{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 704,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007106057914372002,
      "grad_norm": 66.55605725941018,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 2.4888,
      "mean_token_accuracy": 0.6310219071805477,
      "num_tokens": 5923034.0,
      "step": 5
    },
    {
      "epoch": 0.014212115828744005,
      "grad_norm": 31.27595336039531,
      "learning_rate": 8.181818181818183e-06,
      "loss": 1.6522,
      "mean_token_accuracy": 0.6847351841628552,
      "num_tokens": 11832986.0,
      "step": 10
    },
    {
      "epoch": 0.021318173743116006,
      "grad_norm": 6.505788538701925,
      "learning_rate": 1.2727272727272728e-05,
      "loss": 1.0489,
      "mean_token_accuracy": 0.7373726107180119,
      "num_tokens": 17770055.0,
      "step": 15
    },
    {
      "epoch": 0.02842423165748801,
      "grad_norm": 3.4652580364487653,
      "learning_rate": 1.7272727272727274e-05,
      "loss": 0.8759,
      "mean_token_accuracy": 0.7604092583060265,
      "num_tokens": 23710854.0,
      "step": 20
    },
    {
      "epoch": 0.03553028957186001,
      "grad_norm": 3.073564258123934,
      "learning_rate": 1.999961805535155e-05,
      "loss": 0.8341,
      "mean_token_accuracy": 0.7634279936552048,
      "num_tokens": 29669291.0,
      "step": 25
    },
    {
      "epoch": 0.04263634748623201,
      "grad_norm": 2.908944153808897,
      "learning_rate": 1.9995321550350065e-05,
      "loss": 0.7871,
      "mean_token_accuracy": 0.7735667265951633,
      "num_tokens": 35600851.0,
      "step": 30
    },
    {
      "epoch": 0.04974240540060401,
      "grad_norm": 3.8874402850017673,
      "learning_rate": 1.998625339625423e-05,
      "loss": 0.7597,
      "mean_token_accuracy": 0.7763237416744232,
      "num_tokens": 41530042.0,
      "step": 35
    },
    {
      "epoch": 0.05684846331497602,
      "grad_norm": 2.5449731597634697,
      "learning_rate": 1.9972418403347817e-05,
      "loss": 0.7219,
      "mean_token_accuracy": 0.7790968000888825,
      "num_tokens": 47452248.0,
      "step": 40
    },
    {
      "epoch": 0.06395452122934801,
      "grad_norm": 1.7369247244872457,
      "learning_rate": 1.9953823910527057e-05,
      "loss": 0.6924,
      "mean_token_accuracy": 0.7809838131070137,
      "num_tokens": 53373989.0,
      "step": 45
    },
    {
      "epoch": 0.07106057914372002,
      "grad_norm": 1.619526678453925,
      "learning_rate": 1.993047978140764e-05,
      "loss": 0.6759,
      "mean_token_accuracy": 0.7882510013878345,
      "num_tokens": 59280737.0,
      "step": 50
    },
    {
      "epoch": 0.07816663705809203,
      "grad_norm": 1.4050312043738487,
      "learning_rate": 1.9902398399092494e-05,
      "loss": 0.6604,
      "mean_token_accuracy": 0.7918689742684364,
      "num_tokens": 65195627.0,
      "step": 55
    },
    {
      "epoch": 0.08527269497246402,
      "grad_norm": 3.4085768530761555,
      "learning_rate": 1.9869594659603032e-05,
      "loss": 0.6644,
      "mean_token_accuracy": 0.7887415766716004,
      "num_tokens": 71142550.0,
      "step": 60
    },
    {
      "epoch": 0.09237875288683603,
      "grad_norm": 2.1575833755380707,
      "learning_rate": 1.9832085963977445e-05,
      "loss": 0.6732,
      "mean_token_accuracy": 0.786578668653965,
      "num_tokens": 77081694.0,
      "step": 65
    },
    {
      "epoch": 0.09948481080120802,
      "grad_norm": 2.1880249716642743,
      "learning_rate": 1.978989220904016e-05,
      "loss": 0.6647,
      "mean_token_accuracy": 0.7869649574160575,
      "num_tokens": 83044225.0,
      "step": 70
    },
    {
      "epoch": 0.10659086871558003,
      "grad_norm": 3.500718658694026,
      "learning_rate": 1.9743035776847377e-05,
      "loss": 0.6618,
      "mean_token_accuracy": 0.7884958483278751,
      "num_tokens": 88988372.0,
      "step": 75
    },
    {
      "epoch": 0.11369692662995204,
      "grad_norm": 3.5435216425879,
      "learning_rate": 1.9691541522814327e-05,
      "loss": 0.6538,
      "mean_token_accuracy": 0.7917183473706245,
      "num_tokens": 94931302.0,
      "step": 80
    },
    {
      "epoch": 0.12080298454432403,
      "grad_norm": 3.3524381829737675,
      "learning_rate": 1.963543676253048e-05,
      "loss": 0.668,
      "mean_token_accuracy": 0.7873365215957164,
      "num_tokens": 100848612.0,
      "step": 85
    },
    {
      "epoch": 0.12790904245869603,
      "grad_norm": 2.5235394583310655,
      "learning_rate": 1.9574751257269748e-05,
      "loss": 0.6632,
      "mean_token_accuracy": 0.7879712589085102,
      "num_tokens": 106754467.0,
      "step": 90
    },
    {
      "epoch": 0.13501510037306805,
      "grad_norm": 2.1804248444209673,
      "learning_rate": 1.950951719820335e-05,
      "loss": 0.6519,
      "mean_token_accuracy": 0.7906477533280849,
      "num_tokens": 112664258.0,
      "step": 95
    },
    {
      "epoch": 0.14212115828744004,
      "grad_norm": 2.1726552942080497,
      "learning_rate": 1.9439769189323727e-05,
      "loss": 0.654,
      "mean_token_accuracy": 0.7900555059313774,
      "num_tokens": 118606688.0,
      "step": 100
    },
    {
      "epoch": 0.14212115828744004,
      "eval_loss": 0.6459761261940002,
      "eval_mean_token_accuracy": 0.7921642295101232,
      "eval_num_tokens": 118606688.0,
      "eval_runtime": 295.44,
      "eval_samples_per_second": 12.317,
      "eval_steps_per_second": 0.386,
      "step": 100
    },
    {
      "epoch": 0.14922721620181204,
      "grad_norm": 2.0323675408135484,
      "learning_rate": 1.9365544229088517e-05,
      "loss": 0.6521,
      "mean_token_accuracy": 0.7909861005842685,
      "num_tokens": 124512961.0,
      "step": 105
    },
    {
      "epoch": 0.15633327411618406,
      "grad_norm": 5.430519736360723,
      "learning_rate": 1.9286881690794425e-05,
      "loss": 0.6453,
      "mean_token_accuracy": 0.7935434632003308,
      "num_tokens": 130441640.0,
      "step": 110
    },
    {
      "epoch": 0.16343933203055605,
      "grad_norm": 2.6425373488919943,
      "learning_rate": 1.9203823301691272e-05,
      "loss": 0.6683,
      "mean_token_accuracy": 0.7861981622874736,
      "num_tokens": 136394287.0,
      "step": 115
    },
    {
      "epoch": 0.17054538994492804,
      "grad_norm": 3.2273755870117724,
      "learning_rate": 1.9116413120847425e-05,
      "loss": 0.6649,
      "mean_token_accuracy": 0.7885523498058319,
      "num_tokens": 142309319.0,
      "step": 120
    },
    {
      "epoch": 0.17765144785930007,
      "grad_norm": 4.163268789121899,
      "learning_rate": 1.902469751577826e-05,
      "loss": 0.6679,
      "mean_token_accuracy": 0.7868955120444298,
      "num_tokens": 148247490.0,
      "step": 125
    },
    {
      "epoch": 0.18475750577367206,
      "grad_norm": 1.6661402706531616,
      "learning_rate": 1.892872513785008e-05,
      "loss": 0.6394,
      "mean_token_accuracy": 0.7936947368085384,
      "num_tokens": 154195595.0,
      "step": 130
    },
    {
      "epoch": 0.19186356368804405,
      "grad_norm": 1.5794423300632663,
      "learning_rate": 1.88285468964726e-05,
      "loss": 0.6373,
      "mean_token_accuracy": 0.7937119543552399,
      "num_tokens": 160139401.0,
      "step": 135
    },
    {
      "epoch": 0.19896962160241605,
      "grad_norm": 1.9146265900976953,
      "learning_rate": 1.872421593209355e-05,
      "loss": 0.6367,
      "mean_token_accuracy": 0.7936570249497891,
      "num_tokens": 166057697.0,
      "step": 140
    },
    {
      "epoch": 0.20607567951678807,
      "grad_norm": 1.8920022402043013,
      "learning_rate": 1.861578758800989e-05,
      "loss": 0.6448,
      "mean_token_accuracy": 0.7935090765357018,
      "num_tokens": 171980981.0,
      "step": 145
    },
    {
      "epoch": 0.21318173743116006,
      "grad_norm": 1.8829721816030007,
      "learning_rate": 1.8503319381010414e-05,
      "loss": 0.6394,
      "mean_token_accuracy": 0.7929822854697705,
      "num_tokens": 177920811.0,
      "step": 150
    },
    {
      "epoch": 0.22028779534553206,
      "grad_norm": 2.710625896660201,
      "learning_rate": 1.8386870970865488e-05,
      "loss": 0.6297,
      "mean_token_accuracy": 0.795507474988699,
      "num_tokens": 183857404.0,
      "step": 155
    },
    {
      "epoch": 0.22739385325990408,
      "grad_norm": 1.7692011271897554,
      "learning_rate": 1.8266504128679988e-05,
      "loss": 0.6374,
      "mean_token_accuracy": 0.7952376157045364,
      "num_tokens": 189793021.0,
      "step": 160
    },
    {
      "epoch": 0.23449991117427607,
      "grad_norm": 1.226150395650258,
      "learning_rate": 1.814228270412624e-05,
      "loss": 0.6525,
      "mean_token_accuracy": 0.7921756438910961,
      "num_tokens": 195715230.0,
      "step": 165
    },
    {
      "epoch": 0.24160596908864806,
      "grad_norm": 2.1730739897367206,
      "learning_rate": 1.8014272591574405e-05,
      "loss": 0.6452,
      "mean_token_accuracy": 0.7933229982852936,
      "num_tokens": 201644724.0,
      "step": 170
    },
    {
      "epoch": 0.2487120270030201,
      "grad_norm": 1.1273239605881678,
      "learning_rate": 1.7882541695138224e-05,
      "loss": 0.6328,
      "mean_token_accuracy": 0.795696322619915,
      "num_tokens": 207585561.0,
      "step": 175
    },
    {
      "epoch": 0.25581808491739205,
      "grad_norm": 1.094910349166684,
      "learning_rate": 1.7747159892654646e-05,
      "loss": 0.6294,
      "mean_token_accuracy": 0.7952122300863266,
      "num_tokens": 213540458.0,
      "step": 180
    },
    {
      "epoch": 0.2629241428317641,
      "grad_norm": 1.4265158165299685,
      "learning_rate": 1.7608198998616533e-05,
      "loss": 0.6401,
      "mean_token_accuracy": 0.7934339419007301,
      "num_tokens": 219456523.0,
      "step": 185
    },
    {
      "epoch": 0.2700302007461361,
      "grad_norm": 1.1866353031490813,
      "learning_rate": 1.7465732726077993e-05,
      "loss": 0.6376,
      "mean_token_accuracy": 0.7937880590558052,
      "num_tokens": 225385658.0,
      "step": 190
    },
    {
      "epoch": 0.27713625866050806,
      "grad_norm": 1.1122272717695676,
      "learning_rate": 1.731983664755264e-05,
      "loss": 0.6434,
      "mean_token_accuracy": 0.793655838817358,
      "num_tokens": 231315244.0,
      "step": 195
    },
    {
      "epoch": 0.2842423165748801,
      "grad_norm": 0.9655859557982183,
      "learning_rate": 1.717058815492548e-05,
      "loss": 0.641,
      "mean_token_accuracy": 0.792879494279623,
      "num_tokens": 237275441.0,
      "step": 200
    },
    {
      "epoch": 0.2842423165748801,
      "eval_loss": 0.6347336769104004,
      "eval_mean_token_accuracy": 0.7943492432435354,
      "eval_num_tokens": 237275441.0,
      "eval_runtime": 296.27,
      "eval_samples_per_second": 12.283,
      "eval_steps_per_second": 0.385,
      "step": 200
    },
    {
      "epoch": 0.2913483744892521,
      "grad_norm": 0.9237497775331585,
      "learning_rate": 1.701806641839967e-05,
      "loss": 0.6373,
      "mean_token_accuracy": 0.7951489560306072,
      "num_tokens": 243189498.0,
      "step": 205
    },
    {
      "epoch": 0.29845443240362407,
      "grad_norm": 0.953221738403784,
      "learning_rate": 1.6862352344500004e-05,
      "loss": 0.6402,
      "mean_token_accuracy": 0.7936429545283318,
      "num_tokens": 249109887.0,
      "step": 210
    },
    {
      "epoch": 0.3055604903179961,
      "grad_norm": 0.9703848636195304,
      "learning_rate": 1.6703528533155283e-05,
      "loss": 0.6414,
      "mean_token_accuracy": 0.7941999517381191,
      "num_tokens": 255049150.0,
      "step": 215
    },
    {
      "epoch": 0.3126665482323681,
      "grad_norm": 0.96915858246581,
      "learning_rate": 1.6541679233882477e-05,
      "loss": 0.6298,
      "mean_token_accuracy": 0.7961237229406833,
      "num_tokens": 260976846.0,
      "step": 220
    },
    {
      "epoch": 0.3197726061467401,
      "grad_norm": 0.9933766730385373,
      "learning_rate": 1.63768903010958e-05,
      "loss": 0.6356,
      "mean_token_accuracy": 0.795130829513073,
      "num_tokens": 266905427.0,
      "step": 225
    },
    {
      "epoch": 0.3268786640611121,
      "grad_norm": 1.0546956761351753,
      "learning_rate": 1.6209249148564437e-05,
      "loss": 0.6304,
      "mean_token_accuracy": 0.7961406633257866,
      "num_tokens": 272824016.0,
      "step": 230
    },
    {
      "epoch": 0.3339847219754841,
      "grad_norm": 0.9224242489597514,
      "learning_rate": 1.603884470304318e-05,
      "loss": 0.6414,
      "mean_token_accuracy": 0.7928595051169396,
      "num_tokens": 278772935.0,
      "step": 235
    },
    {
      "epoch": 0.3410907798898561,
      "grad_norm": 1.775037599109117,
      "learning_rate": 1.5865767357100383e-05,
      "loss": 0.6247,
      "mean_token_accuracy": 0.7980937138199806,
      "num_tokens": 284699860.0,
      "step": 240
    },
    {
      "epoch": 0.3481968378042281,
      "grad_norm": 3.6986968651179866,
      "learning_rate": 1.5690108921168428e-05,
      "loss": 0.6282,
      "mean_token_accuracy": 0.7979529812932015,
      "num_tokens": 290642638.0,
      "step": 245
    },
    {
      "epoch": 0.35530289571860013,
      "grad_norm": 1.1970583265540296,
      "learning_rate": 1.5511962574842073e-05,
      "loss": 0.6265,
      "mean_token_accuracy": 0.7972055464982987,
      "num_tokens": 296584355.0,
      "step": 250
    },
    {
      "epoch": 0.3624089536329721,
      "grad_norm": 1.3725931963002411,
      "learning_rate": 1.5331422817450485e-05,
      "loss": 0.6316,
      "mean_token_accuracy": 0.7962015710771084,
      "num_tokens": 302514276.0,
      "step": 255
    },
    {
      "epoch": 0.3695150115473441,
      "grad_norm": 1.052505642053235,
      "learning_rate": 1.5148585417929212e-05,
      "loss": 0.6265,
      "mean_token_accuracy": 0.797739926725626,
      "num_tokens": 308442058.0,
      "step": 260
    },
    {
      "epoch": 0.37662106946171614,
      "grad_norm": 1.2266076226995661,
      "learning_rate": 1.4963547364018711e-05,
      "loss": 0.6228,
      "mean_token_accuracy": 0.7978610590100288,
      "num_tokens": 314358386.0,
      "step": 265
    },
    {
      "epoch": 0.3837271273760881,
      "grad_norm": 0.9779433419910959,
      "learning_rate": 1.477640681081632e-05,
      "loss": 0.6274,
      "mean_token_accuracy": 0.797017228603363,
      "num_tokens": 320330186.0,
      "step": 270
    },
    {
      "epoch": 0.39083318529046013,
      "grad_norm": 1.386190197186753,
      "learning_rate": 1.4587263028709013e-05,
      "loss": 0.6243,
      "mean_token_accuracy": 0.7981224037706852,
      "num_tokens": 326254997.0,
      "step": 275
    },
    {
      "epoch": 0.3979392432048321,
      "grad_norm": 2.614420120110884,
      "learning_rate": 1.4396216350714512e-05,
      "loss": 0.6262,
      "mean_token_accuracy": 0.7971527561545372,
      "num_tokens": 332131183.0,
      "step": 280
    },
    {
      "epoch": 0.4050453011192041,
      "grad_norm": 1.042309981896861,
      "learning_rate": 1.4203368119258759e-05,
      "loss": 0.6239,
      "mean_token_accuracy": 0.7973503857851029,
      "num_tokens": 338051217.0,
      "step": 285
    },
    {
      "epoch": 0.41215135903357614,
      "grad_norm": 1.3072882069993246,
      "learning_rate": 1.4008820632417906e-05,
      "loss": 0.6153,
      "mean_token_accuracy": 0.8006650306284427,
      "num_tokens": 343962755.0,
      "step": 290
    },
    {
      "epoch": 0.4192574169479481,
      "grad_norm": 0.9860476843101288,
      "learning_rate": 1.381267708965339e-05,
      "loss": 0.6203,
      "mean_token_accuracy": 0.7994598127901554,
      "num_tokens": 349868870.0,
      "step": 295
    },
    {
      "epoch": 0.4263634748623201,
      "grad_norm": 1.4036170847387148,
      "learning_rate": 1.3615041537068831e-05,
      "loss": 0.6295,
      "mean_token_accuracy": 0.7960586912930012,
      "num_tokens": 355805962.0,
      "step": 300
    },
    {
      "epoch": 0.4263634748623201,
      "eval_loss": 0.6163658499717712,
      "eval_mean_token_accuracy": 0.7995022409840634,
      "eval_num_tokens": 355805962.0,
      "eval_runtime": 296.606,
      "eval_samples_per_second": 12.269,
      "eval_steps_per_second": 0.384,
      "step": 300
    },
    {
      "epoch": 0.43346953277669215,
      "grad_norm": 1.4151324155929559,
      "learning_rate": 1.3416018812217866e-05,
      "loss": 0.6254,
      "mean_token_accuracy": 0.7966083332896232,
      "num_tokens": 361723150.0,
      "step": 305
    },
    {
      "epoch": 0.4405755906910641,
      "grad_norm": 1.2868115337774566,
      "learning_rate": 1.3215714488492121e-05,
      "loss": 0.6078,
      "mean_token_accuracy": 0.8015532568097115,
      "num_tokens": 367658526.0,
      "step": 310
    },
    {
      "epoch": 0.44768164860543613,
      "grad_norm": 1.928461708072047,
      "learning_rate": 1.3014234819118846e-05,
      "loss": 0.606,
      "mean_token_accuracy": 0.8015141606330871,
      "num_tokens": 373569982.0,
      "step": 315
    },
    {
      "epoch": 0.45478770651980815,
      "grad_norm": 1.899128993970621,
      "learning_rate": 1.2811686680797942e-05,
      "loss": 0.6182,
      "mean_token_accuracy": 0.7982937648892403,
      "num_tokens": 379503775.0,
      "step": 320
    },
    {
      "epoch": 0.4618937644341801,
      "grad_norm": 1.4589431652822673,
      "learning_rate": 1.2608177517008268e-05,
      "loss": 0.606,
      "mean_token_accuracy": 0.8024964012205601,
      "num_tokens": 385421798.0,
      "step": 325
    },
    {
      "epoch": 0.46899982234855214,
      "grad_norm": 2.312165029030224,
      "learning_rate": 1.240381528101327e-05,
      "loss": 0.6036,
      "mean_token_accuracy": 0.8019742712378501,
      "num_tokens": 391368024.0,
      "step": 330
    },
    {
      "epoch": 0.47610588026292416,
      "grad_norm": 1.538144906292264,
      "learning_rate": 1.2198708378596198e-05,
      "loss": 0.5993,
      "mean_token_accuracy": 0.804237449914217,
      "num_tokens": 397292736.0,
      "step": 335
    },
    {
      "epoch": 0.48321193817729613,
      "grad_norm": 1.307446043015124,
      "learning_rate": 1.19929656105553e-05,
      "loss": 0.6046,
      "mean_token_accuracy": 0.8020064242184162,
      "num_tokens": 403217739.0,
      "step": 340
    },
    {
      "epoch": 0.49031799609166815,
      "grad_norm": 1.4958823006704178,
      "learning_rate": 1.1786696114989455e-05,
      "loss": 0.6049,
      "mean_token_accuracy": 0.8029085315763951,
      "num_tokens": 409134483.0,
      "step": 345
    },
    {
      "epoch": 0.4974240540060402,
      "grad_norm": 1.3590504704358832,
      "learning_rate": 1.1580009309404887e-05,
      "loss": 0.6068,
      "mean_token_accuracy": 0.8020060114562512,
      "num_tokens": 415063752.0,
      "step": 350
    },
    {
      "epoch": 0.5045301119204122,
      "grad_norm": 1.5361640914741626,
      "learning_rate": 1.1373014832673661e-05,
      "loss": 0.6058,
      "mean_token_accuracy": 0.8034303903579711,
      "num_tokens": 420986440.0,
      "step": 355
    },
    {
      "epoch": 0.5116361698347841,
      "grad_norm": 1.7406253504574596,
      "learning_rate": 1.1165822486874773e-05,
      "loss": 0.6013,
      "mean_token_accuracy": 0.8025040835142135,
      "num_tokens": 426919043.0,
      "step": 360
    },
    {
      "epoch": 0.5187422277491561,
      "grad_norm": 1.4252805000356146,
      "learning_rate": 1.0958542179048637e-05,
      "loss": 0.5975,
      "mean_token_accuracy": 0.8049191392958164,
      "num_tokens": 432835321.0,
      "step": 365
    },
    {
      "epoch": 0.5258482856635281,
      "grad_norm": 1.6635488719233447,
      "learning_rate": 1.0751283862895914e-05,
      "loss": 0.6198,
      "mean_token_accuracy": 0.7999552808701992,
      "num_tokens": 438731328.0,
      "step": 370
    },
    {
      "epoch": 0.5329543435779002,
      "grad_norm": 1.3128810337611407,
      "learning_rate": 1.0544157480451586e-05,
      "loss": 0.6091,
      "mean_token_accuracy": 0.800326906144619,
      "num_tokens": 444678429.0,
      "step": 375
    },
    {
      "epoch": 0.5400604014922722,
      "grad_norm": 1.4889952603189804,
      "learning_rate": 1.033727290376522e-05,
      "loss": 0.6138,
      "mean_token_accuracy": 0.8001590967178345,
      "num_tokens": 450570676.0,
      "step": 380
    },
    {
      "epoch": 0.5471664594066442,
      "grad_norm": 1.466863294864097,
      "learning_rate": 1.013073987661834e-05,
      "loss": 0.6135,
      "mean_token_accuracy": 0.7993282429873944,
      "num_tokens": 456505695.0,
      "step": 385
    },
    {
      "epoch": 0.5542725173210161,
      "grad_norm": 2.0382279212397996,
      "learning_rate": 9.924667956309862e-06,
      "loss": 0.5994,
      "mean_token_accuracy": 0.8032713256776333,
      "num_tokens": 462436441.0,
      "step": 390
    },
    {
      "epoch": 0.5613785752353881,
      "grad_norm": 2.5944556078034373,
      "learning_rate": 9.719166455540437e-06,
      "loss": 0.6081,
      "mean_token_accuracy": 0.801172049343586,
      "num_tokens": 468376688.0,
      "step": 395
    },
    {
      "epoch": 0.5684846331497602,
      "grad_norm": 2.3431282794499113,
      "learning_rate": 9.51434438442655e-06,
      "loss": 0.6024,
      "mean_token_accuracy": 0.8036491274833679,
      "num_tokens": 474296296.0,
      "step": 400
    },
    {
      "epoch": 0.5684846331497602,
      "eval_loss": 0.6026122570037842,
      "eval_mean_token_accuracy": 0.8030629869092974,
      "eval_num_tokens": 474296296.0,
      "eval_runtime": 295.7635,
      "eval_samples_per_second": 12.304,
      "eval_steps_per_second": 0.385,
      "step": 400
    },
    {
      "epoch": 0.5755906910641322,
      "grad_norm": 2.004552904299561,
      "learning_rate": 9.310310392675132e-06,
      "loss": 0.5946,
      "mean_token_accuracy": 0.8059669084846973,
      "num_tokens": 480221232.0,
      "step": 405
    },
    {
      "epoch": 0.5826967489785042,
      "grad_norm": 2.182094919737894,
      "learning_rate": 9.107172711949324e-06,
      "loss": 0.6098,
      "mean_token_accuracy": 0.8020808771252632,
      "num_tokens": 486176414.0,
      "step": 410
    },
    {
      "epoch": 0.5898028068928762,
      "grad_norm": 2.181381325740844,
      "learning_rate": 8.905039098456049e-06,
      "loss": 0.6011,
      "mean_token_accuracy": 0.8030769810080528,
      "num_tokens": 492102670.0,
      "step": 415
    },
    {
      "epoch": 0.5969088648072481,
      "grad_norm": 1.8310617565549334,
      "learning_rate": 8.704016775785742e-06,
      "loss": 0.6047,
      "mean_token_accuracy": 0.8027667418122292,
      "num_tokens": 498034157.0,
      "step": 420
    },
    {
      "epoch": 0.6040149227216202,
      "grad_norm": 1.8408277960821589,
      "learning_rate": 8.50421237803464e-06,
      "loss": 0.6,
      "mean_token_accuracy": 0.8030680187046528,
      "num_tokens": 503966085.0,
      "step": 425
    },
    {
      "epoch": 0.6111209806359922,
      "grad_norm": 1.58781331586126,
      "learning_rate": 8.30573189323978e-06,
      "loss": 0.5923,
      "mean_token_accuracy": 0.8065764397382736,
      "num_tokens": 509912387.0,
      "step": 430
    },
    {
      "epoch": 0.6182270385503642,
      "grad_norm": 2.4917867524381085,
      "learning_rate": 8.108680607156669e-06,
      "loss": 0.6057,
      "mean_token_accuracy": 0.8019099831581116,
      "num_tokens": 515846667.0,
      "step": 435
    },
    {
      "epoch": 0.6253330964647362,
      "grad_norm": 1.4683997587556166,
      "learning_rate": 7.913163047409533e-06,
      "loss": 0.6007,
      "mean_token_accuracy": 0.8031819522380829,
      "num_tokens": 521759712.0,
      "step": 440
    },
    {
      "epoch": 0.6324391543791081,
      "grad_norm": 2.0497193131572993,
      "learning_rate": 7.719282928043688e-06,
      "loss": 0.6026,
      "mean_token_accuracy": 0.8027250319719315,
      "num_tokens": 527693723.0,
      "step": 445
    },
    {
      "epoch": 0.6395452122934802,
      "grad_norm": 1.9866627694997772,
      "learning_rate": 7.527143094509492e-06,
      "loss": 0.5962,
      "mean_token_accuracy": 0.8050611786544323,
      "num_tokens": 533641375.0,
      "step": 450
    },
    {
      "epoch": 0.6466512702078522,
      "grad_norm": 2.0671724220740284,
      "learning_rate": 7.336845469107061e-06,
      "loss": 0.6012,
      "mean_token_accuracy": 0.8034082941710949,
      "num_tokens": 539566506.0,
      "step": 455
    },
    {
      "epoch": 0.6537573281222242,
      "grad_norm": 1.1733461701395618,
      "learning_rate": 7.148490996920661e-06,
      "loss": 0.6005,
      "mean_token_accuracy": 0.8036525435745716,
      "num_tokens": 545484724.0,
      "step": 460
    },
    {
      "epoch": 0.6608633860365962,
      "grad_norm": 1.3593627896085958,
      "learning_rate": 6.9621795922714805e-06,
      "loss": 0.5938,
      "mean_token_accuracy": 0.8044769234955311,
      "num_tokens": 551415017.0,
      "step": 465
    },
    {
      "epoch": 0.6679694439509682,
      "grad_norm": 1.3812452385551917,
      "learning_rate": 6.778010085717202e-06,
      "loss": 0.5998,
      "mean_token_accuracy": 0.8046882562339306,
      "num_tokens": 557320461.0,
      "step": 470
    },
    {
      "epoch": 0.6750755018653402,
      "grad_norm": 1.8099599749434352,
      "learning_rate": 6.596080171626409e-06,
      "loss": 0.6023,
      "mean_token_accuracy": 0.8030730128288269,
      "num_tokens": 563258287.0,
      "step": 475
    },
    {
      "epoch": 0.6821815597797122,
      "grad_norm": 1.4003389577047656,
      "learning_rate": 6.416486356355769e-06,
      "loss": 0.6083,
      "mean_token_accuracy": 0.8014967061579228,
      "num_tokens": 569180910.0,
      "step": 480
    },
    {
      "epoch": 0.6892876176940842,
      "grad_norm": 1.064555140921138,
      "learning_rate": 6.239323907057342e-06,
      "loss": 0.6031,
      "mean_token_accuracy": 0.8031500183045864,
      "num_tokens": 575092087.0,
      "step": 485
    },
    {
      "epoch": 0.6963936756084562,
      "grad_norm": 1.42689507484952,
      "learning_rate": 6.064686801143271e-06,
      "loss": 0.5872,
      "mean_token_accuracy": 0.8078797787427903,
      "num_tokens": 581002946.0,
      "step": 490
    },
    {
      "epoch": 0.7034997335228282,
      "grad_norm": 1.4988954301366828,
      "learning_rate": 5.892667676434633e-06,
      "loss": 0.5969,
      "mean_token_accuracy": 0.805253654718399,
      "num_tokens": 586913668.0,
      "step": 495
    },
    {
      "epoch": 0.7106057914372003,
      "grad_norm": 1.812281786151278,
      "learning_rate": 5.723357782020867e-06,
      "loss": 0.5895,
      "mean_token_accuracy": 0.8048664882779122,
      "num_tokens": 592832114.0,
      "step": 500
    },
    {
      "epoch": 0.7106057914372003,
      "eval_loss": 0.5955030918121338,
      "eval_mean_token_accuracy": 0.804359252515592,
      "eval_num_tokens": 592832114.0,
      "eval_runtime": 295.6444,
      "eval_samples_per_second": 12.309,
      "eval_steps_per_second": 0.386,
      "step": 500
    },
    {
      "epoch": 0.7177118493515722,
      "grad_norm": 1.9139670158434299,
      "learning_rate": 5.556846929855857e-06,
      "loss": 0.5887,
      "mean_token_accuracy": 0.8057731881737709,
      "num_tokens": 598754045.0,
      "step": 505
    },
    {
      "epoch": 0.7248179072659442,
      "grad_norm": 1.5674173121754666,
      "learning_rate": 5.393223447116409e-06,
      "loss": 0.6035,
      "mean_token_accuracy": 0.803417231887579,
      "num_tokens": 604714198.0,
      "step": 510
    },
    {
      "epoch": 0.7319239651803162,
      "grad_norm": 1.4540694573929962,
      "learning_rate": 5.232574129348278e-06,
      "loss": 0.5945,
      "mean_token_accuracy": 0.804961483925581,
      "num_tokens": 610643984.0,
      "step": 515
    },
    {
      "epoch": 0.7390300230946882,
      "grad_norm": 1.5167979309975335,
      "learning_rate": 5.0749841944247e-06,
      "loss": 0.6049,
      "mean_token_accuracy": 0.8024425834417344,
      "num_tokens": 616589217.0,
      "step": 520
    },
    {
      "epoch": 0.7461360810090603,
      "grad_norm": 1.4872845521101352,
      "learning_rate": 4.92053723734182e-06,
      "loss": 0.6022,
      "mean_token_accuracy": 0.8034923203289509,
      "num_tokens": 622495105.0,
      "step": 525
    },
    {
      "epoch": 0.7532421389234323,
      "grad_norm": 2.4122501400628993,
      "learning_rate": 4.769315185874951e-06,
      "loss": 0.5961,
      "mean_token_accuracy": 0.8057848632335662,
      "num_tokens": 628430302.0,
      "step": 530
    },
    {
      "epoch": 0.7603481968378042,
      "grad_norm": 3.0932302891179235,
      "learning_rate": 4.621398257119266e-06,
      "loss": 0.5966,
      "mean_token_accuracy": 0.8043950840830802,
      "num_tokens": 634369704.0,
      "step": 535
    },
    {
      "epoch": 0.7674542547521762,
      "grad_norm": 2.761826923400515,
      "learning_rate": 4.476864914937923e-06,
      "loss": 0.5879,
      "mean_token_accuracy": 0.8060916163027286,
      "num_tokens": 640316507.0,
      "step": 540
    },
    {
      "epoch": 0.7745603126665482,
      "grad_norm": 1.810383003438599,
      "learning_rate": 4.335791828340183e-06,
      "loss": 0.6014,
      "mean_token_accuracy": 0.8044339507818222,
      "num_tokens": 646273380.0,
      "step": 545
    },
    {
      "epoch": 0.7816663705809203,
      "grad_norm": 1.57106874125145,
      "learning_rate": 4.1982538308116775e-06,
      "loss": 0.5933,
      "mean_token_accuracy": 0.8038996756076813,
      "num_tokens": 652201532.0,
      "step": 550
    },
    {
      "epoch": 0.7887724284952923,
      "grad_norm": 1.65882515510847,
      "learning_rate": 4.064323880618279e-06,
      "loss": 0.5979,
      "mean_token_accuracy": 0.8053264081478119,
      "num_tokens": 658135558.0,
      "step": 555
    },
    {
      "epoch": 0.7958784864096642,
      "grad_norm": 1.7520635112367258,
      "learning_rate": 3.934073022104759e-06,
      "loss": 0.5942,
      "mean_token_accuracy": 0.8051018618047238,
      "num_tokens": 664053021.0,
      "step": 560
    },
    {
      "epoch": 0.8029845443240362,
      "grad_norm": 2.747926156722073,
      "learning_rate": 3.807570348008672e-06,
      "loss": 0.5958,
      "mean_token_accuracy": 0.805073781311512,
      "num_tokens": 669989223.0,
      "step": 565
    },
    {
      "epoch": 0.8100906022384082,
      "grad_norm": 1.2910340620000669,
      "learning_rate": 3.684882962809484e-06,
      "loss": 0.6003,
      "mean_token_accuracy": 0.804193302989006,
      "num_tokens": 675921465.0,
      "step": 570
    },
    {
      "epoch": 0.8171966601527803,
      "grad_norm": 2.2556806715561817,
      "learning_rate": 3.5660759471324037e-06,
      "loss": 0.5971,
      "mean_token_accuracy": 0.8039252124726772,
      "num_tokens": 681852307.0,
      "step": 575
    },
    {
      "epoch": 0.8243027180671523,
      "grad_norm": 1.7373756069231372,
      "learning_rate": 3.451212323225786e-06,
      "loss": 0.5925,
      "mean_token_accuracy": 0.8057592801749707,
      "num_tokens": 687769737.0,
      "step": 580
    },
    {
      "epoch": 0.8314087759815243,
      "grad_norm": 3.192756357062325,
      "learning_rate": 3.340353021530409e-06,
      "loss": 0.5902,
      "mean_token_accuracy": 0.8047820582985878,
      "num_tokens": 693702256.0,
      "step": 585
    },
    {
      "epoch": 0.8385148338958962,
      "grad_norm": 2.3339162718389854,
      "learning_rate": 3.2335568483583708e-06,
      "loss": 0.5867,
      "mean_token_accuracy": 0.8085503794252873,
      "num_tokens": 699642883.0,
      "step": 590
    },
    {
      "epoch": 0.8456208918102682,
      "grad_norm": 1.7159764847482752,
      "learning_rate": 3.1308804546987615e-06,
      "loss": 0.5988,
      "mean_token_accuracy": 0.8035182796418667,
      "num_tokens": 705606507.0,
      "step": 595
    },
    {
      "epoch": 0.8527269497246402,
      "grad_norm": 2.8135588347439984,
      "learning_rate": 3.0323783061666307e-06,
      "loss": 0.5943,
      "mean_token_accuracy": 0.805966579914093,
      "num_tokens": 711533271.0,
      "step": 600
    },
    {
      "epoch": 0.8527269497246402,
      "eval_loss": 0.5903548002243042,
      "eval_mean_token_accuracy": 0.8059158665046358,
      "eval_num_tokens": 711533271.0,
      "eval_runtime": 296.5539,
      "eval_samples_per_second": 12.271,
      "eval_steps_per_second": 0.384,
      "step": 600
    },
    {
      "epoch": 0.8598330076390123,
      "grad_norm": 1.620026356285568,
      "learning_rate": 2.9381026541112145e-06,
      "loss": 0.5934,
      "mean_token_accuracy": 0.8051214568316937,
      "num_tokens": 717498751.0,
      "step": 605
    },
    {
      "epoch": 0.8669390655533843,
      "grad_norm": 1.346134723264547,
      "learning_rate": 2.848103507898745e-06,
      "loss": 0.5995,
      "mean_token_accuracy": 0.8026661708950996,
      "num_tokens": 723421175.0,
      "step": 610
    },
    {
      "epoch": 0.8740451234677563,
      "grad_norm": 1.3133061113418305,
      "learning_rate": 2.7624286083845187e-06,
      "loss": 0.5905,
      "mean_token_accuracy": 0.8046399556100369,
      "num_tokens": 729353013.0,
      "step": 615
    },
    {
      "epoch": 0.8811511813821282,
      "grad_norm": 1.7379818179026807,
      "learning_rate": 2.6811234025883457e-06,
      "loss": 0.5876,
      "mean_token_accuracy": 0.8071587301790715,
      "num_tokens": 735289319.0,
      "step": 620
    },
    {
      "epoch": 0.8882572392965002,
      "grad_norm": 4.0244229329833,
      "learning_rate": 2.604231019586761e-06,
      "loss": 0.5908,
      "mean_token_accuracy": 0.8060196414589882,
      "num_tokens": 741217107.0,
      "step": 625
    },
    {
      "epoch": 0.8953632972108723,
      "grad_norm": 1.7622879809387222,
      "learning_rate": 2.5317922476348194e-06,
      "loss": 0.5952,
      "mean_token_accuracy": 0.8054036945104599,
      "num_tokens": 747147403.0,
      "step": 630
    },
    {
      "epoch": 0.9024693551252443,
      "grad_norm": 2.2585412448119007,
      "learning_rate": 2.4638455125296043e-06,
      "loss": 0.5925,
      "mean_token_accuracy": 0.8054537117481232,
      "num_tokens": 753044605.0,
      "step": 635
    },
    {
      "epoch": 0.9095754130396163,
      "grad_norm": 1.5262303547489815,
      "learning_rate": 2.400426857226914e-06,
      "loss": 0.5879,
      "mean_token_accuracy": 0.8063779830932617,
      "num_tokens": 758944457.0,
      "step": 640
    },
    {
      "epoch": 0.9166814709539882,
      "grad_norm": 2.545493717270523,
      "learning_rate": 2.3415699227219517e-06,
      "loss": 0.5992,
      "mean_token_accuracy": 0.8029563590884209,
      "num_tokens": 764871608.0,
      "step": 645
    },
    {
      "epoch": 0.9237875288683602,
      "grad_norm": 1.7376685867634216,
      "learning_rate": 2.2873059302041627e-06,
      "loss": 0.5896,
      "mean_token_accuracy": 0.8063512444496155,
      "num_tokens": 770802003.0,
      "step": 650
    },
    {
      "epoch": 0.9308935867827323,
      "grad_norm": 1.7377001386795896,
      "learning_rate": 2.2376636644956656e-06,
      "loss": 0.5874,
      "mean_token_accuracy": 0.8068042829632759,
      "num_tokens": 776734106.0,
      "step": 655
    },
    {
      "epoch": 0.9379996446971043,
      "grad_norm": 3.942139353532877,
      "learning_rate": 2.192669458782096e-06,
      "loss": 0.5952,
      "mean_token_accuracy": 0.8051416061818599,
      "num_tokens": 782648208.0,
      "step": 660
    },
    {
      "epoch": 0.9451057026114763,
      "grad_norm": 2.1646964014669794,
      "learning_rate": 2.1523471806439205e-06,
      "loss": 0.6057,
      "mean_token_accuracy": 0.8031233668327331,
      "num_tokens": 788582279.0,
      "step": 665
    },
    {
      "epoch": 0.9522117605258483,
      "grad_norm": 2.4342297711554393,
      "learning_rate": 2.1167182193956738e-06,
      "loss": 0.5833,
      "mean_token_accuracy": 0.8084848992526531,
      "num_tokens": 794490250.0,
      "step": 670
    },
    {
      "epoch": 0.9593178184402202,
      "grad_norm": 1.3849504055994097,
      "learning_rate": 2.0858014747397952e-06,
      "loss": 0.5927,
      "mean_token_accuracy": 0.8059051290154458,
      "num_tokens": 800406765.0,
      "step": 675
    },
    {
      "epoch": 0.9664238763545923,
      "grad_norm": 1.8233541738404475,
      "learning_rate": 2.0596133467411213e-06,
      "loss": 0.5866,
      "mean_token_accuracy": 0.8072662524878979,
      "num_tokens": 806349672.0,
      "step": 680
    },
    {
      "epoch": 0.9735299342689643,
      "grad_norm": 1.3071754875862744,
      "learning_rate": 2.0381677271273177e-06,
      "loss": 0.589,
      "mean_token_accuracy": 0.8057045452296734,
      "num_tokens": 812264222.0,
      "step": 685
    },
    {
      "epoch": 0.9806359921833363,
      "grad_norm": 1.5289098269363515,
      "learning_rate": 2.0214759919198904e-06,
      "loss": 0.5831,
      "mean_token_accuracy": 0.8068839557468891,
      "num_tokens": 818190520.0,
      "step": 690
    },
    {
      "epoch": 0.9877420500977083,
      "grad_norm": 2.015373930862521,
      "learning_rate": 2.0095469953996724e-06,
      "loss": 0.5994,
      "mean_token_accuracy": 0.8041414134204388,
      "num_tokens": 824098000.0,
      "step": 695
    },
    {
      "epoch": 0.9948481080120803,
      "grad_norm": 2.2460839306695104,
      "learning_rate": 2.002387065409989e-06,
      "loss": 0.5962,
      "mean_token_accuracy": 0.8041631534695626,
      "num_tokens": 830080703.0,
      "step": 700
    },
    {
      "epoch": 0.9948481080120803,
      "eval_loss": 0.5880739688873291,
      "eval_mean_token_accuracy": 0.8062155152622023,
      "eval_num_tokens": 830080703.0,
      "eval_runtime": 296.6258,
      "eval_samples_per_second": 12.268,
      "eval_steps_per_second": 0.384,
      "step": 700
    },
    {
      "epoch": 1.0,
      "step": 704,
      "total_flos": 1.2125984924893184e+16,
      "train_loss": 0.0,
      "train_runtime": 1.6574,
      "train_samples_per_second": 217359.421,
      "train_steps_per_second": 424.761
    }
  ],
  "logging_steps": 5,
  "max_steps": 704,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2125984924893184e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}