{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6314741035856574,
  "eval_steps": 500,
  "global_step": 317,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0099601593625498,
      "grad_norm": 9.232208251953125,
      "learning_rate": 5.493846153846154e-06,
      "loss": 1.8789,
      "mean_token_accuracy": 0.6641474723815918,
      "num_tokens": 203562.0,
      "step": 5
    },
    {
      "epoch": 0.0199203187250996,
      "grad_norm": 2.1925406455993652,
      "learning_rate": 1.2361153846153847e-05,
      "loss": 1.4688,
      "mean_token_accuracy": 0.6874574899673462,
      "num_tokens": 407351.0,
      "step": 10
    },
    {
      "epoch": 0.029880478087649404,
      "grad_norm": 1.5760976076126099,
      "learning_rate": 1.922846153846154e-05,
      "loss": 1.0239,
      "mean_token_accuracy": 0.7515892744064331,
      "num_tokens": 611227.0,
      "step": 15
    },
    {
      "epoch": 0.0398406374501992,
      "grad_norm": 1.1244546175003052,
      "learning_rate": 2.609576923076923e-05,
      "loss": 0.5959,
      "mean_token_accuracy": 0.8532620668411255,
      "num_tokens": 815823.0,
      "step": 20
    },
    {
      "epoch": 0.049800796812749,
      "grad_norm": 1.8382799625396729,
      "learning_rate": 3.2963076923076926e-05,
      "loss": 0.2636,
      "mean_token_accuracy": 0.9294551372528076,
      "num_tokens": 1019539.0,
      "step": 25
    },
    {
      "epoch": 0.05976095617529881,
      "grad_norm": 0.8567749857902527,
      "learning_rate": 3.570737514520681e-05,
      "loss": 0.1274,
      "mean_token_accuracy": 0.9625859975814819,
      "num_tokens": 1223909.0,
      "step": 30
    },
    {
      "epoch": 0.0697211155378486,
      "grad_norm": 0.7172916531562805,
      "learning_rate": 3.569133809221824e-05,
      "loss": 0.1025,
      "mean_token_accuracy": 0.9676972508430481,
      "num_tokens": 1427958.0,
      "step": 35
    },
    {
      "epoch": 0.0796812749003984,
      "grad_norm": 0.21782852709293365,
      "learning_rate": 3.5660739680414176e-05,
      "loss": 0.0859,
      "mean_token_accuracy": 0.9706499457359314,
      "num_tokens": 1632148.0,
      "step": 40
    },
    {
      "epoch": 0.08964143426294821,
      "grad_norm": 0.18133988976478577,
      "learning_rate": 3.561561322828542e-05,
      "loss": 0.0836,
      "mean_token_accuracy": 0.9714468479156494,
      "num_tokens": 1836039.0,
      "step": 45
    },
    {
      "epoch": 0.099601593625498,
      "grad_norm": 0.16465476155281067,
      "learning_rate": 3.555600787384902e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.977270531654358,
      "num_tokens": 2039906.0,
      "step": 50
    },
    {
      "epoch": 0.10956175298804781,
      "grad_norm": 0.15224771201610565,
      "learning_rate": 3.548198852114205e-05,
      "loss": 0.0589,
      "mean_token_accuracy": 0.9786348581314087,
      "num_tokens": 2244311.0,
      "step": 55
    },
    {
      "epoch": 0.11952191235059761,
      "grad_norm": 0.20119178295135498,
      "learning_rate": 3.539363576954785e-05,
      "loss": 0.0616,
      "mean_token_accuracy": 0.9778944253921509,
      "num_tokens": 2448676.0,
      "step": 60
    },
    {
      "epoch": 0.1294820717131474,
      "grad_norm": 0.5322551727294922,
      "learning_rate": 3.529104582603171e-05,
      "loss": 0.0509,
      "mean_token_accuracy": 0.98151695728302,
      "num_tokens": 2653236.0,
      "step": 65
    },
    {
      "epoch": 0.1394422310756972,
      "grad_norm": 0.20895026624202728,
      "learning_rate": 3.517433040038145e-05,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9769706010818482,
      "num_tokens": 2857368.0,
      "step": 70
    },
    {
      "epoch": 0.14940239043824702,
      "grad_norm": 0.14045538008213043,
      "learning_rate": 3.504361658356707e-05,
      "loss": 0.0595,
      "mean_token_accuracy": 0.9787776470184326,
      "num_tokens": 3061303.0,
      "step": 75
    },
    {
      "epoch": 0.1593625498007968,
      "grad_norm": 0.13880300521850586,
      "learning_rate": 3.489904670935192e-05,
      "loss": 0.0524,
      "mean_token_accuracy": 0.9811155080795289,
      "num_tokens": 3265653.0,
      "step": 80
    },
    {
      "epoch": 0.1693227091633466,
      "grad_norm": 0.14233286678791046,
      "learning_rate": 3.4740778199306e-05,
      "loss": 0.0601,
      "mean_token_accuracy": 0.9784350752830505,
      "num_tokens": 3469613.0,
      "step": 85
    },
    {
      "epoch": 0.17928286852589642,
      "grad_norm": 0.13804133236408234,
      "learning_rate": 3.4568983391390215e-05,
      "loss": 0.0509,
      "mean_token_accuracy": 0.9816131353378296,
      "num_tokens": 3673768.0,
      "step": 90
    },
    {
      "epoch": 0.1892430278884462,
      "grad_norm": 0.12046647071838379,
      "learning_rate": 3.438384935229827e-05,
      "loss": 0.0493,
      "mean_token_accuracy": 0.9822425842285156,
      "num_tokens": 3877279.0,
      "step": 95
    },
    {
      "epoch": 0.199203187250996,
      "grad_norm": 0.14844371378421783,
      "learning_rate": 3.418557767376048e-05,
      "loss": 0.0488,
      "mean_token_accuracy": 0.9821234583854676,
      "num_tokens": 4081610.0,
      "step": 100
    },
    {
      "epoch": 0.20916334661354583,
      "grad_norm": 0.18534263968467712,
      "learning_rate": 3.39743842530313e-05,
      "loss": 0.0554,
      "mean_token_accuracy": 0.9797812938690186,
      "num_tokens": 4285498.0,
      "step": 105
    },
    {
      "epoch": 0.21912350597609562,
      "grad_norm": 0.23441195487976074,
      "learning_rate": 3.375049905779968e-05,
      "loss": 0.0549,
      "mean_token_accuracy": 0.9800987362861633,
      "num_tokens": 4490033.0,
      "step": 110
    },
    {
      "epoch": 0.2290836653386454,
      "grad_norm": 0.17242072522640228,
      "learning_rate": 3.35141658757781e-05,
      "loss": 0.0445,
      "mean_token_accuracy": 0.9836186289787292,
      "num_tokens": 4693503.0,
      "step": 115
    },
    {
      "epoch": 0.23904382470119523,
      "grad_norm": 0.19886986911296844,
      "learning_rate": 3.326564204924316e-05,
      "loss": 0.054,
      "mean_token_accuracy": 0.9802655100822448,
      "num_tokens": 4897874.0,
      "step": 120
    },
    {
      "epoch": 0.24900398406374502,
      "grad_norm": 0.2171323448419571,
      "learning_rate": 3.300519819481651e-05,
      "loss": 0.0531,
      "mean_token_accuracy": 0.9806747913360596,
      "num_tokens": 5101492.0,
      "step": 125
    },
    {
      "epoch": 0.2589641434262948,
      "grad_norm": 0.13506922125816345,
      "learning_rate": 3.2733117908791515e-05,
      "loss": 0.0543,
      "mean_token_accuracy": 0.9801211357116699,
      "num_tokens": 5305671.0,
      "step": 130
    },
    {
      "epoch": 0.2689243027888446,
      "grad_norm": 0.10356555134057999,
      "learning_rate": 3.244969745832635e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.9822631359100342,
      "num_tokens": 5510015.0,
      "step": 135
    },
    {
      "epoch": 0.2788844621513944,
      "grad_norm": 0.14405770599842072,
      "learning_rate": 3.215524545883983e-05,
      "loss": 0.0509,
      "mean_token_accuracy": 0.9813132166862488,
      "num_tokens": 5714564.0,
      "step": 140
    },
    {
      "epoch": 0.28884462151394424,
      "grad_norm": 0.13222414255142212,
      "learning_rate": 3.185008253796133e-05,
      "loss": 0.0491,
      "mean_token_accuracy": 0.9821501255035401,
      "num_tokens": 5918498.0,
      "step": 145
    },
    {
      "epoch": 0.29880478087649404,
      "grad_norm": 0.13528183102607727,
      "learning_rate": 3.153454098640057e-05,
      "loss": 0.0569,
      "mean_token_accuracy": 0.9791349172592163,
      "num_tokens": 6122024.0,
      "step": 150
    },
    {
      "epoch": 0.30876494023904383,
      "grad_norm": 0.1421584039926529,
      "learning_rate": 3.1208964396117594e-05,
      "loss": 0.051,
      "mean_token_accuracy": 0.9815291643142701,
      "num_tokens": 6325768.0,
      "step": 155
    },
    {
      "epoch": 0.3187250996015936,
      "grad_norm": 0.17487601935863495,
      "learning_rate": 3.087370728618683e-05,
      "loss": 0.0433,
      "mean_token_accuracy": 0.9842002630233765,
      "num_tokens": 6529719.0,
      "step": 160
    },
    {
      "epoch": 0.3286852589641434,
      "grad_norm": 0.1082608699798584,
      "learning_rate": 3.052913471676263e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.9817960739135743,
      "num_tokens": 6733822.0,
      "step": 165
    },
    {
      "epoch": 0.3386454183266932,
      "grad_norm": 0.1506233662366867,
      "learning_rate": 3.0175621891566735e-05,
      "loss": 0.049,
      "mean_token_accuracy": 0.9819495201110839,
      "num_tokens": 6938042.0,
      "step": 170
    },
    {
      "epoch": 0.34860557768924305,
      "grad_norm": 0.08547397702932358,
      "learning_rate": 2.981355374933037e-05,
      "loss": 0.0455,
      "mean_token_accuracy": 0.9835866689682007,
      "num_tokens": 7142834.0,
      "step": 175
    },
    {
      "epoch": 0.35856573705179284,
      "grad_norm": 0.13623499870300293,
      "learning_rate": 2.9443324544636004e-05,
      "loss": 0.0564,
      "mean_token_accuracy": 0.9794714450836182,
      "num_tokens": 7347226.0,
      "step": 180
    },
    {
      "epoch": 0.36852589641434264,
      "grad_norm": 0.16147273778915405,
      "learning_rate": 2.906533741861506e-05,
      "loss": 0.0468,
      "mean_token_accuracy": 0.9828415513038635,
      "num_tokens": 7551541.0,
      "step": 185
    },
    {
      "epoch": 0.3784860557768924,
      "grad_norm": 0.11898644268512726,
      "learning_rate": 2.868000395996911e-05,
      "loss": 0.0491,
      "mean_token_accuracy": 0.9818155407905579,
      "num_tokens": 7755893.0,
      "step": 190
    },
    {
      "epoch": 0.3884462151394422,
      "grad_norm": 0.09772292524576187,
      "learning_rate": 2.8287743756792572e-05,
      "loss": 0.0442,
      "mean_token_accuracy": 0.9837021589279175,
      "num_tokens": 7960180.0,
      "step": 195
    },
    {
      "epoch": 0.398406374501992,
      "grad_norm": 0.12020798772573471,
      "learning_rate": 2.7888983939684854e-05,
      "loss": 0.0544,
      "mean_token_accuracy": 0.9800261735916138,
      "num_tokens": 8164754.0,
      "step": 200
    },
    {
      "epoch": 0.40836653386454186,
      "grad_norm": 0.09441148489713669,
      "learning_rate": 2.7484158716649527e-05,
      "loss": 0.0466,
      "mean_token_accuracy": 0.9826923131942749,
      "num_tokens": 8369355.0,
      "step": 205
    },
    {
      "epoch": 0.41832669322709165,
      "grad_norm": 0.13226455450057983,
      "learning_rate": 2.707370890028693e-05,
      "loss": 0.0422,
      "mean_token_accuracy": 0.9842750310897828,
      "num_tokens": 8573886.0,
      "step": 210
    },
    {
      "epoch": 0.42828685258964144,
      "grad_norm": 0.1267005354166031,
      "learning_rate": 2.665808142779505e-05,
      "loss": 0.0468,
      "mean_token_accuracy": 0.9828801274299621,
      "num_tokens": 8778229.0,
      "step": 215
    },
    {
      "epoch": 0.43824701195219123,
      "grad_norm": 0.10507000982761383,
      "learning_rate": 2.6237728874301368e-05,
      "loss": 0.048,
      "mean_token_accuracy": 0.9823834776878357,
      "num_tokens": 8982141.0,
      "step": 220
    },
    {
      "epoch": 0.448207171314741,
      "grad_norm": 0.06319554895162582,
      "learning_rate": 2.5813108960055566e-05,
      "loss": 0.0462,
      "mean_token_accuracy": 0.9830332040786743,
      "num_tokens": 9186458.0,
      "step": 225
    },
    {
      "epoch": 0.4581673306772908,
      "grad_norm": 0.13272365927696228,
      "learning_rate": 2.5384684052019733e-05,
      "loss": 0.046,
      "mean_token_accuracy": 0.9831056952476501,
      "num_tokens": 9390828.0,
      "step": 230
    },
    {
      "epoch": 0.4681274900398406,
      "grad_norm": 0.0800432562828064,
      "learning_rate": 2.495292066039881e-05,
      "loss": 0.0489,
      "mean_token_accuracy": 0.9819565773010254,
      "num_tokens": 9595179.0,
      "step": 235
    },
    {
      "epoch": 0.47808764940239046,
      "grad_norm": 0.0766841396689415,
      "learning_rate": 2.451828893065942e-05,
      "loss": 0.0445,
      "mean_token_accuracy": 0.9836514234542847,
      "num_tokens": 9799928.0,
      "step": 240
    },
    {
      "epoch": 0.48804780876494025,
      "grad_norm": 0.1073063313961029,
      "learning_rate": 2.4081262131590365e-05,
      "loss": 0.0512,
      "mean_token_accuracy": 0.9812036514282226,
      "num_tokens": 10003881.0,
      "step": 245
    },
    {
      "epoch": 0.49800796812749004,
      "grad_norm": 0.09998544305562973,
      "learning_rate": 2.3642316139962043e-05,
      "loss": 0.049,
      "mean_token_accuracy": 0.982014799118042,
      "num_tokens": 10208269.0,
      "step": 250
    },
    {
      "epoch": 0.5079681274900398,
      "grad_norm": 0.09729743003845215,
      "learning_rate": 2.3201928922346115e-05,
      "loss": 0.0461,
      "mean_token_accuracy": 0.9830602169036865,
      "num_tokens": 10411859.0,
      "step": 255
    },
    {
      "epoch": 0.5179282868525896,
      "grad_norm": 0.08515891432762146,
      "learning_rate": 2.2760580014659588e-05,
      "loss": 0.0493,
      "mean_token_accuracy": 0.9815994024276733,
      "num_tokens": 10616234.0,
      "step": 260
    },
    {
      "epoch": 0.5278884462151394,
      "grad_norm": 0.12393739819526672,
      "learning_rate": 2.231875e-05,
      "loss": 0.0479,
      "mean_token_accuracy": 0.9821781516075134,
      "num_tokens": 10820187.0,
      "step": 265
    },
    {
      "epoch": 0.5378486055776892,
      "grad_norm": 0.1080358549952507,
      "learning_rate": 2.187691998534042e-05,
      "loss": 0.0406,
      "mean_token_accuracy": 0.9850728034973144,
      "num_tokens": 11024938.0,
      "step": 270
    },
    {
      "epoch": 0.547808764940239,
      "grad_norm": 0.10089477151632309,
      "learning_rate": 2.1435571077653886e-05,
      "loss": 0.0454,
      "mean_token_accuracy": 0.9832309484481812,
      "num_tokens": 11229126.0,
      "step": 275
    },
    {
      "epoch": 0.5577689243027888,
      "grad_norm": 0.1295860856771469,
      "learning_rate": 2.099518386003796e-05,
      "loss": 0.0477,
      "mean_token_accuracy": 0.9824405550956726,
      "num_tokens": 11433338.0,
      "step": 280
    },
    {
      "epoch": 0.5677290836653387,
      "grad_norm": 0.09834039211273193,
      "learning_rate": 2.055623786840964e-05,
      "loss": 0.0484,
      "mean_token_accuracy": 0.9822781562805176,
      "num_tokens": 11637306.0,
      "step": 285
    },
    {
      "epoch": 0.5776892430278885,
      "grad_norm": 0.17566028237342834,
      "learning_rate": 2.011921106934058e-05,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9825265645980835,
      "num_tokens": 11841258.0,
      "step": 290
    },
    {
      "epoch": 0.5876494023904383,
      "grad_norm": 0.10994154214859009,
      "learning_rate": 1.9684579339601198e-05,
      "loss": 0.0462,
      "mean_token_accuracy": 0.9831053376197815,
      "num_tokens": 12045782.0,
      "step": 295
    },
    {
      "epoch": 0.5976095617529881,
      "grad_norm": 0.09694515913724899,
      "learning_rate": 1.925281594798027e-05,
      "loss": 0.044,
      "mean_token_accuracy": 0.9838001012802124,
      "num_tokens": 12249366.0,
      "step": 300
    },
    {
      "epoch": 0.6075697211155379,
      "grad_norm": 0.10750065743923187,
      "learning_rate": 1.8824391039944435e-05,
      "loss": 0.0483,
      "mean_token_accuracy": 0.9823637962341308,
      "num_tokens": 12453947.0,
      "step": 305
    },
    {
      "epoch": 0.6175298804780877,
      "grad_norm": 0.10110612213611603,
      "learning_rate": 1.8399771125698633e-05,
      "loss": 0.0504,
      "mean_token_accuracy": 0.9815295934677124,
      "num_tokens": 12658554.0,
      "step": 310
    },
    {
      "epoch": 0.6274900398406374,
      "grad_norm": 0.11374131590127945,
      "learning_rate": 1.797941857220495e-05,
      "loss": 0.0479,
      "mean_token_accuracy": 0.9823715090751648,
      "num_tokens": 12862357.0,
      "step": 315
    }
  ],
  "logging_steps": 5,
  "max_steps": 502,
  "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": 3.027118452931297e+17,
  "train_batch_size": 20,
  "trial_name": null,
  "trial_params": null
}