{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9972826086956522,
  "eval_steps": 500,
  "global_step": 367,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01358695652173913,
      "grad_norm": 15.789278984069824,
      "learning_rate": 2.162496240601504e-06,
      "loss": 0.3626,
      "mean_token_accuracy": 0.8835940927267074,
      "num_tokens": 735368.0,
      "step": 5
    },
    {
      "epoch": 0.02717391304347826,
      "grad_norm": 2.6020426750183105,
      "learning_rate": 4.865616541353383e-06,
      "loss": 0.1891,
      "mean_token_accuracy": 0.9505777508020401,
      "num_tokens": 1456323.0,
      "step": 10
    },
    {
      "epoch": 0.04076086956521739,
      "grad_norm": 1.255000352859497,
      "learning_rate": 7.568736842105265e-06,
      "loss": 0.1478,
      "mean_token_accuracy": 0.9613735318183899,
      "num_tokens": 2191201.0,
      "step": 15
    },
    {
      "epoch": 0.05434782608695652,
      "grad_norm": 0.9647696018218994,
      "learning_rate": 1.0271857142857143e-05,
      "loss": 0.1274,
      "mean_token_accuracy": 0.9659419178962707,
      "num_tokens": 2917478.0,
      "step": 20
    },
    {
      "epoch": 0.06793478260869565,
      "grad_norm": 0.6830999255180359,
      "learning_rate": 1.2974977443609024e-05,
      "loss": 0.1047,
      "mean_token_accuracy": 0.9702055871486663,
      "num_tokens": 3653538.0,
      "step": 25
    },
    {
      "epoch": 0.08152173913043478,
      "grad_norm": 1.0783493518829346,
      "learning_rate": 1.5678097744360906e-05,
      "loss": 0.1217,
      "mean_token_accuracy": 0.9679086863994598,
      "num_tokens": 4383120.0,
      "step": 30
    },
    {
      "epoch": 0.09510869565217392,
      "grad_norm": 0.8872610330581665,
      "learning_rate": 1.8381218045112784e-05,
      "loss": 0.1053,
      "mean_token_accuracy": 0.9713874399662018,
      "num_tokens": 5120466.0,
      "step": 35
    },
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 0.826252281665802,
      "learning_rate": 1.89167598509104e-05,
      "loss": 0.0936,
      "mean_token_accuracy": 0.972337806224823,
      "num_tokens": 5858133.0,
      "step": 40
    },
    {
      "epoch": 0.12228260869565218,
      "grad_norm": 0.8232932090759277,
      "learning_rate": 1.889612566940291e-05,
      "loss": 0.1178,
      "mean_token_accuracy": 0.9680493772029877,
      "num_tokens": 6587547.0,
      "step": 45
    },
    {
      "epoch": 0.1358695652173913,
      "grad_norm": 0.8012369275093079,
      "learning_rate": 1.885966806597815e-05,
      "loss": 0.1217,
      "mean_token_accuracy": 0.9657657444477081,
      "num_tokens": 7314668.0,
      "step": 50
    },
    {
      "epoch": 0.14945652173913043,
      "grad_norm": 0.696868360042572,
      "learning_rate": 1.8807468636977533e-05,
      "loss": 0.1081,
      "mean_token_accuracy": 0.9682049512863159,
      "num_tokens": 8027588.0,
      "step": 55
    },
    {
      "epoch": 0.16304347826086957,
      "grad_norm": 0.6226001381874084,
      "learning_rate": 1.8739644210772358e-05,
      "loss": 0.1009,
      "mean_token_accuracy": 0.9705910980701447,
      "num_tokens": 8753739.0,
      "step": 60
    },
    {
      "epoch": 0.1766304347826087,
      "grad_norm": 0.8285146355628967,
      "learning_rate": 1.8656346586288405e-05,
      "loss": 0.1104,
      "mean_token_accuracy": 0.9673073709011077,
      "num_tokens": 9482783.0,
      "step": 65
    },
    {
      "epoch": 0.19021739130434784,
      "grad_norm": 0.7593039274215698,
      "learning_rate": 1.8557762193262327e-05,
      "loss": 0.1054,
      "mean_token_accuracy": 0.9698054105043411,
      "num_tokens": 10207240.0,
      "step": 70
    },
    {
      "epoch": 0.20380434782608695,
      "grad_norm": 0.7482079863548279,
      "learning_rate": 1.844411167499031e-05,
      "loss": 0.1028,
      "mean_token_accuracy": 0.9713084936141968,
      "num_tokens": 10941919.0,
      "step": 75
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 0.6597537398338318,
      "learning_rate": 1.8315649394502792e-05,
      "loss": 0.0791,
      "mean_token_accuracy": 0.9772562921047211,
      "num_tokens": 11674200.0,
      "step": 80
    },
    {
      "epoch": 0.23097826086956522,
      "grad_norm": 0.7344685196876526,
      "learning_rate": 1.817266286527054e-05,
      "loss": 0.0973,
      "mean_token_accuracy": 0.9732755422592163,
      "num_tokens": 12397980.0,
      "step": 85
    },
    {
      "epoch": 0.24456521739130435,
      "grad_norm": 0.7957409024238586,
      "learning_rate": 1.801547210771617e-05,
      "loss": 0.1123,
      "mean_token_accuracy": 0.9699113100767136,
      "num_tokens": 13127159.0,
      "step": 90
    },
    {
      "epoch": 0.25815217391304346,
      "grad_norm": 0.7433070540428162,
      "learning_rate": 1.7844428932971356e-05,
      "loss": 0.0955,
      "mean_token_accuracy": 0.9731246918439865,
      "num_tokens": 13856323.0,
      "step": 95
    },
    {
      "epoch": 0.2717391304347826,
      "grad_norm": 0.6211023926734924,
      "learning_rate": 1.7659916155482757e-05,
      "loss": 0.1159,
      "mean_token_accuracy": 0.9678784847259522,
      "num_tokens": 14586264.0,
      "step": 100
    },
    {
      "epoch": 0.28532608695652173,
      "grad_norm": 0.6008529663085938,
      "learning_rate": 1.7462346736228915e-05,
      "loss": 0.0863,
      "mean_token_accuracy": 0.9759507954120636,
      "num_tokens": 15318134.0,
      "step": 105
    },
    {
      "epoch": 0.29891304347826086,
      "grad_norm": 0.6403423547744751,
      "learning_rate": 1.725216285846573e-05,
      "loss": 0.0736,
      "mean_token_accuracy": 0.977612829208374,
      "num_tokens": 16042121.0,
      "step": 110
    },
    {
      "epoch": 0.3125,
      "grad_norm": 0.6671366095542908,
      "learning_rate": 1.7029834938069084e-05,
      "loss": 0.1122,
      "mean_token_accuracy": 0.9690867394208909,
      "num_tokens": 16768498.0,
      "step": 115
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 0.5051082372665405,
      "learning_rate": 1.6795860570689614e-05,
      "loss": 0.0854,
      "mean_token_accuracy": 0.9759419798851013,
      "num_tokens": 17502754.0,
      "step": 120
    },
    {
      "epoch": 0.33967391304347827,
      "grad_norm": 0.7031633853912354,
      "learning_rate": 1.6550763418075957e-05,
      "loss": 0.0994,
      "mean_token_accuracy": 0.9720090597867965,
      "num_tokens": 18241961.0,
      "step": 125
    },
    {
      "epoch": 0.3532608695652174,
      "grad_norm": 0.4745713174343109,
      "learning_rate": 1.629509203605909e-05,
      "loss": 0.0997,
      "mean_token_accuracy": 0.9720667064189911,
      "num_tokens": 18974222.0,
      "step": 130
    },
    {
      "epoch": 0.36684782608695654,
      "grad_norm": 0.8432019352912903,
      "learning_rate": 1.6029418646820808e-05,
      "loss": 0.1093,
      "mean_token_accuracy": 0.9717468440532684,
      "num_tokens": 19720582.0,
      "step": 135
    },
    {
      "epoch": 0.3804347826086957,
      "grad_norm": 0.6240178942680359,
      "learning_rate": 1.5754337858194152e-05,
      "loss": 0.0886,
      "mean_token_accuracy": 0.9752022594213485,
      "num_tokens": 20445973.0,
      "step": 140
    },
    {
      "epoch": 0.39402173913043476,
      "grad_norm": 0.6496020555496216,
      "learning_rate": 1.5470465332862198e-05,
      "loss": 0.0901,
      "mean_token_accuracy": 0.9729790478944779,
      "num_tokens": 21172879.0,
      "step": 145
    },
    {
      "epoch": 0.4076086956521739,
      "grad_norm": 0.6652824878692627,
      "learning_rate": 1.5178436410433598e-05,
      "loss": 0.1073,
      "mean_token_accuracy": 0.9695830166339874,
      "num_tokens": 21903842.0,
      "step": 150
    },
    {
      "epoch": 0.421195652173913,
      "grad_norm": 0.5234840512275696,
      "learning_rate": 1.4878904685478931e-05,
      "loss": 0.0787,
      "mean_token_accuracy": 0.9746903419494629,
      "num_tokens": 22629291.0,
      "step": 155
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.580998957157135,
      "learning_rate": 1.4572540544710303e-05,
      "loss": 0.091,
      "mean_token_accuracy": 0.9738609373569489,
      "num_tokens": 23365316.0,
      "step": 160
    },
    {
      "epoch": 0.4483695652173913,
      "grad_norm": 0.6344147324562073,
      "learning_rate": 1.426002966657818e-05,
      "loss": 0.0891,
      "mean_token_accuracy": 0.9734716594219208,
      "num_tokens": 24088627.0,
      "step": 165
    },
    {
      "epoch": 0.46195652173913043,
      "grad_norm": 0.5536977648735046,
      "learning_rate": 1.3942071486643592e-05,
      "loss": 0.0941,
      "mean_token_accuracy": 0.9735709875822067,
      "num_tokens": 24806992.0,
      "step": 170
    },
    {
      "epoch": 0.47554347826086957,
      "grad_norm": 0.5197504758834839,
      "learning_rate": 1.3619377632160263e-05,
      "loss": 0.0829,
      "mean_token_accuracy": 0.9760784417390823,
      "num_tokens": 25533051.0,
      "step": 175
    },
    {
      "epoch": 0.4891304347826087,
      "grad_norm": 0.6999825239181519,
      "learning_rate": 1.3292670329370347e-05,
      "loss": 0.1096,
      "mean_token_accuracy": 0.9706821143627167,
      "num_tokens": 26272934.0,
      "step": 180
    },
    {
      "epoch": 0.5027173913043478,
      "grad_norm": 0.46688413619995117,
      "learning_rate": 1.2962680787078433e-05,
      "loss": 0.1082,
      "mean_token_accuracy": 0.9704168438911438,
      "num_tokens": 27012774.0,
      "step": 185
    },
    {
      "epoch": 0.5163043478260869,
      "grad_norm": 0.5378188490867615,
      "learning_rate": 1.2630147560121461e-05,
      "loss": 0.1005,
      "mean_token_accuracy": 0.9725490838289261,
      "num_tokens": 27751723.0,
      "step": 190
    },
    {
      "epoch": 0.529891304347826,
      "grad_norm": 0.5852940678596497,
      "learning_rate": 1.2295814896397441e-05,
      "loss": 0.1163,
      "mean_token_accuracy": 0.96868194937706,
      "num_tokens": 28489723.0,
      "step": 195
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 0.5863273739814758,
      "learning_rate": 1.1960431071152402e-05,
      "loss": 0.0812,
      "mean_token_accuracy": 0.975890439748764,
      "num_tokens": 29215689.0,
      "step": 200
    },
    {
      "epoch": 0.5570652173913043,
      "grad_norm": 0.5419257879257202,
      "learning_rate": 1.1624746712253664e-05,
      "loss": 0.0814,
      "mean_token_accuracy": 0.9752096831798553,
      "num_tokens": 29944570.0,
      "step": 205
    },
    {
      "epoch": 0.5706521739130435,
      "grad_norm": 0.7371279001235962,
      "learning_rate": 1.1289513120197692e-05,
      "loss": 0.0827,
      "mean_token_accuracy": 0.9760116189718246,
      "num_tokens": 30667989.0,
      "step": 210
    },
    {
      "epoch": 0.5842391304347826,
      "grad_norm": 0.3487854599952698,
      "learning_rate": 1.0955480586612521e-05,
      "loss": 0.0865,
      "mean_token_accuracy": 0.9751809239387512,
      "num_tokens": 31390880.0,
      "step": 215
    },
    {
      "epoch": 0.5978260869565217,
      "grad_norm": 0.9608821272850037,
      "learning_rate": 1.0623396715018148e-05,
      "loss": 0.091,
      "mean_token_accuracy": 0.9746360749006271,
      "num_tokens": 32122409.0,
      "step": 220
    },
    {
      "epoch": 0.6114130434782609,
      "grad_norm": 0.46955442428588867,
      "learning_rate": 1.0294004747603201e-05,
      "loss": 0.0879,
      "mean_token_accuracy": 0.9754130244255066,
      "num_tokens": 32845734.0,
      "step": 225
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.5701378583908081,
      "learning_rate": 9.968041901762808e-06,
      "loss": 0.0805,
      "mean_token_accuracy": 0.976205337047577,
      "num_tokens": 33571950.0,
      "step": 230
    },
    {
      "epoch": 0.6385869565217391,
      "grad_norm": 0.5272037982940674,
      "learning_rate": 9.64623772012062e-06,
      "loss": 0.0814,
      "mean_token_accuracy": 0.9782846182584762,
      "num_tokens": 34311831.0,
      "step": 235
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.6672101020812988,
      "learning_rate": 9.329312437727877e-06,
      "loss": 0.0844,
      "mean_token_accuracy": 0.9769539326429367,
      "num_tokens": 35036674.0,
      "step": 240
    },
    {
      "epoch": 0.6657608695652174,
      "grad_norm": 0.47675710916519165,
      "learning_rate": 9.01797537009391e-06,
      "loss": 0.0806,
      "mean_token_accuracy": 0.9768453866243363,
      "num_tokens": 35760656.0,
      "step": 245
    },
    {
      "epoch": 0.6793478260869565,
      "grad_norm": 0.4621396064758301,
      "learning_rate": 8.712923325655807e-06,
      "loss": 0.0719,
      "mean_token_accuracy": 0.9790104866027832,
      "num_tokens": 36493978.0,
      "step": 250
    },
    {
      "epoch": 0.6929347826086957,
      "grad_norm": 0.7226515412330627,
      "learning_rate": 8.414839046240412e-06,
      "loss": 0.0771,
      "mean_token_accuracy": 0.9767259240150452,
      "num_tokens": 37231093.0,
      "step": 255
    },
    {
      "epoch": 0.7065217391304348,
      "grad_norm": 0.5121055841445923,
      "learning_rate": 8.124389679008971e-06,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9812835395336151,
      "num_tokens": 37968432.0,
      "step": 260
    },
    {
      "epoch": 0.720108695652174,
      "grad_norm": 0.5179183483123779,
      "learning_rate": 7.842225283304477e-06,
      "loss": 0.0862,
      "mean_token_accuracy": 0.9736676096916199,
      "num_tokens": 38695589.0,
      "step": 265
    },
    {
      "epoch": 0.7336956521739131,
      "grad_norm": 0.6008281707763672,
      "learning_rate": 7.568977375743532e-06,
      "loss": 0.0732,
      "mean_token_accuracy": 0.9792166531085968,
      "num_tokens": 39426153.0,
      "step": 270
    },
    {
      "epoch": 0.7472826086956522,
      "grad_norm": 0.6731653809547424,
      "learning_rate": 7.305257516808976e-06,
      "loss": 0.0925,
      "mean_token_accuracy": 0.9750073671340942,
      "num_tokens": 40162636.0,
      "step": 275
    },
    {
      "epoch": 0.7608695652173914,
      "grad_norm": 0.45099732279777527,
      "learning_rate": 7.051655942106647e-06,
      "loss": 0.0901,
      "mean_token_accuracy": 0.9750892341136932,
      "num_tokens": 40894335.0,
      "step": 280
    },
    {
      "epoch": 0.7744565217391305,
      "grad_norm": 0.4554596245288849,
      "learning_rate": 6.8087402413497055e-06,
      "loss": 0.0973,
      "mean_token_accuracy": 0.9723783254623413,
      "num_tokens": 41621345.0,
      "step": 285
    },
    {
      "epoch": 0.7880434782608695,
      "grad_norm": 0.43782398104667664,
      "learning_rate": 6.5770540880270724e-06,
      "loss": 0.0594,
      "mean_token_accuracy": 0.9813363730907441,
      "num_tokens": 42358687.0,
      "step": 290
    },
    {
      "epoch": 0.8016304347826086,
      "grad_norm": 0.6034961938858032,
      "learning_rate": 6.3571160225991525e-06,
      "loss": 0.0822,
      "mean_token_accuracy": 0.9760138303041458,
      "num_tokens": 43083234.0,
      "step": 295
    },
    {
      "epoch": 0.8152173913043478,
      "grad_norm": 0.49142637848854065,
      "learning_rate": 6.1494182919442125e-06,
      "loss": 0.0704,
      "mean_token_accuracy": 0.9791740953922272,
      "num_tokens": 43815317.0,
      "step": 300
    },
    {
      "epoch": 0.8288043478260869,
      "grad_norm": 0.46567845344543457,
      "learning_rate": 5.954425747652794e-06,
      "loss": 0.0756,
      "mean_token_accuracy": 0.9773548871278763,
      "num_tokens": 44545099.0,
      "step": 305
    },
    {
      "epoch": 0.842391304347826,
      "grad_norm": 0.5463461875915527,
      "learning_rate": 5.7725748056360095e-06,
      "loss": 0.0732,
      "mean_token_accuracy": 0.9785752356052398,
      "num_tokens": 45265607.0,
      "step": 310
    },
    {
      "epoch": 0.8559782608695652,
      "grad_norm": 0.3473645746707916,
      "learning_rate": 5.604272469376143e-06,
      "loss": 0.0939,
      "mean_token_accuracy": 0.9735148876905442,
      "num_tokens": 45991781.0,
      "step": 315
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.5858165621757507,
      "learning_rate": 5.449895419005693e-06,
      "loss": 0.0913,
      "mean_token_accuracy": 0.9739406198263169,
      "num_tokens": 46720519.0,
      "step": 320
    },
    {
      "epoch": 0.8831521739130435,
      "grad_norm": 0.4453265070915222,
      "learning_rate": 5.309789168253591e-06,
      "loss": 0.0809,
      "mean_token_accuracy": 0.9762261033058166,
      "num_tokens": 47454889.0,
      "step": 325
    },
    {
      "epoch": 0.8967391304347826,
      "grad_norm": 0.5313122868537903,
      "learning_rate": 5.184267291145429e-06,
      "loss": 0.0803,
      "mean_token_accuracy": 0.9782892763614655,
      "num_tokens": 48182843.0,
      "step": 330
    },
    {
      "epoch": 0.9103260869565217,
      "grad_norm": 0.48112359642982483,
      "learning_rate": 5.0736107201884554e-06,
      "loss": 0.0913,
      "mean_token_accuracy": 0.9753382563591003,
      "num_tokens": 48919091.0,
      "step": 335
    },
    {
      "epoch": 0.9239130434782609,
      "grad_norm": 0.42446476221084595,
      "learning_rate": 4.978067117612063e-06,
      "loss": 0.0822,
      "mean_token_accuracy": 0.9762956827878952,
      "num_tokens": 49643351.0,
      "step": 340
    },
    {
      "epoch": 0.9375,
      "grad_norm": 0.32655906677246094,
      "learning_rate": 4.897850321071021e-06,
      "loss": 0.0635,
      "mean_token_accuracy": 0.9822634041309357,
      "num_tokens": 50367550.0,
      "step": 345
    },
    {
      "epoch": 0.9510869565217391,
      "grad_norm": 0.5778966546058655,
      "learning_rate": 4.833139865052024e-06,
      "loss": 0.0739,
      "mean_token_accuracy": 0.979621359705925,
      "num_tokens": 51097628.0,
      "step": 350
    },
    {
      "epoch": 0.9646739130434783,
      "grad_norm": 0.5807848572731018,
      "learning_rate": 4.784080579054707e-06,
      "loss": 0.0769,
      "mean_token_accuracy": 0.9772297650575638,
      "num_tokens": 51833775.0,
      "step": 355
    },
    {
      "epoch": 0.9782608695652174,
      "grad_norm": 0.49904078245162964,
      "learning_rate": 4.750782263446433e-06,
      "loss": 0.0841,
      "mean_token_accuracy": 0.9770821154117584,
      "num_tokens": 52570197.0,
      "step": 360
    },
    {
      "epoch": 0.9918478260869565,
      "grad_norm": 0.6134728193283081,
      "learning_rate": 4.7333194437163565e-06,
      "loss": 0.0938,
      "mean_token_accuracy": 0.9744979768991471,
      "num_tokens": 53304068.0,
      "step": 365
    }
  ],
  "logging_steps": 5,
  "max_steps": 367,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.6547103020726354e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}