{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1368,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.3891168855130672,
      "epoch": 0.043859649122807015,
      "grad_norm": 18.536834716796875,
      "learning_rate": 2e-05,
      "loss": 4.281,
      "mean_token_accuracy": 0.5703265547752381,
      "num_tokens": 29513.0,
      "step": 10
    },
    {
      "entropy": 0.749189381301403,
      "epoch": 0.08771929824561403,
      "grad_norm": 3.8398051261901855,
      "learning_rate": 2e-05,
      "loss": 1.9625,
      "mean_token_accuracy": 0.6897866070270539,
      "num_tokens": 58932.0,
      "step": 20
    },
    {
      "entropy": 1.169826951622963,
      "epoch": 0.13157894736842105,
      "grad_norm": 2.5061917304992676,
      "learning_rate": 2e-05,
      "loss": 1.4255,
      "mean_token_accuracy": 0.7357227891683579,
      "num_tokens": 88352.0,
      "step": 30
    },
    {
      "entropy": 1.1334790736436844,
      "epoch": 0.17543859649122806,
      "grad_norm": 1.9473565816879272,
      "learning_rate": 2e-05,
      "loss": 1.1315,
      "mean_token_accuracy": 0.7862708762288093,
      "num_tokens": 117877.0,
      "step": 40
    },
    {
      "entropy": 0.8736076682806015,
      "epoch": 0.21929824561403508,
      "grad_norm": 1.7826578617095947,
      "learning_rate": 2e-05,
      "loss": 0.8549,
      "mean_token_accuracy": 0.8343304082751274,
      "num_tokens": 147392.0,
      "step": 50
    },
    {
      "entropy": 0.5955807730555535,
      "epoch": 0.2631578947368421,
      "grad_norm": 1.6632131338119507,
      "learning_rate": 2e-05,
      "loss": 0.58,
      "mean_token_accuracy": 0.8876266479492188,
      "num_tokens": 176728.0,
      "step": 60
    },
    {
      "entropy": 0.34382262006402015,
      "epoch": 0.30701754385964913,
      "grad_norm": 1.4843252897262573,
      "learning_rate": 2e-05,
      "loss": 0.3445,
      "mean_token_accuracy": 0.9318736225366593,
      "num_tokens": 206184.0,
      "step": 70
    },
    {
      "entropy": 0.19373956136405468,
      "epoch": 0.3508771929824561,
      "grad_norm": 0.904003918170929,
      "learning_rate": 2e-05,
      "loss": 0.189,
      "mean_token_accuracy": 0.9710342198610306,
      "num_tokens": 235688.0,
      "step": 80
    },
    {
      "entropy": 0.12052652426064014,
      "epoch": 0.39473684210526316,
      "grad_norm": 0.7406700849533081,
      "learning_rate": 2e-05,
      "loss": 0.1056,
      "mean_token_accuracy": 0.9886028110980988,
      "num_tokens": 265172.0,
      "step": 90
    },
    {
      "entropy": 0.0992697212845087,
      "epoch": 0.43859649122807015,
      "grad_norm": 0.606019139289856,
      "learning_rate": 2e-05,
      "loss": 0.0817,
      "mean_token_accuracy": 0.989163076877594,
      "num_tokens": 294687.0,
      "step": 100
    },
    {
      "entropy": 0.08367474023252726,
      "epoch": 0.4824561403508772,
      "grad_norm": 0.4705955386161804,
      "learning_rate": 2e-05,
      "loss": 0.0737,
      "mean_token_accuracy": 0.9905414953827858,
      "num_tokens": 324088.0,
      "step": 110
    },
    {
      "entropy": 0.07386345528066159,
      "epoch": 0.5263157894736842,
      "grad_norm": 0.4300910234451294,
      "learning_rate": 2e-05,
      "loss": 0.0631,
      "mean_token_accuracy": 0.9910139158368111,
      "num_tokens": 353582.0,
      "step": 120
    },
    {
      "entropy": 0.06888655042275786,
      "epoch": 0.5701754385964912,
      "grad_norm": 0.45082923769950867,
      "learning_rate": 2e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9910464301705361,
      "num_tokens": 383080.0,
      "step": 130
    },
    {
      "entropy": 0.06553246006369591,
      "epoch": 0.6140350877192983,
      "grad_norm": 0.4258216619491577,
      "learning_rate": 2e-05,
      "loss": 0.0615,
      "mean_token_accuracy": 0.9914803236722947,
      "num_tokens": 412559.0,
      "step": 140
    },
    {
      "entropy": 0.06352790119126439,
      "epoch": 0.6578947368421053,
      "grad_norm": 0.37360644340515137,
      "learning_rate": 2e-05,
      "loss": 0.0579,
      "mean_token_accuracy": 0.9916400715708733,
      "num_tokens": 442008.0,
      "step": 150
    },
    {
      "entropy": 0.060972847137600185,
      "epoch": 0.7017543859649122,
      "grad_norm": 0.3406882584095001,
      "learning_rate": 2e-05,
      "loss": 0.0576,
      "mean_token_accuracy": 0.9914845108985901,
      "num_tokens": 471495.0,
      "step": 160
    },
    {
      "entropy": 0.05891128294169903,
      "epoch": 0.7456140350877193,
      "grad_norm": 0.2854160964488983,
      "learning_rate": 2e-05,
      "loss": 0.057,
      "mean_token_accuracy": 0.9915445119142532,
      "num_tokens": 500954.0,
      "step": 170
    },
    {
      "entropy": 0.0579888011328876,
      "epoch": 0.7894736842105263,
      "grad_norm": 0.40168505907058716,
      "learning_rate": 2e-05,
      "loss": 0.0507,
      "mean_token_accuracy": 0.9923512250185013,
      "num_tokens": 530401.0,
      "step": 180
    },
    {
      "entropy": 0.0573225624859333,
      "epoch": 0.8333333333333334,
      "grad_norm": 0.4589792490005493,
      "learning_rate": 2e-05,
      "loss": 0.0561,
      "mean_token_accuracy": 0.9916056364774704,
      "num_tokens": 559940.0,
      "step": 190
    },
    {
      "entropy": 0.05970530286431312,
      "epoch": 0.8771929824561403,
      "grad_norm": 0.5275149941444397,
      "learning_rate": 2e-05,
      "loss": 0.0573,
      "mean_token_accuracy": 0.9915869951248169,
      "num_tokens": 589427.0,
      "step": 200
    },
    {
      "entropy": 0.059507152158766986,
      "epoch": 0.9210526315789473,
      "grad_norm": 0.2661688029766083,
      "learning_rate": 2e-05,
      "loss": 0.0515,
      "mean_token_accuracy": 0.9923170626163482,
      "num_tokens": 618838.0,
      "step": 210
    },
    {
      "entropy": 0.05687206219881773,
      "epoch": 0.9649122807017544,
      "grad_norm": 0.35179731249809265,
      "learning_rate": 2e-05,
      "loss": 0.0506,
      "mean_token_accuracy": 0.9918064430356026,
      "num_tokens": 648258.0,
      "step": 220
    },
    {
      "entropy": 0.058181710354983804,
      "epoch": 1.0087719298245614,
      "grad_norm": 0.3514145314693451,
      "learning_rate": 2e-05,
      "loss": 0.0522,
      "mean_token_accuracy": 0.9915239199995994,
      "num_tokens": 677779.0,
      "step": 230
    },
    {
      "entropy": 0.055358363036066296,
      "epoch": 1.0526315789473684,
      "grad_norm": 0.3630446493625641,
      "learning_rate": 2e-05,
      "loss": 0.0482,
      "mean_token_accuracy": 0.9922210231423378,
      "num_tokens": 707217.0,
      "step": 240
    },
    {
      "entropy": 0.0538893367163837,
      "epoch": 1.0964912280701755,
      "grad_norm": 0.3740084767341614,
      "learning_rate": 2e-05,
      "loss": 0.0448,
      "mean_token_accuracy": 0.9927210569381714,
      "num_tokens": 736641.0,
      "step": 250
    },
    {
      "entropy": 0.0544189871288836,
      "epoch": 1.1403508771929824,
      "grad_norm": 0.31057634949684143,
      "learning_rate": 2e-05,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9922840282320976,
      "num_tokens": 766189.0,
      "step": 260
    },
    {
      "entropy": 0.053078283276408914,
      "epoch": 1.1842105263157894,
      "grad_norm": 0.21449895203113556,
      "learning_rate": 2e-05,
      "loss": 0.0468,
      "mean_token_accuracy": 0.9923206493258476,
      "num_tokens": 795639.0,
      "step": 270
    },
    {
      "entropy": 0.05415161233395338,
      "epoch": 1.2280701754385965,
      "grad_norm": 0.30692726373672485,
      "learning_rate": 2e-05,
      "loss": 0.0458,
      "mean_token_accuracy": 0.9922255620360374,
      "num_tokens": 825119.0,
      "step": 280
    },
    {
      "entropy": 0.05160839455202222,
      "epoch": 1.2719298245614035,
      "grad_norm": 0.3166206479072571,
      "learning_rate": 2e-05,
      "loss": 0.0446,
      "mean_token_accuracy": 0.9921553313732148,
      "num_tokens": 854581.0,
      "step": 290
    },
    {
      "entropy": 0.05075615206733346,
      "epoch": 1.3157894736842106,
      "grad_norm": 0.3627457618713379,
      "learning_rate": 2e-05,
      "loss": 0.0417,
      "mean_token_accuracy": 0.9924884453415871,
      "num_tokens": 884020.0,
      "step": 300
    },
    {
      "entropy": 0.0517747713252902,
      "epoch": 1.3596491228070176,
      "grad_norm": 0.37561649084091187,
      "learning_rate": 2e-05,
      "loss": 0.0435,
      "mean_token_accuracy": 0.992647610604763,
      "num_tokens": 913529.0,
      "step": 310
    },
    {
      "entropy": 0.05250948471948504,
      "epoch": 1.4035087719298245,
      "grad_norm": 0.42561790347099304,
      "learning_rate": 2e-05,
      "loss": 0.0431,
      "mean_token_accuracy": 0.9922929123044014,
      "num_tokens": 942986.0,
      "step": 320
    },
    {
      "entropy": 0.05172846736386418,
      "epoch": 1.4473684210526316,
      "grad_norm": 0.2577071785926819,
      "learning_rate": 2e-05,
      "loss": 0.041,
      "mean_token_accuracy": 0.9923748031258584,
      "num_tokens": 972495.0,
      "step": 330
    },
    {
      "entropy": 0.04749757144600153,
      "epoch": 1.4912280701754386,
      "grad_norm": 0.25072360038757324,
      "learning_rate": 2e-05,
      "loss": 0.0402,
      "mean_token_accuracy": 0.9924940422177315,
      "num_tokens": 1001947.0,
      "step": 340
    },
    {
      "entropy": 0.04762007407844067,
      "epoch": 1.5350877192982457,
      "grad_norm": 0.3604894280433655,
      "learning_rate": 2e-05,
      "loss": 0.0392,
      "mean_token_accuracy": 0.992351396381855,
      "num_tokens": 1031488.0,
      "step": 350
    },
    {
      "entropy": 0.047403684537857774,
      "epoch": 1.5789473684210527,
      "grad_norm": 0.2911185920238495,
      "learning_rate": 2e-05,
      "loss": 0.0381,
      "mean_token_accuracy": 0.9927886813879013,
      "num_tokens": 1060887.0,
      "step": 360
    },
    {
      "entropy": 0.04852928835898638,
      "epoch": 1.6228070175438596,
      "grad_norm": 0.2618354856967926,
      "learning_rate": 2e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.9927341029047966,
      "num_tokens": 1090348.0,
      "step": 370
    },
    {
      "entropy": 0.04723916696384549,
      "epoch": 1.6666666666666665,
      "grad_norm": 0.26962020993232727,
      "learning_rate": 2e-05,
      "loss": 0.0383,
      "mean_token_accuracy": 0.9925345599651336,
      "num_tokens": 1119839.0,
      "step": 380
    },
    {
      "entropy": 0.046288730949163436,
      "epoch": 1.7105263157894737,
      "grad_norm": 0.4525667130947113,
      "learning_rate": 2e-05,
      "loss": 0.0352,
      "mean_token_accuracy": 0.9927756577730179,
      "num_tokens": 1149317.0,
      "step": 390
    },
    {
      "entropy": 0.04401880670338869,
      "epoch": 1.7543859649122808,
      "grad_norm": 0.23167090117931366,
      "learning_rate": 2e-05,
      "loss": 0.0354,
      "mean_token_accuracy": 0.9926382765173912,
      "num_tokens": 1178783.0,
      "step": 400
    },
    {
      "entropy": 0.04442735444754362,
      "epoch": 1.7982456140350878,
      "grad_norm": 0.2764274477958679,
      "learning_rate": 2e-05,
      "loss": 0.0375,
      "mean_token_accuracy": 0.9925318494439125,
      "num_tokens": 1208255.0,
      "step": 410
    },
    {
      "entropy": 0.046166476979851725,
      "epoch": 1.8421052631578947,
      "grad_norm": 0.3111913800239563,
      "learning_rate": 2e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.992423489689827,
      "num_tokens": 1237708.0,
      "step": 420
    },
    {
      "entropy": 0.044908494455739856,
      "epoch": 1.8859649122807016,
      "grad_norm": 0.23237602412700653,
      "learning_rate": 2e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.9935551881790161,
      "num_tokens": 1267186.0,
      "step": 430
    },
    {
      "entropy": 0.04562048772349954,
      "epoch": 1.9298245614035088,
      "grad_norm": 0.25153592228889465,
      "learning_rate": 2e-05,
      "loss": 0.033,
      "mean_token_accuracy": 0.9941498264670372,
      "num_tokens": 1296591.0,
      "step": 440
    },
    {
      "entropy": 0.044570739334449175,
      "epoch": 1.973684210526316,
      "grad_norm": 0.23802141845226288,
      "learning_rate": 2e-05,
      "loss": 0.033,
      "mean_token_accuracy": 0.9942585557699204,
      "num_tokens": 1326042.0,
      "step": 450
    },
    {
      "entropy": 0.044205385353416206,
      "epoch": 2.017543859649123,
      "grad_norm": 0.3467763066291809,
      "learning_rate": 2e-05,
      "loss": 0.0331,
      "mean_token_accuracy": 0.9940735891461372,
      "num_tokens": 1355561.0,
      "step": 460
    },
    {
      "entropy": 0.042760218819603325,
      "epoch": 2.06140350877193,
      "grad_norm": 0.2680515646934509,
      "learning_rate": 2e-05,
      "loss": 0.0319,
      "mean_token_accuracy": 0.9945705458521843,
      "num_tokens": 1385048.0,
      "step": 470
    },
    {
      "entropy": 0.04117725370451808,
      "epoch": 2.1052631578947367,
      "grad_norm": 0.39523598551750183,
      "learning_rate": 2e-05,
      "loss": 0.033,
      "mean_token_accuracy": 0.9936558306217194,
      "num_tokens": 1414543.0,
      "step": 480
    },
    {
      "entropy": 0.04356150296516716,
      "epoch": 2.1491228070175437,
      "grad_norm": 0.23617857694625854,
      "learning_rate": 2e-05,
      "loss": 0.0332,
      "mean_token_accuracy": 0.9941312283277511,
      "num_tokens": 1444017.0,
      "step": 490
    },
    {
      "entropy": 0.04058904880657792,
      "epoch": 2.192982456140351,
      "grad_norm": 0.3990024924278259,
      "learning_rate": 2e-05,
      "loss": 0.0285,
      "mean_token_accuracy": 0.9948335066437721,
      "num_tokens": 1473442.0,
      "step": 500
    },
    {
      "entropy": 0.04089017482474446,
      "epoch": 2.236842105263158,
      "grad_norm": 0.22417540848255157,
      "learning_rate": 2e-05,
      "loss": 0.0314,
      "mean_token_accuracy": 0.994497561454773,
      "num_tokens": 1502879.0,
      "step": 510
    },
    {
      "entropy": 0.04202216519042849,
      "epoch": 2.280701754385965,
      "grad_norm": 0.20546036958694458,
      "learning_rate": 2e-05,
      "loss": 0.0321,
      "mean_token_accuracy": 0.9938287988305092,
      "num_tokens": 1532392.0,
      "step": 520
    },
    {
      "entropy": 0.04143659081310034,
      "epoch": 2.324561403508772,
      "grad_norm": 0.26604264974594116,
      "learning_rate": 2e-05,
      "loss": 0.0313,
      "mean_token_accuracy": 0.9944339916110039,
      "num_tokens": 1561839.0,
      "step": 530
    },
    {
      "entropy": 0.03936287453398109,
      "epoch": 2.3684210526315788,
      "grad_norm": 0.23989664018154144,
      "learning_rate": 2e-05,
      "loss": 0.029,
      "mean_token_accuracy": 0.9943608403205871,
      "num_tokens": 1591271.0,
      "step": 540
    },
    {
      "entropy": 0.039195985486730936,
      "epoch": 2.412280701754386,
      "grad_norm": 0.3053726255893707,
      "learning_rate": 2e-05,
      "loss": 0.0306,
      "mean_token_accuracy": 0.9943523585796357,
      "num_tokens": 1620794.0,
      "step": 550
    },
    {
      "entropy": 0.039683265471830965,
      "epoch": 2.456140350877193,
      "grad_norm": 0.2588869631290436,
      "learning_rate": 2e-05,
      "loss": 0.0293,
      "mean_token_accuracy": 0.9944723203778267,
      "num_tokens": 1650276.0,
      "step": 560
    },
    {
      "entropy": 0.03602620945312083,
      "epoch": 2.5,
      "grad_norm": 0.2668582797050476,
      "learning_rate": 2e-05,
      "loss": 0.0292,
      "mean_token_accuracy": 0.9945638462901115,
      "num_tokens": 1679717.0,
      "step": 570
    },
    {
      "entropy": 0.03886530273593962,
      "epoch": 2.543859649122807,
      "grad_norm": 0.2594759464263916,
      "learning_rate": 2e-05,
      "loss": 0.0294,
      "mean_token_accuracy": 0.9944349855184555,
      "num_tokens": 1709202.0,
      "step": 580
    },
    {
      "entropy": 0.03563018930144608,
      "epoch": 2.587719298245614,
      "grad_norm": 0.22012071311473846,
      "learning_rate": 2e-05,
      "loss": 0.0272,
      "mean_token_accuracy": 0.9947109371423721,
      "num_tokens": 1738701.0,
      "step": 590
    },
    {
      "entropy": 0.03517519012093544,
      "epoch": 2.6315789473684212,
      "grad_norm": 0.2758197486400604,
      "learning_rate": 2e-05,
      "loss": 0.03,
      "mean_token_accuracy": 0.9942290917038917,
      "num_tokens": 1768151.0,
      "step": 600
    },
    {
      "entropy": 0.03706208867952228,
      "epoch": 2.675438596491228,
      "grad_norm": 0.3245140016078949,
      "learning_rate": 2e-05,
      "loss": 0.0288,
      "mean_token_accuracy": 0.9942933633923531,
      "num_tokens": 1797583.0,
      "step": 610
    },
    {
      "entropy": 0.034479991812258956,
      "epoch": 2.719298245614035,
      "grad_norm": 0.22120347619056702,
      "learning_rate": 2e-05,
      "loss": 0.0282,
      "mean_token_accuracy": 0.9947970882058144,
      "num_tokens": 1827006.0,
      "step": 620
    },
    {
      "entropy": 0.033304579788818955,
      "epoch": 2.763157894736842,
      "grad_norm": 0.26880016922950745,
      "learning_rate": 2e-05,
      "loss": 0.0277,
      "mean_token_accuracy": 0.994759914278984,
      "num_tokens": 1856428.0,
      "step": 630
    },
    {
      "entropy": 0.0331753586884588,
      "epoch": 2.807017543859649,
      "grad_norm": 0.24720199406147003,
      "learning_rate": 2e-05,
      "loss": 0.0278,
      "mean_token_accuracy": 0.9946068048477172,
      "num_tokens": 1885916.0,
      "step": 640
    },
    {
      "entropy": 0.03345069149509072,
      "epoch": 2.8508771929824563,
      "grad_norm": 0.27891653776168823,
      "learning_rate": 2e-05,
      "loss": 0.0275,
      "mean_token_accuracy": 0.9948126003146172,
      "num_tokens": 1915396.0,
      "step": 650
    },
    {
      "entropy": 0.0333542559761554,
      "epoch": 2.8947368421052633,
      "grad_norm": 0.4004022479057312,
      "learning_rate": 2e-05,
      "loss": 0.0289,
      "mean_token_accuracy": 0.9938531696796418,
      "num_tokens": 1945009.0,
      "step": 660
    },
    {
      "entropy": 0.02968177660368383,
      "epoch": 2.93859649122807,
      "grad_norm": 0.2902744710445404,
      "learning_rate": 2e-05,
      "loss": 0.0241,
      "mean_token_accuracy": 0.9945998504757881,
      "num_tokens": 1974462.0,
      "step": 670
    },
    {
      "entropy": 0.0323303550016135,
      "epoch": 2.982456140350877,
      "grad_norm": 0.2170533686876297,
      "learning_rate": 2e-05,
      "loss": 0.0287,
      "mean_token_accuracy": 0.9943123281002044,
      "num_tokens": 2003840.0,
      "step": 680
    },
    {
      "entropy": 0.0318671815097332,
      "epoch": 3.026315789473684,
      "grad_norm": 0.19683605432510376,
      "learning_rate": 2e-05,
      "loss": 0.027,
      "mean_token_accuracy": 0.9946763277053833,
      "num_tokens": 2033324.0,
      "step": 690
    },
    {
      "entropy": 0.03094636939931661,
      "epoch": 3.0701754385964914,
      "grad_norm": 0.3156299889087677,
      "learning_rate": 2e-05,
      "loss": 0.027,
      "mean_token_accuracy": 0.9943998187780381,
      "num_tokens": 2062796.0,
      "step": 700
    },
    {
      "entropy": 0.029998012352734804,
      "epoch": 3.1140350877192984,
      "grad_norm": 0.19611181318759918,
      "learning_rate": 2e-05,
      "loss": 0.0251,
      "mean_token_accuracy": 0.9945186242461205,
      "num_tokens": 2092349.0,
      "step": 710
    },
    {
      "entropy": 0.02844015813898295,
      "epoch": 3.1578947368421053,
      "grad_norm": 0.3108363151550293,
      "learning_rate": 2e-05,
      "loss": 0.026,
      "mean_token_accuracy": 0.9948573753237724,
      "num_tokens": 2121717.0,
      "step": 720
    },
    {
      "entropy": 0.02917947373352945,
      "epoch": 3.2017543859649122,
      "grad_norm": 0.3223120868206024,
      "learning_rate": 2e-05,
      "loss": 0.0259,
      "mean_token_accuracy": 0.9950737491250038,
      "num_tokens": 2151171.0,
      "step": 730
    },
    {
      "entropy": 0.02822321942076087,
      "epoch": 3.245614035087719,
      "grad_norm": 0.1794469654560089,
      "learning_rate": 2e-05,
      "loss": 0.0258,
      "mean_token_accuracy": 0.9945150166749954,
      "num_tokens": 2180705.0,
      "step": 740
    },
    {
      "entropy": 0.026581059489399195,
      "epoch": 3.2894736842105265,
      "grad_norm": 0.2445470094680786,
      "learning_rate": 2e-05,
      "loss": 0.026,
      "mean_token_accuracy": 0.9945855379104614,
      "num_tokens": 2210233.0,
      "step": 750
    },
    {
      "entropy": 0.02701793306041509,
      "epoch": 3.3333333333333335,
      "grad_norm": 0.28041425347328186,
      "learning_rate": 2e-05,
      "loss": 0.0237,
      "mean_token_accuracy": 0.9952751606702804,
      "num_tokens": 2239637.0,
      "step": 760
    },
    {
      "entropy": 0.03019589218311012,
      "epoch": 3.3771929824561404,
      "grad_norm": 0.25402745604515076,
      "learning_rate": 2e-05,
      "loss": 0.0283,
      "mean_token_accuracy": 0.9943036273121834,
      "num_tokens": 2269135.0,
      "step": 770
    },
    {
      "entropy": 0.030228979233652354,
      "epoch": 3.4210526315789473,
      "grad_norm": 0.22020572423934937,
      "learning_rate": 2e-05,
      "loss": 0.0269,
      "mean_token_accuracy": 0.9944290310144425,
      "num_tokens": 2298575.0,
      "step": 780
    },
    {
      "entropy": 0.03184722135774791,
      "epoch": 3.4649122807017543,
      "grad_norm": 0.2342633605003357,
      "learning_rate": 2e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9947415545582772,
      "num_tokens": 2328042.0,
      "step": 790
    },
    {
      "entropy": 0.02975663202814758,
      "epoch": 3.5087719298245617,
      "grad_norm": 0.46808409690856934,
      "learning_rate": 2e-05,
      "loss": 0.0277,
      "mean_token_accuracy": 0.9943336308002472,
      "num_tokens": 2357539.0,
      "step": 800
    },
    {
      "entropy": 0.02755520197097212,
      "epoch": 3.5526315789473686,
      "grad_norm": 0.2859908938407898,
      "learning_rate": 2e-05,
      "loss": 0.0269,
      "mean_token_accuracy": 0.9948696240782737,
      "num_tokens": 2386963.0,
      "step": 810
    },
    {
      "entropy": 0.0279413893353194,
      "epoch": 3.5964912280701755,
      "grad_norm": 0.2786264419555664,
      "learning_rate": 2e-05,
      "loss": 0.0275,
      "mean_token_accuracy": 0.9943128302693367,
      "num_tokens": 2416501.0,
      "step": 820
    },
    {
      "entropy": 0.02768249998334795,
      "epoch": 3.6403508771929824,
      "grad_norm": 0.22936373949050903,
      "learning_rate": 2e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9947321966290474,
      "num_tokens": 2445941.0,
      "step": 830
    },
    {
      "entropy": 0.02423749384470284,
      "epoch": 3.6842105263157894,
      "grad_norm": 0.2773398756980896,
      "learning_rate": 2e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9952654018998146,
      "num_tokens": 2475299.0,
      "step": 840
    },
    {
      "entropy": 0.02699003741145134,
      "epoch": 3.7280701754385968,
      "grad_norm": 0.23085203766822815,
      "learning_rate": 2e-05,
      "loss": 0.0257,
      "mean_token_accuracy": 0.9947722434997559,
      "num_tokens": 2504764.0,
      "step": 850
    },
    {
      "entropy": 0.029456905997358264,
      "epoch": 3.7719298245614032,
      "grad_norm": 0.2776418924331665,
      "learning_rate": 2e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.9948301285505294,
      "num_tokens": 2534185.0,
      "step": 860
    },
    {
      "entropy": 0.026715041836723685,
      "epoch": 3.8157894736842106,
      "grad_norm": 0.40782320499420166,
      "learning_rate": 2e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9943931043148041,
      "num_tokens": 2563632.0,
      "step": 870
    },
    {
      "entropy": 0.026448413264006376,
      "epoch": 3.8596491228070176,
      "grad_norm": 0.20839615166187286,
      "learning_rate": 2e-05,
      "loss": 0.0253,
      "mean_token_accuracy": 0.9949382901191711,
      "num_tokens": 2593079.0,
      "step": 880
    },
    {
      "entropy": 0.026889733923599123,
      "epoch": 3.9035087719298245,
      "grad_norm": 0.4022772014141083,
      "learning_rate": 2e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.9950519934296608,
      "num_tokens": 2622599.0,
      "step": 890
    },
    {
      "entropy": 0.02558579680044204,
      "epoch": 3.9473684210526314,
      "grad_norm": 0.23025710880756378,
      "learning_rate": 2e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.9947320595383644,
      "num_tokens": 2652027.0,
      "step": 900
    },
    {
      "entropy": 0.026929323840886355,
      "epoch": 3.9912280701754383,
      "grad_norm": 0.558501660823822,
      "learning_rate": 2e-05,
      "loss": 0.0263,
      "mean_token_accuracy": 0.9945844635367393,
      "num_tokens": 2681568.0,
      "step": 910
    },
    {
      "entropy": 0.027245206129737198,
      "epoch": 4.035087719298246,
      "grad_norm": 0.34754958748817444,
      "learning_rate": 2e-05,
      "loss": 0.0267,
      "mean_token_accuracy": 0.9946398049592972,
      "num_tokens": 2711021.0,
      "step": 920
    },
    {
      "entropy": 0.028458662680350244,
      "epoch": 4.078947368421052,
      "grad_norm": 0.2995450496673584,
      "learning_rate": 2e-05,
      "loss": 0.027,
      "mean_token_accuracy": 0.994346709549427,
      "num_tokens": 2740564.0,
      "step": 930
    },
    {
      "entropy": 0.026019252184778453,
      "epoch": 4.12280701754386,
      "grad_norm": 0.18306803703308105,
      "learning_rate": 2e-05,
      "loss": 0.0222,
      "mean_token_accuracy": 0.995404352247715,
      "num_tokens": 2769966.0,
      "step": 940
    },
    {
      "entropy": 0.025571403023786844,
      "epoch": 4.166666666666667,
      "grad_norm": 0.24489884078502655,
      "learning_rate": 2e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.99520303606987,
      "num_tokens": 2799356.0,
      "step": 950
    },
    {
      "entropy": 0.02373389925342053,
      "epoch": 4.2105263157894735,
      "grad_norm": 0.4220126271247864,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.994646181166172,
      "num_tokens": 2828878.0,
      "step": 960
    },
    {
      "entropy": 0.02672711748164147,
      "epoch": 4.254385964912281,
      "grad_norm": 0.20107538998126984,
      "learning_rate": 2e-05,
      "loss": 0.0233,
      "mean_token_accuracy": 0.9949999019503594,
      "num_tokens": 2858288.0,
      "step": 970
    },
    {
      "entropy": 0.02580628953874111,
      "epoch": 4.298245614035087,
      "grad_norm": 0.3406866490840912,
      "learning_rate": 2e-05,
      "loss": 0.0254,
      "mean_token_accuracy": 0.9947706565260888,
      "num_tokens": 2887762.0,
      "step": 980
    },
    {
      "entropy": 0.02430701975245029,
      "epoch": 4.342105263157895,
      "grad_norm": 0.28178682923316956,
      "learning_rate": 2e-05,
      "loss": 0.0229,
      "mean_token_accuracy": 0.9950452461838722,
      "num_tokens": 2917219.0,
      "step": 990
    },
    {
      "entropy": 0.026894421339966355,
      "epoch": 4.385964912280702,
      "grad_norm": 0.28810808062553406,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.994709712266922,
      "num_tokens": 2946727.0,
      "step": 1000
    },
    {
      "entropy": 0.026626669918186963,
      "epoch": 4.4298245614035086,
      "grad_norm": 0.2008417248725891,
      "learning_rate": 2e-05,
      "loss": 0.0247,
      "mean_token_accuracy": 0.9952127397060394,
      "num_tokens": 2976166.0,
      "step": 1010
    },
    {
      "entropy": 0.02615027381107211,
      "epoch": 4.473684210526316,
      "grad_norm": 0.3345981240272522,
      "learning_rate": 2e-05,
      "loss": 0.0229,
      "mean_token_accuracy": 0.9947855144739151,
      "num_tokens": 3005701.0,
      "step": 1020
    },
    {
      "entropy": 0.02315097493119538,
      "epoch": 4.517543859649123,
      "grad_norm": 0.2304411679506302,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9950236231088638,
      "num_tokens": 3035228.0,
      "step": 1030
    },
    {
      "entropy": 0.024475370533764362,
      "epoch": 4.56140350877193,
      "grad_norm": 0.2187710851430893,
      "learning_rate": 2e-05,
      "loss": 0.0237,
      "mean_token_accuracy": 0.994983771443367,
      "num_tokens": 3064721.0,
      "step": 1040
    },
    {
      "entropy": 0.024467523396015167,
      "epoch": 4.605263157894737,
      "grad_norm": 0.2989374101161957,
      "learning_rate": 2e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.994673240184784,
      "num_tokens": 3094215.0,
      "step": 1050
    },
    {
      "entropy": 0.02632226881105453,
      "epoch": 4.649122807017544,
      "grad_norm": 0.33434978127479553,
      "learning_rate": 2e-05,
      "loss": 0.0234,
      "mean_token_accuracy": 0.9951106742024421,
      "num_tokens": 3123674.0,
      "step": 1060
    },
    {
      "entropy": 0.025994484080001712,
      "epoch": 4.692982456140351,
      "grad_norm": 0.3363839089870453,
      "learning_rate": 2e-05,
      "loss": 0.0238,
      "mean_token_accuracy": 0.9950530841946602,
      "num_tokens": 3153200.0,
      "step": 1070
    },
    {
      "entropy": 0.024378301412798464,
      "epoch": 4.7368421052631575,
      "grad_norm": 0.27308163046836853,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.9951689943671227,
      "num_tokens": 3182589.0,
      "step": 1080
    },
    {
      "entropy": 0.02580142463557422,
      "epoch": 4.780701754385965,
      "grad_norm": 0.3396266996860504,
      "learning_rate": 2e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9950757563114166,
      "num_tokens": 3212040.0,
      "step": 1090
    },
    {
      "entropy": 0.024319417704828084,
      "epoch": 4.824561403508772,
      "grad_norm": 0.23770387470722198,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.9949398577213288,
      "num_tokens": 3241483.0,
      "step": 1100
    },
    {
      "entropy": 0.024646017188206314,
      "epoch": 4.868421052631579,
      "grad_norm": 0.18578499555587769,
      "learning_rate": 2e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.995578208565712,
      "num_tokens": 3270879.0,
      "step": 1110
    },
    {
      "entropy": 0.025846811849623917,
      "epoch": 4.912280701754386,
      "grad_norm": 0.2943616509437561,
      "learning_rate": 2e-05,
      "loss": 0.0262,
      "mean_token_accuracy": 0.9946421980857849,
      "num_tokens": 3300390.0,
      "step": 1120
    },
    {
      "entropy": 0.02559350139927119,
      "epoch": 4.956140350877193,
      "grad_norm": 0.3006184995174408,
      "learning_rate": 2e-05,
      "loss": 0.0246,
      "mean_token_accuracy": 0.9950146496295929,
      "num_tokens": 3329878.0,
      "step": 1130
    },
    {
      "entropy": 0.027681316062808037,
      "epoch": 5.0,
      "grad_norm": 0.2197951078414917,
      "learning_rate": 2e-05,
      "loss": 0.0255,
      "mean_token_accuracy": 0.9948058515787125,
      "num_tokens": 3359340.0,
      "step": 1140
    },
    {
      "entropy": 0.025585089693777263,
      "epoch": 5.043859649122807,
      "grad_norm": 0.19763857126235962,
      "learning_rate": 2e-05,
      "loss": 0.0213,
      "mean_token_accuracy": 0.9954831823706627,
      "num_tokens": 3388801.0,
      "step": 1150
    },
    {
      "entropy": 0.02210813742130995,
      "epoch": 5.087719298245614,
      "grad_norm": 0.2954810857772827,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9953773066401481,
      "num_tokens": 3418214.0,
      "step": 1160
    },
    {
      "entropy": 0.026301260688342154,
      "epoch": 5.131578947368421,
      "grad_norm": 0.30119267106056213,
      "learning_rate": 2e-05,
      "loss": 0.0227,
      "mean_token_accuracy": 0.9950450032949447,
      "num_tokens": 3447671.0,
      "step": 1170
    },
    {
      "entropy": 0.0237973083043471,
      "epoch": 5.175438596491228,
      "grad_norm": 0.23260529339313507,
      "learning_rate": 2e-05,
      "loss": 0.0206,
      "mean_token_accuracy": 0.9954913884401322,
      "num_tokens": 3477175.0,
      "step": 1180
    },
    {
      "entropy": 0.022307809442281723,
      "epoch": 5.219298245614035,
      "grad_norm": 0.17688173055648804,
      "learning_rate": 2e-05,
      "loss": 0.0239,
      "mean_token_accuracy": 0.9948815569281578,
      "num_tokens": 3506685.0,
      "step": 1190
    },
    {
      "entropy": 0.022513316967524587,
      "epoch": 5.2631578947368425,
      "grad_norm": 0.3002011775970459,
      "learning_rate": 2e-05,
      "loss": 0.0218,
      "mean_token_accuracy": 0.9951437518000603,
      "num_tokens": 3536154.0,
      "step": 1200
    },
    {
      "entropy": 0.023108458751812576,
      "epoch": 5.307017543859649,
      "grad_norm": 0.30037811398506165,
      "learning_rate": 2e-05,
      "loss": 0.0219,
      "mean_token_accuracy": 0.9951157048344612,
      "num_tokens": 3565636.0,
      "step": 1210
    },
    {
      "entropy": 0.02366964400280267,
      "epoch": 5.350877192982456,
      "grad_norm": 0.2503150701522827,
      "learning_rate": 2e-05,
      "loss": 0.0228,
      "mean_token_accuracy": 0.9948362991213798,
      "num_tokens": 3595079.0,
      "step": 1220
    },
    {
      "entropy": 0.0236843794118613,
      "epoch": 5.394736842105263,
      "grad_norm": 0.3021239638328552,
      "learning_rate": 2e-05,
      "loss": 0.0229,
      "mean_token_accuracy": 0.9950185254216194,
      "num_tokens": 3624589.0,
      "step": 1230
    },
    {
      "entropy": 0.024407316814176738,
      "epoch": 5.43859649122807,
      "grad_norm": 0.2514321506023407,
      "learning_rate": 2e-05,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9953509420156479,
      "num_tokens": 3654087.0,
      "step": 1240
    },
    {
      "entropy": 0.023451847536489368,
      "epoch": 5.482456140350878,
      "grad_norm": 0.3244996964931488,
      "learning_rate": 2e-05,
      "loss": 0.0221,
      "mean_token_accuracy": 0.9955065801739693,
      "num_tokens": 3683485.0,
      "step": 1250
    },
    {
      "entropy": 0.02330639832653105,
      "epoch": 5.526315789473684,
      "grad_norm": 0.2821803689002991,
      "learning_rate": 2e-05,
      "loss": 0.0224,
      "mean_token_accuracy": 0.9953455001115799,
      "num_tokens": 3712932.0,
      "step": 1260
    },
    {
      "entropy": 0.023308307630941272,
      "epoch": 5.5701754385964914,
      "grad_norm": 0.31802570819854736,
      "learning_rate": 2e-05,
      "loss": 0.0231,
      "mean_token_accuracy": 0.9949766904115677,
      "num_tokens": 3742399.0,
      "step": 1270
    },
    {
      "entropy": 0.023460485064424573,
      "epoch": 5.614035087719298,
      "grad_norm": 0.20157690346240997,
      "learning_rate": 2e-05,
      "loss": 0.0236,
      "mean_token_accuracy": 0.9949595555663109,
      "num_tokens": 3771924.0,
      "step": 1280
    },
    {
      "entropy": 0.026444529974833132,
      "epoch": 5.657894736842105,
      "grad_norm": 0.3347353935241699,
      "learning_rate": 2e-05,
      "loss": 0.023,
      "mean_token_accuracy": 0.9946390092372894,
      "num_tokens": 3801411.0,
      "step": 1290
    },
    {
      "entropy": 0.024692195327952506,
      "epoch": 5.701754385964913,
      "grad_norm": 0.23435048758983612,
      "learning_rate": 2e-05,
      "loss": 0.0242,
      "mean_token_accuracy": 0.9948781177401542,
      "num_tokens": 3830901.0,
      "step": 1300
    },
    {
      "entropy": 0.02177093622740358,
      "epoch": 5.745614035087719,
      "grad_norm": 0.25117844343185425,
      "learning_rate": 2e-05,
      "loss": 0.0225,
      "mean_token_accuracy": 0.9955831781029701,
      "num_tokens": 3860316.0,
      "step": 1310
    },
    {
      "entropy": 0.024748310795985162,
      "epoch": 5.7894736842105265,
      "grad_norm": 0.2464597225189209,
      "learning_rate": 2e-05,
      "loss": 0.024,
      "mean_token_accuracy": 0.9947022467851638,
      "num_tokens": 3889778.0,
      "step": 1320
    },
    {
      "entropy": 0.02363416156731546,
      "epoch": 5.833333333333333,
      "grad_norm": 0.1910584568977356,
      "learning_rate": 2e-05,
      "loss": 0.022,
      "mean_token_accuracy": 0.9954767659306526,
      "num_tokens": 3919184.0,
      "step": 1330
    },
    {
      "entropy": 0.024428941844962537,
      "epoch": 5.87719298245614,
      "grad_norm": 0.24225406348705292,
      "learning_rate": 2e-05,
      "loss": 0.0216,
      "mean_token_accuracy": 0.9951212272047997,
      "num_tokens": 3948694.0,
      "step": 1340
    },
    {
      "entropy": 0.023069229116663335,
      "epoch": 5.921052631578947,
      "grad_norm": 0.2355465441942215,
      "learning_rate": 2e-05,
      "loss": 0.0215,
      "mean_token_accuracy": 0.9953619986772537,
      "num_tokens": 3978227.0,
      "step": 1350
    },
    {
      "entropy": 0.02356387039180845,
      "epoch": 5.964912280701754,
      "grad_norm": 0.3358451724052429,
      "learning_rate": 2e-05,
      "loss": 0.0234,
      "mean_token_accuracy": 0.9953474655747414,
      "num_tokens": 4007681.0,
      "step": 1360
    }
  ],
  "logging_steps": 10,
  "max_steps": 1824,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2518449017001984.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}