{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.916666666666667,
  "eval_steps": 500,
  "global_step": 426,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06944444444444445,
      "grad_norm": 1.379091501235962,
      "learning_rate": 6.363636363636364e-06,
      "loss": 1.0771,
      "mean_token_accuracy": 0.8594340324401856,
      "num_tokens": 322079.0,
      "step": 5
    },
    {
      "epoch": 0.1388888888888889,
      "grad_norm": 0.40100836753845215,
      "learning_rate": 1.431818181818182e-05,
      "loss": 0.2901,
      "mean_token_accuracy": 0.9172819375991821,
      "num_tokens": 640460.0,
      "step": 10
    },
    {
      "epoch": 0.20833333333333334,
      "grad_norm": 0.33646833896636963,
      "learning_rate": 2.2272727272727274e-05,
      "loss": 0.2915,
      "mean_token_accuracy": 0.9141990065574646,
      "num_tokens": 961148.0,
      "step": 15
    },
    {
      "epoch": 0.2777777777777778,
      "grad_norm": 0.4926954507827759,
      "learning_rate": 3.0227272727272732e-05,
      "loss": 0.2623,
      "mean_token_accuracy": 0.9213010311126709,
      "num_tokens": 1279782.0,
      "step": 20
    },
    {
      "epoch": 0.3472222222222222,
      "grad_norm": 0.49179142713546753,
      "learning_rate": 3.499845882186083e-05,
      "loss": 0.2894,
      "mean_token_accuracy": 0.9159800171852112,
      "num_tokens": 1603539.0,
      "step": 25
    },
    {
      "epoch": 0.4166666666666667,
      "grad_norm": 0.2617477774620056,
      "learning_rate": 3.4981124724179196e-05,
      "loss": 0.2547,
      "mean_token_accuracy": 0.9256701231002807,
      "num_tokens": 1928745.0,
      "step": 30
    },
    {
      "epoch": 0.4861111111111111,
      "grad_norm": 0.3516538143157959,
      "learning_rate": 3.4944555581151475e-05,
      "loss": 0.2729,
      "mean_token_accuracy": 0.9196312546730041,
      "num_tokens": 2254716.0,
      "step": 35
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.44069457054138184,
      "learning_rate": 3.488880506303787e-05,
      "loss": 0.2521,
      "mean_token_accuracy": 0.9248023390769958,
      "num_tokens": 2573059.0,
      "step": 40
    },
    {
      "epoch": 0.625,
      "grad_norm": 0.3967224657535553,
      "learning_rate": 3.481395499141163e-05,
      "loss": 0.2576,
      "mean_token_accuracy": 0.9215535283088684,
      "num_tokens": 2894529.0,
      "step": 45
    },
    {
      "epoch": 0.6944444444444444,
      "grad_norm": 0.34342455863952637,
      "learning_rate": 3.472011521907455e-05,
      "loss": 0.2383,
      "mean_token_accuracy": 0.9299270510673523,
      "num_tokens": 3212853.0,
      "step": 50
    },
    {
      "epoch": 0.7638888888888888,
      "grad_norm": 0.33602777123451233,
      "learning_rate": 3.4607423468832905e-05,
      "loss": 0.262,
      "mean_token_accuracy": 0.9218739867210388,
      "num_tokens": 3534023.0,
      "step": 55
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.3001266121864319,
      "learning_rate": 3.4476045131370165e-05,
      "loss": 0.2475,
      "mean_token_accuracy": 0.9232696294784546,
      "num_tokens": 3854972.0,
      "step": 60
    },
    {
      "epoch": 0.9027777777777778,
      "grad_norm": 0.29761236906051636,
      "learning_rate": 3.432617302251345e-05,
      "loss": 0.2771,
      "mean_token_accuracy": 0.9182257652282715,
      "num_tokens": 4179982.0,
      "step": 65
    },
    {
      "epoch": 0.9722222222222222,
      "grad_norm": 0.3231615722179413,
      "learning_rate": 3.4158027100249596e-05,
      "loss": 0.2577,
      "mean_token_accuracy": 0.9226336240768432,
      "num_tokens": 4506234.0,
      "step": 70
    },
    {
      "epoch": 1.0416666666666667,
      "grad_norm": 0.25277966260910034,
      "learning_rate": 3.3971854141906594e-05,
      "loss": 0.1924,
      "mean_token_accuracy": 0.9404735565185547,
      "num_tokens": 4828503.0,
      "step": 75
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.37335672974586487,
      "learning_rate": 3.376792738197375e-05,
      "loss": 0.1659,
      "mean_token_accuracy": 0.9473877549171448,
      "num_tokens": 5148951.0,
      "step": 80
    },
    {
      "epoch": 1.1805555555555556,
      "grad_norm": 0.30353403091430664,
      "learning_rate": 3.3546546111092444e-05,
      "loss": 0.1576,
      "mean_token_accuracy": 0.9477113008499145,
      "num_tokens": 5464934.0,
      "step": 85
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.30610179901123047,
      "learning_rate": 3.330803523680581e-05,
      "loss": 0.1568,
      "mean_token_accuracy": 0.9483184695243836,
      "num_tokens": 5789007.0,
      "step": 90
    },
    {
      "epoch": 1.3194444444444444,
      "grad_norm": 0.44495099782943726,
      "learning_rate": 3.305274480671212e-05,
      "loss": 0.1681,
      "mean_token_accuracy": 0.9451550841331482,
      "num_tokens": 6108510.0,
      "step": 95
    },
    {
      "epoch": 1.3888888888888888,
      "grad_norm": 0.24649260938167572,
      "learning_rate": 3.2781049494721686e-05,
      "loss": 0.1586,
      "mean_token_accuracy": 0.9489330053329468,
      "num_tokens": 6427675.0,
      "step": 100
    },
    {
      "epoch": 1.4583333333333333,
      "grad_norm": 0.2683154344558716,
      "learning_rate": 3.2493348051171186e-05,
      "loss": 0.1496,
      "mean_token_accuracy": 0.9500372529029846,
      "num_tokens": 6751676.0,
      "step": 105
    },
    {
      "epoch": 1.5277777777777777,
      "grad_norm": 0.333688884973526,
      "learning_rate": 3.2190062717602604e-05,
      "loss": 0.1402,
      "mean_token_accuracy": 0.9503811240196228,
      "num_tokens": 7077808.0,
      "step": 110
    },
    {
      "epoch": 1.5972222222222223,
      "grad_norm": 0.3892817199230194,
      "learning_rate": 3.18716386070655e-05,
      "loss": 0.1534,
      "mean_token_accuracy": 0.9488339424133301,
      "num_tokens": 7395936.0,
      "step": 115
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.4228910505771637,
      "learning_rate": 3.153854305085226e-05,
      "loss": 0.1725,
      "mean_token_accuracy": 0.943362832069397,
      "num_tokens": 7718474.0,
      "step": 120
    },
    {
      "epoch": 1.7361111111111112,
      "grad_norm": 0.34297773241996765,
      "learning_rate": 3.119126491262496e-05,
      "loss": 0.1594,
      "mean_token_accuracy": 0.9471348404884339,
      "num_tokens": 8037998.0,
      "step": 125
    },
    {
      "epoch": 1.8055555555555556,
      "grad_norm": 0.4999421238899231,
      "learning_rate": 3.0830313870940544e-05,
      "loss": 0.171,
      "mean_token_accuracy": 0.9443796873092651,
      "num_tokens": 8365206.0,
      "step": 130
    },
    {
      "epoch": 1.875,
      "grad_norm": 0.39533600211143494,
      "learning_rate": 3.0456219671227244e-05,
      "loss": 0.167,
      "mean_token_accuracy": 0.9463403940200805,
      "num_tokens": 8689834.0,
      "step": 135
    },
    {
      "epoch": 1.9444444444444444,
      "grad_norm": 0.340938538312912,
      "learning_rate": 3.0069531348310138e-05,
      "loss": 0.148,
      "mean_token_accuracy": 0.9518447160720825,
      "num_tokens": 9011971.0,
      "step": 140
    },
    {
      "epoch": 2.013888888888889,
      "grad_norm": 0.2577916383743286,
      "learning_rate": 2.967081642062678e-05,
      "loss": 0.1432,
      "mean_token_accuracy": 0.9500368356704711,
      "num_tokens": 9334082.0,
      "step": 145
    },
    {
      "epoch": 2.0833333333333335,
      "grad_norm": 0.23916266858577728,
      "learning_rate": 2.92606600573157e-05,
      "loss": 0.0867,
      "mean_token_accuracy": 0.9711397290229797,
      "num_tokens": 9653974.0,
      "step": 150
    },
    {
      "epoch": 2.1527777777777777,
      "grad_norm": 0.3292118310928345,
      "learning_rate": 2.8839664219399954e-05,
      "loss": 0.0983,
      "mean_token_accuracy": 0.9649060964584351,
      "num_tokens": 9979477.0,
      "step": 155
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.2571089565753937,
      "learning_rate": 2.8408446776326326e-05,
      "loss": 0.0861,
      "mean_token_accuracy": 0.9679484009742737,
      "num_tokens": 10304270.0,
      "step": 160
    },
    {
      "epoch": 2.2916666666666665,
      "grad_norm": 0.37951090931892395,
      "learning_rate": 2.7967640599156713e-05,
      "loss": 0.0971,
      "mean_token_accuracy": 0.9654392242431641,
      "num_tokens": 10629691.0,
      "step": 165
    },
    {
      "epoch": 2.361111111111111,
      "grad_norm": 0.4137238562107086,
      "learning_rate": 2.7517892631742567e-05,
      "loss": 0.0993,
      "mean_token_accuracy": 0.9639935731887818,
      "num_tokens": 10950996.0,
      "step": 170
    },
    {
      "epoch": 2.4305555555555554,
      "grad_norm": 0.3094857335090637,
      "learning_rate": 2.7059862941245587e-05,
      "loss": 0.0955,
      "mean_token_accuracy": 0.9673921585083007,
      "num_tokens": 11269636.0,
      "step": 175
    },
    {
      "epoch": 2.5,
      "grad_norm": 0.37508368492126465,
      "learning_rate": 2.659422374939813e-05,
      "loss": 0.0932,
      "mean_token_accuracy": 0.9663854956626892,
      "num_tokens": 11587991.0,
      "step": 180
    },
    {
      "epoch": 2.5694444444444446,
      "grad_norm": 0.41950029134750366,
      "learning_rate": 2.6121658445925123e-05,
      "loss": 0.1081,
      "mean_token_accuracy": 0.9646688818931579,
      "num_tokens": 11904143.0,
      "step": 185
    },
    {
      "epoch": 2.638888888888889,
      "grad_norm": 0.3261002004146576,
      "learning_rate": 2.56428605855754e-05,
      "loss": 0.1028,
      "mean_token_accuracy": 0.9633297085762024,
      "num_tokens": 12224809.0,
      "step": 190
    },
    {
      "epoch": 2.7083333333333335,
      "grad_norm": 0.31264740228652954,
      "learning_rate": 2.515853287023447e-05,
      "loss": 0.0897,
      "mean_token_accuracy": 0.9685159087181091,
      "num_tokens": 12546717.0,
      "step": 195
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.35580959916114807,
      "learning_rate": 2.4669386117612615e-05,
      "loss": 0.0904,
      "mean_token_accuracy": 0.96531001329422,
      "num_tokens": 12871646.0,
      "step": 200
    },
    {
      "epoch": 2.8472222222222223,
      "grad_norm": 0.2896708548069,
      "learning_rate": 2.4176138218021843e-05,
      "loss": 0.0951,
      "mean_token_accuracy": 0.9645703315734864,
      "num_tokens": 13190258.0,
      "step": 205
    },
    {
      "epoch": 2.9166666666666665,
      "grad_norm": 0.45572149753570557,
      "learning_rate": 2.367951308077289e-05,
      "loss": 0.1009,
      "mean_token_accuracy": 0.9611859679222107,
      "num_tokens": 13515024.0,
      "step": 210
    },
    {
      "epoch": 2.986111111111111,
      "grad_norm": 0.3449590802192688,
      "learning_rate": 2.318023957173847e-05,
      "loss": 0.1011,
      "mean_token_accuracy": 0.9611627697944641,
      "num_tokens": 13841722.0,
      "step": 215
    },
    {
      "epoch": 3.0555555555555554,
      "grad_norm": 0.2528570592403412,
      "learning_rate": 2.2679050443642106e-05,
      "loss": 0.0707,
      "mean_token_accuracy": 0.9746357083320618,
      "num_tokens": 14164958.0,
      "step": 220
    },
    {
      "epoch": 3.125,
      "grad_norm": 0.36420342326164246,
      "learning_rate": 2.2176681260642438e-05,
      "loss": 0.0634,
      "mean_token_accuracy": 0.9762553215026856,
      "num_tokens": 14485491.0,
      "step": 225
    },
    {
      "epoch": 3.1944444444444446,
      "grad_norm": 0.2783300578594208,
      "learning_rate": 2.1673869318791425e-05,
      "loss": 0.0621,
      "mean_token_accuracy": 0.9781804919242859,
      "num_tokens": 14813368.0,
      "step": 230
    },
    {
      "epoch": 3.263888888888889,
      "grad_norm": 0.35492244362831116,
      "learning_rate": 2.117135256395068e-05,
      "loss": 0.0616,
      "mean_token_accuracy": 0.9777081608772278,
      "num_tokens": 15134125.0,
      "step": 235
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.42209485173225403,
      "learning_rate": 2.0669868508754178e-05,
      "loss": 0.0546,
      "mean_token_accuracy": 0.978722333908081,
      "num_tokens": 15454632.0,
      "step": 240
    },
    {
      "epoch": 3.4027777777777777,
      "grad_norm": 0.4834480583667755,
      "learning_rate": 2.0170153150206755e-05,
      "loss": 0.0655,
      "mean_token_accuracy": 0.978796124458313,
      "num_tokens": 15773802.0,
      "step": 245
    },
    {
      "epoch": 3.4722222222222223,
      "grad_norm": 0.2575581967830658,
      "learning_rate": 1.967293988950703e-05,
      "loss": 0.0594,
      "mean_token_accuracy": 0.9778955817222595,
      "num_tokens": 16098651.0,
      "step": 250
    },
    {
      "epoch": 3.5416666666666665,
      "grad_norm": 0.3081090748310089,
      "learning_rate": 1.9178958455679977e-05,
      "loss": 0.0568,
      "mean_token_accuracy": 0.9780836462974548,
      "num_tokens": 16423293.0,
      "step": 255
    },
    {
      "epoch": 3.611111111111111,
      "grad_norm": 0.22595608234405518,
      "learning_rate": 1.868893383459897e-05,
      "loss": 0.0597,
      "mean_token_accuracy": 0.9756160259246827,
      "num_tokens": 16751938.0,
      "step": 260
    },
    {
      "epoch": 3.6805555555555554,
      "grad_norm": 0.397429496049881,
      "learning_rate": 1.820358520496905e-05,
      "loss": 0.0608,
      "mean_token_accuracy": 0.9785622119903564,
      "num_tokens": 17070894.0,
      "step": 265
    },
    {
      "epoch": 3.75,
      "grad_norm": 0.28394079208374023,
      "learning_rate": 1.772362488283299e-05,
      "loss": 0.0595,
      "mean_token_accuracy": 0.9793779969215393,
      "num_tokens": 17386694.0,
      "step": 270
    },
    {
      "epoch": 3.8194444444444446,
      "grad_norm": 0.27272671461105347,
      "learning_rate": 1.724975727614928e-05,
      "loss": 0.0613,
      "mean_token_accuracy": 0.9769317388534546,
      "num_tokens": 17712906.0,
      "step": 275
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 0.2596876919269562,
      "learning_rate": 1.6782677850976294e-05,
      "loss": 0.0592,
      "mean_token_accuracy": 0.9769395232200623,
      "num_tokens": 18034448.0,
      "step": 280
    },
    {
      "epoch": 3.9583333333333335,
      "grad_norm": 0.37678053975105286,
      "learning_rate": 1.632307211077998e-05,
      "loss": 0.058,
      "mean_token_accuracy": 0.9776347041130066,
      "num_tokens": 18351515.0,
      "step": 285
    },
    {
      "epoch": 4.027777777777778,
      "grad_norm": 0.1913955807685852,
      "learning_rate": 1.5871614590362954e-05,
      "loss": 0.0494,
      "mean_token_accuracy": 0.9802444696426391,
      "num_tokens": 18669130.0,
      "step": 290
    },
    {
      "epoch": 4.097222222222222,
      "grad_norm": 0.27026042342185974,
      "learning_rate": 1.5428967865891696e-05,
      "loss": 0.0372,
      "mean_token_accuracy": 0.9868873596191406,
      "num_tokens": 18996467.0,
      "step": 295
    },
    {
      "epoch": 4.166666666666667,
      "grad_norm": 0.17194271087646484,
      "learning_rate": 1.4995781582474628e-05,
      "loss": 0.0379,
      "mean_token_accuracy": 0.9847923636436462,
      "num_tokens": 19323281.0,
      "step": 300
    },
    {
      "epoch": 4.236111111111111,
      "grad_norm": 0.448048859834671,
      "learning_rate": 1.4572691500718359e-05,
      "loss": 0.0418,
      "mean_token_accuracy": 0.9859732508659362,
      "num_tokens": 19652166.0,
      "step": 305
    },
    {
      "epoch": 4.305555555555555,
      "grad_norm": 0.18234789371490479,
      "learning_rate": 1.4160318563661293e-05,
      "loss": 0.035,
      "mean_token_accuracy": 0.9862746596336365,
      "num_tokens": 19976780.0,
      "step": 310
    },
    {
      "epoch": 4.375,
      "grad_norm": 0.28801319003105164,
      "learning_rate": 1.3759267985454126e-05,
      "loss": 0.04,
      "mean_token_accuracy": 0.9853678345680237,
      "num_tokens": 20294045.0,
      "step": 315
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.20915500819683075,
      "learning_rate": 1.3370128363124604e-05,
      "loss": 0.0424,
      "mean_token_accuracy": 0.9826555609703064,
      "num_tokens": 20620135.0,
      "step": 320
    },
    {
      "epoch": 4.513888888888889,
      "grad_norm": 0.17628608644008636,
      "learning_rate": 1.2993470812730217e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.9821347594261169,
      "num_tokens": 20941710.0,
      "step": 325
    },
    {
      "epoch": 4.583333333333333,
      "grad_norm": 0.23610952496528625,
      "learning_rate": 1.2629848131166632e-05,
      "loss": 0.0333,
      "mean_token_accuracy": 0.9875306367874146,
      "num_tokens": 21256388.0,
      "step": 330
    },
    {
      "epoch": 4.652777777777778,
      "grad_norm": 0.18080389499664307,
      "learning_rate": 1.2279793984862026e-05,
      "loss": 0.0371,
      "mean_token_accuracy": 0.9843253016471862,
      "num_tokens": 21579073.0,
      "step": 335
    },
    {
      "epoch": 4.722222222222222,
      "grad_norm": 0.20831552147865295,
      "learning_rate": 1.1943822126548e-05,
      "loss": 0.0378,
      "mean_token_accuracy": 0.9837713360786438,
      "num_tokens": 21904012.0,
      "step": 340
    },
    {
      "epoch": 4.791666666666667,
      "grad_norm": 0.31704482436180115,
      "learning_rate": 1.1622425641256605e-05,
      "loss": 0.0428,
      "mean_token_accuracy": 0.9829642415046692,
      "num_tokens": 22224825.0,
      "step": 345
    },
    {
      "epoch": 4.861111111111111,
      "grad_norm": 0.2652062773704529,
      "learning_rate": 1.1316076222650062e-05,
      "loss": 0.0343,
      "mean_token_accuracy": 0.9869701027870178,
      "num_tokens": 22536397.0,
      "step": 350
    },
    {
      "epoch": 4.930555555555555,
      "grad_norm": 0.3284396529197693,
      "learning_rate": 1.102522348074526e-05,
      "loss": 0.0368,
      "mean_token_accuracy": 0.9868231415748596,
      "num_tokens": 22852514.0,
      "step": 355
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12725569307804108,
      "learning_rate": 1.0750294282049076e-05,
      "loss": 0.0375,
      "mean_token_accuracy": 0.9843321442604065,
      "num_tokens": 23174822.0,
      "step": 360
    },
    {
      "epoch": 5.069444444444445,
      "grad_norm": 0.12493331730365753,
      "learning_rate": 1.0491692123072876e-05,
      "loss": 0.0264,
      "mean_token_accuracy": 0.9897792339324951,
      "num_tokens": 23497420.0,
      "step": 365
    },
    {
      "epoch": 5.138888888888889,
      "grad_norm": 0.19263431429862976,
      "learning_rate": 1.0249796538145782e-05,
      "loss": 0.0296,
      "mean_token_accuracy": 0.9884816765785217,
      "num_tokens": 23825885.0,
      "step": 370
    },
    {
      "epoch": 5.208333333333333,
      "grad_norm": 0.25486940145492554,
      "learning_rate": 1.002496254239568e-05,
      "loss": 0.0247,
      "mean_token_accuracy": 0.9909913301467895,
      "num_tokens": 24147773.0,
      "step": 375
    },
    {
      "epoch": 5.277777777777778,
      "grad_norm": 0.23300352692604065,
      "learning_rate": 9.817520110715585e-06,
      "loss": 0.0197,
      "mean_token_accuracy": 0.992484164237976,
      "num_tokens": 24469253.0,
      "step": 380
    },
    {
      "epoch": 5.347222222222222,
      "grad_norm": 0.291847288608551,
      "learning_rate": 9.627773693479954e-06,
      "loss": 0.0254,
      "mean_token_accuracy": 0.9914757132530212,
      "num_tokens": 24790062.0,
      "step": 385
    },
    {
      "epoch": 5.416666666666667,
      "grad_norm": 0.2282291054725647,
      "learning_rate": 9.45600176972181e-06,
      "loss": 0.0232,
      "mean_token_accuracy": 0.9907638549804687,
      "num_tokens": 25109161.0,
      "step": 390
    },
    {
      "epoch": 5.486111111111111,
      "grad_norm": 0.22575975954532623,
      "learning_rate": 9.302456438426336e-06,
      "loss": 0.0239,
      "mean_token_accuracy": 0.991183590888977,
      "num_tokens": 25429823.0,
      "step": 395
    },
    {
      "epoch": 5.555555555555555,
      "grad_norm": 0.2374185472726822,
      "learning_rate": 9.167363048540858e-06,
      "loss": 0.0256,
      "mean_token_accuracy": 0.9888947367668152,
      "num_tokens": 25755989.0,
      "step": 400
    },
    {
      "epoch": 5.625,
      "grad_norm": 0.14167584478855133,
      "learning_rate": 9.0509198682442e-06,
      "loss": 0.0233,
      "mean_token_accuracy": 0.9902876257896424,
      "num_tokens": 26069656.0,
      "step": 405
    },
    {
      "epoch": 5.694444444444445,
      "grad_norm": 0.19081205129623413,
      "learning_rate": 8.953297793960764e-06,
      "loss": 0.0228,
      "mean_token_accuracy": 0.99054856300354,
      "num_tokens": 26386466.0,
      "step": 410
    },
    {
      "epoch": 5.763888888888889,
      "grad_norm": 0.2015998363494873,
      "learning_rate": 8.874640099546494e-06,
      "loss": 0.0304,
      "mean_token_accuracy": 0.9869679093360901,
      "num_tokens": 26716123.0,
      "step": 415
    },
    {
      "epoch": 5.833333333333333,
      "grad_norm": 0.2091236114501953,
      "learning_rate": 8.815062226014707e-06,
      "loss": 0.0266,
      "mean_token_accuracy": 0.9901867628097534,
      "num_tokens": 27032695.0,
      "step": 420
    },
    {
      "epoch": 5.902777777777778,
      "grad_norm": 0.30149850249290466,
      "learning_rate": 8.774651612110506e-06,
      "loss": 0.0315,
      "mean_token_accuracy": 0.9872185707092285,
      "num_tokens": 27354910.0,
      "step": 425
    }
  ],
  "logging_steps": 5,
  "max_steps": 432,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 1.1898096215025254e+18,
  "train_batch_size": 28,
  "trial_name": null,
  "trial_params": null
}