,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048315593518873757,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16226291954517363,
      "num_tokens": 30161520.0,
      "step": 14540
    },
    {
      "entropy": 5.672103071212769,
      "epoch": 1.3193940493468794,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004831436494968268,
      "loss": 5.4047,
      "mean_token_accuracy": 0.16836910843849182,
      "num_tokens": 30171490.0,
      "step": 14545
    },
    {
      "entropy": 5.72885046005249,
      "epoch": 1.3198476052249637,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048313135950045234,
      "loss": 5.448,
      "mean_token_accuracy": 0.16764464676380159,
      "num_tokens": 30182093.0,
      "step": 14550
    },
    {
      "entropy": 5.681404638290405,
      "epoch": 1.320301161103048,
      "grad_norm": 3.765625,
      "learning_rate": 0.00048311906519986824,
      "loss": 5.4873,
      "mean_token_accuracy": 0.1588644191622734,
      "num_tokens": 30192427.0,
      "step": 14555
    },
    {
      "entropy": 5.784791755676269,
      "epoch": 1.320754716981132,
      "grad_norm": 3.046875,
      "learning_rate": 0.000483106766595329,
      "loss": 5.4893,
      "mean_token_accuracy": 0.16909823417663575,
      "num_tokens": 30202044.0,
      "step": 14560
    },
    {
      "entropy": 5.757778358459473,
      "epoch": 1.3212082728592163,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004830944636870887,
      "loss": 5.4726,
      "mean_token_accuracy": 0.16885027885437012,
      "num_tokens": 30212633.0,
      "step": 14565
    },
    {
      "entropy": 5.717832374572754,
      "epoch": 1.3216618287373003,
      "grad_norm": 3.625,
      "learning_rate": 0.00048308215647540183,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1692071557044983,
      "num_tokens": 30223665.0,
      "step": 14570
    },
    {
      "entropy": 5.690496206283569,
      "epoch": 1.3221153846153846,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048306984496052297,
      "loss": 5.4939,
      "mean_token_accuracy": 0.16762909144163132,
      "num_tokens": 30232840.0,
      "step": 14575
    },
    {
      "entropy": 5.745239400863648,
      "epoch": 1.3225689404934688,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048305752914270665,
      "loss": 5.4658,
      "mean_token_accuracy": 0.1637365221977234,
      "num_tokens": 30244486.0,
      "step": 14580
    },
    {
      "entropy": 5.73780574798584,
      "epoch": 1.3230224963715531,
      "grad_norm": 2.1875,
      "learning_rate": 0.00048304520902220767,
      "loss": 5.5505,
      "mean_token_accuracy": 0.16238621473312378,
      "num_tokens": 30255035.0,
      "step": 14585
    },
    {
      "entropy": 5.728442668914795,
      "epoch": 1.3234760522496372,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004830328845992808,
      "loss": 5.4778,
      "mean_token_accuracy": 0.16768175065517427,
      "num_tokens": 30265522.0,
      "step": 14590
    },
    {
      "entropy": 5.677857303619385,
      "epoch": 1.3239296081277212,
      "grad_norm": 2.546875,
      "learning_rate": 0.00048302055587418074,
      "loss": 5.4235,
      "mean_token_accuracy": 0.16768425405025483,
      "num_tokens": 30275335.0,
      "step": 14595
    },
    {
      "entropy": 5.70911226272583,
      "epoch": 1.3243831640058055,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004830082228471626,
      "loss": 5.4691,
      "mean_token_accuracy": 0.16422940492630006,
      "num_tokens": 30284930.0,
      "step": 14600
    },
    {
      "entropy": 5.6697744846344,
      "epoch": 1.3248367198838897,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048299588551848146,
      "loss": 5.4287,
      "mean_token_accuracy": 0.16553070545196533,
      "num_tokens": 30294962.0,
      "step": 14605
    },
    {
      "entropy": 5.6385057926177975,
      "epoch": 1.325290275761974,
      "grad_norm": 3.0625,
      "learning_rate": 0.00048298354388839237,
      "loss": 5.5203,
      "mean_token_accuracy": 0.17136533111333846,
      "num_tokens": 30305424.0,
      "step": 14610
    },
    {
      "entropy": 5.705431795120239,
      "epoch": 1.325743831640058,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004829711979571506,
      "loss": 5.494,
      "mean_token_accuracy": 0.16207024604082107,
      "num_tokens": 30315671.0,
      "step": 14615
    },
    {
      "entropy": 5.760168790817261,
      "epoch": 1.3261973875181423,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004829588477250115,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1571798086166382,
      "num_tokens": 30326556.0,
      "step": 14620
    },
    {
      "entropy": 5.813874626159668,
      "epoch": 1.3266509433962264,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004829464931922304,
      "loss": 5.4635,
      "mean_token_accuracy": 0.16163260638713836,
      "num_tokens": 30337131.0,
      "step": 14625
    },
    {
      "entropy": 5.686849784851074,
      "epoch": 1.3271044992743106,
      "grad_norm": 2.8125,
      "learning_rate": 0.00048293413435906285,
      "loss": 5.4526,
      "mean_token_accuracy": 0.161766941845417,
      "num_tokens": 30347554.0,
      "step": 14630
    },
    {
      "entropy": 5.695045471191406,
      "epoch": 1.327558055152395,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004829217712257644,
      "loss": 5.4266,
      "mean_token_accuracy": 0.17082488685846328,
      "num_tokens": 30357078.0,
      "step": 14635
    },
    {
      "entropy": 5.7266899108886715,
      "epoch": 1.328011611030479,
      "grad_norm": 2.078125,
      "learning_rate": 0.00048290940379259073,
      "loss": 5.5014,
      "mean_token_accuracy": 0.16477995216846467,
      "num_tokens": 30367036.0,
      "step": 14640
    },
    {
      "entropy": 5.65107831954956,
      "epoch": 1.3284651669085632,
      "grad_norm": 4.5,
      "learning_rate": 0.00048289703205979756,
      "loss": 5.419,
      "mean_token_accuracy": 0.1727801039814949,
      "num_tokens": 30377737.0,
      "step": 14645
    },
    {
      "entropy": 5.643034172058106,
      "epoch": 1.3289187227866472,
      "grad_norm": 2.15625,
      "learning_rate": 0.00048288465602764084,
      "loss": 5.4715,
      "mean_token_accuracy": 0.1646549329161644,
      "num_tokens": 30387264.0,
      "step": 14650
    },
    {
      "entropy": 5.6062156677246096,
      "epoch": 1.3293722786647315,
      "grad_norm": 3.375,
      "learning_rate": 0.00048287227569637645,
      "loss": 5.3751,
      "mean_token_accuracy": 0.17715267986059188,
      "num_tokens": 30397837.0,
      "step": 14655
    },
    {
      "entropy": 5.75836443901062,
      "epoch": 1.3298258345428158,
      "grad_norm": 3.5625,
      "learning_rate": 0.00048285989106626036,
      "loss": 5.5549,
      "mean_token_accuracy": 0.1578159064054489,
      "num_tokens": 30407655.0,
      "step": 14660
    },
    {
      "entropy": 5.81877293586731,
      "epoch": 1.3302793904208998,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004828475021375489,
      "loss": 5.475,
      "mean_token_accuracy": 0.16522783637046815,
      "num_tokens": 30418066.0,
      "step": 14665
    },
    {
      "entropy": 5.705786848068238,
      "epoch": 1.330732946298984,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004828351089104981,
      "loss": 5.3972,
      "mean_token_accuracy": 0.17123713791370393,
      "num_tokens": 30428836.0,
      "step": 14670
    },
    {
      "entropy": 5.678180313110351,
      "epoch": 1.3311865021770681,
      "grad_norm": 3.296875,
      "learning_rate": 0.00048282271138536424,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1618223026394844,
      "num_tokens": 30439444.0,
      "step": 14675
    },
    {
      "entropy": 5.690893745422363,
      "epoch": 1.3316400580551524,
      "grad_norm": 3.1875,
      "learning_rate": 0.0004828103095624037,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16059460639953613,
      "num_tokens": 30449059.0,
      "step": 14680
    },
    {
      "entropy": 5.711418914794922,
      "epoch": 1.3320936139332367,
      "grad_norm": 2.4375,
      "learning_rate": 0.000482797903441873,
      "loss": 5.3901,
      "mean_token_accuracy": 0.17458081543445586,
      "num_tokens": 30458882.0,
      "step": 14685
    },
    {
      "entropy": 5.642905759811401,
      "epoch": 1.3325471698113207,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004827854930240288,
      "loss": 5.4008,
      "mean_token_accuracy": 0.1729239821434021,
      "num_tokens": 30468622.0,
      "step": 14690
    },
    {
      "entropy": 5.6951478004455565,
      "epoch": 1.333000725689405,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004827730783091276,
      "loss": 5.5386,
      "mean_token_accuracy": 0.15336743444204332,
      "num_tokens": 30479267.0,
      "step": 14695
    },
    {
      "entropy": 5.767401170730591,
      "epoch": 1.333454281567489,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048276065929742616,
      "loss": 5.5025,
      "mean_token_accuracy": 0.16936272531747817,
      "num_tokens": 30489599.0,
      "step": 14700
    },
    {
      "entropy": 5.746648454666138,
      "epoch": 1.3339078374455733,
      "grad_norm": 2.375,
      "learning_rate": 0.0004827482359891814,
      "loss": 5.4883,
      "mean_token_accuracy": 0.16334529370069503,
      "num_tokens": 30500042.0,
      "step": 14705
    },
    {
      "entropy": 5.682076644897461,
      "epoch": 1.3343613933236576,
      "grad_norm": 3.546875,
      "learning_rate": 0.00048273580838465016,
      "loss": 5.4958,
      "mean_token_accuracy": 0.16087376326322556,
      "num_tokens": 30510162.0,
      "step": 14710
    },
    {
      "entropy": 5.680718183517456,
      "epoch": 1.3348149492017416,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004827233764840894,
      "loss": 5.4264,
      "mean_token_accuracy": 0.16821449398994445,
      "num_tokens": 30520839.0,
      "step": 14715
    },
    {
      "entropy": 5.802205896377563,
      "epoch": 1.3352685050798259,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004827109402877564,
      "loss": 5.5949,
      "mean_token_accuracy": 0.1579836219549179,
      "num_tokens": 30532378.0,
      "step": 14720
    },
    {
      "entropy": 5.809734725952149,
      "epoch": 1.33572206095791,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004826984997959082,
      "loss": 5.5345,
      "mean_token_accuracy": 0.16347507536411285,
      "num_tokens": 30542361.0,
      "step": 14725
    },
    {
      "entropy": 5.7229992866516115,
      "epoch": 1.3361756168359942,
      "grad_norm": 3.328125,
      "learning_rate": 0.000482686055008802,
      "loss": 5.3985,
      "mean_token_accuracy": 0.17336736172437667,
      "num_tokens": 30552876.0,
      "step": 14730
    },
    {
      "entropy": 5.653024768829345,
      "epoch": 1.3366291727140784,
      "grad_norm": 3.71875,
      "learning_rate": 0.00048267360592669535,
      "loss": 5.5418,
      "mean_token_accuracy": 0.16626368910074235,
      "num_tokens": 30562586.0,
      "step": 14735
    },
    {
      "entropy": 5.667896699905396,
      "epoch": 1.3370827285921625,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004826611525498456,
      "loss": 5.4271,
      "mean_token_accuracy": 0.16785882264375687,
      "num_tokens": 30572754.0,
      "step": 14740
    },
    {
      "entropy": 5.735977268218994,
      "epoch": 1.3375362844702468,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004826486948785103,
      "loss": 5.517,
      "mean_token_accuracy": 0.16424253582954407,
      "num_tokens": 30582859.0,
      "step": 14745
    },
    {
      "entropy": 5.797349166870117,
      "epoch": 1.3379898403483308,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048263623291294706,
      "loss": 5.5031,
      "mean_token_accuracy": 0.16591225415468216,
      "num_tokens": 30593821.0,
      "step": 14750
    },
    {
      "entropy": 5.69859733581543,
      "epoch": 1.338443396226415,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048262376665341355,
      "loss": 5.4565,
      "mean_token_accuracy": 0.1664735645055771,
      "num_tokens": 30604185.0,
      "step": 14755
    },
    {
      "entropy": 5.758847379684449,
      "epoch": 1.3388969521044993,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004826112961001677,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1608710303902626,
      "num_tokens": 30616177.0,
      "step": 14760
    },
    {
      "entropy": 5.887916135787964,
      "epoch": 1.3393505079825834,
      "grad_norm": 3.3125,
      "learning_rate": 0.00048259882125346735,
      "loss": 5.6141,
      "mean_token_accuracy": 0.16190952211618423,
      "num_tokens": 30626169.0,
      "step": 14765
    },
    {
      "entropy": 5.705976676940918,
      "epoch": 1.3398040638606676,
      "grad_norm": 3.5,
      "learning_rate": 0.00048258634211357045,
      "loss": 5.3463,
      "mean_token_accuracy": 0.16571845412254332,
      "num_tokens": 30635697.0,
      "step": 14770
    },
    {
      "entropy": 5.69506688117981,
      "epoch": 1.3402576197387517,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004825738586807351,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1634306475520134,
      "num_tokens": 30646689.0,
      "step": 14775
    },
    {
      "entropy": 5.644635486602783,
      "epoch": 1.340711175616836,
      "grad_norm": 2.25,
      "learning_rate": 0.00048256137095521945,
      "loss": 5.3486,
      "mean_token_accuracy": 0.17569890171289443,
      "num_tokens": 30656461.0,
      "step": 14780
    },
    {
      "entropy": 5.662866258621216,
      "epoch": 1.3411647314949202,
      "grad_norm": 4.46875,
      "learning_rate": 0.00048254887893728175,
      "loss": 5.4124,
      "mean_token_accuracy": 0.16557274907827377,
      "num_tokens": 30667023.0,
      "step": 14785
    },
    {
      "entropy": 5.715980577468872,
      "epoch": 1.3416182873730045,
      "grad_norm": 3.0,
      "learning_rate": 0.0004825363826271804,
      "loss": 5.4331,
      "mean_token_accuracy": 0.1656156837940216,
      "num_tokens": 30677984.0,
      "step": 14790
    },
    {
      "entropy": 5.715647554397583,
      "epoch": 1.3420718432510885,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004825238820251737,
      "loss": 5.3762,
      "mean_token_accuracy": 0.17237260043621064,
      "num_tokens": 30688156.0,
      "step": 14795
    },
    {
      "entropy": 5.707441234588623,
      "epoch": 1.3425253991291728,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048251137713152027,
      "loss": 5.4573,
      "mean_token_accuracy": 0.16122709810733796,
      "num_tokens": 30699498.0,
      "step": 14800
    },
    {
      "entropy": 5.734258699417114,
      "epoch": 1.3429789550072568,
      "grad_norm": 5.25,
      "learning_rate": 0.0004824988679464787,
      "loss": 5.4306,
      "mean_token_accuracy": 0.17153163403272628,
      "num_tokens": 30710204.0,
      "step": 14805
    },
    {
      "entropy": 5.72209358215332,
      "epoch": 1.343432510885341,
      "grad_norm": 2.265625,
      "learning_rate": 0.00048248635447030767,
      "loss": 5.4357,
      "mean_token_accuracy": 0.16584837436676025,
      "num_tokens": 30721006.0,
      "step": 14810
    },
    {
      "entropy": 5.706082344055176,
      "epoch": 1.3438860667634254,
      "grad_norm": 4.15625,
      "learning_rate": 0.0004824738367032659,
      "loss": 5.4534,
      "mean_token_accuracy": 0.16555455476045608,
      "num_tokens": 30730488.0,
      "step": 14815
    },
    {
      "entropy": 5.6914032936096195,
      "epoch": 1.3443396226415094,
      "grad_norm": 3.921875,
      "learning_rate": 0.0004824613146456123,
      "loss": 5.4699,
      "mean_token_accuracy": 0.1650742143392563,
      "num_tokens": 30740953.0,
      "step": 14820
    },
    {
      "entropy": 5.704112243652344,
      "epoch": 1.3447931785195937,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004824487882976059,
      "loss": 5.4858,
      "mean_token_accuracy": 0.1596449002623558,
      "num_tokens": 30751470.0,
      "step": 14825
    },
    {
      "entropy": 5.780220699310303,
      "epoch": 1.3452467343976777,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048243625765950564,
      "loss": 5.5203,
      "mean_token_accuracy": 0.17007620483636857,
      "num_tokens": 30762927.0,
      "step": 14830
    },
    {
      "entropy": 5.769157600402832,
      "epoch": 1.345700290275762,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004824237227315708,
      "loss": 5.5481,
      "mean_token_accuracy": 0.1636530190706253,
      "num_tokens": 30773361.0,
      "step": 14835
    },
    {
      "entropy": 5.7685647964477536,
      "epoch": 1.3461538461538463,
      "grad_norm": 8.5,
      "learning_rate": 0.0004824111835140604,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1691884219646454,
      "num_tokens": 30784324.0,
      "step": 14840
    },
    {
      "entropy": 5.7190944194793705,
      "epoch": 1.3466074020319303,
      "grad_norm": 3.078125,
      "learning_rate": 0.000482398640007234,
      "loss": 5.5063,
      "mean_token_accuracy": 0.16801877915859223,
      "num_tokens": 30794536.0,
      "step": 14845
    },
    {
      "entropy": 5.704134273529053,
      "epoch": 1.3470609579100146,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048238609221135076,
      "loss": 5.4167,
      "mean_token_accuracy": 0.17277679741382598,
      "num_tokens": 30804753.0,
      "step": 14850
    },
    {
      "entropy": 5.729612588882446,
      "epoch": 1.3475145137880986,
      "grad_norm": 5.46875,
      "learning_rate": 0.0004823735401266704,
      "loss": 5.4934,
      "mean_token_accuracy": 0.1637903317809105,
      "num_tokens": 30815469.0,
      "step": 14855
    },
    {
      "entropy": 5.782698011398315,
      "epoch": 1.3479680696661829,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048236098375345227,
      "loss": 5.5186,
      "mean_token_accuracy": 0.16560146063566208,
      "num_tokens": 30826339.0,
      "step": 14860
    },
    {
      "entropy": 5.713065147399902,
      "epoch": 1.3484216255442671,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004823484230919563,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1662537008523941,
      "num_tokens": 30836064.0,
      "step": 14865
    },
    {
      "entropy": 5.727951383590698,
      "epoch": 1.3488751814223512,
      "grad_norm": 2.0,
      "learning_rate": 0.00048233585814244195,
      "loss": 5.4756,
      "mean_token_accuracy": 0.1625930905342102,
      "num_tokens": 30845979.0,
      "step": 14870
    },
    {
      "entropy": 5.683716773986816,
      "epoch": 1.3493287373004355,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004823232889051693,
      "loss": 5.4252,
      "mean_token_accuracy": 0.16678502708673476,
      "num_tokens": 30855599.0,
      "step": 14875
    },
    {
      "entropy": 5.683196640014648,
      "epoch": 1.3497822931785195,
      "grad_norm": 3.328125,
      "learning_rate": 0.00048231071538039816,
      "loss": 5.456,
      "mean_token_accuracy": 0.16655931621789932,
      "num_tokens": 30866681.0,
      "step": 14880
    },
    {
      "entropy": 5.682762336730957,
      "epoch": 1.3502358490566038,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048229813756838863,
      "loss": 5.512,
      "mean_token_accuracy": 0.16454392075538635,
      "num_tokens": 30877559.0,
      "step": 14885
    },
    {
      "entropy": 5.835018444061279,
      "epoch": 1.350689404934688,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004822855554694007,
      "loss": 5.5102,
      "mean_token_accuracy": 0.16458864361047745,
      "num_tokens": 30888981.0,
      "step": 14890
    },
    {
      "entropy": 5.7138677597045895,
      "epoch": 1.351142960812772,
      "grad_norm": 3.09375,
      "learning_rate": 0.00048227296908369475,
      "loss": 5.4211,
      "mean_token_accuracy": 0.17841555625200273,
      "num_tokens": 30898920.0,
      "step": 14895
    },
    {
      "entropy": 5.649546575546265,
      "epoch": 1.3515965166908563,
      "grad_norm": 2.09375,
      "learning_rate": 0.00048226037841153104,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16606757342815398,
      "num_tokens": 30909041.0,
      "step": 14900
    },
    {
      "entropy": 5.760982370376587,
      "epoch": 1.3520500725689404,
      "grad_norm": 2.375,
      "learning_rate": 0.0004822477834531698,
      "loss": 5.5154,
      "mean_token_accuracy": 0.15795173048973082,
      "num_tokens": 30921204.0,
      "step": 14905
    },
    {
      "entropy": 5.779460191726685,
      "epoch": 1.3525036284470247,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004822351842088716,
      "loss": 5.5094,
      "mean_token_accuracy": 0.165153431892395,
      "num_tokens": 30931936.0,
      "step": 14910
    },
    {
      "entropy": 5.768240928649902,
      "epoch": 1.352957184325109,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004822225806788969,
      "loss": 5.4486,
      "mean_token_accuracy": 0.17087007015943528,
      "num_tokens": 30942312.0,
      "step": 14915
    },
    {
      "entropy": 5.624965858459473,
      "epoch": 1.353410740203193,
      "grad_norm": 2.421875,
      "learning_rate": 0.00048220997286350654,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17492065578699112,
      "num_tokens": 30953725.0,
      "step": 14920
    },
    {
      "entropy": 5.750910377502441,
      "epoch": 1.3538642960812772,
      "grad_norm": 2.75,
      "learning_rate": 0.00048219736076296114,
      "loss": 5.6468,
      "mean_token_accuracy": 0.16170495450496675,
      "num_tokens": 30965091.0,
      "step": 14925
    },
    {
      "entropy": 5.745261335372925,
      "epoch": 1.3543178519593613,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004821847443775214,
      "loss": 5.536,
      "mean_token_accuracy": 0.15648993998765945,
      "num_tokens": 30974847.0,
      "step": 14930
    },
    {
      "entropy": 5.725817489624023,
      "epoch": 1.3547714078374455,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048217212370744853,
      "loss": 5.4218,
      "mean_token_accuracy": 0.16721526235342027,
      "num_tokens": 30984424.0,
      "step": 14935
    },
    {
      "entropy": 5.739688444137573,
      "epoch": 1.3552249637155298,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048215949875300327,
      "loss": 5.576,
      "mean_token_accuracy": 0.15707599371671677,
      "num_tokens": 30995048.0,
      "step": 14940
    },
    {
      "entropy": 5.692620754241943,
      "epoch": 1.3556785195936139,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004821468695144467,
      "loss": 5.4274,
      "mean_token_accuracy": 0.16779949218034745,
      "num_tokens": 31005680.0,
      "step": 14945
    },
    {
      "entropy": 5.721690654754639,
      "epoch": 1.3561320754716981,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048213423599204015,
      "loss": 5.4893,
      "mean_token_accuracy": 0.16291181445121766,
      "num_tokens": 31016283.0,
      "step": 14950
    },
    {
      "entropy": 5.673280858993531,
      "epoch": 1.3565856313497822,
      "grad_norm": 2.96875,
      "learning_rate": 0.00048212159818604477,
      "loss": 5.4228,
      "mean_token_accuracy": 0.1703336641192436,
      "num_tokens": 31026174.0,
      "step": 14955
    },
    {
      "entropy": 5.823092746734619,
      "epoch": 1.3570391872278664,
      "grad_norm": 3.015625,
      "learning_rate": 0.000482108956096722,
      "loss": 5.5647,
      "mean_token_accuracy": 0.16324292719364167,
      "num_tokens": 31036750.0,
      "step": 14960
    },
    {
      "entropy": 5.8408506393432615,
      "epoch": 1.3574927431059507,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048209630972433326,
      "loss": 5.7321,
      "mean_token_accuracy": 0.1493436835706234,
      "num_tokens": 31048204.0,
      "step": 14965
    },
    {
      "entropy": 5.739716243743897,
      "epoch": 1.357946298984035,
      "grad_norm": 4.0625,
      "learning_rate": 0.00048208365906914,
      "loss": 5.4359,
      "mean_token_accuracy": 0.1637223556637764,
      "num_tokens": 31058183.0,
      "step": 14970
    },
    {
      "entropy": 5.695493888854981,
      "epoch": 1.358399854862119,
      "grad_norm": 2.296875,
      "learning_rate": 0.00048207100413140397,
      "loss": 5.492,
      "mean_token_accuracy": 0.16387912184000014,
      "num_tokens": 31068985.0,
      "step": 14975
    },
    {
      "entropy": 5.814902114868164,
      "epoch": 1.3588534107402033,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004820583449113868,
      "loss": 5.5777,
      "mean_token_accuracy": 0.163078211247921,
      "num_tokens": 31079401.0,
      "step": 14980
    },
    {
      "entropy": 5.666174125671387,
      "epoch": 1.3593069666182873,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048204568140935023,
      "loss": 5.3944,
      "mean_token_accuracy": 0.1700096920132637,
      "num_tokens": 31088884.0,
      "step": 14985
    },
    {
      "entropy": 5.628246259689331,
      "epoch": 1.3597605224963716,
      "grad_norm": 4.28125,
      "learning_rate": 0.0004820330136255563,
      "loss": 5.4194,
      "mean_token_accuracy": 0.17062444537878035,
      "num_tokens": 31099684.0,
      "step": 14990
    },
    {
      "entropy": 5.727261257171631,
      "epoch": 1.3602140783744558,
      "grad_norm": 2.5,
      "learning_rate": 0.00048202034156026683,
      "loss": 5.4909,
      "mean_token_accuracy": 0.1664130598306656,
      "num_tokens": 31109962.0,
      "step": 14995
    },
    {
      "entropy": 5.737498474121094,
      "epoch": 1.36066763425254,
      "grad_norm": 3.25,
      "learning_rate": 0.00048200766521374396,
      "loss": 5.4396,
      "mean_token_accuracy": 0.16466503888368605,
      "num_tokens": 31120977.0,
      "step": 15000
    },
    {
      "epoch": 1.36066763425254,
      "eval_entropy": 5.487789913539973,
      "eval_loss": 5.528749942779541,
      "eval_mean_token_accuracy": 0.17131102507284998,
      "eval_num_tokens": 31120977.0,
      "eval_runtime": 26.594,
      "eval_samples_per_second": 1329.625,
      "eval_steps_per_second": 166.203,
      "step": 15000
    },
    {
      "entropy": 5.67296199798584,
      "epoch": 1.3611211901306242,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004819949845862498,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16915711164474487,
      "num_tokens": 31130966.0,
      "step": 15005
    },
    {
      "entropy": 5.707244539260865,
      "epoch": 1.3615747460087082,
      "grad_norm": 2.0,
      "learning_rate": 0.0004819822996780466,
      "loss": 5.5198,
      "mean_token_accuracy": 0.17049406915903093,
      "num_tokens": 31141245.0,
      "step": 15010
    },
    {
      "entropy": 5.727001714706421,
      "epoch": 1.3620283018867925,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004819696104893967,
      "loss": 5.3963,
      "mean_token_accuracy": 0.17270589172840117,
      "num_tokens": 31150936.0,
      "step": 15015
    },
    {
      "entropy": 5.636113834381104,
      "epoch": 1.3624818577648767,
      "grad_norm": 2.0,
      "learning_rate": 0.0004819569170205625,
      "loss": 5.4945,
      "mean_token_accuracy": 0.16242493391036988,
      "num_tokens": 31160941.0,
      "step": 15020
    },
    {
      "entropy": 5.661560487747193,
      "epoch": 1.3629354136429608,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048194421927180654,
      "loss": 5.4202,
      "mean_token_accuracy": 0.16457071304321289,
      "num_tokens": 31171045.0,
      "step": 15025
    },
    {
      "entropy": 5.748689270019531,
      "epoch": 1.363388969521045,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004819315172433914,
      "loss": 5.4618,
      "mean_token_accuracy": 0.16173867881298065,
      "num_tokens": 31182019.0,
      "step": 15030
    },
    {
      "entropy": 5.715936279296875,
      "epoch": 1.363842525399129,
      "grad_norm": 4.15625,
      "learning_rate": 0.00048191881093557977,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1603324219584465,
      "num_tokens": 31192574.0,
      "step": 15035
    },
    {
      "entropy": 5.692886304855347,
      "epoch": 1.3642960812772134,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048190610034863434,
      "loss": 5.5065,
      "mean_token_accuracy": 0.16415584832429886,
      "num_tokens": 31203057.0,
      "step": 15040
    },
    {
      "entropy": 5.718279027938843,
      "epoch": 1.3647496371552976,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004818933854828181,
      "loss": 5.4582,
      "mean_token_accuracy": 0.16424989551305771,
      "num_tokens": 31215205.0,
      "step": 15045
    },
    {
      "entropy": 5.71901388168335,
      "epoch": 1.3652031930333817,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004818806663383939,
      "loss": 5.5522,
      "mean_token_accuracy": 0.16161487400531768,
      "num_tokens": 31225393.0,
      "step": 15050
    },
    {
      "entropy": 5.713966941833496,
      "epoch": 1.365656748911466,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004818679429156248,
      "loss": 5.5081,
      "mean_token_accuracy": 0.162591415643692,
      "num_tokens": 31237005.0,
      "step": 15055
    },
    {
      "entropy": 5.727983903884888,
      "epoch": 1.36611030478955,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048185521521477385,
      "loss": 5.4717,
      "mean_token_accuracy": 0.16932910978794097,
      "num_tokens": 31247635.0,
      "step": 15060
    },
    {
      "entropy": 5.6719073295593265,
      "epoch": 1.3665638606676342,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004818424832361044,
      "loss": 5.3805,
      "mean_token_accuracy": 0.16364826261997223,
      "num_tokens": 31258068.0,
      "step": 15065
    },
    {
      "entropy": 5.762904644012451,
      "epoch": 1.3670174165457185,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004818297469798797,
      "loss": 5.4876,
      "mean_token_accuracy": 0.16461797803640366,
      "num_tokens": 31268944.0,
      "step": 15070
    },
    {
      "entropy": 5.739477491378784,
      "epoch": 1.3674709724238026,
      "grad_norm": 2.59375,
      "learning_rate": 0.00048181700644636314,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1595294654369354,
      "num_tokens": 31278678.0,
      "step": 15075
    },
    {
      "entropy": 5.789193964004516,
      "epoch": 1.3679245283018868,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048180426163581816,
      "loss": 5.5305,
      "mean_token_accuracy": 0.16126444339752197,
      "num_tokens": 31290853.0,
      "step": 15080
    },
    {
      "entropy": 5.7466624736785885,
      "epoch": 1.3683780841799709,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004817915125485083,
      "loss": 5.5352,
      "mean_token_accuracy": 0.1637232407927513,
      "num_tokens": 31301175.0,
      "step": 15085
    },
    {
      "entropy": 5.826459074020386,
      "epoch": 1.3688316400580551,
      "grad_norm": 2.859375,
      "learning_rate": 0.00048177875918469735,
      "loss": 5.5582,
      "mean_token_accuracy": 0.15788597315549852,
      "num_tokens": 31310847.0,
      "step": 15090
    },
    {
      "entropy": 5.721231460571289,
      "epoch": 1.3692851959361394,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048176600154464897,
      "loss": 5.4183,
      "mean_token_accuracy": 0.16985800564289094,
      "num_tokens": 31321257.0,
      "step": 15095
    },
    {
      "entropy": 5.698572301864624,
      "epoch": 1.3697387518142234,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004817532396286269,
      "loss": 5.4711,
      "mean_token_accuracy": 0.1722186580300331,
      "num_tokens": 31331267.0,
      "step": 15100
    },
    {
      "entropy": 5.652106237411499,
      "epoch": 1.3701923076923077,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004817404734368952,
      "loss": 5.4691,
      "mean_token_accuracy": 0.1635959729552269,
      "num_tokens": 31341050.0,
      "step": 15105
    },
    {
      "entropy": 5.759237289428711,
      "epoch": 1.3706458635703918,
      "grad_norm": 3.234375,
      "learning_rate": 0.00048172770296971785,
      "loss": 5.4836,
      "mean_token_accuracy": 0.16445182263851166,
      "num_tokens": 31351322.0,
      "step": 15110
    },
    {
      "entropy": 5.724334955215454,
      "epoch": 1.371099419448476,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004817149282273589,
      "loss": 5.5781,
      "mean_token_accuracy": 0.15886155813932418,
      "num_tokens": 31362266.0,
      "step": 15115
    },
    {
      "entropy": 5.714306354522705,
      "epoch": 1.3715529753265603,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004817021492100825,
      "loss": 5.3987,
      "mean_token_accuracy": 0.170139342546463,
      "num_tokens": 31372253.0,
      "step": 15120
    },
    {
      "entropy": 5.7626865863800045,
      "epoch": 1.3720065312046443,
      "grad_norm": 2.15625,
      "learning_rate": 0.000481689365918153,
      "loss": 5.4724,
      "mean_token_accuracy": 0.16081786751747132,
      "num_tokens": 31382525.0,
      "step": 15125
    },
    {
      "entropy": 5.7925652980804445,
      "epoch": 1.3724600870827286,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004816765783518348,
      "loss": 5.5291,
      "mean_token_accuracy": 0.1660178169608116,
      "num_tokens": 31393678.0,
      "step": 15130
    },
    {
      "entropy": 5.6641288757324215,
      "epoch": 1.3729136429608126,
      "grad_norm": 2.0,
      "learning_rate": 0.0004816637865113923,
      "loss": 5.4041,
      "mean_token_accuracy": 0.16606738269329072,
      "num_tokens": 31403651.0,
      "step": 15135
    },
    {
      "entropy": 5.73638391494751,
      "epoch": 1.373367198838897,
      "grad_norm": 2.734375,
      "learning_rate": 0.00048165099039708994,
      "loss": 5.5461,
      "mean_token_accuracy": 0.16382634714245797,
      "num_tokens": 31414342.0,
      "step": 15140
    },
    {
      "entropy": 5.7211267948150635,
      "epoch": 1.3738207547169812,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004816381900091925,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16622642278671265,
      "num_tokens": 31424101.0,
      "step": 15145
    },
    {
      "entropy": 5.745302009582519,
      "epoch": 1.3742743105950654,
      "grad_norm": 4.0,
      "learning_rate": 0.00048162538534796465,
      "loss": 5.426,
      "mean_token_accuracy": 0.17618341743946075,
      "num_tokens": 31434454.0,
      "step": 15150
    },
    {
      "entropy": 5.7340137481689455,
      "epoch": 1.3747278664731495,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004816125764136711,
      "loss": 5.4923,
      "mean_token_accuracy": 0.16074581891298295,
      "num_tokens": 31445600.0,
      "step": 15155
    },
    {
      "entropy": 5.728594493865967,
      "epoch": 1.3751814223512338,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004815997632065769,
      "loss": 5.5997,
      "mean_token_accuracy": 0.15952298790216446,
      "num_tokens": 31456583.0,
      "step": 15160
    },
    {
      "entropy": 5.710681962966919,
      "epoch": 1.3756349782293178,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048158694572694694,
      "loss": 5.4744,
      "mean_token_accuracy": 0.16368636041879653,
      "num_tokens": 31467323.0,
      "step": 15165
    },
    {
      "entropy": 5.808926439285278,
      "epoch": 1.376088534107402,
      "grad_norm": 5.9375,
      "learning_rate": 0.00048157412397504623,
      "loss": 5.542,
      "mean_token_accuracy": 0.1643250122666359,
      "num_tokens": 31477807.0,
      "step": 15170
    },
    {
      "entropy": 5.7280128479003904,
      "epoch": 1.3765420899854863,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004815612979511401,
      "loss": 5.4425,
      "mean_token_accuracy": 0.16859143376350402,
      "num_tokens": 31487620.0,
      "step": 15175
    },
    {
      "entropy": 5.66163740158081,
      "epoch": 1.3769956458635704,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048154846765549353,
      "loss": 5.4009,
      "mean_token_accuracy": 0.17200081199407577,
      "num_tokens": 31498109.0,
      "step": 15180
    },
    {
      "entropy": 5.567583274841309,
      "epoch": 1.3774492017416546,
      "grad_norm": 2.875,
      "learning_rate": 0.00048153563308837215,
      "loss": 5.3388,
      "mean_token_accuracy": 0.17700877487659455,
      "num_tokens": 31507437.0,
      "step": 15185
    },
    {
      "entropy": 5.738662528991699,
      "epoch": 1.3779027576197387,
      "grad_norm": 1.828125,
      "learning_rate": 0.00048152279425004124,
      "loss": 5.4843,
      "mean_token_accuracy": 0.16981325149536133,
      "num_tokens": 31517333.0,
      "step": 15190
    },
    {
      "entropy": 5.724883604049682,
      "epoch": 1.378356313497823,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004815099511407663,
      "loss": 5.3922,
      "mean_token_accuracy": 0.17464248836040497,
      "num_tokens": 31527556.0,
      "step": 15195
    },
    {
      "entropy": 5.605590343475342,
      "epoch": 1.3788098693759072,
      "grad_norm": 2.40625,
      "learning_rate": 0.000481497103760813,
      "loss": 5.3347,
      "mean_token_accuracy": 0.17365533858537674,
      "num_tokens": 31537119.0,
      "step": 15200
    },
    {
      "entropy": 5.785111808776856,
      "epoch": 1.3792634252539913,
      "grad_norm": 2.609375,
      "learning_rate": 0.00048148425211044684,
      "loss": 5.6003,
      "mean_token_accuracy": 0.15817971676588058,
      "num_tokens": 31548971.0,
      "step": 15205
    },
    {
      "entropy": 5.713584280014038,
      "epoch": 1.3797169811320755,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004814713961899338,
      "loss": 5.3811,
      "mean_token_accuracy": 0.1691209465265274,
      "num_tokens": 31558117.0,
      "step": 15210
    },
    {
      "entropy": 5.715424251556397,
      "epoch": 1.3801705370101596,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048145853599953966,
      "loss": 5.4659,
      "mean_token_accuracy": 0.1640186131000519,
      "num_tokens": 31569225.0,
      "step": 15215
    },
    {
      "entropy": 5.7092150211334225,
      "epoch": 1.3806240928882438,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048144567153953036,
      "loss": 5.4489,
      "mean_token_accuracy": 0.16510704383254052,
      "num_tokens": 31580068.0,
      "step": 15220
    },
    {
      "entropy": 5.628576278686523,
      "epoch": 1.381077648766328,
      "grad_norm": 3.90625,
      "learning_rate": 0.00048143280281017197,
      "loss": 5.433,
      "mean_token_accuracy": 0.1683439716696739,
      "num_tokens": 31589567.0,
      "step": 15225
    },
    {
      "entropy": 5.605720138549804,
      "epoch": 1.3815312046444121,
      "grad_norm": 11.8125,
      "learning_rate": 0.0004814199298117307,
      "loss": 5.3543,
      "mean_token_accuracy": 0.17329955697059632,
      "num_tokens": 31599109.0,
      "step": 15230
    },
    {
      "entropy": 5.7682146549224855,
      "epoch": 1.3819847605224964,
      "grad_norm": 4.65625,
      "learning_rate": 0.00048140705254447256,
      "loss": 5.4688,
      "mean_token_accuracy": 0.17109836190938948,
      "num_tokens": 31610406.0,
      "step": 15235
    },
    {
      "entropy": 5.713457107543945,
      "epoch": 1.3824383164005805,
      "grad_norm": 10.375,
      "learning_rate": 0.00048139417100866405,
      "loss": 5.5112,
      "mean_token_accuracy": 0.15723670944571494,
      "num_tokens": 31620405.0,
      "step": 15240
    },
    {
      "entropy": 5.7135532855987545,
      "epoch": 1.3828918722786647,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048138128520457146,
      "loss": 5.5362,
      "mean_token_accuracy": 0.16042250990867615,
      "num_tokens": 31630884.0,
      "step": 15245
    },
    {
      "entropy": 5.700714588165283,
      "epoch": 1.383345428156749,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004813683951324613,
      "loss": 5.3791,
      "mean_token_accuracy": 0.167238250374794,
      "num_tokens": 31641235.0,
      "step": 15250
    },
    {
      "entropy": 5.682051944732666,
      "epoch": 1.383798984034833,
      "grad_norm": 3.46875,
      "learning_rate": 0.00048135550079260014,
      "loss": 5.4639,
      "mean_token_accuracy": 0.17208099812269212,
      "num_tokens": 31652802.0,
      "step": 15255
    },
    {
      "entropy": 5.701805353164673,
      "epoch": 1.3842525399129173,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004813426021852547,
      "loss": 5.4191,
      "mean_token_accuracy": 0.1602771371603012,
      "num_tokens": 31663148.0,
      "step": 15260
    },
    {
      "entropy": 5.7019868850708,
      "epoch": 1.3847060957910013,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048132969931069156,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16526192128658296,
      "num_tokens": 31672749.0,
      "step": 15265
    },
    {
      "entropy": 5.707476282119751,
      "epoch": 1.3851596516690856,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004813167921691777,
      "loss": 5.4355,
      "mean_token_accuracy": 0.16019868850708008,
      "num_tokens": 31684249.0,
      "step": 15270
    },
    {
      "entropy": 5.686953735351563,
      "epoch": 1.3856132075471699,
      "grad_norm": 4.8125,
      "learning_rate": 0.00048130388076098003,
      "loss": 5.391,
      "mean_token_accuracy": 0.17330639958381652,
      "num_tokens": 31694251.0,
      "step": 15275
    },
    {
      "entropy": 5.627883052825927,
      "epoch": 1.386066763425254,
      "grad_norm": 2.3125,
      "learning_rate": 0.00048129096508636547,
      "loss": 5.378,
      "mean_token_accuracy": 0.16609676629304887,
      "num_tokens": 31705530.0,
      "step": 15280
    },
    {
      "entropy": 5.738698959350586,
      "epoch": 1.3865203193033382,
      "grad_norm": 3.125,
      "learning_rate": 0.00048127804514560126,
      "loss": 5.4383,
      "mean_token_accuracy": 0.16543743163347244,
      "num_tokens": 31716465.0,
      "step": 15285
    },
    {
      "entropy": 5.694364309310913,
      "epoch": 1.3869738751814222,
      "grad_norm": 2.390625,
      "learning_rate": 0.00048126512093895447,
      "loss": 5.4243,
      "mean_token_accuracy": 0.1684693858027458,
      "num_tokens": 31726751.0,
      "step": 15290
    },
    {
      "entropy": 5.666890382766724,
      "epoch": 1.3874274310595065,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004812521924666925,
      "loss": 5.4513,
      "mean_token_accuracy": 0.16758009940385818,
      "num_tokens": 31737175.0,
      "step": 15295
    },
    {
      "entropy": 5.66137924194336,
      "epoch": 1.3878809869375908,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004812392597290825,
      "loss": 5.4137,
      "mean_token_accuracy": 0.16238660216331482,
      "num_tokens": 31747581.0,
      "step": 15300
    },
    {
      "entropy": 5.665886259078979,
      "epoch": 1.3883345428156748,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004812263227263921,
      "loss": 5.5117,
      "mean_token_accuracy": 0.1588585317134857,
      "num_tokens": 31757827.0,
      "step": 15305
    },
    {
      "entropy": 5.698472309112549,
      "epoch": 1.388788098693759,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004812133814588888,
      "loss": 5.4386,
      "mean_token_accuracy": 0.1640047162771225,
      "num_tokens": 31767933.0,
      "step": 15310
    },
    {
      "entropy": 5.688170337677002,
      "epoch": 1.3892416545718431,
      "grad_norm": 4.5625,
      "learning_rate": 0.0004812004359268402,
      "loss": 5.4676,
      "mean_token_accuracy": 0.16569098234176635,
      "num_tokens": 31777821.0,
      "step": 15315
    },
    {
      "entropy": 5.718161916732788,
      "epoch": 1.3896952104499274,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004811874861305141,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17141211330890654,
      "num_tokens": 31788310.0,
      "step": 15320
    },
    {
      "entropy": 5.724776124954223,
      "epoch": 1.3901487663280117,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004811745320701782,
      "loss": 5.4523,
      "mean_token_accuracy": 0.17010384649038315,
      "num_tokens": 31798404.0,
      "step": 15325
    },
    {
      "entropy": 5.7703978538513185,
      "epoch": 1.390602322206096,
      "grad_norm": 3.078125,
      "learning_rate": 0.0004811615737461004,
      "loss": 5.487,
      "mean_token_accuracy": 0.1688493862748146,
      "num_tokens": 31808428.0,
      "step": 15330
    },
    {
      "entropy": 5.802201461791992,
      "epoch": 1.39105587808418,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048114861115854877,
      "loss": 5.5349,
      "mean_token_accuracy": 0.1608268901705742,
      "num_tokens": 31818603.0,
      "step": 15335
    },
    {
      "entropy": 5.752864742279053,
      "epoch": 1.3915094339622642,
      "grad_norm": 2.21875,
      "learning_rate": 0.00048113564430779136,
      "loss": 5.4273,
      "mean_token_accuracy": 0.16420445144176482,
      "num_tokens": 31828550.0,
      "step": 15340
    },
    {
      "entropy": 5.6352509498596195,
      "epoch": 1.3919629898403483,
      "grad_norm": 4.53125,
      "learning_rate": 0.0004811226731940962,
      "loss": 5.3873,
      "mean_token_accuracy": 0.17606679648160933,
      "num_tokens": 31839697.0,
      "step": 15345
    },
    {
      "entropy": 5.698238849639893,
      "epoch": 1.3924165457184325,
      "grad_norm": 3.109375,
      "learning_rate": 0.00048110969781773173,
      "loss": 5.4951,
      "mean_token_accuracy": 0.16472046971321105,
      "num_tokens": 31851172.0,
      "step": 15350
    },
    {
      "entropy": 5.732287883758545,
      "epoch": 1.3928701015965168,
      "grad_norm": 2.125,
      "learning_rate": 0.00048109671817896614,
      "loss": 5.466,
      "mean_token_accuracy": 0.1730425924062729,
      "num_tokens": 31861189.0,
      "step": 15355
    },
    {
      "entropy": 5.741789245605469,
      "epoch": 1.3933236574746009,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048108373427806797,
      "loss": 5.4375,
      "mean_token_accuracy": 0.17017186135053636,
      "num_tokens": 31871798.0,
      "step": 15360
    },
    {
      "entropy": 5.631878089904785,
      "epoch": 1.3937772133526851,
      "grad_norm": 2.40625,
      "learning_rate": 0.00048107074611530555,
      "loss": 5.3728,
      "mean_token_accuracy": 0.17749967873096467,
      "num_tokens": 31881685.0,
      "step": 15365
    },
    {
      "entropy": 5.625518703460694,
      "epoch": 1.3942307692307692,
      "grad_norm": 2.359375,
      "learning_rate": 0.00048105775369094756,
      "loss": 5.4708,
      "mean_token_accuracy": 0.16433674693107606,
      "num_tokens": 31892288.0,
      "step": 15370
    },
    {
      "entropy": 5.700084829330445,
      "epoch": 1.3946843251088534,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004810447570052628,
      "loss": 5.4107,
      "mean_token_accuracy": 0.16978809386491775,
      "num_tokens": 31902169.0,
      "step": 15375
    },
    {
      "entropy": 5.7646559715271,
      "epoch": 1.3951378809869377,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048103175605851997,
      "loss": 5.443,
      "mean_token_accuracy": 0.16675204783678055,
      "num_tokens": 31913311.0,
      "step": 15380
    },
    {
      "entropy": 5.815285968780517,
      "epoch": 1.3955914368650217,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004810187508509879,
      "loss": 5.543,
      "mean_token_accuracy": 0.16381209939718247,
      "num_tokens": 31924177.0,
      "step": 15385
    },
    {
      "entropy": 5.671485710144043,
      "epoch": 1.396044992743106,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004810057413829355,
      "loss": 5.4405,
      "mean_token_accuracy": 0.1677071049809456,
      "num_tokens": 31935033.0,
      "step": 15390
    },
    {
      "entropy": 5.71195878982544,
      "epoch": 1.39649854862119,
      "grad_norm": 2.453125,
      "learning_rate": 0.00048099272765463193,
      "loss": 5.5153,
      "mean_token_accuracy": 0.1681165650486946,
      "num_tokens": 31945196.0,
      "step": 15395
    },
    {
      "entropy": 5.665500783920288,
      "epoch": 1.3969521044992743,
      "grad_norm": 1.875,
      "learning_rate": 0.0004809797096663463,
      "loss": 5.3794,
      "mean_token_accuracy": 0.17383298575878142,
      "num_tokens": 31954891.0,
      "step": 15400
    },
    {
      "entropy": 5.744655752182007,
      "epoch": 1.3974056603773586,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004809666874183477,
      "loss": 5.549,
      "mean_token_accuracy": 0.1620172828435898,
      "num_tokens": 31964590.0,
      "step": 15405
    },
    {
      "entropy": 5.727875280380249,
      "epoch": 1.3978592162554426,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004809536609109056,
      "loss": 5.4467,
      "mean_token_accuracy": 0.16242271512746811,
      "num_tokens": 31975044.0,
      "step": 15410
    },
    {
      "entropy": 5.791713428497315,
      "epoch": 1.398312772133527,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048094063014428925,
      "loss": 5.6396,
      "mean_token_accuracy": 0.1532103896141052,
      "num_tokens": 31987250.0,
      "step": 15415
    },
    {
      "entropy": 5.668324708938599,
      "epoch": 1.398766328011611,
      "grad_norm": 2.5,
      "learning_rate": 0.0004809275951187682,
      "loss": 5.3477,
      "mean_token_accuracy": 0.1742496058344841,
      "num_tokens": 31996902.0,
      "step": 15420
    },
    {
      "entropy": 5.775240325927735,
      "epoch": 1.3992198838896952,
      "grad_norm": 2.5,
      "learning_rate": 0.00048091455583461204,
      "loss": 5.5193,
      "mean_token_accuracy": 0.16251832842826844,
      "num_tokens": 32007075.0,
      "step": 15425
    },
    {
      "entropy": 5.664343929290771,
      "epoch": 1.3996734397677795,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048090151229209036,
      "loss": 5.4042,
      "mean_token_accuracy": 0.17109157890081406,
      "num_tokens": 32017604.0,
      "step": 15430
    },
    {
      "entropy": 5.671636438369751,
      "epoch": 1.4001269956458635,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004808884644914729,
      "loss": 5.5158,
      "mean_token_accuracy": 0.16929546147584915,
      "num_tokens": 32027427.0,
      "step": 15435
    },
    {
      "entropy": 5.725512599945068,
      "epoch": 1.4005805515239478,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004808754124330296,
      "loss": 5.5094,
      "mean_token_accuracy": 0.1589631423354149,
      "num_tokens": 32039348.0,
      "step": 15440
    },
    {
      "entropy": 5.746923589706421,
      "epoch": 1.4010341074020318,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004808623561170302,
      "loss": 5.5383,
      "mean_token_accuracy": 0.17003083676099778,
      "num_tokens": 32050367.0,
      "step": 15445
    },
    {
      "entropy": 5.7523133754730225,
      "epoch": 1.401487663280116,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004808492955437449,
      "loss": 5.4645,
      "mean_token_accuracy": 0.16453548967838288,
      "num_tokens": 32059665.0,
      "step": 15450
    },
    {
      "entropy": 5.6862293720245365,
      "epoch": 1.4019412191582004,
      "grad_norm": 1.7265625,
      "learning_rate": 0.00048083623071344374,
      "loss": 5.3871,
      "mean_token_accuracy": 0.17008482664823532,
      "num_tokens": 32069583.0,
      "step": 15455
    },
    {
      "entropy": 5.760904788970947,
      "epoch": 1.4023947750362844,
      "grad_norm": 2.4375,
      "learning_rate": 0.00048082316162639685,
      "loss": 5.5894,
      "mean_token_accuracy": 0.1567358136177063,
      "num_tokens": 32080127.0,
      "step": 15460
    },
    {
      "entropy": 5.76021032333374,
      "epoch": 1.4028483309143687,
      "grad_norm": 2.125,
      "learning_rate": 0.00048081008828287447,
      "loss": 5.3752,
      "mean_token_accuracy": 0.17696642875671387,
      "num_tokens": 32089408.0,
      "step": 15465
    },
    {
      "entropy": 5.654487037658692,
      "epoch": 1.4033018867924527,
      "grad_norm": 2.765625,
      "learning_rate": 0.00048079701068314706,
      "loss": 5.4376,
      "mean_token_accuracy": 0.16965504586696625,
      "num_tokens": 32100378.0,
      "step": 15470
    },
    {
      "entropy": 5.6347733497619625,
      "epoch": 1.403755442670537,
      "grad_norm": 2.78125,
      "learning_rate": 0.00048078392882748507,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1626123696565628,
      "num_tokens": 32110988.0,
      "step": 15475
    },
    {
      "entropy": 5.714270162582397,
      "epoch": 1.4042089985486212,
      "grad_norm": 3.40625,
      "learning_rate": 0.00048077084271615895,
      "loss": 5.4275,
      "mean_token_accuracy": 0.16717500537633895,
      "num_tokens": 32121693.0,
      "step": 15480
    },
    {
      "entropy": 5.6984082698822025,
      "epoch": 1.4046625544267053,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004807577523494394,
      "loss": 5.4152,
      "mean_token_accuracy": 0.1755129173398018,
      "num_tokens": 32131036.0,
      "step": 15485
    },
    {
      "entropy": 5.7067700862884525,
      "epoch": 1.4051161103047896,
      "grad_norm": 2.5,
      "learning_rate": 0.00048074465772759707,
      "loss": 5.5365,
      "mean_token_accuracy": 0.1592101350426674,
      "num_tokens": 32141633.0,
      "step": 15490
    },
    {
      "entropy": 5.663105773925781,
      "epoch": 1.4055696661828736,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00048073155885090276,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16702721118927003,
      "num_tokens": 32151331.0,
      "step": 15495
    },
    {
      "entropy": 5.729877233505249,
      "epoch": 1.4060232220609579,
      "grad_norm": 2.53125,
      "learning_rate": 0.00048071845571962733,
      "loss": 5.5045,
      "mean_token_accuracy": 0.1671190842986107,
      "num_tokens": 32161730.0,
      "step": 15500
    },
    {
      "entropy": 5.749266147613525,
      "epoch": 1.4064767779390421,
      "grad_norm": 2.0,
      "learning_rate": 0.0004807053483340419,
      "loss": 5.5488,
      "mean_token_accuracy": 0.16743069440126418,
      "num_tokens": 32171970.0,
      "step": 15505
    },
    {
      "entropy": 5.800461673736573,
      "epoch": 1.4069303338171264,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004806922366944175,
      "loss": 5.5449,
      "mean_token_accuracy": 0.1601049154996872,
      "num_tokens": 32182453.0,
      "step": 15510
    },
    {
      "entropy": 5.787008762359619,
      "epoch": 1.4073838896952104,
      "grad_norm": 2.84375,
      "learning_rate": 0.00048067912080102513,
      "loss": 5.6106,
      "mean_token_accuracy": 0.1607410155236721,
      "num_tokens": 32194047.0,
      "step": 15515
    },
    {
      "entropy": 5.791662073135376,
      "epoch": 1.4078374455732947,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004806660006541362,
      "loss": 5.4445,
      "mean_token_accuracy": 0.16857210993766786,
      "num_tokens": 32204471.0,
      "step": 15520
    },
    {
      "entropy": 5.736038112640381,
      "epoch": 1.4082910014513788,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004806528762540219,
      "loss": 5.5004,
      "mean_token_accuracy": 0.16630353331565856,
      "num_tokens": 32213872.0,
      "step": 15525
    },
    {
      "entropy": 5.724391460418701,
      "epoch": 1.408744557329463,
      "grad_norm": 3.484375,
      "learning_rate": 0.00048063974760095374,
      "loss": 5.4194,
      "mean_token_accuracy": 0.16850799918174744,
      "num_tokens": 32223987.0,
      "step": 15530
    },
    {
      "entropy": 5.6427903175354,
      "epoch": 1.4091981132075473,
      "grad_norm": 3.984375,
      "learning_rate": 0.00048062661469520327,
      "loss": 5.3932,
      "mean_token_accuracy": 0.17059637606143951,
      "num_tokens": 32234227.0,
      "step": 15535
    },
    {
      "entropy": 5.653149032592774,
      "epoch": 1.4096516690856313,
      "grad_norm": 3.4375,
      "learning_rate": 0.00048061347753704193,
      "loss": 5.4264,
      "mean_token_accuracy": 0.16465269774198532,
      "num_tokens": 32244276.0,
      "step": 15540
    },
    {
      "entropy": 5.719611978530883,
      "epoch": 1.4101052249637156,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004806003361267415,
      "loss": 5.41,
      "mean_token_accuracy": 0.1719347968697548,
      "num_tokens": 32253745.0,
      "step": 15545
    },
    {
      "entropy": 5.732329368591309,
      "epoch": 1.4105587808417996,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004805871904645737,
      "loss": 5.489,
      "mean_token_accuracy": 0.16701649725437165,
      "num_tokens": 32263848.0,
      "step": 15550
    },
    {
      "entropy": 5.671817302703857,
      "epoch": 1.411012336719884,
      "grad_norm": 2.125,
      "learning_rate": 0.0004805740405508103,
      "loss": 5.4485,
      "mean_token_accuracy": 0.16460975259542465,
      "num_tokens": 32274346.0,
      "step": 15555
    },
    {
      "entropy": 5.676841115951538,
      "epoch": 1.4114658925979682,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004805608863857236,
      "loss": 5.4216,
      "mean_token_accuracy": 0.16542745381593704,
      "num_tokens": 32284622.0,
      "step": 15560
    },
    {
      "entropy": 5.652889537811279,
      "epoch": 1.4119194484760522,
      "grad_norm": 3.015625,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.3517,
      "mean_token_accuracy": 0.17364450246095658,
      "num_tokens": 32294712.0,
      "step": 15565
    },
    {
      "entropy": 5.679360437393188,
      "epoch": 1.4123730043541365,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004805345653026675,
      "loss": 5.435,
      "mean_token_accuracy": 0.17669014781713485,
      "num_tokens": 32304487.0,
      "step": 15570
    },
    {
      "entropy": 5.670064926147461,
      "epoch": 1.4128265602322205,
      "grad_norm": 2.34375,
      "learning_rate": 0.00048052139838524263,
      "loss": 5.4321,
      "mean_token_accuracy": 0.1727903425693512,
      "num_tokens": 32316123.0,
      "step": 15575
    },
    {
      "entropy": 5.768592071533203,
      "epoch": 1.4132801161103048,
      "grad_norm": 2.046875,
      "learning_rate": 0.00048050822721758274,
      "loss": 5.4819,
      "mean_token_accuracy": 0.16563772410154343,
      "num_tokens": 32326298.0,
      "step": 15580
    },
    {
      "entropy": 5.663297176361084,
      "epoch": 1.413733671988389,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004804950517999604,
      "loss": 5.3893,
      "mean_token_accuracy": 0.16355795860290528,
      "num_tokens": 32336321.0,
      "step": 15585
    },
    {
      "entropy": 5.708510255813598,
      "epoch": 1.414187227866473,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004804818721326481,
      "loss": 5.4667,
      "mean_token_accuracy": 0.16346232146024703,
      "num_tokens": 32346807.0,
      "step": 15590
    },
    {
      "entropy": 5.694107151031494,
      "epoch": 1.4146407837445574,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004804686882159183,
      "loss": 5.4203,
      "mean_token_accuracy": 0.16758151352405548,
      "num_tokens": 32357541.0,
      "step": 15595
    },
    {
      "entropy": 5.72988715171814,
      "epoch": 1.4150943396226414,
      "grad_norm": 2.921875,
      "learning_rate": 0.00048045550005004355,
      "loss": 5.4719,
      "mean_token_accuracy": 0.17255567163228988,
      "num_tokens": 32367221.0,
      "step": 15600
    },
    {
      "entropy": 5.65183424949646,
      "epoch": 1.4155478955007257,
      "grad_norm": 7.96875,
      "learning_rate": 0.0004804423076352968,
      "loss": 5.3981,
      "mean_token_accuracy": 0.1648460939526558,
      "num_tokens": 32376954.0,
      "step": 15605
    },
    {
      "entropy": 5.710407733917236,
      "epoch": 1.41600145137881,
      "grad_norm": 3.859375,
      "learning_rate": 0.00048042911097195075,
      "loss": 5.5568,
      "mean_token_accuracy": 0.16776450127363204,
      "num_tokens": 32387349.0,
      "step": 15610
    },
    {
      "entropy": 5.789750719070435,
      "epoch": 1.416455007256894,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004804159100602784,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16646922528743743,
      "num_tokens": 32397236.0,
      "step": 15615
    },
    {
      "entropy": 5.678567838668823,
      "epoch": 1.4169085631349783,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00048040270490055264,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1716088280081749,
      "num_tokens": 32407390.0,
      "step": 15620
    },
    {
      "entropy": 5.61689395904541,
      "epoch": 1.4173621190130623,
      "grad_norm": 3.796875,
      "learning_rate": 0.0004803894954930465,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16887947767972947,
      "num_tokens": 32418065.0,
      "step": 15625
    },
    {
      "entropy": 5.76354193687439,
      "epoch": 1.4178156748911466,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004803762818380334,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1709938019514084,
      "num_tokens": 32428737.0,
      "step": 15630
    },
    {
      "entropy": 5.691916513442993,
      "epoch": 1.4182692307692308,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00048036306393578644,
      "loss": 5.5153,
      "mean_token_accuracy": 0.16232305020093918,
      "num_tokens": 32438503.0,
      "step": 15635
    },
    {
      "entropy": 5.660832309722901,
      "epoch": 1.4187227866473149,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004803498417865789,
      "loss": 5.374,
      "mean_token_accuracy": 0.1727769896388054,
      "num_tokens": 32448435.0,
      "step": 15640
    },
    {
      "entropy": 5.710304212570191,
      "epoch": 1.4191763425253991,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004803366153906844,
      "loss": 5.4474,
      "mean_token_accuracy": 0.17066663652658462,
      "num_tokens": 32458147.0,
      "step": 15645
    },
    {
      "entropy": 5.690779972076416,
      "epoch": 1.4196298984034832,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004803233847483763,
      "loss": 5.4491,
      "mean_token_accuracy": 0.17066556960344315,
      "num_tokens": 32467590.0,
      "step": 15650
    },
    {
      "entropy": 5.718313789367675,
      "epoch": 1.4200834542815675,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00048031014985992825,
      "loss": 5.5064,
      "mean_token_accuracy": 0.16121402829885484,
      "num_tokens": 32479006.0,
      "step": 15655
    },
    {
      "entropy": 5.632800912857055,
      "epoch": 1.4205370101596517,
      "grad_norm": 2.109375,
      "learning_rate": 0.000480296910725614,
      "loss": 5.3979,
      "mean_token_accuracy": 0.1750167578458786,
      "num_tokens": 32489053.0,
      "step": 15660
    },
    {
      "entropy": 5.671632719039917,
      "epoch": 1.4209905660377358,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004802836673457074,
      "loss": 5.4682,
      "mean_token_accuracy": 0.15919609665870665,
      "num_tokens": 32498948.0,
      "step": 15665
    },
    {
      "entropy": 5.765984439849854,
      "epoch": 1.42144412191582,
      "grad_norm": 2.328125,
      "learning_rate": 0.00048027041972048225,
      "loss": 5.4299,
      "mean_token_accuracy": 0.1682679682970047,
      "num_tokens": 32510207.0,
      "step": 15670
    },
    {
      "entropy": 5.7563577651977536,
      "epoch": 1.421897677793904,
      "grad_norm": 5.9375,
      "learning_rate": 0.0004802571678502124,
      "loss": 5.5618,
      "mean_token_accuracy": 0.15649960339069366,
      "num_tokens": 32521083.0,
      "step": 15675
    },
    {
      "entropy": 5.6472546577453615,
      "epoch": 1.4223512336719883,
      "grad_norm": 2.8125,
      "learning_rate": 0.000480243911735172,
      "loss": 5.4183,
      "mean_token_accuracy": 0.16748034954071045,
      "num_tokens": 32530846.0,
      "step": 15680
    },
    {
      "entropy": 5.758372163772583,
      "epoch": 1.4228047895500726,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048023065137563534,
      "loss": 5.5831,
      "mean_token_accuracy": 0.15883456617593766,
      "num_tokens": 32542709.0,
      "step": 15685
    },
    {
      "entropy": 5.7537284851074215,
      "epoch": 1.4232583454281569,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004802173867718764,
      "loss": 5.4406,
      "mean_token_accuracy": 0.168556809425354,
      "num_tokens": 32553457.0,
      "step": 15690
    },
    {
      "entropy": 5.751621341705322,
      "epoch": 1.423711901306241,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004802041179241697,
      "loss": 5.6074,
      "mean_token_accuracy": 0.15810568183660506,
      "num_tokens": 32564670.0,
      "step": 15695
    },
    {
      "entropy": 5.76178011894226,
      "epoch": 1.4241654571843252,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004801908448327895,
      "loss": 5.4567,
      "mean_token_accuracy": 0.17351664900779723,
      "num_tokens": 32575167.0,
      "step": 15700
    },
    {
      "entropy": 5.742387485504151,
      "epoch": 1.4246190130624092,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004801775674980103,
      "loss": 5.4906,
      "mean_token_accuracy": 0.17127618342638015,
      "num_tokens": 32584455.0,
      "step": 15705
    },
    {
      "entropy": 5.570784091949463,
      "epoch": 1.4250725689404935,
      "grad_norm": 2.015625,
      "learning_rate": 0.00048016428592010676,
      "loss": 5.3335,
      "mean_token_accuracy": 0.1759684666991234,
      "num_tokens": 32594802.0,
      "step": 15710
    },
    {
      "entropy": 5.8384199142456055,
      "epoch": 1.4255261248185778,
      "grad_norm": 2.109375,
      "learning_rate": 0.00048015100009935347,
      "loss": 5.5478,
      "mean_token_accuracy": 0.1578245311975479,
      "num_tokens": 32606757.0,
      "step": 15715
    },
    {
      "entropy": 5.755512380599976,
      "epoch": 1.4259796806966618,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004801377100360253,
      "loss": 5.5378,
      "mean_token_accuracy": 0.16057031899690627,
      "num_tokens": 32617111.0,
      "step": 15720
    },
    {
      "entropy": 5.627434778213501,
      "epoch": 1.426433236574746,
      "grad_norm": 2.203125,
      "learning_rate": 0.000480124415730397,
      "loss": 5.3892,
      "mean_token_accuracy": 0.1746115729212761,
      "num_tokens": 32626620.0,
      "step": 15725
    },
    {
      "entropy": 5.70077075958252,
      "epoch": 1.4268867924528301,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004801111171827434,
      "loss": 5.4256,
      "mean_token_accuracy": 0.1705096960067749,
      "num_tokens": 32637284.0,
      "step": 15730
    },
    {
      "entropy": 5.70149245262146,
      "epoch": 1.4273403483309144,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004800978143933398,
      "loss": 5.4833,
      "mean_token_accuracy": 0.16218256056308747,
      "num_tokens": 32647256.0,
      "step": 15735
    },
    {
      "entropy": 5.705234050750732,
      "epoch": 1.4277939042089987,
      "grad_norm": 2.515625,
      "learning_rate": 0.00048008450736246094,
      "loss": 5.4873,
      "mean_token_accuracy": 0.1712659254670143,
      "num_tokens": 32658406.0,
      "step": 15740
    },
    {
      "entropy": 5.638548994064331,
      "epoch": 1.4282474600870827,
      "grad_norm": 2.90625,
      "learning_rate": 0.00048007119609038237,
      "loss": 5.401,
      "mean_token_accuracy": 0.17134318202733995,
      "num_tokens": 32668533.0,
      "step": 15745
    },
    {
      "entropy": 5.691101837158203,
      "epoch": 1.428701015965167,
      "grad_norm": 2.03125,
      "learning_rate": 0.00048005788057737907,
      "loss": 5.4724,
      "mean_token_accuracy": 0.16583883464336396,
      "num_tokens": 32678689.0,
      "step": 15750
    },
    {
      "entropy": 5.745490980148316,
      "epoch": 1.429154571843251,
      "grad_norm": 2.125,
      "learning_rate": 0.0004800445608237266,
      "loss": 5.5226,
      "mean_token_accuracy": 0.15988076776266097,
      "num_tokens": 32690380.0,
      "step": 15755
    },
    {
      "entropy": 5.750646257400513,
      "epoch": 1.4296081277213353,
      "grad_norm": 2.703125,
      "learning_rate": 0.00048003123682970045,
      "loss": 5.4581,
      "mean_token_accuracy": 0.16682433485984802,
      "num_tokens": 32700406.0,
      "step": 15760
    },
    {
      "entropy": 5.664740562438965,
      "epoch": 1.4300616835994195,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00048001790859557596,
      "loss": 5.3857,
      "mean_token_accuracy": 0.16817823499441148,
      "num_tokens": 32710477.0,
      "step": 15765
    },
    {
      "entropy": 5.701738691329956,
      "epoch": 1.4305152394775036,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004800045761216289,
      "loss": 5.4667,
      "mean_token_accuracy": 0.16602648049592972,
      "num_tokens": 32721068.0,
      "step": 15770
    },
    {
      "entropy": 5.665327024459839,
      "epoch": 1.4309687953555879,
      "grad_norm": 2.359375,
      "learning_rate": 0.000479991239408135,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1602112501859665,
      "num_tokens": 32731983.0,
      "step": 15775
    },
    {
      "entropy": 5.7848467350006105,
      "epoch": 1.431422351233672,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047997789845537003,
      "loss": 5.4426,
      "mean_token_accuracy": 0.1657240018248558,
      "num_tokens": 32742052.0,
      "step": 15780
    },
    {
      "entropy": 5.829418468475342,
      "epoch": 1.4318759071117562,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004799645532636098,
      "loss": 5.5529,
      "mean_token_accuracy": 0.15871381163597106,
      "num_tokens": 32752818.0,
      "step": 15785
    },
    {
      "entropy": 5.642741346359253,
      "epoch": 1.4323294629898404,
      "grad_norm": 3.8125,
      "learning_rate": 0.00047995120383313044,
      "loss": 5.4538,
      "mean_token_accuracy": 0.16765670329332352,
      "num_tokens": 32762922.0,
      "step": 15790
    },
    {
      "entropy": 5.71010913848877,
      "epoch": 1.4327830188679245,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047993785016420796,
      "loss": 5.5367,
      "mean_token_accuracy": 0.1612371563911438,
      "num_tokens": 32773554.0,
      "step": 15795
    },
    {
      "entropy": 5.769787216186524,
      "epoch": 1.4332365747460087,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047992449225711854,
      "loss": 5.5161,
      "mean_token_accuracy": 0.16435697227716445,
      "num_tokens": 32784510.0,
      "step": 15800
    },
    {
      "entropy": 5.670501279830932,
      "epoch": 1.4336901306240928,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047991113011213837,
      "loss": 5.3731,
      "mean_token_accuracy": 0.16291718631982804,
      "num_tokens": 32794613.0,
      "step": 15805
    },
    {
      "entropy": 5.631981325149536,
      "epoch": 1.434143686502177,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004798977637295438,
      "loss": 5.3991,
      "mean_token_accuracy": 0.17151408344507219,
      "num_tokens": 32804041.0,
      "step": 15810
    },
    {
      "entropy": 5.663556385040283,
      "epoch": 1.4345972423802613,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004798843931096113,
      "loss": 5.4175,
      "mean_token_accuracy": 0.1664323940873146,
      "num_tokens": 32814101.0,
      "step": 15815
    },
    {
      "entropy": 5.721880674362183,
      "epoch": 1.4350507982583454,
      "grad_norm": 3.578125,
      "learning_rate": 0.0004798710182526173,
      "loss": 5.4142,
      "mean_token_accuracy": 0.17001063972711564,
      "num_tokens": 32824458.0,
      "step": 15820
    },
    {
      "entropy": 5.747647047042847,
      "epoch": 1.4355043541364296,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047985763915883847,
      "loss": 5.4846,
      "mean_token_accuracy": 0.1673012986779213,
      "num_tokens": 32834602.0,
      "step": 15825
    },
    {
      "entropy": 5.688952398300171,
      "epoch": 1.4359579100145137,
      "grad_norm": 4.0,
      "learning_rate": 0.0004798442558285514,
      "loss": 5.4427,
      "mean_token_accuracy": 0.17022891342639923,
      "num_tokens": 32845375.0,
      "step": 15830
    },
    {
      "entropy": 5.661872673034668,
      "epoch": 1.436411465892598,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004798308682620329,
      "loss": 5.4746,
      "mean_token_accuracy": 0.16347875595092773,
      "num_tokens": 32856187.0,
      "step": 15835
    },
    {
      "entropy": 5.703317928314209,
      "epoch": 1.4368650217706822,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047981747645955996,
      "loss": 5.3729,
      "mean_token_accuracy": 0.1673371359705925,
      "num_tokens": 32867238.0,
      "step": 15840
    },
    {
      "entropy": 5.65754919052124,
      "epoch": 1.4373185776487662,
      "grad_norm": 3.25,
      "learning_rate": 0.0004798040804214094,
      "loss": 5.358,
      "mean_token_accuracy": 0.17089327424764633,
      "num_tokens": 32877228.0,
      "step": 15845
    },
    {
      "entropy": 5.6147651195526125,
      "epoch": 1.4377721335268505,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004797906801478582,
      "loss": 5.4202,
      "mean_token_accuracy": 0.15245310366153716,
      "num_tokens": 32887932.0,
      "step": 15850
    },
    {
      "entropy": 5.721226787567138,
      "epoch": 1.4382256894049346,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004797772756391836,
      "loss": 5.491,
      "mean_token_accuracy": 0.16640633940696717,
      "num_tokens": 32898193.0,
      "step": 15855
    },
    {
      "entropy": 5.698588562011719,
      "epoch": 1.4386792452830188,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004797638668956627,
      "loss": 5.4125,
      "mean_token_accuracy": 0.16879950910806657,
      "num_tokens": 32908472.0,
      "step": 15860
    },
    {
      "entropy": 5.676741981506348,
      "epoch": 1.439132801161103,
      "grad_norm": 2.625,
      "learning_rate": 0.0004797504539175729,
      "loss": 5.3915,
      "mean_token_accuracy": 0.17146891206502915,
      "num_tokens": 32918665.0,
      "step": 15865
    },
    {
      "entropy": 5.64249906539917,
      "epoch": 1.4395863570391874,
      "grad_norm": 3.375,
      "learning_rate": 0.0004797370367051915,
      "loss": 5.4995,
      "mean_token_accuracy": 0.16246135234832765,
      "num_tokens": 32930109.0,
      "step": 15870
    },
    {
      "entropy": 5.650494003295899,
      "epoch": 1.4400399129172714,
      "grad_norm": 2.6875,
      "learning_rate": 0.000479723615258796,
      "loss": 5.3706,
      "mean_token_accuracy": 0.16806180775165558,
      "num_tokens": 32939750.0,
      "step": 15875
    },
    {
      "entropy": 5.677830076217651,
      "epoch": 1.4404934687953557,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047971018957866396,
      "loss": 5.411,
      "mean_token_accuracy": 0.1711413085460663,
      "num_tokens": 32949905.0,
      "step": 15880
    },
    {
      "entropy": 5.701831293106079,
      "epoch": 1.4409470246734397,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047969675966507314,
      "loss": 5.551,
      "mean_token_accuracy": 0.16172923967242242,
      "num_tokens": 32961031.0,
      "step": 15885
    },
    {
      "entropy": 5.703682470321655,
      "epoch": 1.441400580551524,
      "grad_norm": 2.0625,
      "learning_rate": 0.000479683325518301,
      "loss": 5.3663,
      "mean_token_accuracy": 0.17102610766887666,
      "num_tokens": 32970818.0,
      "step": 15890
    },
    {
      "entropy": 5.704067945480347,
      "epoch": 1.4418541364296082,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004796698871386257,
      "loss": 5.3863,
      "mean_token_accuracy": 0.17605783194303512,
      "num_tokens": 32979516.0,
      "step": 15895
    },
    {
      "entropy": 5.745927429199218,
      "epoch": 1.4423076923076923,
      "grad_norm": 4.21875,
      "learning_rate": 0.0004796564445263249,
      "loss": 5.5584,
      "mean_token_accuracy": 0.16308927834033965,
      "num_tokens": 32990569.0,
      "step": 15900
    },
    {
      "entropy": 5.703577709197998,
      "epoch": 1.4427612481857766,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047964299768167665,
      "loss": 5.3966,
      "mean_token_accuracy": 0.17020488381385804,
      "num_tokens": 32999585.0,
      "step": 15905
    },
    {
      "entropy": 5.737225580215454,
      "epoch": 1.4432148040638606,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004796295466049591,
      "loss": 5.4813,
      "mean_token_accuracy": 0.1649126246571541,
      "num_tokens": 33010151.0,
      "step": 15910
    },
    {
      "entropy": 5.699170923233032,
      "epoch": 1.4436683599419449,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047961609129645036,
      "loss": 5.4594,
      "mean_token_accuracy": 0.16157076507806778,
      "num_tokens": 33019833.0,
      "step": 15915
    },
    {
      "entropy": 5.63279914855957,
      "epoch": 1.4441219158200291,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047960263175642867,
      "loss": 5.38,
      "mean_token_accuracy": 0.17339448928833007,
      "num_tokens": 33029335.0,
      "step": 15920
    },
    {
      "entropy": 5.687233257293701,
      "epoch": 1.4445754716981132,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004795891679851724,
      "loss": 5.473,
      "mean_token_accuracy": 0.15922760963439941,
      "num_tokens": 33039505.0,
      "step": 15925
    },
    {
      "entropy": 5.8130182266235355,
      "epoch": 1.4450290275761974,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004795756999829601,
      "loss": 5.5178,
      "mean_token_accuracy": 0.16685259342193604,
      "num_tokens": 33050778.0,
      "step": 15930
    },
    {
      "entropy": 5.656160879135132,
      "epoch": 1.4454825834542815,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047956222775007016,
      "loss": 5.4294,
      "mean_token_accuracy": 0.17275117188692093,
      "num_tokens": 33060416.0,
      "step": 15935
    },
    {
      "entropy": 5.657961225509643,
      "epoch": 1.4459361393323658,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004795487512867812,
      "loss": 5.3682,
      "mean_token_accuracy": 0.1678393617272377,
      "num_tokens": 33070630.0,
      "step": 15940
    },
    {
      "entropy": 5.774859476089477,
      "epoch": 1.44638969521045,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047953527059337194,
      "loss": 5.4806,
      "mean_token_accuracy": 0.16394369304180145,
      "num_tokens": 33081533.0,
      "step": 15945
    },
    {
      "entropy": 5.761707592010498,
      "epoch": 1.446843251088534,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004795217856701212,
      "loss": 5.4536,
      "mean_token_accuracy": 0.16268696784973144,
      "num_tokens": 33092256.0,
      "step": 15950
    },
    {
      "entropy": 5.621526575088501,
      "epoch": 1.4472968069666183,
      "grad_norm": 4.78125,
      "learning_rate": 0.00047950829651730774,
      "loss": 5.387,
      "mean_token_accuracy": 0.1679667130112648,
      "num_tokens": 33102956.0,
      "step": 15955
    },
    {
      "entropy": 5.678064775466919,
      "epoch": 1.4477503628447024,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004794948031352107,
      "loss": 5.4536,
      "mean_token_accuracy": 0.17154159396886826,
      "num_tokens": 33112750.0,
      "step": 15960
    },
    {
      "entropy": 5.737695693969727,
      "epoch": 1.4482039187227866,
      "grad_norm": 2.578125,
      "learning_rate": 0.00047948130552410897,
      "loss": 5.5216,
      "mean_token_accuracy": 0.15873547345399858,
      "num_tokens": 33122497.0,
      "step": 15965
    },
    {
      "entropy": 5.721784973144532,
      "epoch": 1.448657474600871,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047946780368428175,
      "loss": 5.4654,
      "mean_token_accuracy": 0.16265515983104706,
      "num_tokens": 33132892.0,
      "step": 15970
    },
    {
      "entropy": 5.711485052108765,
      "epoch": 1.449111030478955,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004794542976160081,
      "loss": 5.4518,
      "mean_token_accuracy": 0.16338601410388948,
      "num_tokens": 33142737.0,
      "step": 15975
    },
    {
      "entropy": 5.645646953582764,
      "epoch": 1.4495645863570392,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047944078731956765,
      "loss": 5.4068,
      "mean_token_accuracy": 0.1679958775639534,
      "num_tokens": 33153352.0,
      "step": 15980
    },
    {
      "entropy": 5.675401878356934,
      "epoch": 1.4500181422351233,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047942727279523957,
      "loss": 5.3864,
      "mean_token_accuracy": 0.16932000666856767,
      "num_tokens": 33164018.0,
      "step": 15985
    },
    {
      "entropy": 5.735835027694702,
      "epoch": 1.4504716981132075,
      "grad_norm": 5.40625,
      "learning_rate": 0.0004794137540433034,
      "loss": 5.5085,
      "mean_token_accuracy": 0.16317780315876007,
      "num_tokens": 33174811.0,
      "step": 15990
    },
    {
      "entropy": 5.6550013542175295,
      "epoch": 1.4509252539912918,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004794002310640386,
      "loss": 5.4261,
      "mean_token_accuracy": 0.16860293745994567,
      "num_tokens": 33185509.0,
      "step": 15995
    },
    {
      "entropy": 5.704597997665405,
      "epoch": 1.4513788098693758,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047938670385772503,
      "loss": 5.4162,
      "mean_token_accuracy": 0.17468949407339096,
      "num_tokens": 33195441.0,
      "step": 16000
    },
    {
      "entropy": 5.671339607238769,
      "epoch": 1.45183236574746,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047937317242464244,
      "loss": 5.3976,
      "mean_token_accuracy": 0.17078361064195632,
      "num_tokens": 33206002.0,
      "step": 16005
    },
    {
      "entropy": 5.6531877517700195,
      "epoch": 1.4522859216255442,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004793596367650704,
      "loss": 5.406,
      "mean_token_accuracy": 0.16664284765720366,
      "num_tokens": 33217408.0,
      "step": 16010
    },
    {
      "entropy": 5.615947628021241,
      "epoch": 1.4527394775036284,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004793460968792891,
      "loss": 5.3863,
      "mean_token_accuracy": 0.1717242032289505,
      "num_tokens": 33227795.0,
      "step": 16015
    },
    {
      "entropy": 5.749507713317871,
      "epoch": 1.4531930333817127,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047933255276757836,
      "loss": 5.4233,
      "mean_token_accuracy": 0.1635444790124893,
      "num_tokens": 33239047.0,
      "step": 16020
    },
    {
      "entropy": 5.68520679473877,
      "epoch": 1.4536465892597967,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004793190044302184,
      "loss": 5.447,
      "mean_token_accuracy": 0.16823295950889589,
      "num_tokens": 33249476.0,
      "step": 16025
    },
    {
      "entropy": 5.658813428878784,
      "epoch": 1.454100145137881,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047930545186748936,
      "loss": 5.3889,
      "mean_token_accuracy": 0.17129696309566497,
      "num_tokens": 33259169.0,
      "step": 16030
    },
    {
      "entropy": 5.71914267539978,
      "epoch": 1.454553701015965,
      "grad_norm": 7.1875,
      "learning_rate": 0.00047929189507967153,
      "loss": 5.5266,
      "mean_token_accuracy": 0.16494582444429398,
      "num_tokens": 33270540.0,
      "step": 16035
    },
    {
      "entropy": 5.704966926574707,
      "epoch": 1.4550072568940493,
      "grad_norm": 1.7109375,
      "learning_rate": 0.00047927833406704516,
      "loss": 5.4375,
      "mean_token_accuracy": 0.1695617511868477,
      "num_tokens": 33281176.0,
      "step": 16040
    },
    {
      "entropy": 5.778288459777832,
      "epoch": 1.4554608127721336,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004792647688298909,
      "loss": 5.5317,
      "mean_token_accuracy": 0.16041023433208465,
      "num_tokens": 33293178.0,
      "step": 16045
    },
    {
      "entropy": 5.6851115226745605,
      "epoch": 1.4559143686502178,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004792511993684891,
      "loss": 5.4056,
      "mean_token_accuracy": 0.16844548285007477,
      "num_tokens": 33303407.0,
      "step": 16050
    },
    {
      "entropy": 5.797557020187378,
      "epoch": 1.4563679245283019,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047923762568312053,
      "loss": 5.5145,
      "mean_token_accuracy": 0.16178034096956254,
      "num_tokens": 33313964.0,
      "step": 16055
    },
    {
      "entropy": 5.6715535640716555,
      "epoch": 1.4568214804063861,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004792240477740657,
      "loss": 5.3922,
      "mean_token_accuracy": 0.1706519529223442,
      "num_tokens": 33323365.0,
      "step": 16060
    },
    {
      "entropy": 5.63593373298645,
      "epoch": 1.4572750362844702,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004792104656416056,
      "loss": 5.4611,
      "mean_token_accuracy": 0.16830355376005174,
      "num_tokens": 33334000.0,
      "step": 16065
    },
    {
      "entropy": 5.785321474075317,
      "epoch": 1.4577285921625545,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047919687928602104,
      "loss": 5.522,
      "mean_token_accuracy": 0.16140753775835037,
      "num_tokens": 33344548.0,
      "step": 16070
    },
    {
      "entropy": 5.71856541633606,
      "epoch": 1.4581821480406387,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047918328870759293,
      "loss": 5.3786,
      "mean_token_accuracy": 0.17567453682422637,
      "num_tokens": 33354017.0,
      "step": 16075
    },
    {
      "entropy": 5.71616005897522,
      "epoch": 1.4586357039187228,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004791696939066024,
      "loss": 5.4881,
      "mean_token_accuracy": 0.1566734716296196,
      "num_tokens": 33364689.0,
      "step": 16080
    },
    {
      "entropy": 5.647443437576294,
      "epoch": 1.459089259796807,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004791560948833306,
      "loss": 5.3361,
      "mean_token_accuracy": 0.17472701370716096,
      "num_tokens": 33374359.0,
      "step": 16085
    },
    {
      "entropy": 5.740912342071534,
      "epoch": 1.459542815674891,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004791424916380586,
      "loss": 5.5185,
      "mean_token_accuracy": 0.1657344028353691,
      "num_tokens": 33384804.0,
      "step": 16090
    },
    {
      "entropy": 5.658357763290406,
      "epoch": 1.4599963715529753,
      "grad_norm": 2.015625,
      "learning_rate": 0.000479128884171068,
      "loss": 5.3476,
      "mean_token_accuracy": 0.17548676282167436,
      "num_tokens": 33394869.0,
      "step": 16095
    },
    {
      "entropy": 5.655737495422363,
      "epoch": 1.4604499274310596,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047911527248264003,
      "loss": 5.5038,
      "mean_token_accuracy": 0.1646803468465805,
      "num_tokens": 33405359.0,
      "step": 16100
    },
    {
      "entropy": 5.6828632831573485,
      "epoch": 1.4609034833091437,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047910165657305613,
      "loss": 5.4473,
      "mean_token_accuracy": 0.16394637972116471,
      "num_tokens": 33415091.0,
      "step": 16105
    },
    {
      "entropy": 5.722692632675171,
      "epoch": 1.461357039187228,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047908803644259804,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16994312405586243,
      "num_tokens": 33425379.0,
      "step": 16110
    },
    {
      "entropy": 5.76458101272583,
      "epoch": 1.461810595065312,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047907441209154726,
      "loss": 5.4934,
      "mean_token_accuracy": 0.1610972359776497,
      "num_tokens": 33436288.0,
      "step": 16115
    },
    {
      "entropy": 5.744371795654297,
      "epoch": 1.4622641509433962,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047906078352018575,
      "loss": 5.5012,
      "mean_token_accuracy": 0.16886803954839708,
      "num_tokens": 33445997.0,
      "step": 16120
    },
    {
      "entropy": 5.718356275558472,
      "epoch": 1.4627177068214805,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004790471507287952,
      "loss": 5.4548,
      "mean_token_accuracy": 0.16952823996543884,
      "num_tokens": 33455627.0,
      "step": 16125
    },
    {
      "entropy": 5.730396270751953,
      "epoch": 1.4631712626995645,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047903351371765757,
      "loss": 5.5216,
      "mean_token_accuracy": 0.158284729719162,
      "num_tokens": 33467288.0,
      "step": 16130
    },
    {
      "entropy": 5.723852682113647,
      "epoch": 1.4636248185776488,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047901987248705484,
      "loss": 5.4999,
      "mean_token_accuracy": 0.1644822970032692,
      "num_tokens": 33477932.0,
      "step": 16135
    },
    {
      "entropy": 5.659976625442505,
      "epoch": 1.4640783744557329,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004790062270372693,
      "loss": 5.3739,
      "mean_token_accuracy": 0.1735864147543907,
      "num_tokens": 33487334.0,
      "step": 16140
    },
    {
      "entropy": 5.671331119537354,
      "epoch": 1.4645319303338171,
      "grad_norm": 4.40625,
      "learning_rate": 0.0004789925773685828,
      "loss": 5.4225,
      "mean_token_accuracy": 0.1732308581471443,
      "num_tokens": 33497334.0,
      "step": 16145
    },
    {
      "entropy": 5.747703218460083,
      "epoch": 1.4649854862119014,
      "grad_norm": 2.78125,
      "learning_rate": 0.0004789789234812779,
      "loss": 5.4329,
      "mean_token_accuracy": 0.16623817533254623,
      "num_tokens": 33508117.0,
      "step": 16150
    },
    {
      "entropy": 5.66282696723938,
      "epoch": 1.4654390420899854,
      "grad_norm": 3.90625,
      "learning_rate": 0.00047896526537563686,
      "loss": 5.4029,
      "mean_token_accuracy": 0.1732719749212265,
      "num_tokens": 33518461.0,
      "step": 16155
    },
    {
      "entropy": 5.676852035522461,
      "epoch": 1.4658925979680697,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047895160305194223,
      "loss": 5.461,
      "mean_token_accuracy": 0.16681208163499833,
      "num_tokens": 33529073.0,
      "step": 16160
    },
    {
      "entropy": 5.652419757843018,
      "epoch": 1.4663461538461537,
      "grad_norm": 3.15625,
      "learning_rate": 0.0004789379365104764,
      "loss": 5.4271,
      "mean_token_accuracy": 0.16931309998035432,
      "num_tokens": 33539038.0,
      "step": 16165
    },
    {
      "entropy": 5.708762216567993,
      "epoch": 1.466799709724238,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004789242657515221,
      "loss": 5.4699,
      "mean_token_accuracy": 0.16933149248361587,
      "num_tokens": 33549574.0,
      "step": 16170
    },
    {
      "entropy": 5.687426471710205,
      "epoch": 1.4672532656023223,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004789105907753621,
      "loss": 5.3552,
      "mean_token_accuracy": 0.17888641804456712,
      "num_tokens": 33559276.0,
      "step": 16175
    },
    {
      "entropy": 5.648719692230225,
      "epoch": 1.4677068214804063,
      "grad_norm": 2.953125,
      "learning_rate": 0.000478896911582279,
      "loss": 5.3878,
      "mean_token_accuracy": 0.1703267976641655,
      "num_tokens": 33569787.0,
      "step": 16180
    },
    {
      "entropy": 5.702033948898316,
      "epoch": 1.4681603773584906,
      "grad_norm": 5.34375,
      "learning_rate": 0.0004788832281725558,
      "loss": 5.6116,
      "mean_token_accuracy": 0.1598723217844963,
      "num_tokens": 33580577.0,
      "step": 16185
    },
    {
      "entropy": 5.774663782119751,
      "epoch": 1.4686139332365746,
      "grad_norm": 3.265625,
      "learning_rate": 0.0004788695405464755,
      "loss": 5.5161,
      "mean_token_accuracy": 0.16154904961585997,
      "num_tokens": 33591708.0,
      "step": 16190
    },
    {
      "entropy": 5.772363233566284,
      "epoch": 1.469067489114659,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047885584870432127,
      "loss": 5.4192,
      "mean_token_accuracy": 0.1745361343026161,
      "num_tokens": 33602553.0,
      "step": 16195
    },
    {
      "entropy": 5.673396968841553,
      "epoch": 1.4695210449927432,
      "grad_norm": 2.8125,
      "learning_rate": 0.00047884215264637606,
      "loss": 5.4381,
      "mean_token_accuracy": 0.17418480813503265,
      "num_tokens": 33613325.0,
      "step": 16200
    },
    {
      "entropy": 5.621952819824219,
      "epoch": 1.4699746008708272,
      "grad_norm": 3.46875,
      "learning_rate": 0.0004788284523729232,
      "loss": 5.4378,
      "mean_token_accuracy": 0.16975034326314925,
      "num_tokens": 33624664.0,
      "step": 16205
    },
    {
      "entropy": 5.779737758636474,
      "epoch": 1.4704281567489115,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047881474788424595,
      "loss": 5.54,
      "mean_token_accuracy": 0.1636625722050667,
      "num_tokens": 33635346.0,
      "step": 16210
    },
    {
      "entropy": 5.741463470458984,
      "epoch": 1.4708817126269955,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047880103918062776,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16869738101959228,
      "num_tokens": 33647008.0,
      "step": 16215
    },
    {
      "entropy": 5.763271522521973,
      "epoch": 1.4713352685050798,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047878732626235225,
      "loss": 5.5334,
      "mean_token_accuracy": 0.16272443383932114,
      "num_tokens": 33656964.0,
      "step": 16220
    },
    {
      "entropy": 5.707889366149902,
      "epoch": 1.471788824383164,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004787736091297028,
      "loss": 5.3738,
      "mean_token_accuracy": 0.16872794181108475,
      "num_tokens": 33667602.0,
      "step": 16225
    },
    {
      "entropy": 5.687304306030273,
      "epoch": 1.4722423802612483,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004787598877829633,
      "loss": 5.4914,
      "mean_token_accuracy": 0.16459726840257644,
      "num_tokens": 33677947.0,
      "step": 16230
    },
    {
      "entropy": 5.6810691356658936,
      "epoch": 1.4726959361393324,
      "grad_norm": 6.3125,
      "learning_rate": 0.0004787461622224173,
      "loss": 5.3875,
      "mean_token_accuracy": 0.17166147530078887,
      "num_tokens": 33687264.0,
      "step": 16235
    },
    {
      "entropy": 5.745103645324707,
      "epoch": 1.4731494920174166,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004787324324483488,
      "loss": 5.4505,
      "mean_token_accuracy": 0.17018003314733504,
      "num_tokens": 33696890.0,
      "step": 16240
    },
    {
      "entropy": 5.749726819992065,
      "epoch": 1.4736030478955007,
      "grad_norm": 5.71875,
      "learning_rate": 0.0004787186984610416,
      "loss": 5.4261,
      "mean_token_accuracy": 0.16269438415765763,
      "num_tokens": 33707243.0,
      "step": 16245
    },
    {
      "entropy": 5.744277715682983,
      "epoch": 1.474056603773585,
      "grad_norm": 3.859375,
      "learning_rate": 0.00047870496026077984,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16141964495182037,
      "num_tokens": 33717973.0,
      "step": 16250
    },
    {
      "entropy": 5.689771747589111,
      "epoch": 1.4745101596516692,
      "grad_norm": 3.9375,
      "learning_rate": 0.0004786912178478476,
      "loss": 5.4218,
      "mean_token_accuracy": 0.17051519602537155,
      "num_tokens": 33727912.0,
      "step": 16255
    },
    {
      "entropy": 5.653558301925659,
      "epoch": 1.4749637155297532,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004786774712225291,
      "loss": 5.3732,
      "mean_token_accuracy": 0.17946792989969254,
      "num_tokens": 33738076.0,
      "step": 16260
    },
    {
      "entropy": 5.710431289672852,
      "epoch": 1.4754172714078375,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004786637203851086,
      "loss": 5.4733,
      "mean_token_accuracy": 0.16299742460250854,
      "num_tokens": 33748790.0,
      "step": 16265
    },
    {
      "entropy": 5.774993371963501,
      "epoch": 1.4758708272859216,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004786499653358705,
      "loss": 5.4499,
      "mean_token_accuracy": 0.1668673798441887,
      "num_tokens": 33758602.0,
      "step": 16270
    },
    {
      "entropy": 5.756646299362183,
      "epoch": 1.4763243831640058,
      "grad_norm": 2.578125,
      "learning_rate": 0.0004786362060750991,
      "loss": 5.5782,
      "mean_token_accuracy": 0.16048363894224166,
      "num_tokens": 33769646.0,
      "step": 16275
    },
    {
      "entropy": 5.700676679611206,
      "epoch": 1.47677793904209,
      "grad_norm": 3.90625,
      "learning_rate": 0.00047862244260307913,
      "loss": 5.4245,
      "mean_token_accuracy": 0.16483661532402039,
      "num_tokens": 33779874.0,
      "step": 16280
    },
    {
      "entropy": 5.729207944869995,
      "epoch": 1.4772314949201741,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004786086749200951,
      "loss": 5.4003,
      "mean_token_accuracy": 0.17341534197330474,
      "num_tokens": 33789730.0,
      "step": 16285
    },
    {
      "entropy": 5.7319653034210205,
      "epoch": 1.4776850507982584,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004785949030264319,
      "loss": 5.5158,
      "mean_token_accuracy": 0.16959185302257537,
      "num_tokens": 33801038.0,
      "step": 16290
    },
    {
      "entropy": 5.7484174251556395,
      "epoch": 1.4781386066763424,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004785811269223741,
      "loss": 5.4667,
      "mean_token_accuracy": 0.16958183646202088,
      "num_tokens": 33813646.0,
      "step": 16295
    },
    {
      "entropy": 5.754130029678345,
      "epoch": 1.4785921625544267,
      "grad_norm": 2.75,
      "learning_rate": 0.0004785673466082068,
      "loss": 5.4482,
      "mean_token_accuracy": 0.167254301905632,
      "num_tokens": 33824580.0,
      "step": 16300
    },
    {
      "entropy": 5.650108861923218,
      "epoch": 1.479045718432511,
      "grad_norm": 1.625,
      "learning_rate": 0.00047855356208421484,
      "loss": 5.4272,
      "mean_token_accuracy": 0.16929324269294738,
      "num_tokens": 33834366.0,
      "step": 16305
    },
    {
      "entropy": 5.641526079177856,
      "epoch": 1.479499274310595,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004785397733506834,
      "loss": 5.3027,
      "mean_token_accuracy": 0.17710887342691423,
      "num_tokens": 33844194.0,
      "step": 16310
    },
    {
      "entropy": 5.731218528747559,
      "epoch": 1.4799528301886793,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004785259804078976,
      "loss": 5.566,
      "mean_token_accuracy": 0.16404974088072777,
      "num_tokens": 33855161.0,
      "step": 16315
    },
    {
      "entropy": 5.747317314147949,
      "epoch": 1.4804063860667633,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004785121832561426,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1692156746983528,
      "num_tokens": 33865737.0,
      "step": 16320
    },
    {
      "entropy": 5.675132083892822,
      "epoch": 1.4808599419448476,
      "grad_norm": 3.53125,
      "learning_rate": 0.00047849838189570375,
      "loss": 5.4074,
      "mean_token_accuracy": 0.1693673923611641,
      "num_tokens": 33875960.0,
      "step": 16325
    },
    {
      "entropy": 5.726508617401123,
      "epoch": 1.4813134978229319,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004784845763268666,
      "loss": 5.4255,
      "mean_token_accuracy": 0.16444397568702698,
      "num_tokens": 33886234.0,
      "step": 16330
    },
    {
      "entropy": 5.761637926101685,
      "epoch": 1.481767053701016,
      "grad_norm": 2.5,
      "learning_rate": 0.0004784707665499165,
      "loss": 5.3769,
      "mean_token_accuracy": 0.1737170100212097,
      "num_tokens": 33896280.0,
      "step": 16335
    },
    {
      "entropy": 5.619371223449707,
      "epoch": 1.4822206095791002,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004784569525651391,
      "loss": 5.3878,
      "mean_token_accuracy": 0.1759323850274086,
      "num_tokens": 33906308.0,
      "step": 16340
    },
    {
      "entropy": 5.682123708724975,
      "epoch": 1.4826741654571842,
      "grad_norm": 4.4375,
      "learning_rate": 0.00047844313437282005,
      "loss": 5.4448,
      "mean_token_accuracy": 0.1656929925084114,
      "num_tokens": 33917175.0,
      "step": 16345
    },
    {
      "entropy": 5.719280529022217,
      "epoch": 1.4831277213352685,
      "grad_norm": 4.09375,
      "learning_rate": 0.00047842931197324525,
      "loss": 5.4561,
      "mean_token_accuracy": 0.16527069807052613,
      "num_tokens": 33927952.0,
      "step": 16350
    },
    {
      "entropy": 5.749084568023681,
      "epoch": 1.4835812772133528,
      "grad_norm": 2.125,
      "learning_rate": 0.00047841548536670033,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16507349237799646,
      "num_tokens": 33938387.0,
      "step": 16355
    },
    {
      "entropy": 5.768767929077148,
      "epoch": 1.4840348330914368,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047840165455347136,
      "loss": 5.471,
      "mean_token_accuracy": 0.166175939142704,
      "num_tokens": 33949232.0,
      "step": 16360
    },
    {
      "entropy": 5.720411205291748,
      "epoch": 1.484488388969521,
      "grad_norm": 2.859375,
      "learning_rate": 0.00047838781953384436,
      "loss": 5.4717,
      "mean_token_accuracy": 0.16409641802310942,
      "num_tokens": 33959643.0,
      "step": 16365
    },
    {
      "entropy": 5.666046524047852,
      "epoch": 1.484941944847605,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004783739803081054,
      "loss": 5.4447,
      "mean_token_accuracy": 0.1677753821015358,
      "num_tokens": 33969981.0,
      "step": 16370
    },
    {
      "entropy": 5.717892122268677,
      "epoch": 1.4853955007256894,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047836013687654075,
      "loss": 5.4236,
      "mean_token_accuracy": 0.16999652236700058,
      "num_tokens": 33980581.0,
      "step": 16375
    },
    {
      "entropy": 5.673377275466919,
      "epoch": 1.4858490566037736,
      "grad_norm": 7.65625,
      "learning_rate": 0.00047834628923943654,
      "loss": 5.4123,
      "mean_token_accuracy": 0.1717682346701622,
      "num_tokens": 33991960.0,
      "step": 16380
    },
    {
      "entropy": 5.593403148651123,
      "epoch": 1.4863026124818577,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004783324373970793,
      "loss": 5.3556,
      "mean_token_accuracy": 0.17315693795681,
      "num_tokens": 34002044.0,
      "step": 16385
    },
    {
      "entropy": 5.685967016220093,
      "epoch": 1.486756168359942,
      "grad_norm": 4.5,
      "learning_rate": 0.0004783185813497555,
      "loss": 5.5104,
      "mean_token_accuracy": 0.1651084303855896,
      "num_tokens": 34013043.0,
      "step": 16390
    },
    {
      "entropy": 5.66929931640625,
      "epoch": 1.487209724238026,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004783047210977516,
      "loss": 5.4367,
      "mean_token_accuracy": 0.17077335268259047,
      "num_tokens": 34023923.0,
      "step": 16395
    },
    {
      "entropy": 5.696388530731201,
      "epoch": 1.4876632801161103,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047829085664135424,
      "loss": 5.4684,
      "mean_token_accuracy": 0.1671503886580467,
      "num_tokens": 34034518.0,
      "step": 16400
    },
    {
      "entropy": 5.636951017379761,
      "epoch": 1.4881168359941945,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047827698798085015,
      "loss": 5.3602,
      "mean_token_accuracy": 0.174439038336277,
      "num_tokens": 34045179.0,
      "step": 16405
    },
    {
      "entropy": 5.700400972366333,
      "epoch": 1.4885703918722788,
      "grad_norm": 7.125,
      "learning_rate": 0.0004782631151165262,
      "loss": 5.4377,
      "mean_token_accuracy": 0.16646807342767717,
      "num_tokens": 34056075.0,
      "step": 16410
    },
    {
      "entropy": 5.654512405395508,
      "epoch": 1.4890239477503628,
      "grad_norm": 1.890625,
      "learning_rate": 0.00047824923804866914,
      "loss": 5.5092,
      "mean_token_accuracy": 0.16986994594335555,
      "num_tokens": 34066500.0,
      "step": 16415
    },
    {
      "entropy": 5.756744003295898,
      "epoch": 1.4894775036284469,
      "grad_norm": 3.140625,
      "learning_rate": 0.00047823535677756607,
      "loss": 5.4675,
      "mean_token_accuracy": 0.1662815183401108,
      "num_tokens": 34077266.0,
      "step": 16420
    },
    {
      "entropy": 5.668088912963867,
      "epoch": 1.4899310595065312,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004782214713035041,
      "loss": 5.3731,
      "mean_token_accuracy": 0.17467007040977478,
      "num_tokens": 34088128.0,
      "step": 16425
    },
    {
      "entropy": 5.644625806808472,
      "epoch": 1.4903846153846154,
      "grad_norm": 3.8125,
      "learning_rate": 0.00047820758162677026,
      "loss": 5.4582,
      "mean_token_accuracy": 0.1631221130490303,
      "num_tokens": 34099174.0,
      "step": 16430
    },
    {
      "entropy": 5.6127396583557125,
      "epoch": 1.4908381712626997,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047819368774765187,
      "loss": 5.3482,
      "mean_token_accuracy": 0.16802698224782944,
      "num_tokens": 34108457.0,
      "step": 16435
    },
    {
      "entropy": 5.73252305984497,
      "epoch": 1.4912917271407837,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047817978966643626,
      "loss": 5.4612,
      "mean_token_accuracy": 0.1692676544189453,
      "num_tokens": 34119932.0,
      "step": 16440
    },
    {
      "entropy": 5.6808148384094235,
      "epoch": 1.491745283018868,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004781658873834108,
      "loss": 5.4306,
      "mean_token_accuracy": 0.16816536486148834,
      "num_tokens": 34129967.0,
      "step": 16445
    },
    {
      "entropy": 5.749107885360718,
      "epoch": 1.492198838896952,
      "grad_norm": 2.875,
      "learning_rate": 0.00047815198089886296,
      "loss": 5.5092,
      "mean_token_accuracy": 0.17141935527324675,
      "num_tokens": 34140236.0,
      "step": 16450
    },
    {
      "entropy": 5.722138833999634,
      "epoch": 1.4926523947750363,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004781380702130805,
      "loss": 5.455,
      "mean_token_accuracy": 0.1666231021285057,
      "num_tokens": 34149874.0,
      "step": 16455
    },
    {
      "entropy": 5.722505140304565,
      "epoch": 1.4931059506531206,
      "grad_norm": 2.375,
      "learning_rate": 0.00047812415532635087,
      "loss": 5.3965,
      "mean_token_accuracy": 0.1691577509045601,
      "num_tokens": 34160272.0,
      "step": 16460
    },
    {
      "entropy": 5.75211992263794,
      "epoch": 1.4935595065312046,
      "grad_norm": 2.09375,
      "learning_rate": 0.000478110236238962,
      "loss": 5.488,
      "mean_token_accuracy": 0.16862668097019196,
      "num_tokens": 34170619.0,
      "step": 16465
    },
    {
      "entropy": 5.777650308609009,
      "epoch": 1.4940130624092889,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047809631295120164,
      "loss": 5.5339,
      "mean_token_accuracy": 0.16410468965768815,
      "num_tokens": 34181686.0,
      "step": 16470
    },
    {
      "entropy": 5.706946992874146,
      "epoch": 1.494466618287373,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047808238546335786,
      "loss": 5.4567,
      "mean_token_accuracy": 0.1678525000810623,
      "num_tokens": 34190846.0,
      "step": 16475
    },
    {
      "entropy": 5.680081129074097,
      "epoch": 1.4949201741654572,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047806845377571856,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1786820337176323,
      "num_tokens": 34200522.0,
      "step": 16480
    },
    {
      "entropy": 5.651381111145019,
      "epoch": 1.4953737300435415,
      "grad_norm": 2.90625,
      "learning_rate": 0.00047805451788857187,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1728083908557892,
      "num_tokens": 34210937.0,
      "step": 16485
    },
    {
      "entropy": 5.6201256275177,
      "epoch": 1.4958272859216255,
      "grad_norm": 2.828125,
      "learning_rate": 0.000478040577802206,
      "loss": 5.3886,
      "mean_token_accuracy": 0.17053886502981186,
      "num_tokens": 34221068.0,
      "step": 16490
    },
    {
      "entropy": 5.68856954574585,
      "epoch": 1.4962808417997098,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004780266335169093,
      "loss": 5.4133,
      "mean_token_accuracy": 0.16817647516727446,
      "num_tokens": 34231051.0,
      "step": 16495
    },
    {
      "entropy": 5.6697413444519045,
      "epoch": 1.4967343976777938,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047801268503297,
      "loss": 5.3889,
      "mean_token_accuracy": 0.16423759162425994,
      "num_tokens": 34241071.0,
      "step": 16500
    },
    {
      "entropy": 5.728441572189331,
      "epoch": 1.497187953555878,
      "grad_norm": 3.390625,
      "learning_rate": 0.00047799873235067667,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16582834124565124,
      "num_tokens": 34251175.0,
      "step": 16505
    },
    {
      "entropy": 5.743089818954468,
      "epoch": 1.4976415094339623,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004779847754703178,
      "loss": 5.5005,
      "mean_token_accuracy": 0.170814149081707,
      "num_tokens": 34260865.0,
      "step": 16510
    },
    {
      "entropy": 5.663301992416382,
      "epoch": 1.4980950653120464,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004779708143921821,
      "loss": 5.3791,
      "mean_token_accuracy": 0.17257791459560395,
      "num_tokens": 34271375.0,
      "step": 16515
    },
    {
      "entropy": 5.659026479721069,
      "epoch": 1.4985486211901307,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047795684911655824,
      "loss": 5.3402,
      "mean_token_accuracy": 0.16996244490146636,
      "num_tokens": 34281419.0,
      "step": 16520
    },
    {
      "entropy": 5.710468339920044,
      "epoch": 1.4990021770682147,
      "grad_norm": 2.0,
      "learning_rate": 0.000477942879643735,
      "loss": 5.5029,
      "mean_token_accuracy": 0.15724467784166335,
      "num_tokens": 34292593.0,
      "step": 16525
    },
    {
      "entropy": 5.756539916992187,
      "epoch": 1.499455732946299,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047792890597400125,
      "loss": 5.526,
      "mean_token_accuracy": 0.16448103860020638,
      "num_tokens": 34303045.0,
      "step": 16530
    },
    {
      "entropy": 5.709659147262573,
      "epoch": 1.4999092888243832,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047791492810764604,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1680973395705223,
      "num_tokens": 34313694.0,
      "step": 16535
    },
    {
      "entropy": 5.647908496856689,
      "epoch": 1.5003628447024675,
      "grad_norm": 5.5625,
      "learning_rate": 0.0004779009460449584,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17112871557474135,
      "num_tokens": 34322957.0,
      "step": 16540
    },
    {
      "entropy": 5.722380447387695,
      "epoch": 1.5008164005805515,
      "grad_norm": 3.28125,
      "learning_rate": 0.00047788695978622756,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1714981973171234,
      "num_tokens": 34333408.0,
      "step": 16545
    },
    {
      "entropy": 5.636384105682373,
      "epoch": 1.5012699564586356,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047787296933174265,
      "loss": 5.2845,
      "mean_token_accuracy": 0.17394156903028488,
      "num_tokens": 34342327.0,
      "step": 16550
    },
    {
      "entropy": 5.594474983215332,
      "epoch": 1.5017235123367199,
      "grad_norm": 2.59375,
      "learning_rate": 0.000477858974681793,
      "loss": 5.3796,
      "mean_token_accuracy": 0.17380569130182266,
      "num_tokens": 34352369.0,
      "step": 16555
    },
    {
      "entropy": 5.603248500823975,
      "epoch": 1.5021770682148041,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004778449758366681,
      "loss": 5.3972,
      "mean_token_accuracy": 0.17071246057748796,
      "num_tokens": 34362757.0,
      "step": 16560
    },
    {
      "entropy": 5.789229822158814,
      "epoch": 1.5026306240928884,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004778309727966574,
      "loss": 5.5723,
      "mean_token_accuracy": 0.15734468251466752,
      "num_tokens": 34373123.0,
      "step": 16565
    },
    {
      "entropy": 5.720470809936524,
      "epoch": 1.5030841799709724,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047781696556205045,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17256492078304292,
      "num_tokens": 34383728.0,
      "step": 16570
    },
    {
      "entropy": 5.6303393840789795,
      "epoch": 1.5035377358490565,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047780295413313696,
      "loss": 5.4545,
      "mean_token_accuracy": 0.16159689873456956,
      "num_tokens": 34393172.0,
      "step": 16575
    },
    {
      "entropy": 5.679799652099609,
      "epoch": 1.5039912917271407,
      "grad_norm": 2.875,
      "learning_rate": 0.00047778893851020676,
      "loss": 5.4341,
      "mean_token_accuracy": 0.16752085238695144,
      "num_tokens": 34403987.0,
      "step": 16580
    },
    {
      "entropy": 5.761058235168457,
      "epoch": 1.504444847605225,
      "grad_norm": 2.25,
      "learning_rate": 0.0004777749186935496,
      "loss": 5.3943,
      "mean_token_accuracy": 0.17355616837739946,
      "num_tokens": 34414155.0,
      "step": 16585
    },
    {
      "entropy": 5.670676565170288,
      "epoch": 1.5048984034833093,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004777608946834555,
      "loss": 5.4593,
      "mean_token_accuracy": 0.16313928365707397,
      "num_tokens": 34423942.0,
      "step": 16590
    },
    {
      "entropy": 5.758786058425903,
      "epoch": 1.5053519593613933,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004777468664802144,
      "loss": 5.5509,
      "mean_token_accuracy": 0.16142828166484832,
      "num_tokens": 34435813.0,
      "step": 16595
    },
    {
      "entropy": 5.773189878463745,
      "epoch": 1.5058055152394774,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047773283408411636,
      "loss": 5.4762,
      "mean_token_accuracy": 0.1669888377189636,
      "num_tokens": 34447248.0,
      "step": 16600
    },
    {
      "entropy": 5.783857917785644,
      "epoch": 1.5062590711175616,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047771879749545175,
      "loss": 5.5219,
      "mean_token_accuracy": 0.16516093760728837,
      "num_tokens": 34456870.0,
      "step": 16605
    },
    {
      "entropy": 5.72568907737732,
      "epoch": 1.506712626995646,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047770475671451073,
      "loss": 5.5122,
      "mean_token_accuracy": 0.16612381786108016,
      "num_tokens": 34467712.0,
      "step": 16610
    },
    {
      "entropy": 5.676752710342408,
      "epoch": 1.5071661828737302,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004776907117415837,
      "loss": 5.3996,
      "mean_token_accuracy": 0.1732640728354454,
      "num_tokens": 34478754.0,
      "step": 16615
    },
    {
      "entropy": 5.638779973983764,
      "epoch": 1.5076197387518142,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047767666257696103,
      "loss": 5.3606,
      "mean_token_accuracy": 0.16984761506319046,
      "num_tokens": 34488892.0,
      "step": 16620
    },
    {
      "entropy": 5.7117027759552,
      "epoch": 1.5080732946298983,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047766260922093337,
      "loss": 5.4582,
      "mean_token_accuracy": 0.16984613239765167,
      "num_tokens": 34499577.0,
      "step": 16625
    },
    {
      "entropy": 5.67826509475708,
      "epoch": 1.5085268505079825,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004776485516737913,
      "loss": 5.4709,
      "mean_token_accuracy": 0.1665346145629883,
      "num_tokens": 34510384.0,
      "step": 16630
    },
    {
      "entropy": 5.6024635314941404,
      "epoch": 1.5089804063860668,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004776344899358256,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1745147153735161,
      "num_tokens": 34520083.0,
      "step": 16635
    },
    {
      "entropy": 5.708416938781738,
      "epoch": 1.509433962264151,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004776204240073269,
      "loss": 5.3661,
      "mean_token_accuracy": 0.17059632241725922,
      "num_tokens": 34529332.0,
      "step": 16640
    },
    {
      "entropy": 5.750630950927734,
      "epoch": 1.509887518142235,
      "grad_norm": 1.75,
      "learning_rate": 0.00047760635388858624,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1665619507431984,
      "num_tokens": 34539982.0,
      "step": 16645
    },
    {
      "entropy": 5.679825353622436,
      "epoch": 1.5103410740203191,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047759227957989465,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1644623801112175,
      "num_tokens": 34549854.0,
      "step": 16650
    },
    {
      "entropy": 5.6800689697265625,
      "epoch": 1.5107946298984034,
      "grad_norm": 2.34375,
      "learning_rate": 0.000477578201081543,
      "loss": 5.3253,
      "mean_token_accuracy": 0.17808547765016555,
      "num_tokens": 34559905.0,
      "step": 16655
    },
    {
      "entropy": 5.784164571762085,
      "epoch": 1.5112481857764877,
      "grad_norm": 3.0,
      "learning_rate": 0.0004775641183938225,
      "loss": 5.5469,
      "mean_token_accuracy": 0.16576920300722123,
      "num_tokens": 34571624.0,
      "step": 16660
    },
    {
      "entropy": 5.754335832595825,
      "epoch": 1.511701741654572,
      "grad_norm": 3.734375,
      "learning_rate": 0.00047755003151702443,
      "loss": 5.499,
      "mean_token_accuracy": 0.16954799741506577,
      "num_tokens": 34582647.0,
      "step": 16665
    },
    {
      "entropy": 5.7047600746154785,
      "epoch": 1.512155297532656,
      "grad_norm": 2.125,
      "learning_rate": 0.00047753594045144015,
      "loss": 5.533,
      "mean_token_accuracy": 0.1681974485516548,
      "num_tokens": 34593070.0,
      "step": 16670
    },
    {
      "entropy": 5.756381273269653,
      "epoch": 1.5126088534107403,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004775218451973609,
      "loss": 5.5256,
      "mean_token_accuracy": 0.1656026765704155,
      "num_tokens": 34602349.0,
      "step": 16675
    },
    {
      "entropy": 5.660476446151733,
      "epoch": 1.5130624092888243,
      "grad_norm": 2.921875,
      "learning_rate": 0.0004775077457550784,
      "loss": 5.438,
      "mean_token_accuracy": 0.1652193248271942,
      "num_tokens": 34611552.0,
      "step": 16680
    },
    {
      "entropy": 5.684771537780762,
      "epoch": 1.5135159651669086,
      "grad_norm": 2.171875,
      "learning_rate": 0.00047749364212488403,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16842350661754607,
      "num_tokens": 34622556.0,
      "step": 16685
    },
    {
      "entropy": 5.763314151763916,
      "epoch": 1.5139695210449928,
      "grad_norm": 3.34375,
      "learning_rate": 0.0004774795343070696,
      "loss": 5.4658,
      "mean_token_accuracy": 0.16854284852743148,
      "num_tokens": 34633240.0,
      "step": 16690
    },
    {
      "entropy": 5.727551889419556,
      "epoch": 1.5144230769230769,
      "grad_norm": 5.09375,
      "learning_rate": 0.0004774654223019268,
      "loss": 5.5061,
      "mean_token_accuracy": 0.16360394656658173,
      "num_tokens": 34643813.0,
      "step": 16695
    },
    {
      "entropy": 5.717952489852905,
      "epoch": 1.5148766328011611,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004774513061097474,
      "loss": 5.4083,
      "mean_token_accuracy": 0.16538808941841127,
      "num_tokens": 34654328.0,
      "step": 16700
    },
    {
      "entropy": 5.632085847854614,
      "epoch": 1.5153301886792452,
      "grad_norm": 2.0,
      "learning_rate": 0.0004774371857308235,
      "loss": 5.3832,
      "mean_token_accuracy": 0.17550472766160966,
      "num_tokens": 34664048.0,
      "step": 16705
    },
    {
      "entropy": 5.760972213745117,
      "epoch": 1.5157837445573294,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047742306116544693,
      "loss": 5.5248,
      "mean_token_accuracy": 0.16964975446462632,
      "num_tokens": 34675136.0,
      "step": 16710
    },
    {
      "entropy": 5.764431047439575,
      "epoch": 1.5162373004354137,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047740893241390994,
      "loss": 5.4269,
      "mean_token_accuracy": 0.16744859665632247,
      "num_tokens": 34686423.0,
      "step": 16715
    },
    {
      "entropy": 5.6521148681640625,
      "epoch": 1.516690856313498,
      "grad_norm": 4.875,
      "learning_rate": 0.0004773947994765046,
      "loss": 5.4363,
      "mean_token_accuracy": 0.17171479761600494,
      "num_tokens": 34696730.0,
      "step": 16720
    },
    {
      "entropy": 5.684759330749512,
      "epoch": 1.517144412191582,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004773806623535232,
      "loss": 5.5031,
      "mean_token_accuracy": 0.1669832155108452,
      "num_tokens": 34705891.0,
      "step": 16725
    },
    {
      "entropy": 5.82213249206543,
      "epoch": 1.517597968069666,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047736652104525816,
      "loss": 5.5295,
      "mean_token_accuracy": 0.167432564496994,
      "num_tokens": 34717098.0,
      "step": 16730
    },
    {
      "entropy": 5.6095939636230465,
      "epoch": 1.5180515239477503,
      "grad_norm": 2.25,
      "learning_rate": 0.0004773523755520019,
      "loss": 5.3372,
      "mean_token_accuracy": 0.17114016711711882,
      "num_tokens": 34727351.0,
      "step": 16735
    },
    {
      "entropy": 5.692414140701294,
      "epoch": 1.5185050798258346,
      "grad_norm": 2.65625,
      "learning_rate": 0.000477338225874047,
      "loss": 5.4927,
      "mean_token_accuracy": 0.163038969039917,
      "num_tokens": 34737733.0,
      "step": 16740
    },
    {
      "entropy": 5.71074481010437,
      "epoch": 1.5189586357039189,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047732407201168595,
      "loss": 5.4143,
      "mean_token_accuracy": 0.17393598854541778,
      "num_tokens": 34748094.0,
      "step": 16745
    },
    {
      "entropy": 5.7302428722381595,
      "epoch": 1.519412191582003,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004773099139652116,
      "loss": 5.4309,
      "mean_token_accuracy": 0.17427827268838883,
      "num_tokens": 34757896.0,
      "step": 16750
    },
    {
      "entropy": 5.731365966796875,
      "epoch": 1.519865747460087,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004772957517349167,
      "loss": 5.4343,
      "mean_token_accuracy": 0.158809694647789,
      "num_tokens": 34768878.0,
      "step": 16755
    },
    {
      "entropy": 5.771361494064331,
      "epoch": 1.5203193033381712,
      "grad_norm": 2.125,
      "learning_rate": 0.0004772815853210939,
      "loss": 5.6066,
      "mean_token_accuracy": 0.16290359795093537,
      "num_tokens": 34780072.0,
      "step": 16760
    },
    {
      "entropy": 5.723509645462036,
      "epoch": 1.5207728592162555,
      "grad_norm": 1.9375,
      "learning_rate": 0.00047726741472403655,
      "loss": 5.5186,
      "mean_token_accuracy": 0.16244877129793167,
      "num_tokens": 34790227.0,
      "step": 16765
    },
    {
      "entropy": 5.712742280960083,
      "epoch": 1.5212264150943398,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047725323994403753,
      "loss": 5.3806,
      "mean_token_accuracy": 0.16430634558200835,
      "num_tokens": 34802839.0,
      "step": 16770
    },
    {
      "entropy": 5.6984233379364015,
      "epoch": 1.5216799709724238,
      "grad_norm": 2.921875,
      "learning_rate": 0.00047723906098138993,
      "loss": 5.3569,
      "mean_token_accuracy": 0.17754784226417542,
      "num_tokens": 34812906.0,
      "step": 16775
    },
    {
      "entropy": 5.622565984725952,
      "epoch": 1.5221335268505078,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047722487783638697,
      "loss": 5.4205,
      "mean_token_accuracy": 0.16609074026346207,
      "num_tokens": 34824349.0,
      "step": 16780
    },
    {
      "entropy": 5.626798391342163,
      "epoch": 1.5225870827285921,
      "grad_norm": 2.515625,
      "learning_rate": 0.000477210690509322,
      "loss": 5.4628,
      "mean_token_accuracy": 0.16527538299560546,
      "num_tokens": 34834355.0,
      "step": 16785
    },
    {
      "entropy": 5.7673022747039795,
      "epoch": 1.5230406386066764,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047719649900048843,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1711397275328636,
      "num_tokens": 34844549.0,
      "step": 16790
    },
    {
      "entropy": 5.659734582901001,
      "epoch": 1.5234941944847606,
      "grad_norm": 3.765625,
      "learning_rate": 0.00047718230331017975,
      "loss": 5.3741,
      "mean_token_accuracy": 0.1733274281024933,
      "num_tokens": 34855592.0,
      "step": 16795
    },
    {
      "entropy": 5.681205558776855,
      "epoch": 1.5239477503628447,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047716810343868956,
      "loss": 5.3912,
      "mean_token_accuracy": 0.1738739565014839,
      "num_tokens": 34865630.0,
      "step": 16800
    },
    {
      "entropy": 5.69542293548584,
      "epoch": 1.5244013062409287,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047715389938631143,
      "loss": 5.4551,
      "mean_token_accuracy": 0.16848929971456528,
      "num_tokens": 34876984.0,
      "step": 16805
    },
    {
      "entropy": 5.707873153686523,
      "epoch": 1.524854862119013,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047713969115333916,
      "loss": 5.4036,
      "mean_token_accuracy": 0.16461460143327714,
      "num_tokens": 34888112.0,
      "step": 16810
    },
    {
      "entropy": 5.73502516746521,
      "epoch": 1.5253084179970973,
      "grad_norm": 2.671875,
      "learning_rate": 0.0004771254787400666,
      "loss": 5.5002,
      "mean_token_accuracy": 0.16869113147258757,
      "num_tokens": 34897565.0,
      "step": 16815
    },
    {
      "entropy": 5.800412321090699,
      "epoch": 1.5257619738751815,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004771112621467876,
      "loss": 5.5895,
      "mean_token_accuracy": 0.16177970468997954,
      "num_tokens": 34908588.0,
      "step": 16820
    },
    {
      "entropy": 5.759751796722412,
      "epoch": 1.5262155297532656,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004770970413737961,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1671668216586113,
      "num_tokens": 34919684.0,
      "step": 16825
    },
    {
      "entropy": 5.697743368148804,
      "epoch": 1.5266690856313496,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004770828164213865,
      "loss": 5.5033,
      "mean_token_accuracy": 0.16309723556041716,
      "num_tokens": 34929680.0,
      "step": 16830
    },
    {
      "entropy": 5.721653652191162,
      "epoch": 1.5271226415094339,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004770685872898525,
      "loss": 5.455,
      "mean_token_accuracy": 0.16253614276647568,
      "num_tokens": 34940394.0,
      "step": 16835
    },
    {
      "entropy": 5.690266275405884,
      "epoch": 1.5275761973875182,
      "grad_norm": 2.125,
      "learning_rate": 0.00047705435397948876,
      "loss": 5.4041,
      "mean_token_accuracy": 0.1635429322719574,
      "num_tokens": 34951000.0,
      "step": 16840
    },
    {
      "entropy": 5.687080812454224,
      "epoch": 1.5280297532656024,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004770401164905895,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1557669922709465,
      "num_tokens": 34961596.0,
      "step": 16845
    },
    {
      "entropy": 5.787659692764282,
      "epoch": 1.5284833091436865,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047702587482344915,
      "loss": 5.4632,
      "mean_token_accuracy": 0.16702454537153244,
      "num_tokens": 34971659.0,
      "step": 16850
    },
    {
      "entropy": 5.6980156898498535,
      "epoch": 1.5289368650217707,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004770116289783622,
      "loss": 5.3992,
      "mean_token_accuracy": 0.17204674929380417,
      "num_tokens": 34982107.0,
      "step": 16855
    },
    {
      "entropy": 5.655812549591064,
      "epoch": 1.5293904208998548,
      "grad_norm": 4.71875,
      "learning_rate": 0.0004769973789556233,
      "loss": 5.378,
      "mean_token_accuracy": 0.17382519841194152,
      "num_tokens": 34992739.0,
      "step": 16860
    },
    {
      "entropy": 5.670818138122558,
      "epoch": 1.529843976777939,
      "grad_norm": 2.625,
      "learning_rate": 0.00047698312475552714,
      "loss": 5.4146,
      "mean_token_accuracy": 0.16607557684183122,
      "num_tokens": 35002860.0,
      "step": 16865
    },
    {
      "entropy": 5.637082195281982,
      "epoch": 1.5302975326560233,
      "grad_norm": 2.75,
      "learning_rate": 0.0004769688663783684,
      "loss": 5.4277,
      "mean_token_accuracy": 0.17069706469774246,
      "num_tokens": 35013761.0,
      "step": 16870
    },
    {
      "entropy": 5.69539966583252,
      "epoch": 1.5307510885341074,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004769546038244421,
      "loss": 5.3885,
      "mean_token_accuracy": 0.1676061898469925,
      "num_tokens": 35023857.0,
      "step": 16875
    },
    {
      "entropy": 5.7057671546936035,
      "epoch": 1.5312046444121916,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047694033709404313,
      "loss": 5.4511,
      "mean_token_accuracy": 0.16422806680202484,
      "num_tokens": 35034471.0,
      "step": 16880
    },
    {
      "entropy": 5.682156276702881,
      "epoch": 1.5316582002902757,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047692606618746655,
      "loss": 5.4075,
      "mean_token_accuracy": 0.1719357967376709,
      "num_tokens": 35044430.0,
      "step": 16885
    },
    {
      "entropy": 5.685761260986328,
      "epoch": 1.53211175616836,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047691179110500735,
      "loss": 5.3705,
      "mean_token_accuracy": 0.17136194854974746,
      "num_tokens": 35054828.0,
      "step": 16890
    },
    {
      "entropy": 5.630874872207642,
      "epoch": 1.5325653120464442,
      "grad_norm": 3.171875,
      "learning_rate": 0.00047689751184696097,
      "loss": 5.3706,
      "mean_token_accuracy": 0.1701495200395584,
      "num_tokens": 35064305.0,
      "step": 16895
    },
    {
      "entropy": 5.68676986694336,
      "epoch": 1.5330188679245285,
      "grad_norm": 8.1875,
      "learning_rate": 0.0004768832284136226,
      "loss": 5.4751,
      "mean_token_accuracy": 0.16468275487422943,
      "num_tokens": 35074778.0,
      "step": 16900
    },
    {
      "entropy": 5.5829380512237545,
      "epoch": 1.5334724238026125,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004768689408052876,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17302821278572084,
      "num_tokens": 35085750.0,
      "step": 16905
    },
    {
      "entropy": 5.675378274917603,
      "epoch": 1.5339259796806965,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047685464902225144,
      "loss": 5.4207,
      "mean_token_accuracy": 0.16876254379749298,
      "num_tokens": 35096624.0,
      "step": 16910
    },
    {
      "entropy": 5.669952821731568,
      "epoch": 1.5343795355587808,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004768403530648097,
      "loss": 5.4622,
      "mean_token_accuracy": 0.1705664202570915,
      "num_tokens": 35106516.0,
      "step": 16915
    },
    {
      "entropy": 5.6536195278167725,
      "epoch": 1.534833091436865,
      "grad_norm": 2.53125,
      "learning_rate": 0.00047682605293325806,
      "loss": 5.3992,
      "mean_token_accuracy": 0.1753559947013855,
      "num_tokens": 35117264.0,
      "step": 16920
    },
    {
      "entropy": 5.724627494812012,
      "epoch": 1.5352866473149493,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047681174862789215,
      "loss": 5.3485,
      "mean_token_accuracy": 0.16864703893661498,
      "num_tokens": 35127364.0,
      "step": 16925
    },
    {
      "entropy": 5.692388200759888,
      "epoch": 1.5357402031930334,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047679744014900795,
      "loss": 5.4234,
      "mean_token_accuracy": 0.17041468620300293,
      "num_tokens": 35138824.0,
      "step": 16930
    },
    {
      "entropy": 5.690200853347778,
      "epoch": 1.5361937590711174,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004767831274969011,
      "loss": 5.3736,
      "mean_token_accuracy": 0.1764376327395439,
      "num_tokens": 35148739.0,
      "step": 16935
    },
    {
      "entropy": 5.785953950881958,
      "epoch": 1.5366473149492017,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047676881067186793,
      "loss": 5.5516,
      "mean_token_accuracy": 0.1664319410920143,
      "num_tokens": 35158549.0,
      "step": 16940
    },
    {
      "entropy": 5.712868499755859,
      "epoch": 1.537100870827286,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047675448967420426,
      "loss": 5.4525,
      "mean_token_accuracy": 0.17490314692258835,
      "num_tokens": 35169355.0,
      "step": 16945
    },
    {
      "entropy": 5.6912370204925535,
      "epoch": 1.5375544267053702,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004767401645042064,
      "loss": 5.4234,
      "mean_token_accuracy": 0.17766340225934982,
      "num_tokens": 35179792.0,
      "step": 16950
    },
    {
      "entropy": 5.659514141082764,
      "epoch": 1.5380079825834543,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047672583516217043,
      "loss": 5.4302,
      "mean_token_accuracy": 0.17204910963773729,
      "num_tokens": 35190607.0,
      "step": 16955
    },
    {
      "entropy": 5.751991176605225,
      "epoch": 1.5384615384615383,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004767115016483928,
      "loss": 5.4297,
      "mean_token_accuracy": 0.16844416558742523,
      "num_tokens": 35201689.0,
      "step": 16960
    },
    {
      "entropy": 5.689592885971069,
      "epoch": 1.5389150943396226,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004766971639631699,
      "loss": 5.415,
      "mean_token_accuracy": 0.16609276384115218,
      "num_tokens": 35211920.0,
      "step": 16965
    },
    {
      "entropy": 5.667367362976075,
      "epoch": 1.5393686502177069,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047668282210679825,
      "loss": 5.4147,
      "mean_token_accuracy": 0.16604392379522323,
      "num_tokens": 35222135.0,
      "step": 16970
    },
    {
      "entropy": 5.728549861907959,
      "epoch": 1.5398222060957911,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004766684760795745,
      "loss": 5.4704,
      "mean_token_accuracy": 0.1678948864340782,
      "num_tokens": 35232753.0,
      "step": 16975
    },
    {
      "entropy": 5.6695249557495115,
      "epoch": 1.5402757619738752,
      "grad_norm": 2.78125,
      "learning_rate": 0.00047665412588179524,
      "loss": 5.3462,
      "mean_token_accuracy": 0.1710069477558136,
      "num_tokens": 35243317.0,
      "step": 16980
    },
    {
      "entropy": 5.760976552963257,
      "epoch": 1.5407293178519592,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004766397715137573,
      "loss": 5.5778,
      "mean_token_accuracy": 0.15609656274318695,
      "num_tokens": 35254448.0,
      "step": 16985
    },
    {
      "entropy": 5.691459989547729,
      "epoch": 1.5411828737300435,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047662541297575743,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1729973316192627,
      "num_tokens": 35265099.0,
      "step": 16990
    },
    {
      "entropy": 5.712511396408081,
      "epoch": 1.5416364296081277,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047661105026809274,
      "loss": 5.4269,
      "mean_token_accuracy": 0.17425011843442917,
      "num_tokens": 35275080.0,
      "step": 16995
    },
    {
      "entropy": 5.644726324081421,
      "epoch": 1.542089985486212,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004765966833910601,
      "loss": 5.4352,
      "mean_token_accuracy": 0.1625493660569191,
      "num_tokens": 35285200.0,
      "step": 17000
    },
    {
      "entropy": 5.721442937850952,
      "epoch": 1.542543541364296,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004765823123449567,
      "loss": 5.4719,
      "mean_token_accuracy": 0.17046553492546082,
      "num_tokens": 35296033.0,
      "step": 17005
    },
    {
      "entropy": 5.802887010574341,
      "epoch": 1.54299709724238,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047656793713007965,
      "loss": 5.5472,
      "mean_token_accuracy": 0.1604733645915985,
      "num_tokens": 35307062.0,
      "step": 17010
    },
    {
      "entropy": 5.753088331222534,
      "epoch": 1.5434506531204644,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047655355774672636,
      "loss": 5.3936,
      "mean_token_accuracy": 0.1739861086010933,
      "num_tokens": 35316509.0,
      "step": 17015
    },
    {
      "entropy": 5.6459160327911375,
      "epoch": 1.5439042089985486,
      "grad_norm": 2.796875,
      "learning_rate": 0.00047653917419519413,
      "loss": 5.4349,
      "mean_token_accuracy": 0.16793774515390397,
      "num_tokens": 35327138.0,
      "step": 17020
    },
    {
      "entropy": 5.670724582672119,
      "epoch": 1.544357764876633,
      "grad_norm": 4.46875,
      "learning_rate": 0.0004765247864757804,
      "loss": 5.3823,
      "mean_token_accuracy": 0.1727299213409424,
      "num_tokens": 35338297.0,
      "step": 17025
    },
    {
      "entropy": 5.628982734680176,
      "epoch": 1.544811320754717,
      "grad_norm": 3.734375,
      "learning_rate": 0.0004765103945887827,
      "loss": 5.3889,
      "mean_token_accuracy": 0.1639770671725273,
      "num_tokens": 35348070.0,
      "step": 17030
    },
    {
      "entropy": 5.781213188171387,
      "epoch": 1.5452648766328012,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004764959985344987,
      "loss": 5.5562,
      "mean_token_accuracy": 0.163591930270195,
      "num_tokens": 35360025.0,
      "step": 17035
    },
    {
      "entropy": 5.654095411300659,
      "epoch": 1.5457184325108853,
      "grad_norm": 4.25,
      "learning_rate": 0.0004764815983132261,
      "loss": 5.3381,
      "mean_token_accuracy": 0.17117571383714675,
      "num_tokens": 35369831.0,
      "step": 17040
    },
    {
      "entropy": 5.651568078994751,
      "epoch": 1.5461719883889695,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004764671939252628,
      "loss": 5.4243,
      "mean_token_accuracy": 0.16662895530462266,
      "num_tokens": 35379878.0,
      "step": 17045
    },
    {
      "entropy": 5.777516508102417,
      "epoch": 1.5466255442670538,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004764527853709064,
      "loss": 5.5172,
      "mean_token_accuracy": 0.16204022467136384,
      "num_tokens": 35390307.0,
      "step": 17050
    },
    {
      "entropy": 5.75959587097168,
      "epoch": 1.5470791001451378,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004764383726504552,
      "loss": 5.4875,
      "mean_token_accuracy": 0.17194852977991104,
      "num_tokens": 35400932.0,
      "step": 17055
    },
    {
      "entropy": 5.647706174850464,
      "epoch": 1.547532656023222,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004764239557642071,
      "loss": 5.3634,
      "mean_token_accuracy": 0.17478299289941787,
      "num_tokens": 35409973.0,
      "step": 17060
    },
    {
      "entropy": 5.717527914047241,
      "epoch": 1.5479862119013061,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004764095347124603,
      "loss": 5.4213,
      "mean_token_accuracy": 0.171760655939579,
      "num_tokens": 35419672.0,
      "step": 17065
    },
    {
      "entropy": 5.688251066207886,
      "epoch": 1.5484397677793904,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004763951094955129,
      "loss": 5.4742,
      "mean_token_accuracy": 0.16905272156000137,
      "num_tokens": 35430481.0,
      "step": 17070
    },
    {
      "entropy": 5.68491940498352,
      "epoch": 1.5488933236574747,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004763806801136634,
      "loss": 5.4033,
      "mean_token_accuracy": 0.1757503405213356,
      "num_tokens": 35440898.0,
      "step": 17075
    },
    {
      "entropy": 5.6837489128112795,
      "epoch": 1.549346879535559,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004763662465672101,
      "loss": 5.4158,
      "mean_token_accuracy": 0.16780422031879424,
      "num_tokens": 35452078.0,
      "step": 17080
    },
    {
      "entropy": 5.801316499710083,
      "epoch": 1.549800435413643,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047635180885645146,
      "loss": 5.478,
      "mean_token_accuracy": 0.16310373917222024,
      "num_tokens": 35461515.0,
      "step": 17085
    },
    {
      "entropy": 5.620781421661377,
      "epoch": 1.550253991291727,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004763373669816862,
      "loss": 5.3297,
      "mean_token_accuracy": 0.1805766150355339,
      "num_tokens": 35471777.0,
      "step": 17090
    },
    {
      "entropy": 5.697113656997681,
      "epoch": 1.5507075471698113,
      "grad_norm": 4.1875,
      "learning_rate": 0.0004763229209432128,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16577430665493012,
      "num_tokens": 35481694.0,
      "step": 17095
    },
    {
      "entropy": 5.7074041843414305,
      "epoch": 1.5511611030478956,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004763084707413301,
      "loss": 5.4487,
      "mean_token_accuracy": 0.16644046306610108,
      "num_tokens": 35492016.0,
      "step": 17100
    },
    {
      "entropy": 5.665201807022095,
      "epoch": 1.5516146589259798,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047629401637633696,
      "loss": 5.4026,
      "mean_token_accuracy": 0.16688399016857147,
      "num_tokens": 35501565.0,
      "step": 17105
    },
    {
      "entropy": 5.660561609268188,
      "epoch": 1.5520682148040639,
      "grad_norm": 3.640625,
      "learning_rate": 0.00047627955784853226,
      "loss": 5.4208,
      "mean_token_accuracy": 0.17635941803455352,
      "num_tokens": 35511389.0,
      "step": 17110
    },
    {
      "entropy": 5.627258586883545,
      "epoch": 1.552521770682148,
      "grad_norm": 2.09375,
      "learning_rate": 0.000476265095158215,
      "loss": 5.3855,
      "mean_token_accuracy": 0.1755094587802887,
      "num_tokens": 35522172.0,
      "step": 17115
    },
    {
      "entropy": 5.778671216964722,
      "epoch": 1.5529753265602322,
      "grad_norm": 2.5,
      "learning_rate": 0.0004762506283056843,
      "loss": 5.4401,
      "mean_token_accuracy": 0.16567622721195222,
      "num_tokens": 35531746.0,
      "step": 17120
    },
    {
      "entropy": 5.68129415512085,
      "epoch": 1.5534288824383164,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004762361572912393,
      "loss": 5.3961,
      "mean_token_accuracy": 0.17190259248018264,
      "num_tokens": 35540664.0,
      "step": 17125
    },
    {
      "entropy": 5.696501207351685,
      "epoch": 1.5538824383164007,
      "grad_norm": 2.953125,
      "learning_rate": 0.00047622168211517926,
      "loss": 5.4776,
      "mean_token_accuracy": 0.161577457934618,
      "num_tokens": 35552538.0,
      "step": 17130
    },
    {
      "entropy": 5.715075540542602,
      "epoch": 1.5543359941944848,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047620720277780353,
      "loss": 5.4922,
      "mean_token_accuracy": 0.16228119283914566,
      "num_tokens": 35562890.0,
      "step": 17135
    },
    {
      "entropy": 5.773793411254883,
      "epoch": 1.5547895500725688,
      "grad_norm": 1.921875,
      "learning_rate": 0.00047619271927941165,
      "loss": 5.4931,
      "mean_token_accuracy": 0.16270262449979783,
      "num_tokens": 35573333.0,
      "step": 17140
    },
    {
      "entropy": 5.665373754501343,
      "epoch": 1.555243105950653,
      "grad_norm": 2.015625,
      "learning_rate": 0.000476178231620303,
      "loss": 5.3888,
      "mean_token_accuracy": 0.16799750477075576,
      "num_tokens": 35582776.0,
      "step": 17145
    },
    {
      "entropy": 5.721207761764527,
      "epoch": 1.5556966618287373,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004761637398007774,
      "loss": 5.479,
      "mean_token_accuracy": 0.16581505984067918,
      "num_tokens": 35593853.0,
      "step": 17150
    },
    {
      "entropy": 5.757103776931762,
      "epoch": 1.5561502177068216,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004761492438211342,
      "loss": 5.5238,
      "mean_token_accuracy": 0.16644657254219056,
      "num_tokens": 35604304.0,
      "step": 17155
    },
    {
      "entropy": 5.757397985458374,
      "epoch": 1.5566037735849056,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047613474368167355,
      "loss": 5.5566,
      "mean_token_accuracy": 0.16076782643795012,
      "num_tokens": 35614624.0,
      "step": 17160
    },
    {
      "entropy": 5.733210945129395,
      "epoch": 1.5570573294629897,
      "grad_norm": 1.890625,
      "learning_rate": 0.000476120239382695,
      "loss": 5.3958,
      "mean_token_accuracy": 0.17180591970682144,
      "num_tokens": 35624446.0,
      "step": 17165
    },
    {
      "entropy": 5.720513248443604,
      "epoch": 1.557510885341074,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047610573092449877,
      "loss": 5.6078,
      "mean_token_accuracy": 0.15969871431589128,
      "num_tokens": 35635046.0,
      "step": 17170
    },
    {
      "entropy": 5.762639093399048,
      "epoch": 1.5579644412191582,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004760912183073847,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16793945282697678,
      "num_tokens": 35645653.0,
      "step": 17175
    },
    {
      "entropy": 5.642586040496826,
      "epoch": 1.5584179970972425,
      "grad_norm": 2.34375,
      "learning_rate": 0.000476076701531653,
      "loss": 5.4048,
      "mean_token_accuracy": 0.17082022726535798,
      "num_tokens": 35655401.0,
      "step": 17180
    },
    {
      "entropy": 5.613077640533447,
      "epoch": 1.5588715529753265,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004760621805976039,
      "loss": 5.3428,
      "mean_token_accuracy": 0.17206007838249207,
      "num_tokens": 35666297.0,
      "step": 17185
    },
    {
      "entropy": 5.6673197746276855,
      "epoch": 1.5593251088534106,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047604765550553766,
      "loss": 5.4101,
      "mean_token_accuracy": 0.1692783236503601,
      "num_tokens": 35676583.0,
      "step": 17190
    },
    {
      "entropy": 5.754846668243408,
      "epoch": 1.5597786647314948,
      "grad_norm": 3.671875,
      "learning_rate": 0.00047603312625575466,
      "loss": 5.4469,
      "mean_token_accuracy": 0.17130614519119264,
      "num_tokens": 35686982.0,
      "step": 17195
    },
    {
      "entropy": 5.710036134719848,
      "epoch": 1.5602322206095791,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047601859284855544,
      "loss": 5.4506,
      "mean_token_accuracy": 0.16865867525339126,
      "num_tokens": 35697292.0,
      "step": 17200
    },
    {
      "entropy": 5.681497812271118,
      "epoch": 1.5606857764876634,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004760040552842403,
      "loss": 5.4157,
      "mean_token_accuracy": 0.17594456374645234,
      "num_tokens": 35707332.0,
      "step": 17205
    },
    {
      "entropy": 5.678307723999024,
      "epoch": 1.5611393323657474,
      "grad_norm": 1.75,
      "learning_rate": 0.00047598951356311035,
      "loss": 5.4517,
      "mean_token_accuracy": 0.16798876076936722,
      "num_tokens": 35716614.0,
      "step": 17210
    },
    {
      "entropy": 5.677727222442627,
      "epoch": 1.5615928882438317,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004759749676854658,
      "loss": 5.3398,
      "mean_token_accuracy": 0.1732966423034668,
      "num_tokens": 35727624.0,
      "step": 17215
    },
    {
      "entropy": 5.756965923309326,
      "epoch": 1.5620464441219157,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004759604176516079,
      "loss": 5.5166,
      "mean_token_accuracy": 0.16142414063215255,
      "num_tokens": 35738134.0,
      "step": 17220
    },
    {
      "entropy": 5.6659383296966555,
      "epoch": 1.5625,
      "grad_norm": 1.875,
      "learning_rate": 0.0004759458634618372,
      "loss": 5.3615,
      "mean_token_accuracy": 0.171909399330616,
      "num_tokens": 35749384.0,
      "step": 17225
    },
    {
      "entropy": 5.682888507843018,
      "epoch": 1.5629535558780843,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004759313051164549,
      "loss": 5.4487,
      "mean_token_accuracy": 0.15994218438863755,
      "num_tokens": 35761603.0,
      "step": 17230
    },
    {
      "entropy": 5.680469274520874,
      "epoch": 1.5634071117561683,
      "grad_norm": 2.921875,
      "learning_rate": 0.000475916742615762,
      "loss": 5.3988,
      "mean_token_accuracy": 0.16703563183546066,
      "num_tokens": 35772376.0,
      "step": 17235
    },
    {
      "entropy": 5.727883625030517,
      "epoch": 1.5638606676342526,
      "grad_norm": 2.953125,
      "learning_rate": 0.0004759021759600597,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1676334857940674,
      "num_tokens": 35782388.0,
      "step": 17240
    },
    {
      "entropy": 5.689809799194336,
      "epoch": 1.5643142235123366,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004758876051496492,
      "loss": 5.3801,
      "mean_token_accuracy": 0.17086065858602523,
      "num_tokens": 35792073.0,
      "step": 17245
    },
    {
      "entropy": 5.640148401260376,
      "epoch": 1.5647677793904209,
      "grad_norm": 2.5,
      "learning_rate": 0.00047587303018483177,
      "loss": 5.3479,
      "mean_token_accuracy": 0.1715352267026901,
      "num_tokens": 35802383.0,
      "step": 17250
    },
    {
      "entropy": 5.633788633346557,
      "epoch": 1.5652213352685052,
      "grad_norm": 2.0,
      "learning_rate": 0.00047585845106590897,
      "loss": 5.4226,
      "mean_token_accuracy": 0.17237827628850938,
      "num_tokens": 35812294.0,
      "step": 17255
    },
    {
      "entropy": 5.7375589370727536,
      "epoch": 1.5656748911465894,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00047584386779318214,
      "loss": 5.488,
      "mean_token_accuracy": 0.16951687037944793,
      "num_tokens": 35821015.0,
      "step": 17260
    },
    {
      "entropy": 5.728954792022705,
      "epoch": 1.5661284470246735,
      "grad_norm": 2.28125,
      "learning_rate": 0.000475829280366953,
      "loss": 5.4136,
      "mean_token_accuracy": 0.16883931905031205,
      "num_tokens": 35831516.0,
      "step": 17265
    },
    {
      "entropy": 5.7147908210754395,
      "epoch": 1.5665820029027575,
      "grad_norm": 2.671875,
      "learning_rate": 0.00047581468878752313,
      "loss": 5.461,
      "mean_token_accuracy": 0.16669767796993257,
      "num_tokens": 35840873.0,
      "step": 17270
    },
    {
      "entropy": 5.687708377838135,
      "epoch": 1.5670355587808418,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047580009305519427,
      "loss": 5.4937,
      "mean_token_accuracy": 0.16910629719495773,
      "num_tokens": 35851305.0,
      "step": 17275
    },
    {
      "entropy": 5.796312427520752,
      "epoch": 1.567489114658926,
      "grad_norm": 2.375,
      "learning_rate": 0.0004757854931702684,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1620602287352085,
      "num_tokens": 35863189.0,
      "step": 17280
    },
    {
      "entropy": 5.631219434738159,
      "epoch": 1.5679426705370103,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047577088913304733,
      "loss": 5.3387,
      "mean_token_accuracy": 0.17487384378910065,
      "num_tokens": 35873414.0,
      "step": 17285
    },
    {
      "entropy": 5.6035912990570065,
      "epoch": 1.5683962264150944,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047575628094383317,
      "loss": 5.3881,
      "mean_token_accuracy": 0.16952622383832933,
      "num_tokens": 35884030.0,
      "step": 17290
    },
    {
      "entropy": 5.67272162437439,
      "epoch": 1.5688497822931784,
      "grad_norm": 2.640625,
      "learning_rate": 0.0004757416686029279,
      "loss": 5.5046,
      "mean_token_accuracy": 0.15673063099384307,
      "num_tokens": 35895963.0,
      "step": 17295
    },
    {
      "entropy": 5.697833013534546,
      "epoch": 1.5693033381712627,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004757270521106338,
      "loss": 5.3647,
      "mean_token_accuracy": 0.17482858896255493,
      "num_tokens": 35906578.0,
      "step": 17300
    },
    {
      "entropy": 5.774962663650513,
      "epoch": 1.569756894049347,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004757124314672531,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16350427716970445,
      "num_tokens": 35917377.0,
      "step": 17305
    },
    {
      "entropy": 5.6587018966674805,
      "epoch": 1.5702104499274312,
      "grad_norm": 1.859375,
      "learning_rate": 0.00047569780667308816,
      "loss": 5.4039,
      "mean_token_accuracy": 0.17096969187259675,
      "num_tokens": 35927269.0,
      "step": 17310
    },
    {
      "entropy": 5.6763194561004635,
      "epoch": 1.5706640058055152,
      "grad_norm": 2.0,
      "learning_rate": 0.00047568317772844147,
      "loss": 5.5104,
      "mean_token_accuracy": 0.16222217977046965,
      "num_tokens": 35938047.0,
      "step": 17315
    },
    {
      "entropy": 5.808438348770141,
      "epoch": 1.5711175616835993,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047566854463361544,
      "loss": 5.4713,
      "mean_token_accuracy": 0.15766119807958603,
      "num_tokens": 35948929.0,
      "step": 17320
    },
    {
      "entropy": 5.786431550979614,
      "epoch": 1.5715711175616836,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004756539073889128,
      "loss": 5.4779,
      "mean_token_accuracy": 0.16770453453063966,
      "num_tokens": 35959918.0,
      "step": 17325
    },
    {
      "entropy": 5.6479331970214846,
      "epoch": 1.5720246734397678,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047563926599463623,
      "loss": 5.4574,
      "mean_token_accuracy": 0.17489692121744155,
      "num_tokens": 35970659.0,
      "step": 17330
    },
    {
      "entropy": 5.645398330688477,
      "epoch": 1.572478229317852,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004756246204510886,
      "loss": 5.3979,
      "mean_token_accuracy": 0.17345380336046218,
      "num_tokens": 35980359.0,
      "step": 17335
    },
    {
      "entropy": 5.8165946960449215,
      "epoch": 1.5729317851959361,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047560997075857253,
      "loss": 5.589,
      "mean_token_accuracy": 0.16083628982305526,
      "num_tokens": 35990521.0,
      "step": 17340
    },
    {
      "entropy": 5.791608381271362,
      "epoch": 1.5733853410740202,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004755953169173913,
      "loss": 5.5276,
      "mean_token_accuracy": 0.16119582802057267,
      "num_tokens": 36000906.0,
      "step": 17345
    },
    {
      "entropy": 5.652381801605225,
      "epoch": 1.5738388969521044,
      "grad_norm": 10.75,
      "learning_rate": 0.0004755806589278477,
      "loss": 5.3703,
      "mean_token_accuracy": 0.16636275202035905,
      "num_tokens": 36012255.0,
      "step": 17350
    },
    {
      "entropy": 5.733880710601807,
      "epoch": 1.5742924528301887,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047556599679024497,
      "loss": 5.374,
      "mean_token_accuracy": 0.171751070022583,
      "num_tokens": 36022563.0,
      "step": 17355
    },
    {
      "entropy": 5.65550274848938,
      "epoch": 1.574746008708273,
      "grad_norm": 2.84375,
      "learning_rate": 0.00047555133050488637,
      "loss": 5.3451,
      "mean_token_accuracy": 0.17373837828636168,
      "num_tokens": 36032567.0,
      "step": 17360
    },
    {
      "entropy": 5.630388975143433,
      "epoch": 1.575199564586357,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047553666007207516,
      "loss": 5.4519,
      "mean_token_accuracy": 0.17493016719818116,
      "num_tokens": 36041747.0,
      "step": 17365
    },
    {
      "entropy": 5.696891927719117,
      "epoch": 1.575653120464441,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047552198549211463,
      "loss": 5.3693,
      "mean_token_accuracy": 0.1764494374394417,
      "num_tokens": 36051577.0,
      "step": 17370
    },
    {
      "entropy": 5.72400131225586,
      "epoch": 1.5761066763425253,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004755073067653084,
      "loss": 5.408,
      "mean_token_accuracy": 0.17131324857473373,
      "num_tokens": 36060865.0,
      "step": 17375
    },
    {
      "entropy": 5.6951353549957275,
      "epoch": 1.5765602322206096,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004754926238919599,
      "loss": 5.4139,
      "mean_token_accuracy": 0.17610306143760682,
      "num_tokens": 36070453.0,
      "step": 17380
    },
    {
      "entropy": 5.833060550689697,
      "epoch": 1.5770137880986939,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004754779368723729,
      "loss": 5.5993,
      "mean_token_accuracy": 0.16284130066633223,
      "num_tokens": 36081254.0,
      "step": 17385
    },
    {
      "entropy": 5.693627500534058,
      "epoch": 1.577467343976778,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004754632457068511,
      "loss": 5.4242,
      "mean_token_accuracy": 0.17603790163993835,
      "num_tokens": 36091602.0,
      "step": 17390
    },
    {
      "entropy": 5.6984906673431395,
      "epoch": 1.5779208998548622,
      "grad_norm": 2.125,
      "learning_rate": 0.0004754485503956982,
      "loss": 5.5201,
      "mean_token_accuracy": 0.16993775367736816,
      "num_tokens": 36100600.0,
      "step": 17395
    },
    {
      "entropy": 5.688639116287232,
      "epoch": 1.5783744557329462,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004754338509392183,
      "loss": 5.3906,
      "mean_token_accuracy": 0.170069220662117,
      "num_tokens": 36110597.0,
      "step": 17400
    },
    {
      "entropy": 5.697175931930542,
      "epoch": 1.5788280116110305,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004754191473377153,
      "loss": 5.426,
      "mean_token_accuracy": 0.17266573756933212,
      "num_tokens": 36120483.0,
      "step": 17405
    },
    {
      "entropy": 5.794662380218506,
      "epoch": 1.5792815674891147,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047540443959149316,
      "loss": 5.5724,
      "mean_token_accuracy": 0.16027238517999648,
      "num_tokens": 36131010.0,
      "step": 17410
    },
    {
      "entropy": 5.7274857521057125,
      "epoch": 1.5797351233671988,
      "grad_norm": 2.625,
      "learning_rate": 0.0004753897277008562,
      "loss": 5.337,
      "mean_token_accuracy": 0.17845412343740463,
      "num_tokens": 36140959.0,
      "step": 17415
    },
    {
      "entropy": 5.699613332748413,
      "epoch": 1.580188679245283,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004753750116661086,
      "loss": 5.3437,
      "mean_token_accuracy": 0.17117252498865126,
      "num_tokens": 36151579.0,
      "step": 17420
    },
    {
      "entropy": 5.688665866851807,
      "epoch": 1.580642235123367,
      "grad_norm": 3.3125,
      "learning_rate": 0.00047536029148755477,
      "loss": 5.4646,
      "mean_token_accuracy": 0.16677668392658235,
      "num_tokens": 36162726.0,
      "step": 17425
    },
    {
      "entropy": 5.6523219585418705,
      "epoch": 1.5810957910014514,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047534556716549893,
      "loss": 5.3563,
      "mean_token_accuracy": 0.17323063611984252,
      "num_tokens": 36173592.0,
      "step": 17430
    },
    {
      "entropy": 5.7279070854187015,
      "epoch": 1.5815493468795356,
      "grad_norm": 3.328125,
      "learning_rate": 0.00047533083870024575,
      "loss": 5.4837,
      "mean_token_accuracy": 0.17100053429603576,
      "num_tokens": 36184526.0,
      "step": 17435
    },
    {
      "entropy": 5.78371229171753,
      "epoch": 1.58200290275762,
      "grad_norm": 4.0625,
      "learning_rate": 0.0004753161060920998,
      "loss": 5.3775,
      "mean_token_accuracy": 0.1758454993367195,
      "num_tokens": 36195468.0,
      "step": 17440
    },
    {
      "entropy": 5.673629474639893,
      "epoch": 1.582456458635704,
      "grad_norm": 2.125,
      "learning_rate": 0.0004753013693413657,
      "loss": 5.4266,
      "mean_token_accuracy": 0.1706330507993698,
      "num_tokens": 36204959.0,
      "step": 17445
    },
    {
      "entropy": 5.660502862930298,
      "epoch": 1.582910014513788,
      "grad_norm": 3.375,
      "learning_rate": 0.00047528662844834827,
      "loss": 5.4041,
      "mean_token_accuracy": 0.17042894959449767,
      "num_tokens": 36214928.0,
      "step": 17450
    },
    {
      "entropy": 5.64857029914856,
      "epoch": 1.5833635703918723,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047527188341335233,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16957102566957474,
      "num_tokens": 36225184.0,
      "step": 17455
    },
    {
      "entropy": 5.715992879867554,
      "epoch": 1.5838171262699565,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004752571342366828,
      "loss": 5.4684,
      "mean_token_accuracy": 0.16979410946369172,
      "num_tokens": 36235692.0,
      "step": 17460
    },
    {
      "entropy": 5.706666898727417,
      "epoch": 1.5842706821480408,
      "grad_norm": 2.765625,
      "learning_rate": 0.0004752423809186446,
      "loss": 5.3893,
      "mean_token_accuracy": 0.1738292843103409,
      "num_tokens": 36246787.0,
      "step": 17465
    },
    {
      "entropy": 5.720863151550293,
      "epoch": 1.5847242380261248,
      "grad_norm": 2.140625,
      "learning_rate": 0.00047522762345954305,
      "loss": 5.3971,
      "mean_token_accuracy": 0.17269164323806763,
      "num_tokens": 36256032.0,
      "step": 17470
    },
    {
      "entropy": 5.649812412261963,
      "epoch": 1.5851777939042089,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004752128618596832,
      "loss": 5.4085,
      "mean_token_accuracy": 0.17043924033641816,
      "num_tokens": 36265999.0,
      "step": 17475
    },
    {
      "entropy": 5.700809144973755,
      "epoch": 1.5856313497822931,
      "grad_norm": 2.375,
      "learning_rate": 0.00047519809611937036,
      "loss": 5.4714,
      "mean_token_accuracy": 0.1701071783900261,
      "num_tokens": 36277301.0,
      "step": 17480
    },
    {
      "entropy": 5.688368797302246,
      "epoch": 1.5860849056603774,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004751833262389098,
      "loss": 5.4066,
      "mean_token_accuracy": 0.17818648666143416,
      "num_tokens": 36287727.0,
      "step": 17485
    },
    {
      "entropy": 5.7520359516143795,
      "epoch": 1.5865384615384617,
      "grad_norm": 2.734375,
      "learning_rate": 0.00047516855221860715,
      "loss": 5.4844,
      "mean_token_accuracy": 0.16276316195726395,
      "num_tokens": 36298798.0,
      "step": 17490
    },
    {
      "entropy": 5.777594900131225,
      "epoch": 1.5869920174165457,
      "grad_norm": 2.484375,
      "learning_rate": 0.00047515377405876774,
      "loss": 5.4843,
      "mean_token_accuracy": 0.17054251283407212,
      "num_tokens": 36309242.0,
      "step": 17495
    },
    {
      "entropy": 5.742448806762695,
      "epoch": 1.5874455732946298,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004751389917596973,
      "loss": 5.4214,
      "mean_token_accuracy": 0.171281898021698,
      "num_tokens": 36318409.0,
      "step": 17500
    },
    {
      "entropy": 5.6712812900543215,
      "epoch": 1.587899129172714,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047512420532170154,
      "loss": 5.4433,
      "mean_token_accuracy": 0.17115895599126815,
      "num_tokens": 36328488.0,
      "step": 17505
    },
    {
      "entropy": 5.753954124450684,
      "epoch": 1.5883526850507983,
      "grad_norm": 2.8125,
      "learning_rate": 0.0004751094147450862,
      "loss": 5.5374,
      "mean_token_accuracy": 0.16238574534654618,
      "num_tokens": 36339066.0,
      "step": 17510
    },
    {
      "entropy": 5.66734881401062,
      "epoch": 1.5888062409288826,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004750946200301571,
      "loss": 5.3795,
      "mean_token_accuracy": 0.17671157121658326,
      "num_tokens": 36349040.0,
      "step": 17515
    },
    {
      "entropy": 5.616712617874145,
      "epoch": 1.5892597968069666,
      "grad_norm": 2.0,
      "learning_rate": 0.00047507982117722035,
      "loss": 5.3571,
      "mean_token_accuracy": 0.16857801377773285,
      "num_tokens": 36359628.0,
      "step": 17520
    },
    {
      "entropy": 5.634967279434204,
      "epoch": 1.5897133526850507,
      "grad_norm": 1.7890625,
      "learning_rate": 0.000475065018186582,
      "loss": 5.3526,
      "mean_token_accuracy": 0.17413289248943328,
      "num_tokens": 36369256.0,
      "step": 17525
    },
    {
      "entropy": 5.711472034454346,
      "epoch": 1.590166908563135,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047505021105854794,
      "loss": 5.4477,
      "mean_token_accuracy": 0.16265323609113694,
      "num_tokens": 36379687.0,
      "step": 17530
    },
    {
      "entropy": 5.7951075553894045,
      "epoch": 1.5906204644412192,
      "grad_norm": 2.046875,
      "learning_rate": 0.00047503539979342456,
      "loss": 5.5425,
      "mean_token_accuracy": 0.1616298109292984,
      "num_tokens": 36390835.0,
      "step": 17535
    },
    {
      "entropy": 5.646544790267944,
      "epoch": 1.5910740203193035,
      "grad_norm": 2.25,
      "learning_rate": 0.0004750205843915182,
      "loss": 5.4512,
      "mean_token_accuracy": 0.16529152244329454,
      "num_tokens": 36401456.0,
      "step": 17540
    },
    {
      "entropy": 5.690965032577514,
      "epoch": 1.5915275761973875,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004750057648531351,
      "loss": 5.393,
      "mean_token_accuracy": 0.16880803555250168,
      "num_tokens": 36411680.0,
      "step": 17545
    },
    {
      "entropy": 5.721958589553833,
      "epoch": 1.5919811320754715,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004749909411785819,
      "loss": 5.4594,
      "mean_token_accuracy": 0.16631752401590347,
      "num_tokens": 36423186.0,
      "step": 17550
    },
    {
      "entropy": 5.6556333065032955,
      "epoch": 1.5924346879535558,
      "grad_norm": 1.921875,
      "learning_rate": 0.000474976113368165,
      "loss": 5.4313,
      "mean_token_accuracy": 0.16528978943824768,
      "num_tokens": 36433225.0,
      "step": 17555
    },
    {
      "entropy": 5.660756349563599,
      "epoch": 1.59288824383164,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004749612814221912,
      "loss": 5.3669,
      "mean_token_accuracy": 0.16635848581790924,
      "num_tokens": 36444393.0,
      "step": 17560
    },
    {
      "entropy": 5.69635648727417,
      "epoch": 1.5933417997097243,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047494644534096703,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1692635163664818,
      "num_tokens": 36454817.0,
      "step": 17565
    },
    {
      "entropy": 5.714459228515625,
      "epoch": 1.5937953555878084,
      "grad_norm": 3.09375,
      "learning_rate": 0.0004749316051247995,
      "loss": 5.4188,
      "mean_token_accuracy": 0.16905315071344376,
      "num_tokens": 36464908.0,
      "step": 17570
    },
    {
      "entropy": 5.657666635513306,
      "epoch": 1.5942489114658926,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047491676077399543,
      "loss": 5.4377,
      "mean_token_accuracy": 0.16991474330425263,
      "num_tokens": 36475586.0,
      "step": 17575
    },
    {
      "entropy": 5.670181226730347,
      "epoch": 1.5947024673439767,
      "grad_norm": 5.875,
      "learning_rate": 0.0004749019122888618,
      "loss": 5.4292,
      "mean_token_accuracy": 0.16401465684175492,
      "num_tokens": 36486063.0,
      "step": 17580
    },
    {
      "entropy": 5.733072900772095,
      "epoch": 1.595156023222061,
      "grad_norm": 3.34375,
      "learning_rate": 0.00047488705966970565,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1668306440114975,
      "num_tokens": 36496273.0,
      "step": 17585
    },
    {
      "entropy": 5.679945802688598,
      "epoch": 1.5956095791001452,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004748722029168342,
      "loss": 5.4723,
      "mean_token_accuracy": 0.1690114662051201,
      "num_tokens": 36506813.0,
      "step": 17590
    },
    {
      "entropy": 5.735303592681885,
      "epoch": 1.5960631349782293,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004748573420305546,
      "loss": 5.4523,
      "mean_token_accuracy": 0.17122161239385605,
      "num_tokens": 36516454.0,
      "step": 17595
    },
    {
      "entropy": 5.781835889816284,
      "epoch": 1.5965166908563135,
      "grad_norm": 2.21875,
      "learning_rate": 0.00047484247701117427,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16506334394216537,
      "num_tokens": 36527738.0,
      "step": 17600
    },
    {
      "entropy": 5.66502103805542,
      "epoch": 1.5969702467343976,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004748276078590006,
      "loss": 5.3612,
      "mean_token_accuracy": 0.17368610352277755,
      "num_tokens": 36537902.0,
      "step": 17605
    },
    {
      "entropy": 5.646692562103271,
      "epoch": 1.5974238026124818,
      "grad_norm": 2.125,
      "learning_rate": 0.0004748127345743411,
      "loss": 5.3614,
      "mean_token_accuracy": 0.17709177881479263,
      "num_tokens": 36548233.0,
      "step": 17610
    },
    {
      "entropy": 5.605039930343628,
      "epoch": 1.5978773584905661,
      "grad_norm": 3.390625,
      "learning_rate": 0.0004747978571575033,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16375805288553238,
      "num_tokens": 36559672.0,
      "step": 17615
    },
    {
      "entropy": 5.726909971237182,
      "epoch": 1.5983309143686504,
      "grad_norm": 4.09375,
      "learning_rate": 0.0004747829756087949,
      "loss": 5.5106,
      "mean_token_accuracy": 0.16338724493980408,
      "num_tokens": 36571209.0,
      "step": 17620
    },
    {
      "entropy": 5.719848346710205,
      "epoch": 1.5987844702467344,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047476808992852364,
      "loss": 5.4285,
      "mean_token_accuracy": 0.16382413506507873,
      "num_tokens": 36581276.0,
      "step": 17625
    },
    {
      "entropy": 5.750822687149048,
      "epoch": 1.5992380261248185,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004747532001169974,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1615006923675537,
      "num_tokens": 36592734.0,
      "step": 17630
    },
    {
      "entropy": 5.693153190612793,
      "epoch": 1.5996915820029027,
      "grad_norm": 3.453125,
      "learning_rate": 0.00047473830617452415,
      "loss": 5.3736,
      "mean_token_accuracy": 0.17687281519174575,
      "num_tokens": 36603668.0,
      "step": 17635
    },
    {
      "entropy": 5.686762285232544,
      "epoch": 1.600145137880987,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047472340810141163,
      "loss": 5.427,
      "mean_token_accuracy": 0.173717899620533,
      "num_tokens": 36615012.0,
      "step": 17640
    },
    {
      "entropy": 5.747913789749146,
      "epoch": 1.6005986937590713,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047470850589796833,
      "loss": 5.4182,
      "mean_token_accuracy": 0.16969034075737,
      "num_tokens": 36625711.0,
      "step": 17645
    },
    {
      "entropy": 5.696483898162842,
      "epoch": 1.6010522496371553,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047469359956450215,
      "loss": 5.3744,
      "mean_token_accuracy": 0.1760958880186081,
      "num_tokens": 36635241.0,
      "step": 17650
    },
    {
      "entropy": 5.719260120391846,
      "epoch": 1.6015058055152394,
      "grad_norm": 3.03125,
      "learning_rate": 0.00047467868910132144,
      "loss": 5.4376,
      "mean_token_accuracy": 0.16851540207862853,
      "num_tokens": 36646123.0,
      "step": 17655
    },
    {
      "entropy": 5.699032306671143,
      "epoch": 1.6019593613933236,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004746637745087346,
      "loss": 5.4444,
      "mean_token_accuracy": 0.16664781421422958,
      "num_tokens": 36655506.0,
      "step": 17660
    },
    {
      "entropy": 5.6903351783752445,
      "epoch": 1.6024129172714079,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047464885578705,
      "loss": 5.3727,
      "mean_token_accuracy": 0.17362165302038193,
      "num_tokens": 36665324.0,
      "step": 17665
    },
    {
      "entropy": 5.635593605041504,
      "epoch": 1.6028664731494922,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047463393293657616,
      "loss": 5.3165,
      "mean_token_accuracy": 0.17474350482225418,
      "num_tokens": 36675711.0,
      "step": 17670
    },
    {
      "entropy": 5.647988891601562,
      "epoch": 1.6033200290275762,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004746190059576217,
      "loss": 5.4262,
      "mean_token_accuracy": 0.17012860774993896,
      "num_tokens": 36686198.0,
      "step": 17675
    },
    {
      "entropy": 5.5940569877624515,
      "epoch": 1.6037735849056602,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004746040748504954,
      "loss": 5.4322,
      "mean_token_accuracy": 0.1677406594157219,
      "num_tokens": 36696667.0,
      "step": 17680
    },
    {
      "entropy": 5.764087820053101,
      "epoch": 1.6042271407837445,
      "grad_norm": 2.71875,
      "learning_rate": 0.00047458913961550593,
      "loss": 5.4765,
      "mean_token_accuracy": 0.16684584617614745,
      "num_tokens": 36708067.0,
      "step": 17685
    },
    {
      "entropy": 5.7646262645721436,
      "epoch": 1.6046806966618288,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004745742002529622,
      "loss": 5.4709,
      "mean_token_accuracy": 0.16698847562074662,
      "num_tokens": 36719638.0,
      "step": 17690
    },
    {
      "entropy": 5.67256498336792,
      "epoch": 1.605134252539913,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004745592567631731,
      "loss": 5.3483,
      "mean_token_accuracy": 0.17482466548681258,
      "num_tokens": 36730105.0,
      "step": 17695
    },
    {
      "entropy": 5.639059829711914,
      "epoch": 1.605587808417997,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047454430914644773,
      "loss": 5.3542,
      "mean_token_accuracy": 0.1710411474108696,
      "num_tokens": 36740325.0,
      "step": 17700
    },
    {
      "entropy": 5.678561639785767,
      "epoch": 1.6060413642960811,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004745293574030952,
      "loss": 5.3775,
      "mean_token_accuracy": 0.16966034471988678,
      "num_tokens": 36750727.0,
      "step": 17705
    },
    {
      "entropy": 5.679276466369629,
      "epoch": 1.6064949201741654,
      "grad_norm": 2.890625,
      "learning_rate": 0.00047451440153342477,
      "loss": 5.407,
      "mean_token_accuracy": 0.1667017474770546,
      "num_tokens": 36761465.0,
      "step": 17710
    },
    {
      "entropy": 5.703038358688355,
      "epoch": 1.6069484760522497,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004744994415377457,
      "loss": 5.4609,
      "mean_token_accuracy": 0.168880595266819,
      "num_tokens": 36772700.0,
      "step": 17715
    },
    {
      "entropy": 5.6972325325012205,
      "epoch": 1.607402031930334,
      "grad_norm": 2.125,
      "learning_rate": 0.00047448447741636724,
      "loss": 5.4836,
      "mean_token_accuracy": 0.16290999203920364,
      "num_tokens": 36783744.0,
      "step": 17720
    },
    {
      "entropy": 5.694417476654053,
      "epoch": 1.607855587808418,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047446950916959905,
      "loss": 5.3373,
      "mean_token_accuracy": 0.17588941305875777,
      "num_tokens": 36793190.0,
      "step": 17725
    },
    {
      "entropy": 5.636773347854614,
      "epoch": 1.608309143686502,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004744545367977505,
      "loss": 5.3885,
      "mean_token_accuracy": 0.1754712164402008,
      "num_tokens": 36803337.0,
      "step": 17730
    },
    {
      "entropy": 5.625230693817139,
      "epoch": 1.6087626995645863,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047443956030113134,
      "loss": 5.3767,
      "mean_token_accuracy": 0.1739637553691864,
      "num_tokens": 36813581.0,
      "step": 17735
    },
    {
      "entropy": 5.717404985427857,
      "epoch": 1.6092162554426706,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047442457968005123,
      "loss": 5.5009,
      "mean_token_accuracy": 0.16693971455097198,
      "num_tokens": 36823057.0,
      "step": 17740
    },
    {
      "entropy": 5.723193120956421,
      "epoch": 1.6096698113207548,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00047440959493481996,
      "loss": 5.4307,
      "mean_token_accuracy": 0.16764925867319108,
      "num_tokens": 36832618.0,
      "step": 17745
    },
    {
      "entropy": 5.584024858474732,
      "epoch": 1.6101233671988389,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047439460606574745,
      "loss": 5.3689,
      "mean_token_accuracy": 0.17482030838727952,
      "num_tokens": 36843132.0,
      "step": 17750
    },
    {
      "entropy": 5.688275527954102,
      "epoch": 1.6105769230769231,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004743796130731437,
      "loss": 5.5219,
      "mean_token_accuracy": 0.16166358590126037,
      "num_tokens": 36855001.0,
      "step": 17755
    },
    {
      "entropy": 5.7181281566619875,
      "epoch": 1.6110304789550072,
      "grad_norm": 2.828125,
      "learning_rate": 0.0004743646159573188,
      "loss": 5.3425,
      "mean_token_accuracy": 0.16983480751514435,
      "num_tokens": 36865357.0,
      "step": 17760
    },
    {
      "entropy": 5.726531171798706,
      "epoch": 1.6114840348330914,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004743496147185828,
      "loss": 5.4457,
      "mean_token_accuracy": 0.16642328947782517,
      "num_tokens": 36876015.0,
      "step": 17765
    },
    {
      "entropy": 5.621378087997437,
      "epoch": 1.6119375907111757,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047433460935724595,
      "loss": 5.4054,
      "mean_token_accuracy": 0.1716793194413185,
      "num_tokens": 36886557.0,
      "step": 17770
    },
    {
      "entropy": 5.688300323486328,
      "epoch": 1.6123911465892597,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004743195998736186,
      "loss": 5.339,
      "mean_token_accuracy": 0.17605369091033934,
      "num_tokens": 36896405.0,
      "step": 17775
    },
    {
      "entropy": 5.712791776657104,
      "epoch": 1.612844702467344,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047430458626801107,
      "loss": 5.4346,
      "mean_token_accuracy": 0.17230097353458404,
      "num_tokens": 36906194.0,
      "step": 17780
    },
    {
      "entropy": 5.666767740249634,
      "epoch": 1.613298258345428,
      "grad_norm": 3.8125,
      "learning_rate": 0.000474289568540734,
      "loss": 5.3766,
      "mean_token_accuracy": 0.17416322529315947,
      "num_tokens": 36916086.0,
      "step": 17785
    },
    {
      "entropy": 5.707764387130737,
      "epoch": 1.6137518142235123,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047427454669209785,
      "loss": 5.371,
      "mean_token_accuracy": 0.17338528335094452,
      "num_tokens": 36925799.0,
      "step": 17790
    },
    {
      "entropy": 5.737970781326294,
      "epoch": 1.6142053701015966,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047425952072241326,
      "loss": 5.4789,
      "mean_token_accuracy": 0.17217821776866912,
      "num_tokens": 36936337.0,
      "step": 17795
    },
    {
      "entropy": 5.698242664337158,
      "epoch": 1.6146589259796809,
      "grad_norm": 1.9375,
      "learning_rate": 0.000474244490631991,
      "loss": 5.4593,
      "mean_token_accuracy": 0.16429720669984818,
      "num_tokens": 36946473.0,
      "step": 17800
    },
    {
      "entropy": 5.757734155654907,
      "epoch": 1.615112481857765,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004742294564211419,
      "loss": 5.4825,
      "mean_token_accuracy": 0.17040537148714066,
      "num_tokens": 36955974.0,
      "step": 17805
    },
    {
      "entropy": 5.69518814086914,
      "epoch": 1.615566037735849,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004742144180901769,
      "loss": 5.4523,
      "mean_token_accuracy": 0.16207925528287886,
      "num_tokens": 36967127.0,
      "step": 17810
    },
    {
      "entropy": 5.782198095321656,
      "epoch": 1.6160195936139332,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00047419937563940697,
      "loss": 5.5156,
      "mean_token_accuracy": 0.16553474664688111,
      "num_tokens": 36977326.0,
      "step": 17815
    },
    {
      "entropy": 5.6907754898071286,
      "epoch": 1.6164731494920175,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047418432906914314,
      "loss": 5.3985,
      "mean_token_accuracy": 0.17389795333147048,
      "num_tokens": 36986776.0,
      "step": 17820
    },
    {
      "entropy": 5.673222589492798,
      "epoch": 1.6169267053701017,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004741692783796967,
      "loss": 5.3893,
      "mean_token_accuracy": 0.17283719331026076,
      "num_tokens": 36996572.0,
      "step": 17825
    },
    {
      "entropy": 5.782290077209472,
      "epoch": 1.6173802612481858,
      "grad_norm": 4.21875,
      "learning_rate": 0.00047415422357137876,
      "loss": 5.4763,
      "mean_token_accuracy": 0.16467594355344772,
      "num_tokens": 37006088.0,
      "step": 17830
    },
    {
      "entropy": 5.706982898712158,
      "epoch": 1.6178338171262698,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004741391646445008,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1703524649143219,
      "num_tokens": 37016760.0,
      "step": 17835
    },
    {
      "entropy": 5.667888927459717,
      "epoch": 1.618287373004354,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004741241015993741,
      "loss": 5.458,
      "mean_token_accuracy": 0.1657585844397545,
      "num_tokens": 37026261.0,
      "step": 17840
    },
    {
      "entropy": 5.632056140899659,
      "epoch": 1.6187409288824384,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047410903443631026,
      "loss": 5.3314,
      "mean_token_accuracy": 0.17940036654472352,
      "num_tokens": 37036442.0,
      "step": 17845
    },
    {
      "entropy": 5.623647165298462,
      "epoch": 1.6191944847605226,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004740939631556209,
      "loss": 5.3157,
      "mean_token_accuracy": 0.1671147510409355,
      "num_tokens": 37047485.0,
      "step": 17850
    },
    {
      "entropy": 5.676955461502075,
      "epoch": 1.6196480406386067,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004740788877576176,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16679666340351104,
      "num_tokens": 37057174.0,
      "step": 17855
    },
    {
      "entropy": 5.791093111038208,
      "epoch": 1.6201015965166907,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004740638082426121,
      "loss": 5.4791,
      "mean_token_accuracy": 0.16661319732666016,
      "num_tokens": 37067062.0,
      "step": 17860
    },
    {
      "entropy": 5.722626399993897,
      "epoch": 1.620555152394775,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047404872461091643,
      "loss": 5.5019,
      "mean_token_accuracy": 0.1614195853471756,
      "num_tokens": 37077750.0,
      "step": 17865
    },
    {
      "entropy": 5.698396968841553,
      "epoch": 1.6210087082728593,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047403363686284236,
      "loss": 5.5351,
      "mean_token_accuracy": 0.16488468647003174,
      "num_tokens": 37088769.0,
      "step": 17870
    },
    {
      "entropy": 5.656366157531738,
      "epoch": 1.6214622641509435,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00047401854499870197,
      "loss": 5.2969,
      "mean_token_accuracy": 0.17632437199354173,
      "num_tokens": 37097812.0,
      "step": 17875
    },
    {
      "entropy": 5.641413354873658,
      "epoch": 1.6219158200290276,
      "grad_norm": 2.0,
      "learning_rate": 0.00047400344901880735,
      "loss": 5.3983,
      "mean_token_accuracy": 0.17687358111143112,
      "num_tokens": 37107932.0,
      "step": 17880
    },
    {
      "entropy": 5.623113012313842,
      "epoch": 1.6223693759071116,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004739883489234707,
      "loss": 5.3776,
      "mean_token_accuracy": 0.1754260778427124,
      "num_tokens": 37119058.0,
      "step": 17885
    },
    {
      "entropy": 5.740437269210815,
      "epoch": 1.6228229317851959,
      "grad_norm": 2.203125,
      "learning_rate": 0.00047397324471300424,
      "loss": 5.4812,
      "mean_token_accuracy": 0.16713945269584657,
      "num_tokens": 37130663.0,
      "step": 17890
    },
    {
      "entropy": 5.693921947479248,
      "epoch": 1.6232764876632801,
      "grad_norm": 2.65625,
      "learning_rate": 0.00047395813638772047,
      "loss": 5.3872,
      "mean_token_accuracy": 0.17034004628658295,
      "num_tokens": 37140960.0,
      "step": 17895
    },
    {
      "entropy": 5.680989837646484,
      "epoch": 1.6237300435413644,
      "grad_norm": 1.984375,
      "learning_rate": 0.00047394302394793164,
      "loss": 5.4554,
      "mean_token_accuracy": 0.16715622544288636,
      "num_tokens": 37151102.0,
      "step": 17900
    },
    {
      "entropy": 5.775130414962769,
      "epoch": 1.6241835994194485,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004739279073939504,
      "loss": 5.4486,
      "mean_token_accuracy": 0.16573382169008255,
      "num_tokens": 37160640.0,
      "step": 17905
    },
    {
      "entropy": 5.826239204406738,
      "epoch": 1.6246371552975325,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004739127867260894,
      "loss": 5.5625,
      "mean_token_accuracy": 0.16191894561052322,
      "num_tokens": 37171565.0,
      "step": 17910
    },
    {
      "entropy": 5.7784089088439945,
      "epoch": 1.6250907111756168,
      "grad_norm": 1.875,
      "learning_rate": 0.0004738976619446613,
      "loss": 5.4904,
      "mean_token_accuracy": 0.16713612675666809,
      "num_tokens": 37182152.0,
      "step": 17915
    },
    {
      "entropy": 5.63674693107605,
      "epoch": 1.625544267053701,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047388253304997874,
      "loss": 5.2999,
      "mean_token_accuracy": 0.18005081564188002,
      "num_tokens": 37192589.0,
      "step": 17920
    },
    {
      "entropy": 5.680987596511841,
      "epoch": 1.6259978229317853,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004738674000423548,
      "loss": 5.5016,
      "mean_token_accuracy": 0.16324781775474548,
      "num_tokens": 37203216.0,
      "step": 17925
    },
    {
      "entropy": 5.738418483734131,
      "epoch": 1.6264513788098693,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004738522629221023,
      "loss": 5.4655,
      "mean_token_accuracy": 0.16703827828168868,
      "num_tokens": 37214535.0,
      "step": 17930
    },
    {
      "entropy": 5.693620681762695,
      "epoch": 1.6269049346879536,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004738371216895344,
      "loss": 5.4126,
      "mean_token_accuracy": 0.17357216477394105,
      "num_tokens": 37225483.0,
      "step": 17935
    },
    {
      "entropy": 5.712025213241577,
      "epoch": 1.6273584905660377,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004738219763449641,
      "loss": 5.4259,
      "mean_token_accuracy": 0.16831829249858857,
      "num_tokens": 37235002.0,
      "step": 17940
    },
    {
      "entropy": 5.728950023651123,
      "epoch": 1.627812046444122,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004738068268887047,
      "loss": 5.3864,
      "mean_token_accuracy": 0.17717328667640686,
      "num_tokens": 37244907.0,
      "step": 17945
    },
    {
      "entropy": 5.741972303390503,
      "epoch": 1.6282656023222062,
      "grad_norm": 8.75,
      "learning_rate": 0.0004737916733210694,
      "loss": 5.4765,
      "mean_token_accuracy": 0.1611965961754322,
      "num_tokens": 37256580.0,
      "step": 17950
    },
    {
      "entropy": 5.618513917922973,
      "epoch": 1.6287191582002902,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047377651564237165,
      "loss": 5.3038,
      "mean_token_accuracy": 0.17875338047742845,
      "num_tokens": 37266881.0,
      "step": 17955
    },
    {
      "entropy": 5.635388994216919,
      "epoch": 1.6291727140783745,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047376135385292494,
      "loss": 5.4109,
      "mean_token_accuracy": 0.17393783777952193,
      "num_tokens": 37276953.0,
      "step": 17960
    },
    {
      "entropy": 5.811871671676636,
      "epoch": 1.6296262699564585,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004737461879530428,
      "loss": 5.4848,
      "mean_token_accuracy": 0.1676352486014366,
      "num_tokens": 37288007.0,
      "step": 17965
    },
    {
      "entropy": 5.682890558242798,
      "epoch": 1.6300798258345428,
      "grad_norm": 3.359375,
      "learning_rate": 0.00047373101794303884,
      "loss": 5.3256,
      "mean_token_accuracy": 0.17048681527376175,
      "num_tokens": 37297955.0,
      "step": 17970
    },
    {
      "entropy": 5.572243404388428,
      "epoch": 1.630533381712627,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004737158438232267,
      "loss": 5.35,
      "mean_token_accuracy": 0.17525724470615386,
      "num_tokens": 37307464.0,
      "step": 17975
    },
    {
      "entropy": 5.649382591247559,
      "epoch": 1.6309869375907113,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047370066559392033,
      "loss": 5.412,
      "mean_token_accuracy": 0.1742629274725914,
      "num_tokens": 37317655.0,
      "step": 17980
    },
    {
      "entropy": 5.735839605331421,
      "epoch": 1.6314404934687954,
      "grad_norm": 2.5,
      "learning_rate": 0.00047368548325543366,
      "loss": 5.4216,
      "mean_token_accuracy": 0.16158686876296996,
      "num_tokens": 37328354.0,
      "step": 17985
    },
    {
      "entropy": 5.653699445724487,
      "epoch": 1.6318940493468794,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004736702968080804,
      "loss": 5.4061,
      "mean_token_accuracy": 0.17497183680534362,
      "num_tokens": 37339165.0,
      "step": 17990
    },
    {
      "entropy": 5.637998819351196,
      "epoch": 1.6323476052249637,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00047365510625217483,
      "loss": 5.3406,
      "mean_token_accuracy": 0.17502062469720842,
      "num_tokens": 37348919.0,
      "step": 17995
    },
    {
      "entropy": 5.693705606460571,
      "epoch": 1.632801161103048,
      "grad_norm": 3.9375,
      "learning_rate": 0.00047363991158803117,
      "loss": 5.3958,
      "mean_token_accuracy": 0.17545763701200484,
      "num_tokens": 37358946.0,
      "step": 18000
    },
    {
      "epoch": 1.632801161103048,
      "eval_entropy": 5.472744574050558,
      "eval_loss": 5.4727091789245605,
      "eval_mean_token_accuracy": 0.1758232282862337,
      "eval_num_tokens": 37358946.0,
      "eval_runtime": 26.6486,
      "eval_samples_per_second": 1326.899,
      "eval_steps_per_second": 165.862,
      "step": 18000
    },
    {
      "entropy": 5.666504192352295,
      "epoch": 1.6332547169811322,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047362471281596334,
      "loss": 5.4387,
      "mean_token_accuracy": 0.17819511890411377,
      "num_tokens": 37369033.0,
      "step": 18005
    },
    {
      "entropy": 5.720816326141358,
      "epoch": 1.6337082728592163,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047360950993628585,
      "loss": 5.4044,
      "mean_token_accuracy": 0.16944689452648162,
      "num_tokens": 37379146.0,
      "step": 18010
    },
    {
      "entropy": 5.715740060806274,
      "epoch": 1.6341618287373003,
      "grad_norm": 3.5625,
      "learning_rate": 0.0004735943029493132,
      "loss": 5.3939,
      "mean_token_accuracy": 0.17168546617031097,
      "num_tokens": 37389161.0,
      "step": 18015
    },
    {
      "entropy": 5.672096061706543,
      "epoch": 1.6346153846153846,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047357909185535965,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17571175694465638,
      "num_tokens": 37398854.0,
      "step": 18020
    },
    {
      "entropy": 5.641301918029785,
      "epoch": 1.6350689404934688,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004735638766547399,
      "loss": 5.4143,
      "mean_token_accuracy": 0.1709200322628021,
      "num_tokens": 37410186.0,
      "step": 18025
    },
    {
      "entropy": 5.658245849609375,
      "epoch": 1.6355224963715531,
      "grad_norm": 2.859375,
      "learning_rate": 0.00047354865734776857,
      "loss": 5.3712,
      "mean_token_accuracy": 0.17330546379089357,
      "num_tokens": 37419428.0,
      "step": 18030
    },
    {
      "entropy": 5.719909429550171,
      "epoch": 1.6359760522496372,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004735334339347605,
      "loss": 5.3977,
      "mean_token_accuracy": 0.17021696418523788,
      "num_tokens": 37429919.0,
      "step": 18035
    },
    {
      "entropy": 5.75457911491394,
      "epoch": 1.6364296081277212,
      "grad_norm": 1.875,
      "learning_rate": 0.00047351820641603035,
      "loss": 5.4459,
      "mean_token_accuracy": 0.16403694152832032,
      "num_tokens": 37440264.0,
      "step": 18040
    },
    {
      "entropy": 5.639838647842407,
      "epoch": 1.6368831640058055,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004735029747918931,
      "loss": 5.3307,
      "mean_token_accuracy": 0.17506682276725768,
      "num_tokens": 37450148.0,
      "step": 18045
    },
    {
      "entropy": 5.694638967514038,
      "epoch": 1.6373367198838897,
      "grad_norm": 3.015625,
      "learning_rate": 0.00047348773906266374,
      "loss": 5.4832,
      "mean_token_accuracy": 0.16690635681152344,
      "num_tokens": 37459641.0,
      "step": 18050
    },
    {
      "entropy": 5.703060007095337,
      "epoch": 1.637790275761974,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004734724992286574,
      "loss": 5.3836,
      "mean_token_accuracy": 0.17708779126405716,
      "num_tokens": 37470566.0,
      "step": 18055
    },
    {
      "entropy": 5.5941417694091795,
      "epoch": 1.638243831640058,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004734572552901892,
      "loss": 5.3327,
      "mean_token_accuracy": 0.18304276764392852,
      "num_tokens": 37479825.0,
      "step": 18060
    },
    {
      "entropy": 5.75871262550354,
      "epoch": 1.638697387518142,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004734420072475743,
      "loss": 5.4743,
      "mean_token_accuracy": 0.17124216109514237,
      "num_tokens": 37490542.0,
      "step": 18065
    },
    {
      "entropy": 5.697717571258545,
      "epoch": 1.6391509433962264,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004734267551011282,
      "loss": 5.3832,
      "mean_token_accuracy": 0.17315845489501952,
      "num_tokens": 37501549.0,
      "step": 18070
    },
    {
      "entropy": 5.640230751037597,
      "epoch": 1.6396044992743106,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004734114988511662,
      "loss": 5.4243,
      "mean_token_accuracy": 0.17019160836935043,
      "num_tokens": 37513876.0,
      "step": 18075
    },
    {
      "entropy": 5.669102048873901,
      "epoch": 1.640058055152395,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004733962384980039,
      "loss": 5.3875,
      "mean_token_accuracy": 0.16879075467586518,
      "num_tokens": 37524325.0,
      "step": 18080
    },
    {
      "entropy": 5.6780811786651615,
      "epoch": 1.640511611030479,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004733809740419568,
      "loss": 5.415,
      "mean_token_accuracy": 0.167652128636837,
      "num_tokens": 37535097.0,
      "step": 18085
    },
    {
      "entropy": 5.679126596450805,
      "epoch": 1.640965166908563,
      "grad_norm": 1.8125,
      "learning_rate": 0.00047336570548334053,
      "loss": 5.3533,
      "mean_token_accuracy": 0.175538332760334,
      "num_tokens": 37546084.0,
      "step": 18090
    },
    {
      "entropy": 5.735858964920044,
      "epoch": 1.6414187227866472,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047335043282247105,
      "loss": 5.4752,
      "mean_token_accuracy": 0.16672195345163346,
      "num_tokens": 37556951.0,
      "step": 18095
    },
    {
      "entropy": 5.647474670410157,
      "epoch": 1.6418722786647315,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047333515605966396,
      "loss": 5.3927,
      "mean_token_accuracy": 0.16980786770582199,
      "num_tokens": 37566647.0,
      "step": 18100
    },
    {
      "entropy": 5.718963575363159,
      "epoch": 1.6423258345428158,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047331987519523533,
      "loss": 5.4076,
      "mean_token_accuracy": 0.16800110191106796,
      "num_tokens": 37576674.0,
      "step": 18105
    },
    {
      "entropy": 5.641329002380371,
      "epoch": 1.6427793904208998,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004733045902295012,
      "loss": 5.3624,
      "mean_token_accuracy": 0.16901486366987228,
      "num_tokens": 37586416.0,
      "step": 18110
    },
    {
      "entropy": 5.681328821182251,
      "epoch": 1.643232946298984,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047328930116277755,
      "loss": 5.4,
      "mean_token_accuracy": 0.16832200139760972,
      "num_tokens": 37595493.0,
      "step": 18115
    },
    {
      "entropy": 5.71546425819397,
      "epoch": 1.6436865021770681,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004732740079953806,
      "loss": 5.3459,
      "mean_token_accuracy": 0.17812617868185043,
      "num_tokens": 37605106.0,
      "step": 18120
    },
    {
      "entropy": 5.648204374313354,
      "epoch": 1.6441400580551524,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004732587107276267,
      "loss": 5.4502,
      "mean_token_accuracy": 0.16812787801027299,
      "num_tokens": 37615068.0,
      "step": 18125
    },
    {
      "entropy": 5.720009231567383,
      "epoch": 1.6445936139332367,
      "grad_norm": 2.078125,
      "learning_rate": 0.00047324340935983225,
      "loss": 5.4949,
      "mean_token_accuracy": 0.16820484399795532,
      "num_tokens": 37625713.0,
      "step": 18130
    },
    {
      "entropy": 5.805350828170776,
      "epoch": 1.6450471698113207,
      "grad_norm": 4.25,
      "learning_rate": 0.00047322810389231333,
      "loss": 5.5035,
      "mean_token_accuracy": 0.16415413916110994,
      "num_tokens": 37635714.0,
      "step": 18135
    },
    {
      "entropy": 5.685881900787353,
      "epoch": 1.645500725689405,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004732127943253868,
      "loss": 5.3621,
      "mean_token_accuracy": 0.17606481313705444,
      "num_tokens": 37646089.0,
      "step": 18140
    },
    {
      "entropy": 5.700388956069946,
      "epoch": 1.645954281567489,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004731974806593693,
      "loss": 5.4479,
      "mean_token_accuracy": 0.16616871058940888,
      "num_tokens": 37656070.0,
      "step": 18145
    },
    {
      "entropy": 5.673432350158691,
      "epoch": 1.6464078374455733,
      "grad_norm": 2.828125,
      "learning_rate": 0.00047318216289457727,
      "loss": 5.3406,
      "mean_token_accuracy": 0.1708567813038826,
      "num_tokens": 37666841.0,
      "step": 18150
    },
    {
      "entropy": 5.678367710113525,
      "epoch": 1.6468613933236576,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047316684103132766,
      "loss": 5.414,
      "mean_token_accuracy": 0.17460250556468965,
      "num_tokens": 37676789.0,
      "step": 18155
    },
    {
      "entropy": 5.695118522644043,
      "epoch": 1.6473149492017418,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004731515150699372,
      "loss": 5.4345,
      "mean_token_accuracy": 0.17088278979063035,
      "num_tokens": 37687250.0,
      "step": 18160
    },
    {
      "entropy": 5.690443468093872,
      "epoch": 1.6477685050798259,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047313618501072303,
      "loss": 5.3394,
      "mean_token_accuracy": 0.17676150798797607,
      "num_tokens": 37697006.0,
      "step": 18165
    },
    {
      "entropy": 5.61077127456665,
      "epoch": 1.64822206095791,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047312085085400194,
      "loss": 5.3971,
      "mean_token_accuracy": 0.17307000756263732,
      "num_tokens": 37708043.0,
      "step": 18170
    },
    {
      "entropy": 5.714343118667602,
      "epoch": 1.6486756168359942,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047310551260009126,
      "loss": 5.4606,
      "mean_token_accuracy": 0.16519297659397125,
      "num_tokens": 37718624.0,
      "step": 18175
    },
    {
      "entropy": 5.671770477294922,
      "epoch": 1.6491291727140784,
      "grad_norm": 1.9140625,
      "learning_rate": 0.000473090170249308,
      "loss": 5.3884,
      "mean_token_accuracy": 0.1747024819254875,
      "num_tokens": 37729778.0,
      "step": 18180
    },
    {
      "entropy": 5.68512110710144,
      "epoch": 1.6495827285921627,
      "grad_norm": 3.359375,
      "learning_rate": 0.0004730748238019696,
      "loss": 5.418,
      "mean_token_accuracy": 0.1725754827260971,
      "num_tokens": 37739638.0,
      "step": 18185
    },
    {
      "entropy": 5.75724835395813,
      "epoch": 1.6500362844702468,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047305947325839337,
      "loss": 5.4601,
      "mean_token_accuracy": 0.17105008512735367,
      "num_tokens": 37750062.0,
      "step": 18190
    },
    {
      "entropy": 5.687964248657226,
      "epoch": 1.6504898403483308,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004730441186188968,
      "loss": 5.3922,
      "mean_token_accuracy": 0.1664698988199234,
      "num_tokens": 37760891.0,
      "step": 18195
    },
    {
      "entropy": 5.7306890964508055,
      "epoch": 1.650943396226415,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047302875988379734,
      "loss": 5.5724,
      "mean_token_accuracy": 0.16045384332537652,
      "num_tokens": 37772531.0,
      "step": 18200
    },
    {
      "entropy": 5.688921546936035,
      "epoch": 1.6513969521044993,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047301339705341266,
      "loss": 5.2924,
      "mean_token_accuracy": 0.1834925666451454,
      "num_tokens": 37782563.0,
      "step": 18205
    },
    {
      "entropy": 5.691710948944092,
      "epoch": 1.6518505079825836,
      "grad_norm": 2.28125,
      "learning_rate": 0.00047299803012806056,
      "loss": 5.3328,
      "mean_token_accuracy": 0.17597973197698594,
      "num_tokens": 37792775.0,
      "step": 18210
    },
    {
      "entropy": 5.6413780689239506,
      "epoch": 1.6523040638606676,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004729826591080586,
      "loss": 5.3768,
      "mean_token_accuracy": 0.17547038197517395,
      "num_tokens": 37802865.0,
      "step": 18215
    },
    {
      "entropy": 5.628079986572265,
      "epoch": 1.6527576197387517,
      "grad_norm": 1.625,
      "learning_rate": 0.00047296728399372483,
      "loss": 5.2534,
      "mean_token_accuracy": 0.17941755950450897,
      "num_tokens": 37813074.0,
      "step": 18220
    },
    {
      "entropy": 5.650305843353271,
      "epoch": 1.653211175616836,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047295190478537725,
      "loss": 5.4786,
      "mean_token_accuracy": 0.169679257273674,
      "num_tokens": 37823317.0,
      "step": 18225
    },
    {
      "entropy": 5.819519090652466,
      "epoch": 1.6536647314949202,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00047293652148333376,
      "loss": 5.5779,
      "mean_token_accuracy": 0.16478619426488877,
      "num_tokens": 37834240.0,
      "step": 18230
    },
    {
      "entropy": 5.77411789894104,
      "epoch": 1.6541182873730045,
      "grad_norm": 5.21875,
      "learning_rate": 0.00047292113408791266,
      "loss": 5.4122,
      "mean_token_accuracy": 0.16773648113012313,
      "num_tokens": 37845348.0,
      "step": 18235
    },
    {
      "entropy": 5.624904632568359,
      "epoch": 1.6545718432510885,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047290574259943195,
      "loss": 5.3803,
      "mean_token_accuracy": 0.1698829635977745,
      "num_tokens": 37854927.0,
      "step": 18240
    },
    {
      "entropy": 5.683340692520142,
      "epoch": 1.6550253991291726,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047289034701821013,
      "loss": 5.3713,
      "mean_token_accuracy": 0.16806883811950685,
      "num_tokens": 37864918.0,
      "step": 18245
    },
    {
      "entropy": 5.73704719543457,
      "epoch": 1.6554789550072568,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047287494734456545,
      "loss": 5.4273,
      "mean_token_accuracy": 0.1736760050058365,
      "num_tokens": 37875415.0,
      "step": 18250
    },
    {
      "entropy": 5.639870071411133,
      "epoch": 1.655932510885341,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047285954357881646,
      "loss": 5.4747,
      "mean_token_accuracy": 0.17123980075120926,
      "num_tokens": 37886128.0,
      "step": 18255
    },
    {
      "entropy": 5.662854909896851,
      "epoch": 1.6563860667634254,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004728441357212817,
      "loss": 5.3472,
      "mean_token_accuracy": 0.1720072090625763,
      "num_tokens": 37896597.0,
      "step": 18260
    },
    {
      "entropy": 5.768948698043824,
      "epoch": 1.6568396226415094,
      "grad_norm": 2.125,
      "learning_rate": 0.0004728287237722798,
      "loss": 5.534,
      "mean_token_accuracy": 0.16282083839178085,
      "num_tokens": 37908356.0,
      "step": 18265
    },
    {
      "entropy": 5.726126670837402,
      "epoch": 1.6572931785195935,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004728133077321295,
      "loss": 5.4084,
      "mean_token_accuracy": 0.1651185557246208,
      "num_tokens": 37919351.0,
      "step": 18270
    },
    {
      "entropy": 5.739133262634278,
      "epoch": 1.6577467343976777,
      "grad_norm": 2.359375,
      "learning_rate": 0.00047279788760114954,
      "loss": 5.4785,
      "mean_token_accuracy": 0.16504713892936707,
      "num_tokens": 37930309.0,
      "step": 18275
    },
    {
      "entropy": 5.700391817092895,
      "epoch": 1.658200290275762,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004727824633796589,
      "loss": 5.4063,
      "mean_token_accuracy": 0.16720059961080552,
      "num_tokens": 37940144.0,
      "step": 18280
    },
    {
      "entropy": 5.652299451828003,
      "epoch": 1.6586538461538463,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004727670350679765,
      "loss": 5.3237,
      "mean_token_accuracy": 0.17685857713222503,
      "num_tokens": 37950537.0,
      "step": 18285
    },
    {
      "entropy": 5.73731484413147,
      "epoch": 1.6591074020319303,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004727516026664214,
      "loss": 5.4866,
      "mean_token_accuracy": 0.1604531452059746,
      "num_tokens": 37960848.0,
      "step": 18290
    },
    {
      "entropy": 5.723875379562378,
      "epoch": 1.6595609579100146,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004727361661753128,
      "loss": 5.5205,
      "mean_token_accuracy": 0.16886977702379227,
      "num_tokens": 37971099.0,
      "step": 18295
    },
    {
      "entropy": 5.683856201171875,
      "epoch": 1.6600145137880986,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004727207255949699,
      "loss": 5.4026,
      "mean_token_accuracy": 0.17242141366004943,
      "num_tokens": 37981421.0,
      "step": 18300
    },
    {
      "entropy": 5.677627849578857,
      "epoch": 1.6604680696661829,
      "grad_norm": 2.390625,
      "learning_rate": 0.00047270528092571197,
      "loss": 5.4079,
      "mean_token_accuracy": 0.17382007837295532,
      "num_tokens": 37990912.0,
      "step": 18305
    },
    {
      "entropy": 5.626044511795044,
      "epoch": 1.6609216255442671,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004726898321678584,
      "loss": 5.4054,
      "mean_token_accuracy": 0.17488448619842528,
      "num_tokens": 38001279.0,
      "step": 18310
    },
    {
      "entropy": 5.721967792510986,
      "epoch": 1.6613751814223512,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004726743793217288,
      "loss": 5.497,
      "mean_token_accuracy": 0.17008460462093353,
      "num_tokens": 38012273.0,
      "step": 18315
    },
    {
      "entropy": 5.685835981369019,
      "epoch": 1.6618287373004355,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047265892238764265,
      "loss": 5.3982,
      "mean_token_accuracy": 0.1708304613828659,
      "num_tokens": 38023528.0,
      "step": 18320
    },
    {
      "entropy": 5.660630226135254,
      "epoch": 1.6622822931785195,
      "grad_norm": 2.75,
      "learning_rate": 0.0004726434613659196,
      "loss": 5.4249,
      "mean_token_accuracy": 0.1675271660089493,
      "num_tokens": 38034329.0,
      "step": 18325
    },
    {
      "entropy": 5.681183815002441,
      "epoch": 1.6627358490566038,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004726279962568794,
      "loss": 5.3955,
      "mean_token_accuracy": 0.17179032564163207,
      "num_tokens": 38044084.0,
      "step": 18330
    },
    {
      "entropy": 5.712741327285767,
      "epoch": 1.663189404934688,
      "grad_norm": 3.875,
      "learning_rate": 0.0004726125270608419,
      "loss": 5.5211,
      "mean_token_accuracy": 0.1646760955452919,
      "num_tokens": 38055399.0,
      "step": 18335
    },
    {
      "entropy": 5.786197662353516,
      "epoch": 1.6636429608127723,
      "grad_norm": 2.0625,
      "learning_rate": 0.00047259705377812703,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1617826595902443,
      "num_tokens": 38065442.0,
      "step": 18340
    },
    {
      "entropy": 5.707278442382813,
      "epoch": 1.6640965166908563,
      "grad_norm": 1.75,
      "learning_rate": 0.0004725815764090546,
      "loss": 5.4113,
      "mean_token_accuracy": 0.17536486685276031,
      "num_tokens": 38075416.0,
      "step": 18345
    },
    {
      "entropy": 5.710947275161743,
      "epoch": 1.6645500725689404,
      "grad_norm": 2.0,
      "learning_rate": 0.00047256609495394494,
      "loss": 5.4142,
      "mean_token_accuracy": 0.1761839359998703,
      "num_tokens": 38086168.0,
      "step": 18350
    },
    {
      "entropy": 5.680384826660156,
      "epoch": 1.6650036284470247,
      "grad_norm": 2.171875,
      "learning_rate": 0.000472550609413118,
      "loss": 5.3244,
      "mean_token_accuracy": 0.17612555027008056,
      "num_tokens": 38096173.0,
      "step": 18355
    },
    {
      "entropy": 5.580692911148072,
      "epoch": 1.665457184325109,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004725351197868942,
      "loss": 5.3471,
      "mean_token_accuracy": 0.17414311319589615,
      "num_tokens": 38106607.0,
      "step": 18360
    },
    {
      "entropy": 5.68683385848999,
      "epoch": 1.6659107402031932,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00047251962607559373,
      "loss": 5.3975,
      "mean_token_accuracy": 0.17683885395526885,
      "num_tokens": 38116317.0,
      "step": 18365
    },
    {
      "entropy": 5.661891126632691,
      "epoch": 1.6663642960812772,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047250412827953717,
      "loss": 5.3904,
      "mean_token_accuracy": 0.1704545333981514,
      "num_tokens": 38126338.0,
      "step": 18370
    },
    {
      "entropy": 5.684895610809326,
      "epoch": 1.6668178519593613,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004724886263990448,
      "loss": 5.3336,
      "mean_token_accuracy": 0.17971688359975815,
      "num_tokens": 38135471.0,
      "step": 18375
    },
    {
      "entropy": 5.679488849639893,
      "epoch": 1.6672714078374455,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004724731204344372,
      "loss": 5.4543,
      "mean_token_accuracy": 0.17279306352138518,
      "num_tokens": 38144824.0,
      "step": 18380
    },
    {
      "entropy": 5.6909936428070065,
      "epoch": 1.6677249637155298,
      "grad_norm": 2.1875,
      "learning_rate": 0.00047245761038603536,
      "loss": 5.4502,
      "mean_token_accuracy": 0.16216072887182237,
      "num_tokens": 38155443.0,
      "step": 18385
    },
    {
      "entropy": 5.70778226852417,
      "epoch": 1.668178519593614,
      "grad_norm": 2.0,
      "learning_rate": 0.00047244209625415975,
      "loss": 5.3956,
      "mean_token_accuracy": 0.16992000639438629,
      "num_tokens": 38166459.0,
      "step": 18390
    },
    {
      "entropy": 5.742944717407227,
      "epoch": 1.6686320754716981,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004724265780391312,
      "loss": 5.4527,
      "mean_token_accuracy": 0.17050793915987014,
      "num_tokens": 38176998.0,
      "step": 18395
    },
    {
      "entropy": 5.755467081069947,
      "epoch": 1.6690856313497822,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004724110557412708,
      "loss": 5.5196,
      "mean_token_accuracy": 0.16543755829334258,
      "num_tokens": 38188117.0,
      "step": 18400
    },
    {
      "entropy": 5.706991672515869,
      "epoch": 1.6695391872278664,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00047239552936089943,
      "loss": 5.3676,
      "mean_token_accuracy": 0.1724191427230835,
      "num_tokens": 38197622.0,
      "step": 18405
    },
    {
      "entropy": 5.742121887207031,
      "epoch": 1.6699927431059507,
      "grad_norm": 2.09375,
      "learning_rate": 0.00047237999889833816,
      "loss": 5.4457,
      "mean_token_accuracy": 0.16761334985494614,
      "num_tokens": 38208175.0,
      "step": 18410
    },
    {
      "entropy": 5.743998241424561,
      "epoch": 1.670446298984035,
      "grad_norm": 2.9375,
      "learning_rate": 0.00047236446435390826,
      "loss": 5.503,
      "mean_token_accuracy": 0.16686204224824905,
      "num_tokens": 38219095.0,
      "step": 18415
    },
    {
      "entropy": 5.66981315612793,
      "epoch": 1.670899854862119,
      "grad_norm": 2.234375,
      "learning_rate": 0.00047234892572793093,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17972978204488754,
      "num_tokens": 38229401.0,
      "step": 18420
    },
    {
      "entropy": 5.662116003036499,
      "epoch": 1.671353410740203,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004723333830207275,
      "loss": 5.4456,
      "mean_token_accuracy": 0.16907822489738464,
      "num_tokens": 38239928.0,
      "step": 18425
    },
    {
      "entropy": 5.704838466644287,
      "epoch": 1.6718069666182873,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047231783623261954,
      "loss": 5.3994,
      "mean_token_accuracy": 0.17386069893836975,
      "num_tokens": 38249911.0,
      "step": 18430
    },
    {
      "entropy": 5.667321491241455,
      "epoch": 1.6722605224963716,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004723022853639283,
      "loss": 5.4464,
      "mean_token_accuracy": 0.17605025172233582,
      "num_tokens": 38259587.0,
      "step": 18435
    },
    {
      "entropy": 5.681142139434814,
      "epoch": 1.6727140783744558,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004722867304149756,
      "loss": 5.5682,
      "mean_token_accuracy": 0.16499134749174119,
      "num_tokens": 38269979.0,
      "step": 18440
    },
    {
      "entropy": 5.8147725582122805,
      "epoch": 1.67316763425254,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004722711713860831,
      "loss": 5.4452,
      "mean_token_accuracy": 0.1681773766875267,
      "num_tokens": 38280109.0,
      "step": 18445
    },
    {
      "entropy": 5.71650881767273,
      "epoch": 1.673621190130624,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004722556082775724,
      "loss": 5.3783,
      "mean_token_accuracy": 0.16767581999301912,
      "num_tokens": 38290975.0,
      "step": 18450
    },
    {
      "entropy": 5.603062963485717,
      "epoch": 1.6740747460087082,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004722400410897655,
      "loss": 5.3792,
      "mean_token_accuracy": 0.17703232318162918,
      "num_tokens": 38301221.0,
      "step": 18455
    },
    {
      "entropy": 5.664209318161011,
      "epoch": 1.6745283018867925,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004722244698229842,
      "loss": 5.4077,
      "mean_token_accuracy": 0.16771954596042632,
      "num_tokens": 38311605.0,
      "step": 18460
    },
    {
      "entropy": 5.6579760074615475,
      "epoch": 1.6749818577648767,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047220889447755073,
      "loss": 5.3398,
      "mean_token_accuracy": 0.17419948875904084,
      "num_tokens": 38322121.0,
      "step": 18465
    },
    {
      "entropy": 5.732490968704224,
      "epoch": 1.6754354136429608,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004721933150537869,
      "loss": 5.429,
      "mean_token_accuracy": 0.16825764775276184,
      "num_tokens": 38333068.0,
      "step": 18470
    },
    {
      "entropy": 5.749762964248657,
      "epoch": 1.675888969521045,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047217773155201526,
      "loss": 5.4608,
      "mean_token_accuracy": 0.16745549142360688,
      "num_tokens": 38343548.0,
      "step": 18475
    },
    {
      "entropy": 5.691002416610718,
      "epoch": 1.676342525399129,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047216214397255776,
      "loss": 5.4587,
      "mean_token_accuracy": 0.1693328395485878,
      "num_tokens": 38352711.0,
      "step": 18480
    },
    {
      "entropy": 5.719344234466552,
      "epoch": 1.6767960812772134,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047214655231573693,
      "loss": 5.3959,
      "mean_token_accuracy": 0.17080460488796234,
      "num_tokens": 38363563.0,
      "step": 18485
    },
    {
      "entropy": 5.690249347686768,
      "epoch": 1.6772496371552976,
      "grad_norm": 2.0,
      "learning_rate": 0.0004721309565818752,
      "loss": 5.4306,
      "mean_token_accuracy": 0.17482383102178572,
      "num_tokens": 38374305.0,
      "step": 18490
    },
    {
      "entropy": 5.602746295928955,
      "epoch": 1.6777031930333817,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047211535677129496,
      "loss": 5.3772,
      "mean_token_accuracy": 0.17664432674646377,
      "num_tokens": 38384019.0,
      "step": 18495
    },
    {
      "entropy": 5.6821857452392575,
      "epoch": 1.678156748911466,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00047209975288431893,
      "loss": 5.4502,
      "mean_token_accuracy": 0.16617521047592163,
      "num_tokens": 38394804.0,
      "step": 18500
    },
    {
      "entropy": 5.653756237030029,
      "epoch": 1.67861030478955,
      "grad_norm": 2.640625,
      "learning_rate": 0.00047208414492126986,
      "loss": 5.3865,
      "mean_token_accuracy": 0.17486755549907684,
      "num_tokens": 38405096.0,
      "step": 18505
    },
    {
      "entropy": 5.69540696144104,
      "epoch": 1.6790638606676342,
      "grad_norm": 2.03125,
      "learning_rate": 0.00047206853288247043,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1752585530281067,
      "num_tokens": 38415181.0,
      "step": 18510
    },
    {
      "entropy": 5.698315715789795,
      "epoch": 1.6795174165457185,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047205291676824345,
      "loss": 5.3412,
      "mean_token_accuracy": 0.1734205186367035,
      "num_tokens": 38425299.0,
      "step": 18515
    },
    {
      "entropy": 5.689258050918579,
      "epoch": 1.6799709724238028,
      "grad_norm": 3.296875,
      "learning_rate": 0.000472037296578912,
      "loss": 5.4123,
      "mean_token_accuracy": 0.17675835639238358,
      "num_tokens": 38435779.0,
      "step": 18520
    },
    {
      "entropy": 5.705479669570923,
      "epoch": 1.6804245283018868,
      "grad_norm": 2.171875,
      "learning_rate": 0.000472021672314799,
      "loss": 5.5259,
      "mean_token_accuracy": 0.15896397307515145,
      "num_tokens": 38448470.0,
      "step": 18525
    },
    {
      "entropy": 5.6936986446380615,
      "epoch": 1.6808780841799709,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047200604397622763,
      "loss": 5.403,
      "mean_token_accuracy": 0.16784698963165284,
      "num_tokens": 38459938.0,
      "step": 18530
    },
    {
      "entropy": 5.654676342010498,
      "epoch": 1.6813316400580551,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004719904115635211,
      "loss": 5.3839,
      "mean_token_accuracy": 0.1740487262606621,
      "num_tokens": 38470119.0,
      "step": 18535
    },
    {
      "entropy": 5.741268634796143,
      "epoch": 1.6817851959361394,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004719747750770026,
      "loss": 5.4927,
      "mean_token_accuracy": 0.17022154182195665,
      "num_tokens": 38481629.0,
      "step": 18540
    },
    {
      "entropy": 5.6342555522918705,
      "epoch": 1.6822387518142237,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004719591345169956,
      "loss": 5.2712,
      "mean_token_accuracy": 0.17591495662927628,
      "num_tokens": 38491215.0,
      "step": 18545
    },
    {
      "entropy": 5.644246768951416,
      "epoch": 1.6826923076923077,
      "grad_norm": 4.65625,
      "learning_rate": 0.00047194348988382345,
      "loss": 5.2853,
      "mean_token_accuracy": 0.17928629219532013,
      "num_tokens": 38501761.0,
      "step": 18550
    },
    {
      "entropy": 5.689867830276489,
      "epoch": 1.6831458635703918,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004719278411778097,
      "loss": 5.4389,
      "mean_token_accuracy": 0.1734817862510681,
      "num_tokens": 38511755.0,
      "step": 18555
    },
    {
      "entropy": 5.67216682434082,
      "epoch": 1.683599419448476,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047191218839927817,
      "loss": 5.2603,
      "mean_token_accuracy": 0.18525208085775374,
      "num_tokens": 38522870.0,
      "step": 18560
    },
    {
      "entropy": 5.62730393409729,
      "epoch": 1.6840529753265603,
      "grad_norm": 5.375,
      "learning_rate": 0.0004718965315485523,
      "loss": 5.3595,
      "mean_token_accuracy": 0.1752931222319603,
      "num_tokens": 38533568.0,
      "step": 18565
    },
    {
      "entropy": 5.67253007888794,
      "epoch": 1.6845065312046446,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004718808706259561,
      "loss": 5.4716,
      "mean_token_accuracy": 0.17240085899829866,
      "num_tokens": 38543865.0,
      "step": 18570
    },
    {
      "entropy": 5.695382452011108,
      "epoch": 1.6849600870827286,
      "grad_norm": 3.90625,
      "learning_rate": 0.00047186520563181325,
      "loss": 5.3413,
      "mean_token_accuracy": 0.17095973193645478,
      "num_tokens": 38553585.0,
      "step": 18575
    },
    {
      "entropy": 5.644854354858398,
      "epoch": 1.6854136429608126,
      "grad_norm": 3.28125,
      "learning_rate": 0.0004718495365664477,
      "loss": 5.3793,
      "mean_token_accuracy": 0.17340226471424103,
      "num_tokens": 38563909.0,
      "step": 18580
    },
    {
      "entropy": 5.764812278747558,
      "epoch": 1.685867198838897,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047183386343018364,
      "loss": 5.5542,
      "mean_token_accuracy": 0.16160224080085756,
      "num_tokens": 38574787.0,
      "step": 18585
    },
    {
      "entropy": 5.759872341156006,
      "epoch": 1.6863207547169812,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004718181862233451,
      "loss": 5.3907,
      "mean_token_accuracy": 0.16919050216674805,
      "num_tokens": 38584897.0,
      "step": 18590
    },
    {
      "entropy": 5.642881870269775,
      "epoch": 1.6867743105950654,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004718025049462563,
      "loss": 5.4191,
      "mean_token_accuracy": 0.17234448939561844,
      "num_tokens": 38594292.0,
      "step": 18595
    },
    {
      "entropy": 5.6739426612854,
      "epoch": 1.6872278664731495,
      "grad_norm": 2.75,
      "learning_rate": 0.0004717868195992416,
      "loss": 5.3725,
      "mean_token_accuracy": 0.17971548140048982,
      "num_tokens": 38604875.0,
      "step": 18600
    },
    {
      "entropy": 5.761136913299561,
      "epoch": 1.6876814223512335,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004717711301826253,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1706874817609787,
      "num_tokens": 38615379.0,
      "step": 18605
    },
    {
      "entropy": 5.615822458267212,
      "epoch": 1.6881349782293178,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00047175543669673186,
      "loss": 5.3749,
      "mean_token_accuracy": 0.17192001342773439,
      "num_tokens": 38625150.0,
      "step": 18610
    },
    {
      "entropy": 5.690691947937012,
      "epoch": 1.688588534107402,
      "grad_norm": 1.875,
      "learning_rate": 0.00047173973914188587,
      "loss": 5.3538,
      "mean_token_accuracy": 0.17429824620485307,
      "num_tokens": 38634878.0,
      "step": 18615
    },
    {
      "entropy": 5.668873500823975,
      "epoch": 1.6890420899854863,
      "grad_norm": 2.125,
      "learning_rate": 0.00047172403751841196,
      "loss": 5.3446,
      "mean_token_accuracy": 0.1775931403040886,
      "num_tokens": 38644677.0,
      "step": 18620
    },
    {
      "entropy": 5.686873817443848,
      "epoch": 1.6894956458635704,
      "grad_norm": 2.59375,
      "learning_rate": 0.00047170833182663485,
      "loss": 5.4709,
      "mean_token_accuracy": 0.16693421751260756,
      "num_tokens": 38654918.0,
      "step": 18625
    },
    {
      "entropy": 5.686710548400879,
      "epoch": 1.6899492017416544,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004716926220668793,
      "loss": 5.4303,
      "mean_token_accuracy": 0.17145712226629256,
      "num_tokens": 38665477.0,
      "step": 18630
    },
    {
      "entropy": 5.740883731842041,
      "epoch": 1.6904027576197387,
      "grad_norm": 2.265625,
      "learning_rate": 0.00047167690823947015,
      "loss": 5.4241,
      "mean_token_accuracy": 0.1704484388232231,
      "num_tokens": 38676322.0,
      "step": 18635
    },
    {
      "entropy": 5.740987730026245,
      "epoch": 1.690856313497823,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004716611903447325,
      "loss": 5.4109,
      "mean_token_accuracy": 0.16829974055290223,
      "num_tokens": 38686673.0,
      "step": 18640
    },
    {
      "entropy": 5.648611688613892,
      "epoch": 1.6913098693759072,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004716454683829912,
      "loss": 5.44,
      "mean_token_accuracy": 0.16939992606639862,
      "num_tokens": 38696944.0,
      "step": 18645
    },
    {
      "entropy": 5.677216196060181,
      "epoch": 1.6917634252539913,
      "grad_norm": 3.125,
      "learning_rate": 0.00047162974235457163,
      "loss": 5.3073,
      "mean_token_accuracy": 0.17421704530715942,
      "num_tokens": 38707571.0,
      "step": 18650
    },
    {
      "entropy": 5.709779405593872,
      "epoch": 1.6922169811320755,
      "grad_norm": 2.0,
      "learning_rate": 0.0004716140122597988,
      "loss": 5.4032,
      "mean_token_accuracy": 0.16973363757133483,
      "num_tokens": 38717896.0,
      "step": 18655
    },
    {
      "entropy": 5.591789484024048,
      "epoch": 1.6926705370101596,
      "grad_norm": 2.0,
      "learning_rate": 0.0004715982780989981,
      "loss": 5.3437,
      "mean_token_accuracy": 0.1771623283624649,
      "num_tokens": 38727874.0,
      "step": 18660
    },
    {
      "entropy": 5.635274267196655,
      "epoch": 1.6931240928882438,
      "grad_norm": 2.15625,
      "learning_rate": 0.00047158253987249487,
      "loss": 5.439,
      "mean_token_accuracy": 0.16978606432676316,
      "num_tokens": 38737943.0,
      "step": 18665
    },
    {
      "entropy": 5.6656615257263185,
      "epoch": 1.693577648766328,
      "grad_norm": 2.515625,
      "learning_rate": 0.00047156679758061465,
      "loss": 5.4029,
      "mean_token_accuracy": 0.17496798634529115,
      "num_tokens": 38749388.0,
      "step": 18670
    },
    {
      "entropy": 5.681450605392456,
      "epoch": 1.6940312046444121,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047155105122368297,
      "loss": 5.3487,
      "mean_token_accuracy": 0.16702446341514587,
      "num_tokens": 38759839.0,
      "step": 18675
    },
    {
      "entropy": 5.631108331680298,
      "epoch": 1.6944847605224964,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004715353008020255,
      "loss": 5.4089,
      "mean_token_accuracy": 0.17313552051782607,
      "num_tokens": 38770623.0,
      "step": 18680
    },
    {
      "entropy": 5.659900808334351,
      "epoch": 1.6949383164005805,
      "grad_norm": 5.03125,
      "learning_rate": 0.00047151954631596794,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1643161028623581,
      "num_tokens": 38780878.0,
      "step": 18685
    },
    {
      "entropy": 5.677348232269287,
      "epoch": 1.6953918722786647,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047150378776583603,
      "loss": 5.3351,
      "mean_token_accuracy": 0.1757463276386261,
      "num_tokens": 38791083.0,
      "step": 18690
    },
    {
      "entropy": 5.630480766296387,
      "epoch": 1.695845428156749,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004714880251519557,
      "loss": 5.3436,
      "mean_token_accuracy": 0.17544830441474915,
      "num_tokens": 38801203.0,
      "step": 18695
    },
    {
      "entropy": 5.642328119277954,
      "epoch": 1.6962989840348333,
      "grad_norm": 2.28125,
      "learning_rate": 0.000471472258474653,
      "loss": 5.3844,
      "mean_token_accuracy": 0.17874170690774918,
      "num_tokens": 38811276.0,
      "step": 18700
    },
    {
      "entropy": 5.703634929656983,
      "epoch": 1.6967525399129173,
      "grad_norm": 2.703125,
      "learning_rate": 0.00047145648773425384,
      "loss": 5.5177,
      "mean_token_accuracy": 0.1631750151515007,
      "num_tokens": 38822980.0,
      "step": 18705
    },
    {
      "entropy": 5.710088062286377,
      "epoch": 1.6972060957910013,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047144071293108456,
      "loss": 5.3728,
      "mean_token_accuracy": 0.17340612560510635,
      "num_tokens": 38833510.0,
      "step": 18710
    },
    {
      "entropy": 5.66200819015503,
      "epoch": 1.6976596516690856,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004714249340654712,
      "loss": 5.3693,
      "mean_token_accuracy": 0.1692073479294777,
      "num_tokens": 38843844.0,
      "step": 18715
    },
    {
      "entropy": 5.632929754257202,
      "epoch": 1.6981132075471699,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004714091511377402,
      "loss": 5.404,
      "mean_token_accuracy": 0.17124924063682556,
      "num_tokens": 38854998.0,
      "step": 18720
    },
    {
      "entropy": 5.6701513767242435,
      "epoch": 1.6985667634252541,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00047139336414821784,
      "loss": 5.3695,
      "mean_token_accuracy": 0.1753828778862953,
      "num_tokens": 38865395.0,
      "step": 18725
    },
    {
      "entropy": 5.579617023468018,
      "epoch": 1.6990203193033382,
      "grad_norm": 2.46875,
      "learning_rate": 0.00047137757309723073,
      "loss": 5.2903,
      "mean_token_accuracy": 0.18436947613954544,
      "num_tokens": 38874933.0,
      "step": 18730
    },
    {
      "entropy": 5.665982055664062,
      "epoch": 1.6994738751814222,
      "grad_norm": 3.03125,
      "learning_rate": 0.0004713617779851053,
      "loss": 5.3975,
      "mean_token_accuracy": 0.17812635600566865,
      "num_tokens": 38885399.0,
      "step": 18735
    },
    {
      "entropy": 5.774350357055664,
      "epoch": 1.6999274310595065,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004713459788121683,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1678643435239792,
      "num_tokens": 38895576.0,
      "step": 18740
    },
    {
      "entropy": 5.681569957733155,
      "epoch": 1.7003809869375908,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004713301755787464,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17076998054981232,
      "num_tokens": 38906069.0,
      "step": 18745
    },
    {
      "entropy": 5.705448341369629,
      "epoch": 1.700834542815675,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004713143682851665,
      "loss": 5.3608,
      "mean_token_accuracy": 0.1760892853140831,
      "num_tokens": 38915759.0,
      "step": 18750
    },
    {
      "entropy": 5.595787191390992,
      "epoch": 1.701288098693759,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004712985569317553,
      "loss": 5.2698,
      "mean_token_accuracy": 0.17973075658082963,
      "num_tokens": 38925514.0,
      "step": 18755
    },
    {
      "entropy": 5.613077068328858,
      "epoch": 1.7017416545718431,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004712827415188401,
      "loss": 5.4433,
      "mean_token_accuracy": 0.16732046008110046,
      "num_tokens": 38936433.0,
      "step": 18760
    },
    {
      "entropy": 5.705737018585205,
      "epoch": 1.7021952104499274,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004712669220467476,
      "loss": 5.4167,
      "mean_token_accuracy": 0.16792113929986954,
      "num_tokens": 38947044.0,
      "step": 18765
    },
    {
      "entropy": 5.691990709304809,
      "epoch": 1.7026487663280117,
      "grad_norm": 2.328125,
      "learning_rate": 0.00047125109851580533,
      "loss": 5.4355,
      "mean_token_accuracy": 0.1669414684176445,
      "num_tokens": 38957121.0,
      "step": 18770
    },
    {
      "entropy": 5.7733087062835695,
      "epoch": 1.703102322206096,
      "grad_norm": 1.71875,
      "learning_rate": 0.00047123527092634025,
      "loss": 5.4587,
      "mean_token_accuracy": 0.17347999960184096,
      "num_tokens": 38967120.0,
      "step": 18775
    },
    {
      "entropy": 5.650178527832031,
      "epoch": 1.70355587808418,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004712194392786797,
      "loss": 5.3751,
      "mean_token_accuracy": 0.17327772229909896,
      "num_tokens": 38978012.0,
      "step": 18780
    },
    {
      "entropy": 5.633814764022827,
      "epoch": 1.704009433962264,
      "grad_norm": 5.78125,
      "learning_rate": 0.00047120360357315127,
      "loss": 5.3781,
      "mean_token_accuracy": 0.17078096568584442,
      "num_tokens": 38988308.0,
      "step": 18785
    },
    {
      "entropy": 5.666831684112549,
      "epoch": 1.7044629898403483,
      "grad_norm": 2.015625,
      "learning_rate": 0.00047118776381008216,
      "loss": 5.3667,
      "mean_token_accuracy": 0.1753005012869835,
      "num_tokens": 38998163.0,
      "step": 18790
    },
    {
      "entropy": 5.721426153182984,
      "epoch": 1.7049165457184325,
      "grad_norm": 1.53125,
      "learning_rate": 0.00047117191998980026,
      "loss": 5.4135,
      "mean_token_accuracy": 0.1713220700621605,
      "num_tokens": 39009065.0,
      "step": 18795
    },
    {
      "entropy": 5.66545000076294,
      "epoch": 1.7053701015965168,
      "grad_norm": 1.96875,
      "learning_rate": 0.00047115607211263295,
      "loss": 5.3807,
      "mean_token_accuracy": 0.17337886095046998,
      "num_tokens": 39019708.0,
      "step": 18800
    },
    {
      "entropy": 5.605272626876831,
      "epoch": 1.7058236574746009,
      "grad_norm": 2.453125,
      "learning_rate": 0.00047114022017890814,
      "loss": 5.4014,
      "mean_token_accuracy": 0.17380050718784332,
      "num_tokens": 39029755.0,
      "step": 18805
    },
    {
      "entropy": 5.595337200164795,
      "epoch": 1.706277213352685,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004711243641889536,
      "loss": 5.3306,
      "mean_token_accuracy": 0.17734959572553635,
      "num_tokens": 39040811.0,
      "step": 18810
    },
    {
      "entropy": 5.659038925170899,
      "epoch": 1.7067307692307692,
      "grad_norm": 1.90625,
      "learning_rate": 0.00047110850414309714,
      "loss": 5.2866,
      "mean_token_accuracy": 0.1813986286520958,
      "num_tokens": 39050598.0,
      "step": 18815
    },
    {
      "entropy": 5.623749828338623,
      "epoch": 1.7071843251088534,
      "grad_norm": 2.4375,
      "learning_rate": 0.00047109264004166696,
      "loss": 5.3442,
      "mean_token_accuracy": 0.1714172288775444,
      "num_tokens": 39061756.0,
      "step": 18820
    },
    {
      "entropy": 5.729853296279908,
      "epoch": 1.7076378809869377,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004710767718849909,
      "loss": 5.463,
      "mean_token_accuracy": 0.17219043672084808,
      "num_tokens": 39071929.0,
      "step": 18825
    },
    {
      "entropy": 5.652404260635376,
      "epoch": 1.7080914368650217,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004710608996733973,
      "loss": 5.4752,
      "mean_token_accuracy": 0.17130574882030486,
      "num_tokens": 39082260.0,
      "step": 18830
    },
    {
      "entropy": 5.6313660621643065,
      "epoch": 1.708544992743106,
      "grad_norm": 2.125,
      "learning_rate": 0.00047104502340721433,
      "loss": 5.3752,
      "mean_token_accuracy": 0.17230693399906158,
      "num_tokens": 39091697.0,
      "step": 18835
    },
    {
      "entropy": 5.743632221221924,
      "epoch": 1.70899854862119,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00047102914308677033,
      "loss": 5.4567,
      "mean_token_accuracy": 0.16853480488061906,
      "num_tokens": 39101716.0,
      "step": 18840
    },
    {
      "entropy": 5.709078407287597,
      "epoch": 1.7094521044992743,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00047101325871239363,
      "loss": 5.393,
      "mean_token_accuracy": 0.17120738029479982,
      "num_tokens": 39111573.0,
      "step": 18845
    },
    {
      "entropy": 5.657770776748658,
      "epoch": 1.7099056603773586,
      "grad_norm": 2.71875,
      "learning_rate": 0.0004709973702844128,
      "loss": 5.4008,
      "mean_token_accuracy": 0.1722239598631859,
      "num_tokens": 39122711.0,
      "step": 18850
    },
    {
      "entropy": 5.732843112945557,
      "epoch": 1.7103592162554426,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00047098147780315645,
      "loss": 5.4411,
      "mean_token_accuracy": 0.1714389517903328,
      "num_tokens": 39132438.0,
      "step": 18855
    },
    {
      "entropy": 5.68795075416565,
      "epoch": 1.710812772133527,
      "grad_norm": 2.328125,
      "learning_rate": 0.0004709655812689532,
      "loss": 5.3644,
      "mean_token_accuracy": 0.17164712250232697,
      "num_tokens": 39143082.0,
      "step": 18860
    },
    {
      "entropy": 5.633641195297241,
      "epoch": 1.711266328011611,
      "grad_norm": 2.734375,
      "learning_rate": 0.0004709496806821318,
      "loss": 5.3365,
      "mean_token_accuracy": 0.1763573557138443,
      "num_tokens": 39153290.0,
      "step": 18865
    },
    {
      "entropy": 5.686541223526001,
      "epoch": 1.7117198838896952,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00047093377604302106,
      "loss": 5.4125,
      "mean_token_accuracy": 0.17098568230867386,
      "num_tokens": 39163151.0,
      "step": 18870
    },
    {
      "entropy": 5.656773900985717,
      "epoch": 1.7121734397677795,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004709178673519499,
      "loss": 5.3429,
      "mean_token_accuracy": 0.17405557334423066,
      "num_tokens": 39173412.0,
      "step": 18875
    },
    {
      "entropy": 5.6808959484100345,
      "epoch": 1.7126269956458637,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004709019546092473,
      "loss": 5.3646,
      "mean_token_accuracy": 0.16804643720388412,
      "num_tokens": 39184310.0,
      "step": 18880
    },
    {
      "entropy": 5.747192430496216,
      "epoch": 1.7130805515239478,
      "grad_norm": 1.734375,
      "learning_rate": 0.00047088603781524237,
      "loss": 5.5579,
      "mean_token_accuracy": 0.15856632739305496,
      "num_tokens": 39194224.0,
      "step": 18885
    },
    {
      "entropy": 5.7005867004394535,
      "epoch": 1.7135341074020318,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004708701169702642,
      "loss": 5.4135,
      "mean_token_accuracy": 0.17177740782499312,
      "num_tokens": 39204993.0,
      "step": 18890
    },
    {
      "entropy": 5.703170442581177,
      "epoch": 1.713987663280116,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004708541920746422,
      "loss": 5.4195,
      "mean_token_accuracy": 0.16379446387290955,
      "num_tokens": 39215958.0,
      "step": 18895
    },
    {
      "entropy": 5.679533815383911,
      "epoch": 1.7144412191582004,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004708382631287055,
      "loss": 5.4556,
      "mean_token_accuracy": 0.16958072483539582,
      "num_tokens": 39226617.0,
      "step": 18900
    },
    {
      "entropy": 5.667512845993042,
      "epoch": 1.7148947750362846,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004708223301327837,
      "loss": 5.4086,
      "mean_token_accuracy": 0.16738865077495574,
      "num_tokens": 39237801.0,
      "step": 18905
    },
    {
      "entropy": 5.726097345352173,
      "epoch": 1.7153483309143687,
      "grad_norm": 2.109375,
      "learning_rate": 0.00047080639308720617,
      "loss": 5.5114,
      "mean_token_accuracy": 0.16872479170560836,
      "num_tokens": 39247582.0,
      "step": 18910
    },
    {
      "entropy": 5.740476274490357,
      "epoch": 1.7158018867924527,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047079045199230254,
      "loss": 5.3972,
      "mean_token_accuracy": 0.17636481374502183,
      "num_tokens": 39257891.0,
      "step": 18915
    },
    {
      "entropy": 5.655204343795776,
      "epoch": 1.716255442670537,
      "grad_norm": 1.765625,
      "learning_rate": 0.00047077450684840247,
      "loss": 5.3689,
      "mean_token_accuracy": 0.17481369525194168,
      "num_tokens": 39267483.0,
      "step": 18920
    },
    {
      "entropy": 5.82027587890625,
      "epoch": 1.7167089985486212,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004707585576558357,
      "loss": 5.5599,
      "mean_token_accuracy": 0.16796689927577974,
      "num_tokens": 39278450.0,
      "step": 18925
    },
    {
      "entropy": 5.7432512760162355,
      "epoch": 1.7171625544267055,
      "grad_norm": 2.375,
      "learning_rate": 0.0004707426044149321,
      "loss": 5.4476,
      "mean_token_accuracy": 0.16782187819480895,
      "num_tokens": 39288779.0,
      "step": 18930
    },
    {
      "entropy": 5.670406913757324,
      "epoch": 1.7176161103047896,
      "grad_norm": 2.125,
      "learning_rate": 0.00047072664712602153,
      "loss": 5.4302,
      "mean_token_accuracy": 0.17338635474443437,
      "num_tokens": 39298474.0,
      "step": 18935
    },
    {
      "entropy": 5.596939277648926,
      "epoch": 1.7180696661828736,
      "grad_norm": 2.859375,
      "learning_rate": 0.000470710685789434,
      "loss": 5.324,
      "mean_token_accuracy": 0.18023664355278016,
      "num_tokens": 39307947.0,
      "step": 18940
    },
    {
      "entropy": 5.726638841629028,
      "epoch": 1.7185232220609579,
      "grad_norm": 2.984375,
      "learning_rate": 0.00047069472040549965,
      "loss": 5.4267,
      "mean_token_accuracy": 0.17514025717973708,
      "num_tokens": 39318512.0,
      "step": 18945
    },
    {
      "entropy": 5.715106534957886,
      "epoch": 1.7189767779390421,
      "grad_norm": 2.0,
      "learning_rate": 0.0004706787509745486,
      "loss": 5.3761,
      "mean_token_accuracy": 0.16833135336637498,
      "num_tokens": 39329200.0,
      "step": 18950
    },
    {
      "entropy": 5.6895232677459715,
      "epoch": 1.7194303338171264,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00047066277749691115,
      "loss": 5.3722,
      "mean_token_accuracy": 0.17169544100761414,
      "num_tokens": 39340583.0,
      "step": 18955
    },
    {
      "entropy": 5.599074459075927,
      "epoch": 1.7198838896952104,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004706467999729175,
      "loss": 5.3345,
      "mean_token_accuracy": 0.17773206084966658,
      "num_tokens": 39350083.0,
      "step": 18960
    },
    {
      "entropy": 5.618015289306641,
      "epoch": 1.7203374455732945,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004706308184028983,
      "loss": 5.3132,
      "mean_token_accuracy": 0.18091168701648713,
      "num_tokens": 39360652.0,
      "step": 18965
    },
    {
      "entropy": 5.599251413345337,
      "epoch": 1.7207910014513788,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004706148327871839,
      "loss": 5.449,
      "mean_token_accuracy": 0.16626174002885818,
      "num_tokens": 39370773.0,
      "step": 18970
    },
    {
      "entropy": 5.65184178352356,
      "epoch": 1.721244557329463,
      "grad_norm": 1.75,
      "learning_rate": 0.00047059884312610487,
      "loss": 5.3803,
      "mean_token_accuracy": 0.17386310994625093,
      "num_tokens": 39380786.0,
      "step": 18975
    },
    {
      "entropy": 5.547905015945434,
      "epoch": 1.7216981132075473,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00047058284941999185,
      "loss": 5.2852,
      "mean_token_accuracy": 0.17704043984413148,
      "num_tokens": 39389886.0,
      "step": 18980
    },
    {
      "entropy": 5.656164169311523,
      "epoch": 1.7221516690856313,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004705668516691758,
      "loss": 5.4003,
      "mean_token_accuracy": 0.1785064309835434,
      "num_tokens": 39399761.0,
      "step": 18985
    },
    {
      "entropy": 5.703855657577515,
      "epoch": 1.7226052249637154,
      "grad_norm": 1.578125,
      "learning_rate": 0.00047055084987398735,
      "loss": 5.4522,
      "mean_token_accuracy": 0.16916029304265975,
      "num_tokens": 39410449.0,
      "step": 18990
    },
    {
      "entropy": 5.708486413955688,
      "epoch": 1.7230587808417996,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004705348440347575,
      "loss": 5.4071,
      "mean_token_accuracy": 0.17505107671022416,
      "num_tokens": 39421029.0,
      "step": 18995
    },
    {
      "entropy": 5.655918264389038,
      "epoch": 1.723512336719884,
      "grad_norm": 2.421875,
      "learning_rate": 0.00047051883415181724,
      "loss": 5.3762,
      "mean_token_accuracy": 0.16671344339847566,
      "num_tokens": 39431306.0,
      "step": 19000
    },
    {
      "entropy": 5.715918445587159,
      "epoch": 1.7239658925979682,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004705028202254977,
      "loss": 5.368,
      "mean_token_accuracy": 0.17576221376657486,
      "num_tokens": 39441509.0,
      "step": 19005
    },
    {
      "entropy": 5.62521767616272,
      "epoch": 1.7244194484760522,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047048680225612996,
      "loss": 5.3411,
      "mean_token_accuracy": 0.17799898087978364,
      "num_tokens": 39451115.0,
      "step": 19010
    },
    {
      "entropy": 5.625340747833252,
      "epoch": 1.7248730043541363,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004704707802440454,
      "loss": 5.3367,
      "mean_token_accuracy": 0.1740262284874916,
      "num_tokens": 39461523.0,
      "step": 19015
    },
    {
      "entropy": 5.589021348953247,
      "epoch": 1.7253265602322205,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047045475418957524,
      "loss": 5.3359,
      "mean_token_accuracy": 0.17487259954214096,
      "num_tokens": 39470974.0,
      "step": 19020
    },
    {
      "entropy": 5.660117340087891,
      "epoch": 1.7257801161103048,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004704387240930509,
      "loss": 5.4251,
      "mean_token_accuracy": 0.17514911741018296,
      "num_tokens": 39481152.0,
      "step": 19025
    },
    {
      "entropy": 5.704448986053467,
      "epoch": 1.726233671988389,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004704226899548041,
      "loss": 5.4698,
      "mean_token_accuracy": 0.16403055638074876,
      "num_tokens": 39492086.0,
      "step": 19030
    },
    {
      "entropy": 5.681482362747192,
      "epoch": 1.726687227866473,
      "grad_norm": 2.546875,
      "learning_rate": 0.00047040665177516616,
      "loss": 5.4045,
      "mean_token_accuracy": 0.17013614922761916,
      "num_tokens": 39502505.0,
      "step": 19035
    },
    {
      "entropy": 5.785604572296142,
      "epoch": 1.7271407837445574,
      "grad_norm": 2.609375,
      "learning_rate": 0.00047039060955446886,
      "loss": 5.5489,
      "mean_token_accuracy": 0.16352784335613252,
      "num_tokens": 39513101.0,
      "step": 19040
    },
    {
      "entropy": 5.629688024520874,
      "epoch": 1.7275943396226414,
      "grad_norm": 2.0625,
      "learning_rate": 0.000470374563293044,
      "loss": 5.366,
      "mean_token_accuracy": 0.17199003100395202,
      "num_tokens": 39524110.0,
      "step": 19045
    },
    {
      "entropy": 5.7686182975769045,
      "epoch": 1.7280478955007257,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004703585129912234,
      "loss": 5.5022,
      "mean_token_accuracy": 0.16652651280164718,
      "num_tokens": 39535043.0,
      "step": 19050
    },
    {
      "entropy": 5.755824851989746,
      "epoch": 1.72850145137881,
      "grad_norm": 3.109375,
      "learning_rate": 0.00047034245864933894,
      "loss": 5.3999,
      "mean_token_accuracy": 0.17477021664381026,
      "num_tokens": 39545262.0,
      "step": 19055
    },
    {
      "entropy": 5.654735136032104,
      "epoch": 1.728955007256894,
      "grad_norm": 2.5625,
      "learning_rate": 0.00047032640026772274,
      "loss": 5.4101,
      "mean_token_accuracy": 0.17097359150648117,
      "num_tokens": 39555544.0,
      "step": 19060
    },
    {
      "entropy": 5.717685222625732,
      "epoch": 1.7294085631349783,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004703103378467067,
      "loss": 5.4741,
      "mean_token_accuracy": 0.17517699003219606,
      "num_tokens": 39564839.0,
      "step": 19065
    },
    {
      "entropy": 5.753581714630127,
      "epoch": 1.7298621190130623,
      "grad_norm": 2.40625,
      "learning_rate": 0.00047029427138662314,
      "loss": 5.493,
      "mean_token_accuracy": 0.1711267501115799,
      "num_tokens": 39575864.0,
      "step": 19070
    },
    {
      "entropy": 5.6841637134552006,
      "epoch": 1.7303156748911466,
      "grad_norm": 2.796875,
      "learning_rate": 0.0004702782008878042,
      "loss": 5.4178,
      "mean_token_accuracy": 0.17441748976707458,
      "num_tokens": 39585868.0,
      "step": 19075
    },
    {
      "entropy": 5.657657146453857,
      "epoch": 1.7307692307692308,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004702621263505824,
      "loss": 5.3382,
      "mean_token_accuracy": 0.16884157806634903,
      "num_tokens": 39596957.0,
      "step": 19080
    },
    {
      "entropy": 5.719500207901001,
      "epoch": 1.731222786647315,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004702460477752901,
      "loss": 5.3898,
      "mean_token_accuracy": 0.17398527562618255,
      "num_tokens": 39607544.0,
      "step": 19085
    },
    {
      "entropy": 5.718913555145264,
      "epoch": 1.7316763425253991,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004702299651622597,
      "loss": 5.4412,
      "mean_token_accuracy": 0.16733229756355286,
      "num_tokens": 39617362.0,
      "step": 19090
    },
    {
      "entropy": 5.6435404300689695,
      "epoch": 1.7321298984034832,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004702138785118239,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1719333589076996,
      "num_tokens": 39628334.0,
      "step": 19095
    },
    {
      "entropy": 5.687148475646973,
      "epoch": 1.7325834542815675,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004701977878243153,
      "loss": 5.4763,
      "mean_token_accuracy": 0.17610419541597366,
      "num_tokens": 39638034.0,
      "step": 19100
    },
    {
      "entropy": 5.703787040710449,
      "epoch": 1.7330370101596517,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004701816931000668,
      "loss": 5.4119,
      "mean_token_accuracy": 0.16771456599235535,
      "num_tokens": 39647994.0,
      "step": 19105
    },
    {
      "entropy": 5.7236714363098145,
      "epoch": 1.733490566037736,
      "grad_norm": 1.78125,
      "learning_rate": 0.00047016559433941106,
      "loss": 5.4413,
      "mean_token_accuracy": 0.17255171686410903,
      "num_tokens": 39658468.0,
      "step": 19110
    },
    {
      "entropy": 5.710063743591308,
      "epoch": 1.73394412191582,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004701494915426812,
      "loss": 5.3597,
      "mean_token_accuracy": 0.17649462819099426,
      "num_tokens": 39669363.0,
      "step": 19115
    },
    {
      "entropy": 5.69317512512207,
      "epoch": 1.734397677793904,
      "grad_norm": 3.765625,
      "learning_rate": 0.00047013338471021,
      "loss": 5.4321,
      "mean_token_accuracy": 0.1676270470023155,
      "num_tokens": 39680624.0,
      "step": 19120
    },
    {
      "entropy": 5.63647780418396,
      "epoch": 1.7348512336719883,
      "grad_norm": 2.3125,
      "learning_rate": 0.00047011727384233074,
      "loss": 5.3693,
      "mean_token_accuracy": 0.17547402083873748,
      "num_tokens": 39690257.0,
      "step": 19125
    },
    {
      "entropy": 5.661195230484009,
      "epoch": 1.7353047895500726,
      "grad_norm": 3.0625,
      "learning_rate": 0.0004701011589393765,
      "loss": 5.3289,
      "mean_token_accuracy": 0.18081333488225937,
      "num_tokens": 39700014.0,
      "step": 19130
    },
    {
      "entropy": 5.64784574508667,
      "epoch": 1.7357583454281569,
      "grad_norm": 3.4375,
      "learning_rate": 0.00047008504000168054,
      "loss": 5.3153,
      "mean_token_accuracy": 0.1773527592420578,
      "num_tokens": 39709915.0,
      "step": 19135
    },
    {
      "entropy": 5.713722848892212,
      "epoch": 1.736211901306241,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004700689170295763,
      "loss": 5.4155,
      "mean_token_accuracy": 0.17315206825733184,
      "num_tokens": 39720867.0,
      "step": 19140
    },
    {
      "entropy": 5.6741396427154545,
      "epoch": 1.736665457184325,
      "grad_norm": 3.078125,
      "learning_rate": 0.00047005279002339717,
      "loss": 5.4257,
      "mean_token_accuracy": 0.1735614314675331,
      "num_tokens": 39731798.0,
      "step": 19145
    },
    {
      "entropy": 5.650100612640381,
      "epoch": 1.7371190130624092,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047003665898347663,
      "loss": 5.4045,
      "mean_token_accuracy": 0.1675522357225418,
      "num_tokens": 39741786.0,
      "step": 19150
    },
    {
      "entropy": 5.662878465652466,
      "epoch": 1.7375725689404935,
      "grad_norm": 3.109375,
      "learning_rate": 0.0004700205239101483,
      "loss": 5.2898,
      "mean_token_accuracy": 0.17185100615024568,
      "num_tokens": 39752421.0,
      "step": 19155
    },
    {
      "entropy": 5.677902555465698,
      "epoch": 1.7380261248185778,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004700043848037458,
      "loss": 5.4365,
      "mean_token_accuracy": 0.16978211849927902,
      "num_tokens": 39762734.0,
      "step": 19160
    },
    {
      "entropy": 5.658264827728272,
      "epoch": 1.7384796806966618,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004699882416646029,
      "loss": 5.3217,
      "mean_token_accuracy": 0.1758035272359848,
      "num_tokens": 39773288.0,
      "step": 19165
    },
    {
      "entropy": 5.712057828903198,
      "epoch": 1.7389332365747459,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004699720944930535,
      "loss": 5.4774,
      "mean_token_accuracy": 0.16997111290693284,
      "num_tokens": 39783801.0,
      "step": 19170
    },
    {
      "entropy": 5.723344326019287,
      "epoch": 1.7393867924528301,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046995594328943147,
      "loss": 5.4244,
      "mean_token_accuracy": 0.16960949301719666,
      "num_tokens": 39794396.0,
      "step": 19175
    },
    {
      "entropy": 5.706939792633056,
      "epoch": 1.7398403483309144,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004699397880540709,
      "loss": 5.4048,
      "mean_token_accuracy": 0.1667855441570282,
      "num_tokens": 39805368.0,
      "step": 19180
    },
    {
      "entropy": 5.608292055130005,
      "epoch": 1.7402939042089987,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004699236287873058,
      "loss": 5.3647,
      "mean_token_accuracy": 0.16819716691970826,
      "num_tokens": 39815707.0,
      "step": 19185
    },
    {
      "entropy": 5.701311826705933,
      "epoch": 1.7407474600870827,
      "grad_norm": 2.125,
      "learning_rate": 0.0004699074654894705,
      "loss": 5.4304,
      "mean_token_accuracy": 0.17460198253393172,
      "num_tokens": 39825308.0,
      "step": 19190
    },
    {
      "entropy": 5.60925030708313,
      "epoch": 1.7412010159651667,
      "grad_norm": 1.59375,
      "learning_rate": 0.000469891298160899,
      "loss": 5.2866,
      "mean_token_accuracy": 0.17659933716058732,
      "num_tokens": 39835200.0,
      "step": 19195
    },
    {
      "entropy": 5.7552412986755375,
      "epoch": 1.741654571843251,
      "grad_norm": 3.453125,
      "learning_rate": 0.00046987512680192586,
      "loss": 5.5126,
      "mean_token_accuracy": 0.17015811204910278,
      "num_tokens": 39846610.0,
      "step": 19200
    },
    {
      "entropy": 5.700116348266602,
      "epoch": 1.7421081277213353,
      "grad_norm": 2.890625,
      "learning_rate": 0.00046985895141288546,
      "loss": 5.3291,
      "mean_token_accuracy": 0.1731235221028328,
      "num_tokens": 39857541.0,
      "step": 19205
    },
    {
      "entropy": 5.650954818725586,
      "epoch": 1.7425616835994195,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046984277199411224,
      "loss": 5.445,
      "mean_token_accuracy": 0.16735921204090118,
      "num_tokens": 39868942.0,
      "step": 19210
    },
    {
      "entropy": 5.61114821434021,
      "epoch": 1.7430152394775036,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046982658854594083,
      "loss": 5.3191,
      "mean_token_accuracy": 0.17624438256025315,
      "num_tokens": 39879976.0,
      "step": 19215
    },
    {
      "entropy": 5.635677433013916,
      "epoch": 1.7434687953555879,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046981040106870595,
      "loss": 5.318,
      "mean_token_accuracy": 0.17549935430288316,
      "num_tokens": 39889698.0,
      "step": 19220
    },
    {
      "entropy": 5.563587236404419,
      "epoch": 1.743922351233672,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004697942095627423,
      "loss": 5.294,
      "mean_token_accuracy": 0.17744928151369094,
      "num_tokens": 39899904.0,
      "step": 19225
    },
    {
      "entropy": 5.656528806686401,
      "epoch": 1.7443759071117562,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004697780140283848,
      "loss": 5.3739,
      "mean_token_accuracy": 0.17538526952266692,
      "num_tokens": 39910090.0,
      "step": 19230
    },
    {
      "entropy": 5.7019635200500485,
      "epoch": 1.7448294629898404,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004697618144659683,
      "loss": 5.4065,
      "mean_token_accuracy": 0.17450280785560607,
      "num_tokens": 39920142.0,
      "step": 19235
    },
    {
      "entropy": 5.721332740783692,
      "epoch": 1.7452830188679245,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004697456108758278,
      "loss": 5.4062,
      "mean_token_accuracy": 0.17752865254878997,
      "num_tokens": 39929974.0,
      "step": 19240
    },
    {
      "entropy": 5.691800498962403,
      "epoch": 1.7457365747460087,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004697294032582986,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1725692093372345,
      "num_tokens": 39940590.0,
      "step": 19245
    },
    {
      "entropy": 5.672344398498535,
      "epoch": 1.7461901306240928,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046971319161371564,
      "loss": 5.3818,
      "mean_token_accuracy": 0.17206360548734664,
      "num_tokens": 39951351.0,
      "step": 19250
    },
    {
      "entropy": 5.692066621780396,
      "epoch": 1.746643686502177,
      "grad_norm": 1.453125,
      "learning_rate": 0.00046969697594241426,
      "loss": 5.4909,
      "mean_token_accuracy": 0.16929427832365035,
      "num_tokens": 39961306.0,
      "step": 19255
    },
    {
      "entropy": 5.707557964324951,
      "epoch": 1.7470972423802613,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004696807562447298,
      "loss": 5.4052,
      "mean_token_accuracy": 0.16785972267389299,
      "num_tokens": 39972220.0,
      "step": 19260
    },
    {
      "entropy": 5.741218328475952,
      "epoch": 1.7475507982583456,
      "grad_norm": 2.375,
      "learning_rate": 0.00046966453252099763,
      "loss": 5.3815,
      "mean_token_accuracy": 0.1689988061785698,
      "num_tokens": 39981833.0,
      "step": 19265
    },
    {
      "entropy": 5.58036699295044,
      "epoch": 1.7480043541364296,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00046964830477155335,
      "loss": 5.3019,
      "mean_token_accuracy": 0.17547763884067535,
      "num_tokens": 39992416.0,
      "step": 19270
    },
    {
      "entropy": 5.700848150253296,
      "epoch": 1.7484579100145137,
      "grad_norm": 1.625,
      "learning_rate": 0.00046963207299673257,
      "loss": 5.4665,
      "mean_token_accuracy": 0.16929229944944382,
      "num_tokens": 40002368.0,
      "step": 19275
    },
    {
      "entropy": 5.775940990447998,
      "epoch": 1.748911465892598,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046961583719687085,
      "loss": 5.4597,
      "mean_token_accuracy": 0.16667312681674956,
      "num_tokens": 40012699.0,
      "step": 19280
    },
    {
      "entropy": 5.635675239562988,
      "epoch": 1.7493650217706822,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046959959737230414,
      "loss": 5.3681,
      "mean_token_accuracy": 0.17212467789649963,
      "num_tokens": 40022942.0,
      "step": 19285
    },
    {
      "entropy": 5.7600545406341555,
      "epoch": 1.7498185776487665,
      "grad_norm": 2.0,
      "learning_rate": 0.00046958335352336806,
      "loss": 5.4545,
      "mean_token_accuracy": 0.17119124978780748,
      "num_tokens": 40033138.0,
      "step": 19290
    },
    {
      "entropy": 5.719373893737793,
      "epoch": 1.7502721335268505,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004695671056503987,
      "loss": 5.3775,
      "mean_token_accuracy": 0.16999207884073259,
      "num_tokens": 40043874.0,
      "step": 19295
    },
    {
      "entropy": 5.75408706665039,
      "epoch": 1.7507256894049346,
      "grad_norm": 2.359375,
      "learning_rate": 0.000469550853753732,
      "loss": 5.5241,
      "mean_token_accuracy": 0.16760282218456268,
      "num_tokens": 40054164.0,
      "step": 19300
    },
    {
      "entropy": 5.6890871047973635,
      "epoch": 1.7511792452830188,
      "grad_norm": 2.3125,
      "learning_rate": 0.000469534597833704,
      "loss": 5.3865,
      "mean_token_accuracy": 0.17264487445354462,
      "num_tokens": 40064937.0,
      "step": 19305
    },
    {
      "entropy": 5.717933082580567,
      "epoch": 1.751632801161103,
      "grad_norm": 1.625,
      "learning_rate": 0.000469518337890651,
      "loss": 5.4482,
      "mean_token_accuracy": 0.16705314964056014,
      "num_tokens": 40074693.0,
      "step": 19310
    },
    {
      "entropy": 5.619909286499023,
      "epoch": 1.7520863570391874,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046950207392490917,
      "loss": 5.3833,
      "mean_token_accuracy": 0.1703799456357956,
      "num_tokens": 40085906.0,
      "step": 19315
    },
    {
      "entropy": 5.664641809463501,
      "epoch": 1.7525399129172714,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046948580593681496,
      "loss": 5.3928,
      "mean_token_accuracy": 0.17215481102466584,
      "num_tokens": 40096278.0,
      "step": 19320
    },
    {
      "entropy": 5.713938903808594,
      "epoch": 1.7529934687953554,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004694695339267047,
      "loss": 5.4302,
      "mean_token_accuracy": 0.17210033982992173,
      "num_tokens": 40106118.0,
      "step": 19325
    },
    {
      "entropy": 5.760051584243774,
      "epoch": 1.7534470246734397,
      "grad_norm": 2.671875,
      "learning_rate": 0.00046945325789491483,
      "loss": 5.4766,
      "mean_token_accuracy": 0.16487178206443787,
      "num_tokens": 40115765.0,
      "step": 19330
    },
    {
      "entropy": 5.703781652450561,
      "epoch": 1.753900580551524,
      "grad_norm": 2.25,
      "learning_rate": 0.0004694369778417822,
      "loss": 5.3656,
      "mean_token_accuracy": 0.17295841723680497,
      "num_tokens": 40126419.0,
      "step": 19335
    },
    {
      "entropy": 5.6655535221099855,
      "epoch": 1.7543541364296082,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004694206937676431,
      "loss": 5.3827,
      "mean_token_accuracy": 0.16516033858060836,
      "num_tokens": 40137411.0,
      "step": 19340
    },
    {
      "entropy": 5.649045181274414,
      "epoch": 1.7548076923076923,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004694044056728347,
      "loss": 5.4015,
      "mean_token_accuracy": 0.1747464880347252,
      "num_tokens": 40147424.0,
      "step": 19345
    },
    {
      "entropy": 5.655058813095093,
      "epoch": 1.7552612481857763,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046938811355769363,
      "loss": 5.4084,
      "mean_token_accuracy": 0.17384084761142732,
      "num_tokens": 40157549.0,
      "step": 19350
    },
    {
      "entropy": 5.75583963394165,
      "epoch": 1.7557148040638606,
      "grad_norm": 2.296875,
      "learning_rate": 0.00046937181742255673,
      "loss": 5.4485,
      "mean_token_accuracy": 0.16902389675378798,
      "num_tokens": 40168517.0,
      "step": 19355
    },
    {
      "entropy": 5.633741664886474,
      "epoch": 1.7561683599419449,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004693555172677612,
      "loss": 5.4012,
      "mean_token_accuracy": 0.1791708678007126,
      "num_tokens": 40177844.0,
      "step": 19360
    },
    {
      "entropy": 5.668799781799317,
      "epoch": 1.7566219158200291,
      "grad_norm": 2.875,
      "learning_rate": 0.000469339213093644,
      "loss": 5.3657,
      "mean_token_accuracy": 0.1700764402747154,
      "num_tokens": 40188829.0,
      "step": 19365
    },
    {
      "entropy": 5.737525939941406,
      "epoch": 1.7570754716981132,
      "grad_norm": 2.5,
      "learning_rate": 0.00046932290490054235,
      "loss": 5.484,
      "mean_token_accuracy": 0.17314303964376448,
      "num_tokens": 40199893.0,
      "step": 19370
    },
    {
      "entropy": 5.712117052078247,
      "epoch": 1.7575290275761972,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004693065926887936,
      "loss": 5.504,
      "mean_token_accuracy": 0.16475009620189668,
      "num_tokens": 40210602.0,
      "step": 19375
    },
    {
      "entropy": 5.683561563491821,
      "epoch": 1.7579825834542815,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046929027645873484,
      "loss": 5.3779,
      "mean_token_accuracy": 0.16759658753871917,
      "num_tokens": 40221799.0,
      "step": 19380
    },
    {
      "entropy": 5.704042530059814,
      "epoch": 1.7584361393323658,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004692739562107038,
      "loss": 5.3393,
      "mean_token_accuracy": 0.1800952672958374,
      "num_tokens": 40231881.0,
      "step": 19385
    },
    {
      "entropy": 5.697684621810913,
      "epoch": 1.75888969521045,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046925763194503775,
      "loss": 5.3353,
      "mean_token_accuracy": 0.1769966557621956,
      "num_tokens": 40241582.0,
      "step": 19390
    },
    {
      "entropy": 5.682764387130737,
      "epoch": 1.759343251088534,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004692413036620744,
      "loss": 5.3403,
      "mean_token_accuracy": 0.17713758051395417,
      "num_tokens": 40251469.0,
      "step": 19395
    },
    {
      "entropy": 5.626249933242798,
      "epoch": 1.7597968069666183,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004692249713621514,
      "loss": 5.422,
      "mean_token_accuracy": 0.17162647247314453,
      "num_tokens": 40261165.0,
      "step": 19400
    },
    {
      "entropy": 5.681667375564575,
      "epoch": 1.7602503628447024,
      "grad_norm": 1.7890625,
      "learning_rate": 0.00046920863504560647,
      "loss": 5.4166,
      "mean_token_accuracy": 0.16581457257270812,
      "num_tokens": 40271163.0,
      "step": 19405
    },
    {
      "entropy": 5.656208419799805,
      "epoch": 1.7607039187227866,
      "grad_norm": 3.234375,
      "learning_rate": 0.0004691922947127775,
      "loss": 5.4087,
      "mean_token_accuracy": 0.17079609930515288,
      "num_tokens": 40280896.0,
      "step": 19410
    },
    {
      "entropy": 5.777688598632812,
      "epoch": 1.761157474600871,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004691759503640024,
      "loss": 5.4162,
      "mean_token_accuracy": 0.1724862664937973,
      "num_tokens": 40291431.0,
      "step": 19415
    },
    {
      "entropy": 5.74890627861023,
      "epoch": 1.761611030478955,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004691596019996191,
      "loss": 5.4042,
      "mean_token_accuracy": 0.18186096101999283,
      "num_tokens": 40301224.0,
      "step": 19420
    },
    {
      "entropy": 5.675418329238892,
      "epoch": 1.7620645863570392,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004691432496199658,
      "loss": 5.3637,
      "mean_token_accuracy": 0.17987738102674483,
      "num_tokens": 40310294.0,
      "step": 19425
    },
    {
      "entropy": 5.6965416431427,
      "epoch": 1.7625181422351233,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046912689322538064,
      "loss": 5.4868,
      "mean_token_accuracy": 0.16619589924812317,
      "num_tokens": 40320462.0,
      "step": 19430
    },
    {
      "entropy": 5.699446487426758,
      "epoch": 1.7629716981132075,
      "grad_norm": 2.40625,
      "learning_rate": 0.00046911053281620187,
      "loss": 5.4071,
      "mean_token_accuracy": 0.1727581575512886,
      "num_tokens": 40330878.0,
      "step": 19435
    },
    {
      "entropy": 5.65647029876709,
      "epoch": 1.7634252539912918,
      "grad_norm": 2.75,
      "learning_rate": 0.00046909416839276775,
      "loss": 5.2976,
      "mean_token_accuracy": 0.17551178336143494,
      "num_tokens": 40341101.0,
      "step": 19440
    },
    {
      "entropy": 5.696866035461426,
      "epoch": 1.763878809869376,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004690777999554168,
      "loss": 5.4649,
      "mean_token_accuracy": 0.16926110684871673,
      "num_tokens": 40353535.0,
      "step": 19445
    },
    {
      "entropy": 5.70576581954956,
      "epoch": 1.76433236574746,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004690614275044875,
      "loss": 5.4548,
      "mean_token_accuracy": 0.17006197720766067,
      "num_tokens": 40364078.0,
      "step": 19450
    },
    {
      "entropy": 5.724061775207519,
      "epoch": 1.7647859216255442,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004690450510403184,
      "loss": 5.4528,
      "mean_token_accuracy": 0.16460948586463928,
      "num_tokens": 40374790.0,
      "step": 19455
    },
    {
      "entropy": 5.688197660446167,
      "epoch": 1.7652394775036284,
      "grad_norm": 2.9375,
      "learning_rate": 0.0004690286705632483,
      "loss": 5.374,
      "mean_token_accuracy": 0.17935332506895066,
      "num_tokens": 40384377.0,
      "step": 19460
    },
    {
      "entropy": 5.618489599227905,
      "epoch": 1.7656930333817127,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004690122860736157,
      "loss": 5.2189,
      "mean_token_accuracy": 0.1866292804479599,
      "num_tokens": 40395055.0,
      "step": 19465
    },
    {
      "entropy": 5.634308671951294,
      "epoch": 1.766146589259797,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046899589757175964,
      "loss": 5.3932,
      "mean_token_accuracy": 0.1750578388571739,
      "num_tokens": 40405367.0,
      "step": 19470
    },
    {
      "entropy": 5.672655439376831,
      "epoch": 1.766600145137881,
      "grad_norm": 3.140625,
      "learning_rate": 0.000468979505058019,
      "loss": 5.4466,
      "mean_token_accuracy": 0.16654159277677535,
      "num_tokens": 40416727.0,
      "step": 19475
    },
    {
      "entropy": 5.6397181987762455,
      "epoch": 1.767053701015965,
      "grad_norm": 2.703125,
      "learning_rate": 0.00046896310853273274,
      "loss": 5.4132,
      "mean_token_accuracy": 0.178445003926754,
      "num_tokens": 40427518.0,
      "step": 19480
    },
    {
      "entropy": 5.637136316299438,
      "epoch": 1.7675072568940493,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046894670799624,
      "loss": 5.3551,
      "mean_token_accuracy": 0.1741405874490738,
      "num_tokens": 40437874.0,
      "step": 19485
    },
    {
      "entropy": 5.6629385471344,
      "epoch": 1.7679608127721336,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046893030344887995,
      "loss": 5.2559,
      "mean_token_accuracy": 0.17584324330091478,
      "num_tokens": 40447649.0,
      "step": 19490
    },
    {
      "entropy": 5.698254203796386,
      "epoch": 1.7684143686502178,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00046891389489099166,
      "loss": 5.497,
      "mean_token_accuracy": 0.15941113531589507,
      "num_tokens": 40458855.0,
      "step": 19495
    },
    {
      "entropy": 5.701327896118164,
      "epoch": 1.7688679245283019,
      "grad_norm": 1.96875,
      "learning_rate": 0.00046889748232291476,
      "loss": 5.4353,
      "mean_token_accuracy": 0.17047600001096724,
      "num_tokens": 40468654.0,
      "step": 19500
    },
    {
      "entropy": 5.66054334640503,
      "epoch": 1.769321480406386,
      "grad_norm": 3.71875,
      "learning_rate": 0.00046888106574498843,
      "loss": 5.3548,
      "mean_token_accuracy": 0.17324077785015107,
      "num_tokens": 40479028.0,
      "step": 19505
    },
    {
      "entropy": 5.640240573883057,
      "epoch": 1.7697750362844702,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004688646451575522,
      "loss": 5.3853,
      "mean_token_accuracy": 0.17467578053474425,
      "num_tokens": 40489374.0,
      "step": 19510
    },
    {
      "entropy": 5.681777811050415,
      "epoch": 1.7702285921625545,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046884822056094587,
      "loss": 5.3786,
      "mean_token_accuracy": 0.1749932959675789,
      "num_tokens": 40499382.0,
      "step": 19515
    },
    {
      "entropy": 5.66882209777832,
      "epoch": 1.7706821480406387,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046883179195550883,
      "loss": 5.3161,
      "mean_token_accuracy": 0.1802353471517563,
      "num_tokens": 40507940.0,
      "step": 19520
    },
    {
      "entropy": 5.664459848403931,
      "epoch": 1.7711357039187228,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046881535934158094,
      "loss": 5.3996,
      "mean_token_accuracy": 0.17174231112003327,
      "num_tokens": 40518085.0,
      "step": 19525
    },
    {
      "entropy": 5.605812358856201,
      "epoch": 1.7715892597968068,
      "grad_norm": 1.90625,
      "learning_rate": 0.0004687989227195021,
      "loss": 5.3162,
      "mean_token_accuracy": 0.19040293544530867,
      "num_tokens": 40527490.0,
      "step": 19530
    },
    {
      "entropy": 5.7062366008758545,
      "epoch": 1.772042815674891,
      "grad_norm": 2.25,
      "learning_rate": 0.00046878248208961204,
      "loss": 5.4011,
      "mean_token_accuracy": 0.17433261275291442,
      "num_tokens": 40536722.0,
      "step": 19535
    },
    {
      "entropy": 5.682359266281128,
      "epoch": 1.7724963715529753,
      "grad_norm": 1.6875,
      "learning_rate": 0.000468766037452251,
      "loss": 5.3678,
      "mean_token_accuracy": 0.16797128617763518,
      "num_tokens": 40547962.0,
      "step": 19540
    },
    {
      "entropy": 5.667661380767822,
      "epoch": 1.7729499274310596,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046874958880775886,
      "loss": 5.4175,
      "mean_token_accuracy": 0.1668878063559532,
      "num_tokens": 40559779.0,
      "step": 19545
    },
    {
      "entropy": 5.720099258422851,
      "epoch": 1.7734034833091437,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004687331361564758,
      "loss": 5.4721,
      "mean_token_accuracy": 0.16728415936231614,
      "num_tokens": 40571196.0,
      "step": 19550
    },
    {
      "entropy": 5.759532403945923,
      "epoch": 1.7738570391872277,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004687166794987422,
      "loss": 5.3515,
      "mean_token_accuracy": 0.1680346503853798,
      "num_tokens": 40581878.0,
      "step": 19555
    },
    {
      "entropy": 5.66039924621582,
      "epoch": 1.774310595065312,
      "grad_norm": 2.28125,
      "learning_rate": 0.0004687002188348983,
      "loss": 5.4418,
      "mean_token_accuracy": 0.16688452064990997,
      "num_tokens": 40593617.0,
      "step": 19560
    },
    {
      "entropy": 5.7009885787963865,
      "epoch": 1.7747641509433962,
      "grad_norm": 2.0,
      "learning_rate": 0.00046868375416528443,
      "loss": 5.485,
      "mean_token_accuracy": 0.1690539687871933,
      "num_tokens": 40604486.0,
      "step": 19565
    },
    {
      "entropy": 5.725631284713745,
      "epoch": 1.7752177068214805,
      "grad_norm": 1.984375,
      "learning_rate": 0.00046866728549024123,
      "loss": 5.3877,
      "mean_token_accuracy": 0.17243834882974624,
      "num_tokens": 40616291.0,
      "step": 19570
    },
    {
      "entropy": 5.723520565032959,
      "epoch": 1.7756712626995645,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004686508128101092,
      "loss": 5.4373,
      "mean_token_accuracy": 0.17782945334911346,
      "num_tokens": 40625255.0,
      "step": 19575
    },
    {
      "entropy": 5.690884304046631,
      "epoch": 1.7761248185776488,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000468634336125229,
      "loss": 5.3354,
      "mean_token_accuracy": 0.1718562886118889,
      "num_tokens": 40635974.0,
      "step": 19580
    },
    {
      "entropy": 5.677248811721801,
      "epoch": 1.7765783744557329,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046861785543594143,
      "loss": 5.38,
      "mean_token_accuracy": 0.1775345280766487,
      "num_tokens": 40645450.0,
      "step": 19585
    },
    {
      "entropy": 5.715335512161255,
      "epoch": 1.7770319303338171,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004686013707425873,
      "loss": 5.4839,
      "mean_token_accuracy": 0.16663559377193451,
      "num_tokens": 40656198.0,
      "step": 19590
    },
    {
      "entropy": 5.717867231369018,
      "epoch": 1.7774854862119014,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046858488204550735,
      "loss": 5.4054,
      "mean_token_accuracy": 0.1797608256340027,
      "num_tokens": 40667232.0,
      "step": 19595
    },
    {
      "entropy": 5.689809846878052,
      "epoch": 1.7779390420899854,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004685683893450428,
      "loss": 5.3816,
      "mean_token_accuracy": 0.17585025876760482,
      "num_tokens": 40677599.0,
      "step": 19600
    },
    {
      "entropy": 5.648499917984009,
      "epoch": 1.7783925979680697,
      "grad_norm": 2.390625,
      "learning_rate": 0.0004685518926415346,
      "loss": 5.3857,
      "mean_token_accuracy": 0.17316477745771408,
      "num_tokens": 40687380.0,
      "step": 19605
    },
    {
      "entropy": 5.699211835861206,
      "epoch": 1.7788461538461537,
      "grad_norm": 2.046875,
      "learning_rate": 0.000468535391935324,
      "loss": 5.4407,
      "mean_token_accuracy": 0.16909227073192595,
      "num_tokens": 40697841.0,
      "step": 19610
    },
    {
      "entropy": 5.685452270507812,
      "epoch": 1.779299709724238,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004685188872267521,
      "loss": 5.3986,
      "mean_token_accuracy": 0.16552644968032837,
      "num_tokens": 40708605.0,
      "step": 19615
    },
    {
      "entropy": 5.660304164886474,
      "epoch": 1.7797532656023223,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004685023785161603,
      "loss": 5.4221,
      "mean_token_accuracy": 0.17617201805114746,
      "num_tokens": 40719330.0,
      "step": 19620
    },
    {
      "entropy": 5.650123453140258,
      "epoch": 1.7802068214804065,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004684858658038901,
      "loss": 5.4278,
      "mean_token_accuracy": 0.17194269597530365,
      "num_tokens": 40729508.0,
      "step": 19625
    },
    {
      "entropy": 5.649344825744629,
      "epoch": 1.7806603773584906,
      "grad_norm": 2.375,
      "learning_rate": 0.0004684693490902828,
      "loss": 5.3894,
      "mean_token_accuracy": 0.18024425357580184,
      "num_tokens": 40740254.0,
      "step": 19630
    },
    {
      "entropy": 5.658716535568237,
      "epoch": 1.7811139332365746,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046845282837568,
      "loss": 5.3605,
      "mean_token_accuracy": 0.1781076967716217,
      "num_tokens": 40750487.0,
      "step": 19635
    },
    {
      "entropy": 5.691123533248901,
      "epoch": 1.781567489114659,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046843630366042355,
      "loss": 5.4634,
      "mean_token_accuracy": 0.16753043383359909,
      "num_tokens": 40762016.0,
      "step": 19640
    },
    {
      "entropy": 5.69201021194458,
      "epoch": 1.7820210449927432,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046841977494485506,
      "loss": 5.3793,
      "mean_token_accuracy": 0.18149294406175615,
      "num_tokens": 40772038.0,
      "step": 19645
    },
    {
      "entropy": 5.633561754226685,
      "epoch": 1.7824746008708274,
      "grad_norm": 3.015625,
      "learning_rate": 0.0004684032422293162,
      "loss": 5.3328,
      "mean_token_accuracy": 0.17628367990255356,
      "num_tokens": 40782386.0,
      "step": 19650
    },
    {
      "entropy": 5.625001621246338,
      "epoch": 1.7829281567489115,
      "grad_norm": 5.28125,
      "learning_rate": 0.000468386705514149,
      "loss": 5.3471,
      "mean_token_accuracy": 0.1834060177206993,
      "num_tokens": 40792116.0,
      "step": 19655
    },
    {
      "entropy": 5.594267272949219,
      "epoch": 1.7833817126269955,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004683701647996956,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1762518584728241,
      "num_tokens": 40802237.0,
      "step": 19660
    },
    {
      "entropy": 5.598761701583863,
      "epoch": 1.7838352685050798,
      "grad_norm": 2.984375,
      "learning_rate": 0.0004683536200862978,
      "loss": 5.2997,
      "mean_token_accuracy": 0.1739782452583313,
      "num_tokens": 40811969.0,
      "step": 19665
    },
    {
      "entropy": 5.654552412033081,
      "epoch": 1.784288824383164,
      "grad_norm": 2.75,
      "learning_rate": 0.0004683370713742979,
      "loss": 5.3573,
      "mean_token_accuracy": 0.17584001570940017,
      "num_tokens": 40822345.0,
      "step": 19670
    },
    {
      "entropy": 5.70405216217041,
      "epoch": 1.7847423802612483,
      "grad_norm": 1.6875,
      "learning_rate": 0.00046832051866403814,
      "loss": 5.4879,
      "mean_token_accuracy": 0.16865310221910476,
      "num_tokens": 40832665.0,
      "step": 19675
    },
    {
      "entropy": 5.704152011871338,
      "epoch": 1.7851959361393324,
      "grad_norm": 4.34375,
      "learning_rate": 0.0004683039619558607,
      "loss": 5.3683,
      "mean_token_accuracy": 0.17716557830572127,
      "num_tokens": 40843682.0,
      "step": 19680
    },
    {
      "entropy": 5.650419855117798,
      "epoch": 1.7856494920174164,
      "grad_norm": 2.875,
      "learning_rate": 0.00046828740125010813,
      "loss": 5.4059,
      "mean_token_accuracy": 0.17674280405044557,
      "num_tokens": 40853490.0,
      "step": 19685
    },
    {
      "entropy": 5.7187248229980465,
      "epoch": 1.7861030478955007,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004682708365471228,
      "loss": 5.4115,
      "mean_token_accuracy": 0.17210618555545806,
      "num_tokens": 40863789.0,
      "step": 19690
    },
    {
      "entropy": 5.65013370513916,
      "epoch": 1.786556603773585,
      "grad_norm": 1.8125,
      "learning_rate": 0.00046825426784724734,
      "loss": 5.4197,
      "mean_token_accuracy": 0.17295124977827073,
      "num_tokens": 40873628.0,
      "step": 19695
    },
    {
      "entropy": 5.706695890426635,
      "epoch": 1.7870101596516692,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004682376951508244,
      "loss": 5.4207,
      "mean_token_accuracy": 0.17073485106229783,
      "num_tokens": 40883710.0,
      "step": 19700
    },
    {
      "entropy": 5.760124158859253,
      "epoch": 1.7874637155297532,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046822111845819665,
      "loss": 5.4311,
      "mean_token_accuracy": 0.16811173260211945,
      "num_tokens": 40895284.0,
      "step": 19705
    },
    {
      "entropy": 5.669731950759887,
      "epoch": 1.7879172714078373,
      "grad_norm": 2.15625,
      "learning_rate": 0.00046820453776970697,
      "loss": 5.3692,
      "mean_token_accuracy": 0.1749686598777771,
      "num_tokens": 40905949.0,
      "step": 19710
    },
    {
      "entropy": 5.632780933380127,
      "epoch": 1.7883708272859216,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046818795308569815,
      "loss": 5.4016,
      "mean_token_accuracy": 0.17153728008270264,
      "num_tokens": 40915869.0,
      "step": 19715
    },
    {
      "entropy": 5.629489326477051,
      "epoch": 1.7888243831640058,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004681713644065133,
      "loss": 5.3774,
      "mean_token_accuracy": 0.1746656894683838,
      "num_tokens": 40925168.0,
      "step": 19720
    },
    {
      "entropy": 5.628101539611817,
      "epoch": 1.78927793904209,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004681547717324954,
      "loss": 5.28,
      "mean_token_accuracy": 0.18135300725698472,
      "num_tokens": 40935496.0,
      "step": 19725
    },
    {
      "entropy": 5.715478897094727,
      "epoch": 1.7897314949201741,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046813817506398755,
      "loss": 5.4191,
      "mean_token_accuracy": 0.16598139554262162,
      "num_tokens": 40946772.0,
      "step": 19730
    },
    {
      "entropy": 5.656552934646607,
      "epoch": 1.7901850507982582,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046812157440133303,
      "loss": 5.4055,
      "mean_token_accuracy": 0.171905018389225,
      "num_tokens": 40957639.0,
      "step": 19735
    },
    {
      "entropy": 5.709648466110229,
      "epoch": 1.7906386066763424,
      "grad_norm": 2.859375,
      "learning_rate": 0.0004681049697448752,
      "loss": 5.3552,
      "mean_token_accuracy": 0.17924804538488387,
      "num_tokens": 40968255.0,
      "step": 19740
    },
    {
      "entropy": 5.74591646194458,
      "epoch": 1.7910921625544267,
      "grad_norm": 2.0625,
      "learning_rate": 0.00046808836109495734,
      "loss": 5.4694,
      "mean_token_accuracy": 0.16457413882017136,
      "num_tokens": 40979335.0,
      "step": 19745
    },
    {
      "entropy": 5.645269155502319,
      "epoch": 1.791545718432511,
      "grad_norm": 4.5,
      "learning_rate": 0.0004680717484519229,
      "loss": 5.2974,
      "mean_token_accuracy": 0.17502287030220032,
      "num_tokens": 40989716.0,
      "step": 19750
    },
    {
      "entropy": 5.661801528930664,
      "epoch": 1.791999274310595,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004680551318161156,
      "loss": 5.386,
      "mean_token_accuracy": 0.17411947548389434,
      "num_tokens": 41000304.0,
      "step": 19755
    },
    {
      "entropy": 5.703804016113281,
      "epoch": 1.7924528301886793,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046803851118787897,
      "loss": 5.3804,
      "mean_token_accuracy": 0.1770551785826683,
      "num_tokens": 41012229.0,
      "step": 19760
    },
    {
      "entropy": 5.596820878982544,
      "epoch": 1.7929063860667633,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004680218865675567,
      "loss": 5.3086,
      "mean_token_accuracy": 0.17036328315734864,
      "num_tokens": 41022732.0,
      "step": 19765
    },
    {
      "entropy": 5.643392658233642,
      "epoch": 1.7933599419448476,
      "grad_norm": 3.171875,
      "learning_rate": 0.0004680052579554926,
      "loss": 5.4597,
      "mean_token_accuracy": 0.17069420367479324,
      "num_tokens": 41033054.0,
      "step": 19770
    },
    {
      "entropy": 5.634865140914917,
      "epoch": 1.7938134978229319,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004679886253520307,
      "loss": 5.296,
      "mean_token_accuracy": 0.17533819675445556,
      "num_tokens": 41042371.0,
      "step": 19775
    },
    {
      "entropy": 5.677905130386352,
      "epoch": 1.794267053701016,
      "grad_norm": 2.75,
      "learning_rate": 0.00046797198875751474,
      "loss": 5.3435,
      "mean_token_accuracy": 0.18337418884038925,
      "num_tokens": 41052405.0,
      "step": 19780
    },
    {
      "entropy": 5.594333171844482,
      "epoch": 1.7947206095791002,
      "grad_norm": 3.21875,
      "learning_rate": 0.0004679553481722889,
      "loss": 5.2877,
      "mean_token_accuracy": 0.17781112492084503,
      "num_tokens": 41062863.0,
      "step": 19785
    },
    {
      "entropy": 5.689164161682129,
      "epoch": 1.7951741654571842,
      "grad_norm": 2.25,
      "learning_rate": 0.00046793870359669733,
      "loss": 5.4579,
      "mean_token_accuracy": 0.16446109414100646,
      "num_tokens": 41073447.0,
      "step": 19790
    },
    {
      "entropy": 5.711205005645752,
      "epoch": 1.7956277213352685,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004679220550310842,
      "loss": 5.412,
      "mean_token_accuracy": 0.1708745002746582,
      "num_tokens": 41085506.0,
      "step": 19795
    },
    {
      "entropy": 5.727150535583496,
      "epoch": 1.7960812772133528,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004679054024757937,
      "loss": 5.3744,
      "mean_token_accuracy": 0.17888532280921937,
      "num_tokens": 41097128.0,
      "step": 19800
    },
    {
      "entropy": 5.695698642730713,
      "epoch": 1.796534833091437,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004678887459311704,
      "loss": 5.4449,
      "mean_token_accuracy": 0.1702427625656128,
      "num_tokens": 41107848.0,
      "step": 19805
    },
    {
      "entropy": 5.737817668914795,
      "epoch": 1.796988388969521,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004678720853975587,
      "loss": 5.5087,
      "mean_token_accuracy": 0.16977541595697404,
      "num_tokens": 41118582.0,
      "step": 19810
    },
    {
      "entropy": 5.780579900741577,
      "epoch": 1.797441944847605,
      "grad_norm": 3.0625,
      "learning_rate": 0.00046785542087530307,
      "loss": 5.4089,
      "mean_token_accuracy": 0.16659294962882995,
      "num_tokens": 41129387.0,
      "step": 19815
    },
    {
      "entropy": 5.687614822387696,
      "epoch": 1.7978955007256894,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004678387523647483,
      "loss": 5.3089,
      "mean_token_accuracy": 0.17357004284858704,
      "num_tokens": 41139584.0,
      "step": 19820
    },
    {
      "entropy": 5.549474573135376,
      "epoch": 1.7983490566037736,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046782207986623885,
      "loss": 5.2501,
      "mean_token_accuracy": 0.17942317724227905,
      "num_tokens": 41149629.0,
      "step": 19825
    },
    {
      "entropy": 5.691339159011841,
      "epoch": 1.798802612481858,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046780540338011974,
      "loss": 5.5135,
      "mean_token_accuracy": 0.16899208575487137,
      "num_tokens": 41159728.0,
      "step": 19830
    },
    {
      "entropy": 5.793948602676392,
      "epoch": 1.799256168359942,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004677887229067358,
      "loss": 5.4132,
      "mean_token_accuracy": 0.1683789998292923,
      "num_tokens": 41170059.0,
      "step": 19835
    },
    {
      "entropy": 5.709729337692261,
      "epoch": 1.799709724238026,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046777203844643184,
      "loss": 5.3599,
      "mean_token_accuracy": 0.1764975905418396,
      "num_tokens": 41178708.0,
      "step": 19840
    },
    {
      "entropy": 5.5862242698669435,
      "epoch": 1.8001632801161103,
      "grad_norm": 9.625,
      "learning_rate": 0.00046775534999955303,
      "loss": 5.3137,
      "mean_token_accuracy": 0.1800222545862198,
      "num_tokens": 41189121.0,
      "step": 19845
    },
    {
      "entropy": 5.665577411651611,
      "epoch": 1.8006168359941945,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004677386575664445,
      "loss": 5.3814,
      "mean_token_accuracy": 0.1705608993768692,
      "num_tokens": 41199598.0,
      "step": 19850
    },
    {
      "entropy": 5.737516832351685,
      "epoch": 1.8010703918722788,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046772196114745133,
      "loss": 5.44,
      "mean_token_accuracy": 0.1726987674832344,
      "num_tokens": 41209839.0,
      "step": 19855
    },
    {
      "entropy": 5.697774076461792,
      "epoch": 1.8015239477503628,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004677052607429189,
      "loss": 5.3771,
      "mean_token_accuracy": 0.17209954857826232,
      "num_tokens": 41219948.0,
      "step": 19860
    },
    {
      "entropy": 5.708835506439209,
      "epoch": 1.8019775036284469,
      "grad_norm": 3.375,
      "learning_rate": 0.0004676885563531927,
      "loss": 5.4529,
      "mean_token_accuracy": 0.17167501151561737,
      "num_tokens": 41229791.0,
      "step": 19865
    },
    {
      "entropy": 5.748013019561768,
      "epoch": 1.8024310595065312,
      "grad_norm": 3.640625,
      "learning_rate": 0.00046767184797861795,
      "loss": 5.4759,
      "mean_token_accuracy": 0.16724979132413864,
      "num_tokens": 41241169.0,
      "step": 19870
    },
    {
      "entropy": 5.813654279708862,
      "epoch": 1.8028846153846154,
      "grad_norm": 2.984375,
      "learning_rate": 0.00046765513561954026,
      "loss": 5.4745,
      "mean_token_accuracy": 0.16236065775156022,
      "num_tokens": 41252520.0,
      "step": 19875
    },
    {
      "entropy": 5.661193466186523,
      "epoch": 1.8033381712626997,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046763841927630534,
      "loss": 5.3779,
      "mean_token_accuracy": 0.18399600833654403,
      "num_tokens": 41262133.0,
      "step": 19880
    },
    {
      "entropy": 5.683443880081176,
      "epoch": 1.8037917271407837,
      "grad_norm": 4.03125,
      "learning_rate": 0.00046762169894925876,
      "loss": 5.4098,
      "mean_token_accuracy": 0.17390918731689453,
      "num_tokens": 41272568.0,
      "step": 19885
    },
    {
      "entropy": 5.692693424224854,
      "epoch": 1.8042452830188678,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004676049746387464,
      "loss": 5.4293,
      "mean_token_accuracy": 0.17508018761873245,
      "num_tokens": 41282181.0,
      "step": 19890
    },
    {
      "entropy": 5.612255191802978,
      "epoch": 1.804698838896952,
      "grad_norm": 2.53125,
      "learning_rate": 0.0004675882463451141,
      "loss": 5.2779,
      "mean_token_accuracy": 0.17489370852708816,
      "num_tokens": 41292764.0,
      "step": 19895
    },
    {
      "entropy": 5.642804145812988,
      "epoch": 1.8051523947750363,
      "grad_norm": 2.0,
      "learning_rate": 0.0004675715140687078,
      "loss": 5.3358,
      "mean_token_accuracy": 0.18027858883142472,
      "num_tokens": 41303306.0,
      "step": 19900
    },
    {
      "entropy": 5.600942802429199,
      "epoch": 1.8056059506531206,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046755477780987344,
      "loss": 5.223,
      "mean_token_accuracy": 0.18202665448188782,
      "num_tokens": 41312651.0,
      "step": 19905
    },
    {
      "entropy": 5.631641292572022,
      "epoch": 1.8060595065312046,
      "grad_norm": 2.8125,
      "learning_rate": 0.00046753803756895726,
      "loss": 5.2759,
      "mean_token_accuracy": 0.18179862648248674,
      "num_tokens": 41322405.0,
      "step": 19910
    },
    {
      "entropy": 5.686446475982666,
      "epoch": 1.8065130624092887,
      "grad_norm": 1.875,
      "learning_rate": 0.0004675212933463054,
      "loss": 5.4462,
      "mean_token_accuracy": 0.1643253445625305,
      "num_tokens": 41332989.0,
      "step": 19915
    },
    {
      "entropy": 5.694095277786255,
      "epoch": 1.806966618287373,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004675045451422641,
      "loss": 5.441,
      "mean_token_accuracy": 0.16239337176084517,
      "num_tokens": 41343526.0,
      "step": 19920
    },
    {
      "entropy": 5.658193206787109,
      "epoch": 1.8074201741654572,
      "grad_norm": 2.1875,
      "learning_rate": 0.00046748779295717973,
      "loss": 5.348,
      "mean_token_accuracy": 0.17746989279985428,
      "num_tokens": 41353467.0,
      "step": 19925
    },
    {
      "entropy": 5.723032093048095,
      "epoch": 1.8078737300435415,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004674710367913989,
      "loss": 5.4463,
      "mean_token_accuracy": 0.17195288836956024,
      "num_tokens": 41364253.0,
      "step": 19930
    },
    {
      "entropy": 5.8597184181213375,
      "epoch": 1.8083272859216255,
      "grad_norm": 5.25,
      "learning_rate": 0.0004674542766452678,
      "loss": 5.5245,
      "mean_token_accuracy": 0.1663799375295639,
      "num_tokens": 41374884.0,
      "step": 19935
    },
    {
      "entropy": 5.705721139907837,
      "epoch": 1.8087808417997098,
      "grad_norm": 2.4375,
      "learning_rate": 0.00046743751251913327,
      "loss": 5.3288,
      "mean_token_accuracy": 0.17324694246053696,
      "num_tokens": 41385034.0,
      "step": 19940
    },
    {
      "entropy": 5.6469238758087155,
      "epoch": 1.8092343976777938,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046742074441334197,
      "loss": 5.4141,
      "mean_token_accuracy": 0.16844134777784348,
      "num_tokens": 41396248.0,
      "step": 19945
    },
    {
      "entropy": 5.707080411911011,
      "epoch": 1.809687953555878,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004674039723282406,
      "loss": 5.4736,
      "mean_token_accuracy": 0.16496036797761918,
      "num_tokens": 41406486.0,
      "step": 19950
    },
    {
      "entropy": 5.720289659500122,
      "epoch": 1.8101415094339623,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046738719626417607,
      "loss": 5.4336,
      "mean_token_accuracy": 0.1669118359684944,
      "num_tokens": 41416562.0,
      "step": 19955
    },
    {
      "entropy": 5.726631832122803,
      "epoch": 1.8105950653120464,
      "grad_norm": 2.078125,
      "learning_rate": 0.00046737041622149536,
      "loss": 5.4279,
      "mean_token_accuracy": 0.16400882005691528,
      "num_tokens": 41427620.0,
      "step": 19960
    },
    {
      "entropy": 5.69576063156128,
      "epoch": 1.8110486211901307,
      "grad_norm": 2.515625,
      "learning_rate": 0.0004673536322005454,
      "loss": 5.3219,
      "mean_token_accuracy": 0.17510611116886138,
      "num_tokens": 41437450.0,
      "step": 19965
    },
    {
      "entropy": 5.714027976989746,
      "epoch": 1.8115021770682147,
      "grad_norm": 3.3125,
      "learning_rate": 0.0004673368442016732,
      "loss": 5.4947,
      "mean_token_accuracy": 0.16664799004793168,
      "num_tokens": 41447494.0,
      "step": 19970
    },
    {
      "entropy": 5.7228288650512695,
      "epoch": 1.811955732946299,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004673200522252261,
      "loss": 5.3863,
      "mean_token_accuracy": 0.17400291711091995,
      "num_tokens": 41457740.0,
      "step": 19975
    },
    {
      "entropy": 5.702834415435791,
      "epoch": 1.8124092888243832,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004673032562715514,
      "loss": 5.3753,
      "mean_token_accuracy": 0.17080771774053574,
      "num_tokens": 41467262.0,
      "step": 19980
    },
    {
      "entropy": 5.673035526275635,
      "epoch": 1.8128628447024675,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046728645634099627,
      "loss": 5.4415,
      "mean_token_accuracy": 0.17408981770277024,
      "num_tokens": 41477480.0,
      "step": 19985
    },
    {
      "entropy": 5.7407301902771,
      "epoch": 1.8133164005805515,
      "grad_norm": 1.734375,
      "learning_rate": 0.00046726965243390825,
      "loss": 5.4453,
      "mean_token_accuracy": 0.1718200832605362,
      "num_tokens": 41488292.0,
      "step": 19990
    },
    {
      "entropy": 5.741639232635498,
      "epoch": 1.8137699564586356,
      "grad_norm": 3.953125,
      "learning_rate": 0.0004672528445506348,
      "loss": 5.4219,
      "mean_token_accuracy": 0.17023264467716218,
      "num_tokens": 41498718.0,
      "step": 19995
    },
    {
      "entropy": 5.64872465133667,
      "epoch": 1.8142235123367199,
      "grad_norm": 3.515625,
      "learning_rate": 0.0004672360326915236,
      "loss": 5.2751,
      "mean_token_accuracy": 0.17687819004058838,
      "num_tokens": 41507893.0,
      "step": 20000
    },
    {
      "entropy": 5.648960733413697,
      "epoch": 1.8146770682148041,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004672192168569222,
      "loss": 5.3812,
      "mean_token_accuracy": 0.1734568476676941,
      "num_tokens": 41518334.0,
      "step": 20005
    },
    {
      "entropy": 5.688988256454468,
      "epoch": 1.8151306240928884,
      "grad_norm": 2.125,
      "learning_rate": 0.00046720239704717845,
      "loss": 5.3185,
      "mean_token_accuracy": 0.1756274253129959,
      "num_tokens": 41528655.0,
      "step": 20010
    },
    {
      "entropy": 5.579896736145019,
      "epoch": 1.8155841799709724,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046718557326264,
      "loss": 5.2757,
      "mean_token_accuracy": 0.1871892899274826,
      "num_tokens": 41538713.0,
      "step": 20015
    },
    {
      "entropy": 5.622099304199219,
      "epoch": 1.8160377358490565,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004671687455036551,
      "loss": 5.4291,
      "mean_token_accuracy": 0.17268739640712738,
      "num_tokens": 41551157.0,
      "step": 20020
    },
    {
      "entropy": 5.693927907943726,
      "epoch": 1.8164912917271407,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004671519137705715,
      "loss": 5.3763,
      "mean_token_accuracy": 0.17566147893667222,
      "num_tokens": 41560873.0,
      "step": 20025
    },
    {
      "entropy": 5.696670866012573,
      "epoch": 1.816944847605225,
      "grad_norm": 2.828125,
      "learning_rate": 0.00046713507806373736,
      "loss": 5.4646,
      "mean_token_accuracy": 0.1651209220290184,
      "num_tokens": 41572220.0,
      "step": 20030
    },
    {
      "entropy": 5.71622257232666,
      "epoch": 1.8173984034833093,
      "grad_norm": 1.671875,
      "learning_rate": 0.00046711823838350086,
      "loss": 5.4156,
      "mean_token_accuracy": 0.17106860280036926,
      "num_tokens": 41583148.0,
      "step": 20035
    },
    {
      "entropy": 5.6921630859375,
      "epoch": 1.8178519593613933,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00046710139473021025,
      "loss": 5.3942,
      "mean_token_accuracy": 0.1773748829960823,
      "num_tokens": 41593774.0,
      "step": 20040
    },
    {
      "entropy": 5.625674438476563,
      "epoch": 1.8183055152394774,
      "grad_norm": 4.78125,
      "learning_rate": 0.00046708454710421376,
      "loss": 5.3458,
      "mean_token_accuracy": 0.17498193085193633,
      "num_tokens": 41603556.0,
      "step": 20045
    },
    {
      "entropy": 5.730724048614502,
      "epoch": 1.8187590711175616,
      "grad_norm": 3.3125,
      "learning_rate": 0.00046706769550586,
      "loss": 5.3691,
      "mean_token_accuracy": 0.17491924911737441,
      "num_tokens": 41613803.0,
      "step": 20050
    },
    {
      "entropy": 5.653136777877807,
      "epoch": 1.819212626995646,
      "grad_norm": 3.0,
      "learning_rate": 0.00046705083993549723,
      "loss": 5.3135,
      "mean_token_accuracy": 0.17564015686511994,
      "num_tokens": 41624576.0,
      "step": 20055
    },
    {
      "entropy": 5.667943859100342,
      "epoch": 1.8196661828737302,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004670339803934743,
      "loss": 5.413,
      "mean_token_accuracy": 0.1698056235909462,
      "num_tokens": 41635293.0,
      "step": 20060
    },
    {
      "entropy": 5.65444278717041,
      "epoch": 1.8201197387518142,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004670171168801396,
      "loss": 5.348,
      "mean_token_accuracy": 0.17135924398899077,
      "num_tokens": 41645202.0,
      "step": 20065
    },
    {
      "entropy": 5.695136499404907,
      "epoch": 1.8205732946298983,
      "grad_norm": 2.21875,
      "learning_rate": 0.000467000249395842,
      "loss": 5.3588,
      "mean_token_accuracy": 0.17028220668435096,
      "num_tokens": 41655205.0,
      "step": 20070
    },
    {
      "entropy": 5.611849308013916,
      "epoch": 1.8210268505079825,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004669833779409303,
      "loss": 5.3638,
      "mean_token_accuracy": 0.1747521162033081,
      "num_tokens": 41665684.0,
      "step": 20075
    },
    {
      "entropy": 5.619455623626709,
      "epoch": 1.8214804063860668,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004669665025157535,
      "loss": 5.3298,
      "mean_token_accuracy": 0.17793879956007003,
      "num_tokens": 41675651.0,
      "step": 20080
    },
    {
      "entropy": 5.694954967498779,
      "epoch": 1.821933962264151,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046694962312066046,
      "loss": 5.4014,
      "mean_token_accuracy": 0.17449382096529006,
      "num_tokens": 41688124.0,
      "step": 20085
    },
    {
      "entropy": 5.6402161598205565,
      "epoch": 1.822387518142235,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046693273975600036,
      "loss": 5.4364,
      "mean_token_accuracy": 0.16780110001564025,
      "num_tokens": 41698237.0,
      "step": 20090
    },
    {
      "entropy": 5.7276115894317625,
      "epoch": 1.8228410740203191,
      "grad_norm": 1.5,
      "learning_rate": 0.0004669158524221222,
      "loss": 5.4131,
      "mean_token_accuracy": 0.17172733396291734,
      "num_tokens": 41708919.0,
      "step": 20095
    },
    {
      "entropy": 5.699565839767456,
      "epoch": 1.8232946298984034,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046689896111937537,
      "loss": 5.4746,
      "mean_token_accuracy": 0.1642145335674286,
      "num_tokens": 41720167.0,
      "step": 20100
    },
    {
      "entropy": 5.707230091094971,
      "epoch": 1.8237481857764877,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004668820658481091,
      "loss": 5.414,
      "mean_token_accuracy": 0.16994061768054963,
      "num_tokens": 41730049.0,
      "step": 20105
    },
    {
      "entropy": 5.638854742050171,
      "epoch": 1.824201741654572,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046686516660867283,
      "loss": 5.3284,
      "mean_token_accuracy": 0.17365269958972931,
      "num_tokens": 41739027.0,
      "step": 20110
    },
    {
      "entropy": 5.633011388778686,
      "epoch": 1.824655297532656,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004668482634014159,
      "loss": 5.382,
      "mean_token_accuracy": 0.1707365944981575,
      "num_tokens": 41748780.0,
      "step": 20115
    },
    {
      "entropy": 5.7083593845367435,
      "epoch": 1.8251088534107403,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004668313562266882,
      "loss": 5.429,
      "mean_token_accuracy": 0.17127667516469955,
      "num_tokens": 41759203.0,
      "step": 20120
    },
    {
      "entropy": 5.580803251266479,
      "epoch": 1.8255624092888243,
      "grad_norm": 1.8359375,
      "learning_rate": 0.000466814445084839,
      "loss": 5.2063,
      "mean_token_accuracy": 0.1780097022652626,
      "num_tokens": 41769622.0,
      "step": 20125
    },
    {
      "entropy": 5.675507354736328,
      "epoch": 1.8260159651669086,
      "grad_norm": 1.625,
      "learning_rate": 0.0004667975299762183,
      "loss": 5.3584,
      "mean_token_accuracy": 0.1747714564204216,
      "num_tokens": 41780053.0,
      "step": 20130
    },
    {
      "entropy": 5.695764064788818,
      "epoch": 1.8264695210449928,
      "grad_norm": 1.875,
      "learning_rate": 0.00046678061090117576,
      "loss": 5.4585,
      "mean_token_accuracy": 0.17050983607769013,
      "num_tokens": 41790021.0,
      "step": 20135
    },
    {
      "entropy": 5.653466844558716,
      "epoch": 1.8269230769230769,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004667636878600613,
      "loss": 5.3677,
      "mean_token_accuracy": 0.16832229793071746,
      "num_tokens": 41800215.0,
      "step": 20140
    },
    {
      "entropy": 5.670646238327026,
      "epoch": 1.8273766328011611,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004667467608532249,
      "loss": 5.3348,
      "mean_token_accuracy": 0.17505489736795427,
      "num_tokens": 41810185.0,
      "step": 20145
    },
    {
      "entropy": 5.656306552886963,
      "epoch": 1.8278301886792452,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004667298298810167,
      "loss": 5.4352,
      "mean_token_accuracy": 0.1683538004755974,
      "num_tokens": 41820325.0,
      "step": 20150
    },
    {
      "entropy": 5.63390040397644,
      "epoch": 1.8282837445573294,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004667128949437867,
      "loss": 5.3313,
      "mean_token_accuracy": 0.1771603226661682,
      "num_tokens": 41830121.0,
      "step": 20155
    },
    {
      "entropy": 5.7078736305236815,
      "epoch": 1.8287373004354137,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004666959560418852,
      "loss": 5.3245,
      "mean_token_accuracy": 0.17641376852989196,
      "num_tokens": 41840479.0,
      "step": 20160
    },
    {
      "entropy": 5.705859422683716,
      "epoch": 1.829190856313498,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046667901317566245,
      "loss": 5.3668,
      "mean_token_accuracy": 0.17884127646684647,
      "num_tokens": 41850326.0,
      "step": 20165
    },
    {
      "entropy": 5.664911413192749,
      "epoch": 1.829644412191582,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004666620663454689,
      "loss": 5.4488,
      "mean_token_accuracy": 0.16705051958560943,
      "num_tokens": 41860376.0,
      "step": 20170
    },
    {
      "entropy": 5.6734575748443605,
      "epoch": 1.830097968069666,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046664511555165496,
      "loss": 5.3519,
      "mean_token_accuracy": 0.1769927531480789,
      "num_tokens": 41869987.0,
      "step": 20175
    },
    {
      "entropy": 5.636085557937622,
      "epoch": 1.8305515239477503,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046662816079457114,
      "loss": 5.426,
      "mean_token_accuracy": 0.17808954119682313,
      "num_tokens": 41880576.0,
      "step": 20180
    },
    {
      "entropy": 5.67034502029419,
      "epoch": 1.8310050798258346,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004666112020745682,
      "loss": 5.3503,
      "mean_token_accuracy": 0.17820678353309632,
      "num_tokens": 41890711.0,
      "step": 20185
    },
    {
      "entropy": 5.635022735595703,
      "epoch": 1.8314586357039189,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004665942393919967,
      "loss": 5.294,
      "mean_token_accuracy": 0.1827741414308548,
      "num_tokens": 41901156.0,
      "step": 20190
    },
    {
      "entropy": 5.623656511306763,
      "epoch": 1.831912191582003,
      "grad_norm": 2.6875,
      "learning_rate": 0.00046657727274720755,
      "loss": 5.389,
      "mean_token_accuracy": 0.17274575680494308,
      "num_tokens": 41911842.0,
      "step": 20195
    },
    {
      "entropy": 5.734349250793457,
      "epoch": 1.832365747460087,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004665603021405516,
      "loss": 5.5095,
      "mean_token_accuracy": 0.167895083129406,
      "num_tokens": 41922534.0,
      "step": 20200
    },
    {
      "entropy": 5.670120620727539,
      "epoch": 1.8328193033381712,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004665433275723798,
      "loss": 5.4233,
      "mean_token_accuracy": 0.17332451343536376,
      "num_tokens": 41932219.0,
      "step": 20205
    },
    {
      "entropy": 5.74638934135437,
      "epoch": 1.8332728592162555,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004665263490430431,
      "loss": 5.4957,
      "mean_token_accuracy": 0.16869926303625107,
      "num_tokens": 41943220.0,
      "step": 20210
    },
    {
      "entropy": 5.665520906448364,
      "epoch": 1.8337264150943398,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004665093665528928,
      "loss": 5.3721,
      "mean_token_accuracy": 0.17322715669870375,
      "num_tokens": 41954096.0,
      "step": 20215
    },
    {
      "entropy": 5.652259063720703,
      "epoch": 1.8341799709724238,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046649238010227983,
      "loss": 5.3189,
      "mean_token_accuracy": 0.17467793077230453,
      "num_tokens": 41964919.0,
      "step": 20220
    },
    {
      "entropy": 5.608958721160889,
      "epoch": 1.8346335268505078,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004664753896915558,
      "loss": 5.3621,
      "mean_token_accuracy": 0.17702565491199493,
      "num_tokens": 41974790.0,
      "step": 20225
    },
    {
      "entropy": 5.655884313583374,
      "epoch": 1.8350870827285921,
      "grad_norm": 3.96875,
      "learning_rate": 0.0004664583953210719,
      "loss": 5.3684,
      "mean_token_accuracy": 0.1738762825727463,
      "num_tokens": 41985637.0,
      "step": 20230
    },
    {
      "entropy": 5.666739463806152,
      "epoch": 1.8355406386066764,
      "grad_norm": 1.625,
      "learning_rate": 0.0004664413969911795,
      "loss": 5.2718,
      "mean_token_accuracy": 0.1825869232416153,
      "num_tokens": 41996091.0,
      "step": 20235
    },
    {
      "entropy": 5.620964479446411,
      "epoch": 1.8359941944847606,
      "grad_norm": 2.1875,
      "learning_rate": 0.0004664243947022303,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1692447230219841,
      "num_tokens": 42006935.0,
      "step": 20240
    },
    {
      "entropy": 5.586961984634399,
      "epoch": 1.8364477503628447,
      "grad_norm": 2.578125,
      "learning_rate": 0.00046640738845457586,
      "loss": 5.3663,
      "mean_token_accuracy": 0.17836857438087464,
      "num_tokens": 42018386.0,
      "step": 20245
    },
    {
      "entropy": 5.72585015296936,
      "epoch": 1.8369013062409287,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004663903782485678,
      "loss": 5.4407,
      "mean_token_accuracy": 0.16871878653764724,
      "num_tokens": 42028036.0,
      "step": 20250
    },
    {
      "entropy": 5.644705533981323,
      "epoch": 1.837354862119013,
      "grad_norm": 3.046875,
      "learning_rate": 0.00046637336408455793,
      "loss": 5.3503,
      "mean_token_accuracy": 0.17195575684309006,
      "num_tokens": 42037535.0,
      "step": 20255
    },
    {
      "entropy": 5.626899194717407,
      "epoch": 1.8378084179970973,
      "grad_norm": 2.640625,
      "learning_rate": 0.00046635634596289816,
      "loss": 5.2944,
      "mean_token_accuracy": 0.1819162830710411,
      "num_tokens": 42047924.0,
      "step": 20260
    },
    {
      "entropy": 5.611018466949463,
      "epoch": 1.8382619738751815,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004663393238839404,
      "loss": 5.2882,
      "mean_token_accuracy": 0.18486265689134598,
      "num_tokens": 42058734.0,
      "step": 20265
    },
    {
      "entropy": 5.651136302947998,
      "epoch": 1.8387155297532656,
      "grad_norm": 2.25,
      "learning_rate": 0.0004663222978480366,
      "loss": 5.4408,
      "mean_token_accuracy": 0.17486929893493652,
      "num_tokens": 42069929.0,
      "step": 20270
    },
    {
      "entropy": 5.753873729705811,
      "epoch": 1.8391690856313496,
      "grad_norm": 1.71875,
      "learning_rate": 0.000466305267855539,
      "loss": 5.4604,
      "mean_token_accuracy": 0.16942733973264695,
      "num_tokens": 42080812.0,
      "step": 20275
    },
    {
      "entropy": 5.711077928543091,
      "epoch": 1.8396226415094339,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004662882339067996,
      "loss": 5.3577,
      "mean_token_accuracy": 0.17804041504859924,
      "num_tokens": 42091094.0,
      "step": 20280
    },
    {
      "entropy": 5.678379535675049,
      "epoch": 1.8400761973875182,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004662711960021709,
      "loss": 5.4053,
      "mean_token_accuracy": 0.1758732408285141,
      "num_tokens": 42100955.0,
      "step": 20285
    },
    {
      "entropy": 5.697181844711304,
      "epoch": 1.8405297532656024,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00046625415414200513,
      "loss": 5.5033,
      "mean_token_accuracy": 0.1740601494908333,
      "num_tokens": 42112250.0,
      "step": 20290
    },
    {
      "entropy": 5.729772424697876,
      "epoch": 1.8409833091436865,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004662371083266546,
      "loss": 5.5273,
      "mean_token_accuracy": 0.17085374891757965,
      "num_tokens": 42123619.0,
      "step": 20295
    },
    {
      "entropy": 5.708144235610962,
      "epoch": 1.8414368650217707,
      "grad_norm": 2.53125,
      "learning_rate": 0.000466220058556472,
      "loss": 5.3525,
      "mean_token_accuracy": 0.17291635721921922,
      "num_tokens": 42133645.0,
      "step": 20300
    },
    {
      "entropy": 5.749553680419922,
      "epoch": 1.8418904208998548,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046620300483180986,
      "loss": 5.4422,
      "mean_token_accuracy": 0.17099908143281936,
      "num_tokens": 42143860.0,
      "step": 20305
    },
    {
      "entropy": 5.698175525665283,
      "epoch": 1.842343976777939,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004661859471530209,
      "loss": 5.425,
      "mean_token_accuracy": 0.1718703642487526,
      "num_tokens": 42153921.0,
      "step": 20310
    },
    {
      "entropy": 5.640582513809204,
      "epoch": 1.8427975326560233,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004661688855204578,
      "loss": 5.3679,
      "mean_token_accuracy": 0.17819705605506897,
      "num_tokens": 42163999.0,
      "step": 20315
    },
    {
      "entropy": 5.7128682136535645,
      "epoch": 1.8432510885341074,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046615181993447344,
      "loss": 5.4642,
      "mean_token_accuracy": 0.16553137451410294,
      "num_tokens": 42174669.0,
      "step": 20320
    },
    {
      "entropy": 5.670054006576538,
      "epoch": 1.8437046444121916,
      "grad_norm": 3.640625,
      "learning_rate": 0.0004661347503954207,
      "loss": 5.3972,
      "mean_token_accuracy": 0.17055997997522354,
      "num_tokens": 42185287.0,
      "step": 20325
    },
    {
      "entropy": 5.667064428329468,
      "epoch": 1.8441582002902757,
      "grad_norm": 3.84375,
      "learning_rate": 0.00046611767690365265,
      "loss": 5.3965,
      "mean_token_accuracy": 0.16996079087257385,
      "num_tokens": 42195646.0,
      "step": 20330
    },
    {
      "entropy": 5.668585968017578,
      "epoch": 1.84461175616836,
      "grad_norm": 5.25,
      "learning_rate": 0.0004661005994595223,
      "loss": 5.3531,
      "mean_token_accuracy": 0.17339830100536346,
      "num_tokens": 42205193.0,
      "step": 20335
    },
    {
      "entropy": 5.710688686370849,
      "epoch": 1.8450653120464442,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004660835180633829,
      "loss": 5.3894,
      "mean_token_accuracy": 0.17475088983774184,
      "num_tokens": 42215873.0,
      "step": 20340
    },
    {
      "entropy": 5.640176105499267,
      "epoch": 1.8455188679245285,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004660664327155876,
      "loss": 5.3956,
      "mean_token_accuracy": 0.1790688917040825,
      "num_tokens": 42226724.0,
      "step": 20345
    },
    {
      "entropy": 5.717349100112915,
      "epoch": 1.8459724238026125,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004660493434164898,
      "loss": 5.4325,
      "mean_token_accuracy": 0.16667297035455703,
      "num_tokens": 42236749.0,
      "step": 20350
    },
    {
      "entropy": 5.667660140991211,
      "epoch": 1.8464259796806965,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004660322501664429,
      "loss": 5.3401,
      "mean_token_accuracy": 0.17285132259130478,
      "num_tokens": 42246800.0,
      "step": 20355
    },
    {
      "entropy": 5.594280624389649,
      "epoch": 1.8468795355587808,
      "grad_norm": 2.890625,
      "learning_rate": 0.0004660151529658004,
      "loss": 5.25,
      "mean_token_accuracy": 0.1818648725748062,
      "num_tokens": 42257113.0,
      "step": 20360
    },
    {
      "entropy": 5.461200332641601,
      "epoch": 1.847333091436865,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004659980518149158,
      "loss": 5.2059,
      "mean_token_accuracy": 0.18386257141828538,
      "num_tokens": 42266614.0,
      "step": 20365
    },
    {
      "entropy": 5.644856691360474,
      "epoch": 1.8477866473149493,
      "grad_norm": 2.328125,
      "learning_rate": 0.00046598094671414285,
      "loss": 5.3906,
      "mean_token_accuracy": 0.16989048719406127,
      "num_tokens": 42278064.0,
      "step": 20370
    },
    {
      "entropy": 5.6906177520751955,
      "epoch": 1.8482402031930334,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046596383766383533,
      "loss": 5.2672,
      "mean_token_accuracy": 0.1778758242726326,
      "num_tokens": 42288915.0,
      "step": 20375
    },
    {
      "entropy": 5.688180446624756,
      "epoch": 1.8486937590711174,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004659467246643469,
      "loss": 5.3761,
      "mean_token_accuracy": 0.17170279026031493,
      "num_tokens": 42298547.0,
      "step": 20380
    },
    {
      "entropy": 5.651880979537964,
      "epoch": 1.8491473149492017,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046592960771603154,
      "loss": 5.3481,
      "mean_token_accuracy": 0.180756051838398,
      "num_tokens": 42307402.0,
      "step": 20385
    },
    {
      "entropy": 5.626698589324951,
      "epoch": 1.849600870827286,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004659124868192433,
      "loss": 5.3434,
      "mean_token_accuracy": 0.18166282773017883,
      "num_tokens": 42317920.0,
      "step": 20390
    },
    {
      "entropy": 5.604750633239746,
      "epoch": 1.8500544267053702,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004658953619743361,
      "loss": 5.3277,
      "mean_token_accuracy": 0.17666488587856294,
      "num_tokens": 42328098.0,
      "step": 20395
    },
    {
      "entropy": 5.611264085769653,
      "epoch": 1.8505079825834543,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046587823318166427,
      "loss": 5.2927,
      "mean_token_accuracy": 0.1802089422941208,
      "num_tokens": 42338454.0,
      "step": 20400
    },
    {
      "entropy": 5.683953046798706,
      "epoch": 1.8509615384615383,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046586110044158184,
      "loss": 5.3491,
      "mean_token_accuracy": 0.17684970945119857,
      "num_tokens": 42348760.0,
      "step": 20405
    },
    {
      "entropy": 5.787009000778198,
      "epoch": 1.8514150943396226,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046584396375444335,
      "loss": 5.515,
      "mean_token_accuracy": 0.1626497060060501,
      "num_tokens": 42358919.0,
      "step": 20410
    },
    {
      "entropy": 5.6022905826568605,
      "epoch": 1.8518686502177069,
      "grad_norm": 2.109375,
      "learning_rate": 0.00046582682312060295,
      "loss": 5.4033,
      "mean_token_accuracy": 0.17201340496540068,
      "num_tokens": 42368270.0,
      "step": 20415
    },
    {
      "entropy": 5.655784463882446,
      "epoch": 1.8523222060957911,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004658096785404152,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17106322050094605,
      "num_tokens": 42378703.0,
      "step": 20420
    },
    {
      "entropy": 5.601116943359375,
      "epoch": 1.8527757619738752,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046579253001423473,
      "loss": 5.3122,
      "mean_token_accuracy": 0.1846354842185974,
      "num_tokens": 42388251.0,
      "step": 20425
    },
    {
      "entropy": 5.733778285980224,
      "epoch": 1.8532293178519592,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00046577537754241604,
      "loss": 5.5569,
      "mean_token_accuracy": 0.1633085548877716,
      "num_tokens": 42399186.0,
      "step": 20430
    },
    {
      "entropy": 5.712880611419678,
      "epoch": 1.8536828737300435,
      "grad_norm": 2.0625,
      "learning_rate": 0.000465758221125314,
      "loss": 5.4187,
      "mean_token_accuracy": 0.1736907944083214,
      "num_tokens": 42409472.0,
      "step": 20435
    },
    {
      "entropy": 5.706215238571167,
      "epoch": 1.8541364296081277,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00046574106076328333,
      "loss": 5.4356,
      "mean_token_accuracy": 0.16916139125823976,
      "num_tokens": 42419908.0,
      "step": 20440
    },
    {
      "entropy": 5.718715333938599,
      "epoch": 1.854589985486212,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004657238964566789,
      "loss": 5.3507,
      "mean_token_accuracy": 0.17050822973251342,
      "num_tokens": 42429448.0,
      "step": 20445
    },
    {
      "entropy": 5.635370540618896,
      "epoch": 1.855043541364296,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004657067282058557,
      "loss": 5.3136,
      "mean_token_accuracy": 0.18014246672391893,
      "num_tokens": 42439337.0,
      "step": 20450
    },
    {
      "entropy": 5.611591434478759,
      "epoch": 1.85549709724238,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004656895560111687,
      "loss": 5.3654,
      "mean_token_accuracy": 0.1753848060965538,
      "num_tokens": 42450794.0,
      "step": 20455
    },
    {
      "entropy": 5.611173868179321,
      "epoch": 1.8559506531204644,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046567237987297323,
      "loss": 5.3291,
      "mean_token_accuracy": 0.17189863324165344,
      "num_tokens": 42461128.0,
      "step": 20460
    },
    {
      "entropy": 5.666402387619018,
      "epoch": 1.8564042089985486,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004656551997916243,
      "loss": 5.4218,
      "mean_token_accuracy": 0.1774281308054924,
      "num_tokens": 42471578.0,
      "step": 20465
    },
    {
      "entropy": 5.640766334533692,
      "epoch": 1.856857764876633,
      "grad_norm": 4.59375,
      "learning_rate": 0.00046563801576747716,
      "loss": 5.3725,
      "mean_token_accuracy": 0.17644034773111344,
      "num_tokens": 42482013.0,
      "step": 20470
    },
    {
      "entropy": 5.722328424453735,
      "epoch": 1.857311320754717,
      "grad_norm": 1.875,
      "learning_rate": 0.00046562082780088733,
      "loss": 5.4363,
      "mean_token_accuracy": 0.17173457145690918,
      "num_tokens": 42492577.0,
      "step": 20475
    },
    {
      "entropy": 5.661021995544433,
      "epoch": 1.8577648766328012,
      "grad_norm": 4.9375,
      "learning_rate": 0.0004656036358922103,
      "loss": 5.3629,
      "mean_token_accuracy": 0.17590501457452773,
      "num_tokens": 42502319.0,
      "step": 20480
    },
    {
      "entropy": 5.6875927448272705,
      "epoch": 1.8582184325108853,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004655864400418014,
      "loss": 5.2916,
      "mean_token_accuracy": 0.1795194387435913,
      "num_tokens": 42512723.0,
      "step": 20485
    },
    {
      "entropy": 5.646891498565674,
      "epoch": 1.8586719883889695,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004655692402500163,
      "loss": 5.3391,
      "mean_token_accuracy": 0.17237386107444763,
      "num_tokens": 42522933.0,
      "step": 20490
    },
    {
      "entropy": 5.5848997116088865,
      "epoch": 1.8591255442670538,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046555203651721085,
      "loss": 5.3574,
      "mean_token_accuracy": 0.1785918653011322,
      "num_tokens": 42532767.0,
      "step": 20495
    },
    {
      "entropy": 5.6453314304351805,
      "epoch": 1.8595791001451378,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004655348288437407,
      "loss": 5.2675,
      "mean_token_accuracy": 0.17896240949630737,
      "num_tokens": 42543344.0,
      "step": 20500
    },
    {
      "entropy": 5.59553713798523,
      "epoch": 1.860032656023222,
      "grad_norm": 2.90625,
      "learning_rate": 0.0004655176172299618,
      "loss": 5.3678,
      "mean_token_accuracy": 0.17198372930288314,
      "num_tokens": 42553370.0,
      "step": 20505
    },
    {
      "entropy": 5.671284866333008,
      "epoch": 1.8604862119013061,
      "grad_norm": 5.40625,
      "learning_rate": 0.00046550040167622986,
      "loss": 5.4591,
      "mean_token_accuracy": 0.16712542176246642,
      "num_tokens": 42564299.0,
      "step": 20510
    },
    {
      "entropy": 5.741981840133667,
      "epoch": 1.8609397677793904,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046548318218290124,
      "loss": 5.4174,
      "mean_token_accuracy": 0.1685711994767189,
      "num_tokens": 42574742.0,
      "step": 20515
    },
    {
      "entropy": 5.650928211212158,
      "epoch": 1.8613933236574747,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004654659587503318,
      "loss": 5.4064,
      "mean_token_accuracy": 0.16906213015317917,
      "num_tokens": 42584733.0,
      "step": 20520
    },
    {
      "entropy": 5.739599084854126,
      "epoch": 1.861846879535559,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004654487313788778,
      "loss": 5.4323,
      "mean_token_accuracy": 0.16812880635261535,
      "num_tokens": 42595765.0,
      "step": 20525
    },
    {
      "entropy": 5.746594429016113,
      "epoch": 1.862300435413643,
      "grad_norm": 2.625,
      "learning_rate": 0.00046543150006889546,
      "loss": 5.3804,
      "mean_token_accuracy": 0.17978963404893875,
      "num_tokens": 42607038.0,
      "step": 20530
    },
    {
      "entropy": 5.6420125484466555,
      "epoch": 1.862753991291727,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004654142648207412,
      "loss": 5.323,
      "mean_token_accuracy": 0.18254159688949584,
      "num_tokens": 42616598.0,
      "step": 20535
    },
    {
      "entropy": 5.603188753128052,
      "epoch": 1.8632075471698113,
      "grad_norm": 2.015625,
      "learning_rate": 0.0004653970256347715,
      "loss": 5.3344,
      "mean_token_accuracy": 0.18407076746225357,
      "num_tokens": 42625713.0,
      "step": 20540
    },
    {
      "entropy": 5.7472617626190186,
      "epoch": 1.8636611030478956,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046537978251134265,
      "loss": 5.4441,
      "mean_token_accuracy": 0.16824159771203995,
      "num_tokens": 42637173.0,
      "step": 20545
    },
    {
      "entropy": 5.708297967910767,
      "epoch": 1.8641146589259798,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004653625354508115,
      "loss": 5.3794,
      "mean_token_accuracy": 0.174301777780056,
      "num_tokens": 42646916.0,
      "step": 20550
    },
    {
      "entropy": 5.651949739456176,
      "epoch": 1.8645682148040639,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004653452844535346,
      "loss": 5.4709,
      "mean_token_accuracy": 0.16798167824745178,
      "num_tokens": 42657680.0,
      "step": 20555
    },
    {
      "entropy": 5.6015543937683105,
      "epoch": 1.865021770682148,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046532802951986863,
      "loss": 5.2412,
      "mean_token_accuracy": 0.18428541719913483,
      "num_tokens": 42669368.0,
      "step": 20560
    },
    {
      "entropy": 5.617056179046631,
      "epoch": 1.8654753265602322,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004653107706501705,
      "loss": 5.303,
      "mean_token_accuracy": 0.18147781491279602,
      "num_tokens": 42678865.0,
      "step": 20565
    },
    {
      "entropy": 5.652208518981934,
      "epoch": 1.8659288824383164,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004652935078447973,
      "loss": 5.4352,
      "mean_token_accuracy": 0.1690854087471962,
      "num_tokens": 42690040.0,
      "step": 20570
    },
    {
      "entropy": 5.713571405410766,
      "epoch": 1.8663824383164007,
      "grad_norm": 1.9375,
      "learning_rate": 0.00046527624110410574,
      "loss": 5.2931,
      "mean_token_accuracy": 0.18704282939434053,
      "num_tokens": 42700207.0,
      "step": 20575
    },
    {
      "entropy": 5.6390880107879635,
      "epoch": 1.8668359941944848,
      "grad_norm": 2.421875,
      "learning_rate": 0.000465258970428453,
      "loss": 5.3611,
      "mean_token_accuracy": 0.17294020801782609,
      "num_tokens": 42710514.0,
      "step": 20580
    },
    {
      "entropy": 5.666609811782837,
      "epoch": 1.8672895500725688,
      "grad_norm": 2.390625,
      "learning_rate": 0.00046524169581819633,
      "loss": 5.478,
      "mean_token_accuracy": 0.16641386598348618,
      "num_tokens": 42719687.0,
      "step": 20585
    },
    {
      "entropy": 5.681258821487427,
      "epoch": 1.867743105950653,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004652244172736929,
      "loss": 5.2929,
      "mean_token_accuracy": 0.17680273503065108,
      "num_tokens": 42729444.0,
      "step": 20590
    },
    {
      "entropy": 5.705546140670776,
      "epoch": 1.8681966618287373,
      "grad_norm": 1.9453125,
      "learning_rate": 0.0004652071347953,
      "loss": 5.3923,
      "mean_token_accuracy": 0.16899392902851104,
      "num_tokens": 42739441.0,
      "step": 20595
    },
    {
      "entropy": 5.634426498413086,
      "epoch": 1.8686502177068216,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046518984838337513,
      "loss": 5.3187,
      "mean_token_accuracy": 0.17389596551656722,
      "num_tokens": 42749008.0,
      "step": 20600
    },
    {
      "entropy": 5.613066816329956,
      "epoch": 1.8691037735849056,
      "grad_norm": 2.296875,
      "learning_rate": 0.0004651725580382757,
      "loss": 5.2905,
      "mean_token_accuracy": 0.17876540571451188,
      "num_tokens": 42759611.0,
      "step": 20605
    },
    {
      "entropy": 5.612070083618164,
      "epoch": 1.8695573294629897,
      "grad_norm": 1.890625,
      "learning_rate": 0.00046515526376035934,
      "loss": 5.3442,
      "mean_token_accuracy": 0.17607614398002625,
      "num_tokens": 42770822.0,
      "step": 20610
    },
    {
      "entropy": 5.764590167999268,
      "epoch": 1.870010885341074,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004651379655499837,
      "loss": 5.4561,
      "mean_token_accuracy": 0.17055740505456923,
      "num_tokens": 42782818.0,
      "step": 20615
    },
    {
      "entropy": 5.6732673168182375,
      "epoch": 1.8704644412191582,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00046512066340750636,
      "loss": 5.4099,
      "mean_token_accuracy": 0.17329320460557937,
      "num_tokens": 42793188.0,
      "step": 20620
    },
    {
      "entropy": 5.6622727394104,
      "epoch": 1.8709179970972425,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004651033573332853,
      "loss": 5.4031,
      "mean_token_accuracy": 0.17083307951688767,
      "num_tokens": 42803059.0,
      "step": 20625
    },
    {
      "entropy": 5.703035593032837,
      "epoch": 1.8713715529753265,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004650860473276784,
      "loss": 5.3629,
      "mean_token_accuracy": 0.1767849698662758,
      "num_tokens": 42812956.0,
      "step": 20630
    },
    {
      "entropy": 5.6179687023162845,
      "epoch": 1.8718251088534106,
      "grad_norm": 3.0,
      "learning_rate": 0.0004650687333910436,
      "loss": 5.3553,
      "mean_token_accuracy": 0.17864009141921997,
      "num_tokens": 42824184.0,
      "step": 20635
    },
    {
      "entropy": 5.58851432800293,
      "epoch": 1.8722786647314948,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046505141552373897,
      "loss": 5.3599,
      "mean_token_accuracy": 0.18072095066308974,
      "num_tokens": 42835064.0,
      "step": 20640
    },
    {
      "entropy": 5.7012840747833256,
      "epoch": 1.8727322206095791,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004650340937261226,
      "loss": 5.4122,
      "mean_token_accuracy": 0.17333813309669494,
      "num_tokens": 42845605.0,
      "step": 20645
    },
    {
      "entropy": 5.677788877487183,
      "epoch": 1.8731857764876634,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004650167679985528,
      "loss": 5.2384,
      "mean_token_accuracy": 0.18692701756954194,
      "num_tokens": 42855271.0,
      "step": 20650
    },
    {
      "entropy": 5.615408229827881,
      "epoch": 1.8736393323657474,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004649994383413877,
      "loss": 5.3327,
      "mean_token_accuracy": 0.17635460942983627,
      "num_tokens": 42865763.0,
      "step": 20655
    },
    {
      "entropy": 5.64029541015625,
      "epoch": 1.8740928882438317,
      "grad_norm": 2.625,
      "learning_rate": 0.00046498210475498595,
      "loss": 5.4028,
      "mean_token_accuracy": 0.17061852514743805,
      "num_tokens": 42877488.0,
      "step": 20660
    },
    {
      "entropy": 5.697337436676025,
      "epoch": 1.8745464441219157,
      "grad_norm": 2.3125,
      "learning_rate": 0.00046496476723970574,
      "loss": 5.405,
      "mean_token_accuracy": 0.17076704502105713,
      "num_tokens": 42887946.0,
      "step": 20665
    },
    {
      "entropy": 5.687584304809571,
      "epoch": 1.875,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004649474257959058,
      "loss": 5.3217,
      "mean_token_accuracy": 0.1769715040922165,
      "num_tokens": 42897809.0,
      "step": 20670
    },
    {
      "entropy": 5.745008707046509,
      "epoch": 1.8754535558780843,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004649300804239447,
      "loss": 5.4368,
      "mean_token_accuracy": 0.17400169223546982,
      "num_tokens": 42907657.0,
      "step": 20675
    },
    {
      "entropy": 5.730487298965454,
      "epoch": 1.8759071117561683,
      "grad_norm": 3.765625,
      "learning_rate": 0.00046491273112418105,
      "loss": 5.433,
      "mean_token_accuracy": 0.16334341764450072,
      "num_tokens": 42919380.0,
      "step": 20680
    },
    {
      "entropy": 5.693149280548096,
      "epoch": 1.8763606676342526,
      "grad_norm": 1.9765625,
      "learning_rate": 0.00046489537789697386,
      "loss": 5.4601,
      "mean_token_accuracy": 0.16949280127882957,
      "num_tokens": 42930740.0,
      "step": 20685
    },
    {
      "entropy": 5.628039598464966,
      "epoch": 1.8768142235123366,
      "grad_norm": 2.046875,
      "learning_rate": 0.00046487802074268177,
      "loss": 5.3637,
      "mean_token_accuracy": 0.18182642310857772,
      "num_tokens": 42941814.0,
      "step": 20690
    },
    {
      "entropy": 5.62661600112915,
      "epoch": 1.8772677793904209,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046486065966166377,
      "loss": 5.3347,
      "mean_token_accuracy": 0.16985590010881424,
      "num_tokens": 42952442.0,
      "step": 20695
    },
    {
      "entropy": 5.647927856445312,
      "epoch": 1.8777213352685052,
      "grad_norm": 2.0,
      "learning_rate": 0.0004648432946542791,
      "loss": 5.3066,
      "mean_token_accuracy": 0.17665124237537383,
      "num_tokens": 42963134.0,
      "step": 20700
    },
    {
      "entropy": 5.622011470794678,
      "epoch": 1.8781748911465894,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046482592572088654,
      "loss": 5.3388,
      "mean_token_accuracy": 0.17682987302541733,
      "num_tokens": 42973526.0,
      "step": 20705
    },
    {
      "entropy": 5.664530801773071,
      "epoch": 1.8786284470246735,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046480855286184565,
      "loss": 5.4083,
      "mean_token_accuracy": 0.17350216209888458,
      "num_tokens": 42983362.0,
      "step": 20710
    },
    {
      "entropy": 5.635238170623779,
      "epoch": 1.8790820029027575,
      "grad_norm": 1.8359375,
      "learning_rate": 0.00046479117607751535,
      "loss": 5.3322,
      "mean_token_accuracy": 0.17420356571674347,
      "num_tokens": 42993241.0,
      "step": 20715
    },
    {
      "entropy": 5.738344717025757,
      "epoch": 1.8795355587808418,
      "grad_norm": 1.625,
      "learning_rate": 0.00046477379536825534,
      "loss": 5.3699,
      "mean_token_accuracy": 0.1728500857949257,
      "num_tokens": 43002447.0,
      "step": 20720
    },
    {
      "entropy": 5.65714168548584,
      "epoch": 1.879989114658926,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004647564107344247,
      "loss": 5.3805,
      "mean_token_accuracy": 0.17768396884202958,
      "num_tokens": 43012016.0,
      "step": 20725
    },
    {
      "entropy": 5.622313070297241,
      "epoch": 1.8804426705370103,
      "grad_norm": 2.015625,
      "learning_rate": 0.00046473902217638327,
      "loss": 5.3144,
      "mean_token_accuracy": 0.17977253049612046,
      "num_tokens": 43021781.0,
      "step": 20730
    },
    {
      "entropy": 5.717486715316772,
      "epoch": 1.8808962264150944,
      "grad_norm": 2.46875,
      "learning_rate": 0.00046472162969449037,
      "loss": 5.4535,
      "mean_token_accuracy": 0.17684924751520156,
      "num_tokens": 43033400.0,
      "step": 20735
    },
    {
      "entropy": 5.733179807662964,
      "epoch": 1.8813497822931784,
      "grad_norm": 2.203125,
      "learning_rate": 0.00046470423328910594,
      "loss": 5.4557,
      "mean_token_accuracy": 0.1727564141154289,
      "num_tokens": 43043506.0,
      "step": 20740
    },
    {
      "entropy": 5.686896944046021,
      "epoch": 1.8818033381712627,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046468683296058957,
      "loss": 5.3682,
      "mean_token_accuracy": 0.17695869505405426,
      "num_tokens": 43054888.0,
      "step": 20745
    },
    {
      "entropy": 5.656042051315308,
      "epoch": 1.882256894049347,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00046466942870930126,
      "loss": 5.4258,
      "mean_token_accuracy": 0.1657623827457428,
      "num_tokens": 43066305.0,
      "step": 20750
    },
    {
      "entropy": 5.634464406967163,
      "epoch": 1.8827104499274312,
      "grad_norm": 2.15625,
      "learning_rate": 0.0004646520205356007,
      "loss": 5.3263,
      "mean_token_accuracy": 0.17152570933103561,
      "num_tokens": 43076937.0,
      "step": 20755
    },
    {
      "entropy": 5.636438369750977,
      "epoch": 1.8831640058055152,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00046463460843984805,
      "loss": 5.3914,
      "mean_token_accuracy": 0.17345686107873917,
      "num_tokens": 43086759.0,
      "step": 20760
    },
    {
      "entropy": 5.62012677192688,
      "epoch": 1.8836175616835993,
      "grad_norm": 2.96875,
      "learning_rate": 0.0004646171924224034,
      "loss": 5.3733,
      "mean_token_accuracy": 0.16977958083152772,
      "num_tokens": 43097603.0,
      "step": 20765
    },
    {
      "entropy": 5.629866313934326,
      "epoch": 1.8840711175616836,
      "grad_norm": 2.78125,
      "learning_rate": 0.00046459977248362687,
      "loss": 5.3117,
      "mean_token_accuracy": 0.17724286913871765,
      "num_tokens": 43107222.0,
      "step": 20770
    },
    {
      "entropy": 5.591308498382569,
      "epoch": 1.8845246734397678,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004645823486238788,
      "loss": 5.3163,
      "mean_token_accuracy": 0.19049601554870604,
      "num_tokens": 43117259.0,
      "step": 20775
    },
    {
      "entropy": 5.632685041427612,
      "epoch": 1.884978229317852,
      "grad_norm": 2.375,
      "learning_rate": 0.0004645649208435193,
      "loss": 5.4027,
      "mean_token_accuracy": 0.17530108541250228,
      "num_tokens": 43127647.0,
      "step": 20780
    },
    {
      "entropy": 5.721476650238037,
      "epoch": 1.8854317851959361,
      "grad_norm": 2.28125,
      "learning_rate": 0.00046454748914290906,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1723230853676796,
      "num_tokens": 43138338.0,
      "step": 20785
    },
    {
      "entropy": 5.752392721176148,
      "epoch": 1.8858853410740202,
      "grad_norm": 2.234375,
      "learning_rate": 0.00046453005352240835,
      "loss": 5.4711,
      "mean_token_accuracy": 0.16796962767839432,
      "num_tokens": 43148985.0,
      "step": 20790
    },
    {
      "entropy": 5.735848331451416,
      "epoch": 1.8863388969521044,
      "grad_norm": 1.890625,
      "learning_rate": 0.000464512613982378,
      "loss": 5.4472,
      "mean_token_accuracy": 0.172340552508831,
      "num_tokens": 43158754.0,
      "step": 20795
    },
    {
      "entropy": 5.742519855499268,
      "epoch": 1.8867924528301887,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004644951705231783,
      "loss": 5.4204,
      "mean_token_accuracy": 0.1671513020992279,
      "num_tokens": 43169923.0,
      "step": 20800
    },
    {
      "entropy": 5.666316318511963,
      "epoch": 1.887246008708273,
      "grad_norm": 1.8984375,
      "learning_rate": 0.00046447772314517024,
      "loss": 5.3366,
      "mean_token_accuracy": 0.1732471764087677,
      "num_tokens": 43179859.0,
      "step": 20805
    },
    {
      "entropy": 5.649026107788086,
      "epoch": 1.887699564586357,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046446027184871464,
      "loss": 5.3883,
      "mean_token_accuracy": 0.17691343426704406,
      "num_tokens": 43189715.0,
      "step": 20810
    },
    {
      "entropy": 5.608745765686035,
      "epoch": 1.888153120464441,
      "grad_norm": 2.140625,
      "learning_rate": 0.00046444281663417236,
      "loss": 5.3159,
      "mean_token_accuracy": 0.1719416618347168,
      "num_tokens": 43199939.0,
      "step": 20815
    },
    {
      "entropy": 5.725190258026123,
      "epoch": 1.8886066763425253,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004644253575019043,
      "loss": 5.4391,
      "mean_token_accuracy": 0.17553533911705016,
      "num_tokens": 43209688.0,
      "step": 20820
    },
    {
      "entropy": 5.675483703613281,
      "epoch": 1.8890602322206096,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004644078944522716,
      "loss": 5.4367,
      "mean_token_accuracy": 0.17167377173900605,
      "num_tokens": 43220180.0,
      "step": 20825
    },
    {
      "entropy": 5.656000995635987,
      "epoch": 1.8895137880986939,
      "grad_norm": 2.59375,
      "learning_rate": 0.0004643904274856353,
      "loss": 5.4539,
      "mean_token_accuracy": 0.16917806565761567,
      "num_tokens": 43231671.0,
      "step": 20830
    },
    {
      "entropy": 5.718230772018432,
      "epoch": 1.889967343976778,
      "grad_norm": 2.484375,
      "learning_rate": 0.00046437295660235674,
      "loss": 5.3616,
      "mean_token_accuracy": 0.1793511167168617,
      "num_tokens": 43242228.0,
      "step": 20835
    },
    {
      "entropy": 5.609744548797607,
      "epoch": 1.8904208998548622,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004643554818027972,
      "loss": 5.4032,
      "mean_token_accuracy": 0.17403829991817474,
      "num_tokens": 43252406.0,
      "step": 20840
    },
    {
      "entropy": 5.685324859619141,
      "epoch": 1.8908744557329462,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004643380030873181,
      "loss": 5.3572,
      "mean_token_accuracy": 0.16907366812229158,
      "num_tokens": 43264170.0,
      "step": 20845
    },
    {
      "entropy": 5.724819707870483,
      "epoch": 1.8913280116110305,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004643205204562808,
      "loss": 5.3864,
      "mean_token_accuracy": 0.17024774849414825,
      "num_tokens": 43273841.0,
      "step": 20850
    },
    {
      "entropy": 5.7200014114379885,
      "epoch": 1.8917815674891147,
      "grad_norm": 1.84375,
      "learning_rate": 0.00046430303391004687,
      "loss": 5.4768,
      "mean_token_accuracy": 0.16591163277626036,
      "num_tokens": 43283785.0,
      "step": 20855
    },
    {
      "entropy": 5.713426017761231,
      "epoch": 1.8922351233671988,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004642855434489779,
      "loss": 5.4351,
      "mean_token_accuracy": 0.1684142455458641,
      "num_tokens": 43294665.0,
      "step": 20860
    },
    {
      "entropy": 5.6577435493469235,
      "epoch": 1.892688679245283,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00046426804907343574,
      "loss": 5.3727,
      "mean_token_accuracy": 0.177417354285717,
      "num_tokens": 43305457.0,
      "step": 20865
    },
    {
      "entropy": 5.670142698287964,
      "epoch": 1.893142235123367,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046425055078378206,
      "loss": 5.352,
      "mean_token_accuracy": 0.17906032502651215,
      "num_tokens": 43316180.0,
      "step": 20870
    },
    {
      "entropy": 5.757786846160888,
      "epoch": 1.8935957910014514,
      "grad_norm": 2.0,
      "learning_rate": 0.00046423304858037885,
      "loss": 5.4342,
      "mean_token_accuracy": 0.1690910205245018,
      "num_tokens": 43326390.0,
      "step": 20875
    },
    {
      "entropy": 5.637454891204834,
      "epoch": 1.8940493468795356,
      "grad_norm": 2.4375,
      "learning_rate": 0.0004642155424635878,
      "loss": 5.311,
      "mean_token_accuracy": 0.1747751459479332,
      "num_tokens": 43335834.0,
      "step": 20880
    },
    {
      "entropy": 5.616539001464844,
      "epoch": 1.89450290275762,
      "grad_norm": 2.34375,
      "learning_rate": 0.00046419803243377125,
      "loss": 5.3478,
      "mean_token_accuracy": 0.17001776099205018,
      "num_tokens": 43346350.0,
      "step": 20885
    },
    {
      "entropy": 5.584320831298828,
      "epoch": 1.894956458635704,
      "grad_norm": 1.71875,
      "learning_rate": 0.00046418051849129115,
      "loss": 5.291,
      "mean_token_accuracy": 0.1734038770198822,
      "num_tokens": 43357066.0,
      "step": 20890
    },
    {
      "entropy": 5.565988683700562,
      "epoch": 1.895410014513788,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004641630006365097,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1803300589323044,
      "num_tokens": 43366915.0,
      "step": 20895
    },
    {
      "entropy": 5.586270952224732,
      "epoch": 1.8958635703918723,
      "grad_norm": 2.21875,
      "learning_rate": 0.00046414547886978926,
      "loss": 5.2516,
      "mean_token_accuracy": 0.1837461471557617,
      "num_tokens": 43376892.0,
      "step": 20900
    },
    {
      "entropy": 5.6851232051849365,
      "epoch": 1.8963171262699565,
      "grad_norm": 1.8671875,
      "learning_rate": 0.00046412795319149204,
      "loss": 5.3525,
      "mean_token_accuracy": 0.17639155089855194,
      "num_tokens": 43387349.0,
      "step": 20905
    },
    {
      "entropy": 5.72498140335083,
      "epoch": 1.8967706821480408,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046411042360198053,
      "loss": 5.4297,
      "mean_token_accuracy": 0.16603329479694368,
      "num_tokens": 43397386.0,
      "step": 20910
    },
    {
      "entropy": 5.604062271118164,
      "epoch": 1.8972242380261248,
      "grad_norm": 3.6875,
      "learning_rate": 0.0004640928901016173,
      "loss": 5.3884,
      "mean_token_accuracy": 0.1708528444170952,
      "num_tokens": 43407447.0,
      "step": 20915
    },
    {
      "entropy": 5.646574544906616,
      "epoch": 1.8976777939042089,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046407535269076493,
      "loss": 5.392,
      "mean_token_accuracy": 0.17666592001914977,
      "num_tokens": 43418361.0,
      "step": 20920
    },
    {
      "entropy": 5.648053121566773,
      "epoch": 1.8981313497822931,
      "grad_norm": 1.578125,
      "learning_rate": 0.00046405781136978606,
      "loss": 5.3375,
      "mean_token_accuracy": 0.17418869882822036,
      "num_tokens": 43429273.0,
      "step": 20925
    },
    {
      "entropy": 5.74165678024292,
      "epoch": 1.8985849056603774,
      "grad_norm": 1.953125,
      "learning_rate": 0.00046404026613904357,
      "loss": 5.3958,
      "mean_token_accuracy": 0.1737823784351349,
      "num_tokens": 43440479.0,
      "step": 20930
    },
    {
      "entropy": 5.583006763458252,
      "epoch": 1.8990384615384617,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004640227169989001,
      "loss": 5.3291,
      "mean_token_accuracy": 0.17837791740894318,
      "num_tokens": 43450479.0,
      "step": 20935
    },
    {
      "entropy": 5.656941652297974,
      "epoch": 1.8994920174165457,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004640051639497187,
      "loss": 5.33,
      "mean_token_accuracy": 0.17845285385847093,
      "num_tokens": 43460294.0,
      "step": 20940
    },
    {
      "entropy": 5.558904981613159,
      "epoch": 1.8999455732946298,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004639876069918624,
      "loss": 5.2841,
      "mean_token_accuracy": 0.1805598571896553,
      "num_tokens": 43471016.0,
      "step": 20945
    },
    {
      "entropy": 5.6601104736328125,
      "epoch": 1.900399129172714,
      "grad_norm": 1.640625,
      "learning_rate": 0.00046397004612569416,
      "loss": 5.2984,
      "mean_token_accuracy": 0.17779153287410737,
      "num_tokens": 43480937.0,
      "step": 20950
    },
    {
      "entropy": 5.798142194747925,
      "epoch": 1.9008526850507983,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00046395248135157715,
      "loss": 5.4829,
      "mean_token_accuracy": 0.1665995806455612,
      "num_tokens": 43491726.0,
      "step": 20955
    },
    {
      "entropy": 5.696995067596435,
      "epoch": 1.9013062409288826,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046393491266987484,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1776529312133789,
      "num_tokens": 43502392.0,
      "step": 20960
    },
    {
      "entropy": 5.6151978969573975,
      "epoch": 1.9017597968069666,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004639173400809502,
      "loss": 5.3233,
      "mean_token_accuracy": 0.17552799880504608,
      "num_tokens": 43512673.0,
      "step": 20965
    },
    {
      "entropy": 5.580069637298584,
      "epoch": 1.9022133526850507,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004638997635851669,
      "loss": 5.334,
      "mean_token_accuracy": 0.16825908422470093,
      "num_tokens": 43523178.0,
      "step": 20970
    },
    {
      "entropy": 5.5599109649658205,
      "epoch": 1.902666908563135,
      "grad_norm": 2.625,
      "learning_rate": 0.00046388218318288844,
      "loss": 5.2391,
      "mean_token_accuracy": 0.1786363884806633,
      "num_tokens": 43532065.0,
      "step": 20975
    },
    {
      "entropy": 5.713760232925415,
      "epoch": 1.9031204644412192,
      "grad_norm": 3.4375,
      "learning_rate": 0.0004638645988744782,
      "loss": 5.5145,
      "mean_token_accuracy": 0.16787322163581847,
      "num_tokens": 43543736.0,
      "step": 20980
    },
    {
      "entropy": 5.727261257171631,
      "epoch": 1.9035740203193035,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004638470106603,
      "loss": 5.3666,
      "mean_token_accuracy": 0.17254180312156678,
      "num_tokens": 43553033.0,
      "step": 20985
    },
    {
      "entropy": 5.7016438961029055,
      "epoch": 1.9040275761973875,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004638294185407174,
      "loss": 5.3642,
      "mean_token_accuracy": 0.17831367552280425,
      "num_tokens": 43563105.0,
      "step": 20990
    },
    {
      "entropy": 5.578889274597168,
      "epoch": 1.9044811320754715,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004638118225160944,
      "loss": 5.2324,
      "mean_token_accuracy": 0.18929712027311324,
      "num_tokens": 43572600.0,
      "step": 20995
    },
    {
      "entropy": 5.608998537063599,
      "epoch": 1.9049346879535558,
      "grad_norm": 2.171875,
      "learning_rate": 0.00046379422258679477,
      "loss": 5.3516,
      "mean_token_accuracy": 0.17793585658073424,
      "num_tokens": 43582390.0,
      "step": 21000
    },
    {
      "epoch": 1.9049346879535558,
      "eval_entropy": 5.431988091382506,
      "eval_loss": 5.428965091705322,
      "eval_mean_token_accuracy": 0.1799008139876893,
      "eval_num_tokens": 43582390.0,
      "eval_runtime": 26.43,
      "eval_samples_per_second": 1337.875,
      "eval_steps_per_second": 167.234,
      "step": 21000
    }
  ],
  "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": 6.3753034825728e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}