{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9979123173277662,
  "eval_steps": 500,
  "global_step": 478,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010438413361169102,
      "grad_norm": 26.391578674316406,
      "learning_rate": 9.704121543300004e-06,
      "loss": 3.1296,
      "mean_token_accuracy": 0.5274454951286316,
      "num_tokens": 168029.0,
      "step": 5
    },
    {
      "epoch": 0.020876826722338204,
      "grad_norm": 2.0678582191467285,
      "learning_rate": 2.1834273472425008e-05,
      "loss": 0.5452,
      "mean_token_accuracy": 0.8023583650588989,
      "num_tokens": 336172.0,
      "step": 10
    },
    {
      "epoch": 0.031315240083507306,
      "grad_norm": 0.2009689062833786,
      "learning_rate": 3.3964425401550014e-05,
      "loss": 0.1935,
      "mean_token_accuracy": 0.8816015601158143,
      "num_tokens": 502617.0,
      "step": 15
    },
    {
      "epoch": 0.04175365344467641,
      "grad_norm": 0.252365380525589,
      "learning_rate": 4.6094577330675016e-05,
      "loss": 0.1771,
      "mean_token_accuracy": 0.8812037944793701,
      "num_tokens": 667062.0,
      "step": 20
    },
    {
      "epoch": 0.05219206680584551,
      "grad_norm": 0.12022583931684494,
      "learning_rate": 5.8224729259800026e-05,
      "loss": 0.1725,
      "mean_token_accuracy": 0.8987160563468933,
      "num_tokens": 836730.0,
      "step": 25
    },
    {
      "epoch": 0.06263048016701461,
      "grad_norm": 0.15564362704753876,
      "learning_rate": 7.035488118892503e-05,
      "loss": 0.1798,
      "mean_token_accuracy": 0.8868848085403442,
      "num_tokens": 1002377.0,
      "step": 30
    },
    {
      "epoch": 0.07306889352818371,
      "grad_norm": 0.09313672035932541,
      "learning_rate": 8.248503311805003e-05,
      "loss": 0.1748,
      "mean_token_accuracy": 0.8814776778221131,
      "num_tokens": 1169134.0,
      "step": 35
    },
    {
      "epoch": 0.08350730688935282,
      "grad_norm": 0.11390068382024765,
      "learning_rate": 8.489831115018415e-05,
      "loss": 0.1773,
      "mean_token_accuracy": 0.8929869174957276,
      "num_tokens": 1335296.0,
      "step": 40
    },
    {
      "epoch": 0.09394572025052192,
      "grad_norm": 0.11824589967727661,
      "learning_rate": 8.484652221903538e-05,
      "loss": 0.1816,
      "mean_token_accuracy": 0.8852614164352417,
      "num_tokens": 1502149.0,
      "step": 45
    },
    {
      "epoch": 0.10438413361169102,
      "grad_norm": 0.07780276983976364,
      "learning_rate": 8.475496445194093e-05,
      "loss": 0.1773,
      "mean_token_accuracy": 0.8905007004737854,
      "num_tokens": 1667759.0,
      "step": 50
    },
    {
      "epoch": 0.11482254697286012,
      "grad_norm": 0.052690450102090836,
      "learning_rate": 8.462375243282933e-05,
      "loss": 0.1758,
      "mean_token_accuracy": 0.891989552974701,
      "num_tokens": 1834499.0,
      "step": 55
    },
    {
      "epoch": 0.12526096033402923,
      "grad_norm": 0.13043910264968872,
      "learning_rate": 8.445305037265836e-05,
      "loss": 0.1827,
      "mean_token_accuracy": 0.88782799243927,
      "num_tokens": 2002043.0,
      "step": 60
    },
    {
      "epoch": 0.13569937369519833,
      "grad_norm": 0.11741600185632706,
      "learning_rate": 8.424307190390606e-05,
      "loss": 0.1785,
      "mean_token_accuracy": 0.8852924227714538,
      "num_tokens": 2166640.0,
      "step": 65
    },
    {
      "epoch": 0.14613778705636743,
      "grad_norm": 0.08231513947248459,
      "learning_rate": 8.399407981321114e-05,
      "loss": 0.1764,
      "mean_token_accuracy": 0.8852210521697998,
      "num_tokens": 2335194.0,
      "step": 70
    },
    {
      "epoch": 0.15657620041753653,
      "grad_norm": 0.06105300411581993,
      "learning_rate": 8.37063857124973e-05,
      "loss": 0.1775,
      "mean_token_accuracy": 0.8854288697242737,
      "num_tokens": 2503295.0,
      "step": 75
    },
    {
      "epoch": 0.16701461377870563,
      "grad_norm": 0.14096473157405853,
      "learning_rate": 8.3380349648993e-05,
      "loss": 0.1744,
      "mean_token_accuracy": 0.8955572128295899,
      "num_tokens": 2669896.0,
      "step": 80
    },
    {
      "epoch": 0.17745302713987474,
      "grad_norm": 0.09351199120283127,
      "learning_rate": 8.301637965463476e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.8935379385948181,
      "num_tokens": 2834340.0,
      "step": 85
    },
    {
      "epoch": 0.18789144050104384,
      "grad_norm": 0.056742168962955475,
      "learning_rate": 8.261493123541793e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.8908402442932128,
      "num_tokens": 3001552.0,
      "step": 90
    },
    {
      "epoch": 0.19832985386221294,
      "grad_norm": 0.07304312288761139,
      "learning_rate": 8.217650680133402e-05,
      "loss": 0.1762,
      "mean_token_accuracy": 0.8894022226333618,
      "num_tokens": 3168476.0,
      "step": 95
    },
    {
      "epoch": 0.20876826722338204,
      "grad_norm": 0.07051249593496323,
      "learning_rate": 8.170165503760803e-05,
      "loss": 0.1787,
      "mean_token_accuracy": 0.8864586472511291,
      "num_tokens": 3337021.0,
      "step": 100
    },
    {
      "epoch": 0.21920668058455114,
      "grad_norm": 0.041761696338653564,
      "learning_rate": 8.11909702180225e-05,
      "loss": 0.1787,
      "mean_token_accuracy": 0.8881038904190064,
      "num_tokens": 3501275.0,
      "step": 105
    },
    {
      "epoch": 0.22964509394572025,
      "grad_norm": 0.05747745558619499,
      "learning_rate": 8.064509146118796e-05,
      "loss": 0.1772,
      "mean_token_accuracy": 0.8871283888816833,
      "num_tokens": 3665322.0,
      "step": 110
    },
    {
      "epoch": 0.24008350730688935,
      "grad_norm": 0.08522920310497284,
      "learning_rate": 8.006470193069023e-05,
      "loss": 0.1775,
      "mean_token_accuracy": 0.8836086511611938,
      "num_tokens": 3828738.0,
      "step": 115
    },
    {
      "epoch": 0.25052192066805845,
      "grad_norm": 0.08257102966308594,
      "learning_rate": 7.945052798011581e-05,
      "loss": 0.1754,
      "mean_token_accuracy": 0.8929376482963562,
      "num_tokens": 3997471.0,
      "step": 120
    },
    {
      "epoch": 0.2609603340292276,
      "grad_norm": 0.09331203252077103,
      "learning_rate": 7.880333824402533e-05,
      "loss": 0.1733,
      "mean_token_accuracy": 0.8952223777770996,
      "num_tokens": 4163468.0,
      "step": 125
    },
    {
      "epoch": 0.27139874739039666,
      "grad_norm": 0.07619410753250122,
      "learning_rate": 7.812394267601246e-05,
      "loss": 0.1753,
      "mean_token_accuracy": 0.8914909482002258,
      "num_tokens": 4329641.0,
      "step": 130
    },
    {
      "epoch": 0.2818371607515658,
      "grad_norm": 0.13225223124027252,
      "learning_rate": 7.74131915350525e-05,
      "loss": 0.1723,
      "mean_token_accuracy": 0.8927483916282654,
      "num_tokens": 4496630.0,
      "step": 135
    },
    {
      "epoch": 0.29227557411273486,
      "grad_norm": 0.21763747930526733,
      "learning_rate": 7.667197432140909e-05,
      "loss": 0.1805,
      "mean_token_accuracy": 0.8857357025146484,
      "num_tokens": 4663350.0,
      "step": 140
    },
    {
      "epoch": 0.302713987473904,
      "grad_norm": 0.07913567870855331,
      "learning_rate": 7.590121866343044e-05,
      "loss": 0.1709,
      "mean_token_accuracy": 0.8957363724708557,
      "num_tokens": 4830451.0,
      "step": 145
    },
    {
      "epoch": 0.31315240083507306,
      "grad_norm": 0.07224448025226593,
      "learning_rate": 7.510188915662886e-05,
      "loss": 0.1805,
      "mean_token_accuracy": 0.8855878591537476,
      "num_tokens": 4997687.0,
      "step": 150
    },
    {
      "epoch": 0.3235908141962422,
      "grad_norm": 0.07714155316352844,
      "learning_rate": 7.427498615649591e-05,
      "loss": 0.178,
      "mean_token_accuracy": 0.883197569847107,
      "num_tokens": 5163538.0,
      "step": 155
    },
    {
      "epoch": 0.33402922755741127,
      "grad_norm": 0.10624797642230988,
      "learning_rate": 7.342154452656425e-05,
      "loss": 0.1776,
      "mean_token_accuracy": 0.8813526868820191,
      "num_tokens": 5328740.0,
      "step": 160
    },
    {
      "epoch": 0.3444676409185804,
      "grad_norm": 0.06445207446813583,
      "learning_rate": 7.25426323432829e-05,
      "loss": 0.1742,
      "mean_token_accuracy": 0.8918550252914429,
      "num_tokens": 5496305.0,
      "step": 165
    },
    {
      "epoch": 0.35490605427974947,
      "grad_norm": 0.0771849974989891,
      "learning_rate": 7.163934955932678e-05,
      "loss": 0.1697,
      "mean_token_accuracy": 0.8902682423591614,
      "num_tokens": 5666101.0,
      "step": 170
    },
    {
      "epoch": 0.3653444676409186,
      "grad_norm": 0.12977905571460724,
      "learning_rate": 7.071282662701331e-05,
      "loss": 0.1805,
      "mean_token_accuracy": 0.8896605968475342,
      "num_tokens": 5832805.0,
      "step": 175
    },
    {
      "epoch": 0.3757828810020877,
      "grad_norm": 0.10428526252508163,
      "learning_rate": 6.976422308354902e-05,
      "loss": 0.1745,
      "mean_token_accuracy": 0.9001333594322205,
      "num_tokens": 5997403.0,
      "step": 180
    },
    {
      "epoch": 0.3862212943632568,
      "grad_norm": 0.0962299257516861,
      "learning_rate": 6.879472609987647e-05,
      "loss": 0.1745,
      "mean_token_accuracy": 0.8933409571647644,
      "num_tokens": 6164320.0,
      "step": 185
    },
    {
      "epoch": 0.3966597077244259,
      "grad_norm": 0.06445040553808212,
      "learning_rate": 6.780554899493784e-05,
      "loss": 0.1633,
      "mean_token_accuracy": 0.902362859249115,
      "num_tokens": 6334749.0,
      "step": 190
    },
    {
      "epoch": 0.407098121085595,
      "grad_norm": 0.0730392336845398,
      "learning_rate": 6.679792971721452e-05,
      "loss": 0.1699,
      "mean_token_accuracy": 0.8901161074638366,
      "num_tokens": 6502501.0,
      "step": 195
    },
    {
      "epoch": 0.4175365344467641,
      "grad_norm": 0.09375865012407303,
      "learning_rate": 6.577312929544282e-05,
      "loss": 0.1744,
      "mean_token_accuracy": 0.8900301694869995,
      "num_tokens": 6670606.0,
      "step": 200
    },
    {
      "epoch": 0.4279749478079332,
      "grad_norm": 0.043036509305238724,
      "learning_rate": 6.473243026044507e-05,
      "loss": 0.1692,
      "mean_token_accuracy": 0.8939194440841675,
      "num_tokens": 6840912.0,
      "step": 205
    },
    {
      "epoch": 0.4384133611691023,
      "grad_norm": 0.055392514914274216,
      "learning_rate": 6.367713504005093e-05,
      "loss": 0.1699,
      "mean_token_accuracy": 0.8984872937202454,
      "num_tokens": 7007035.0,
      "step": 210
    },
    {
      "epoch": 0.4488517745302714,
      "grad_norm": 0.07093252241611481,
      "learning_rate": 6.26085643291177e-05,
      "loss": 0.17,
      "mean_token_accuracy": 0.897310209274292,
      "num_tokens": 7174655.0,
      "step": 215
    },
    {
      "epoch": 0.4592901878914405,
      "grad_norm": 0.08197320252656937,
      "learning_rate": 6.152805543668965e-05,
      "loss": 0.1675,
      "mean_token_accuracy": 0.904119336605072,
      "num_tokens": 7341565.0,
      "step": 220
    },
    {
      "epoch": 0.4697286012526096,
      "grad_norm": 0.06568682938814163,
      "learning_rate": 6.043696061236478e-05,
      "loss": 0.1725,
      "mean_token_accuracy": 0.8928499341011047,
      "num_tokens": 7509056.0,
      "step": 225
    },
    {
      "epoch": 0.4801670146137787,
      "grad_norm": 0.03814328834414482,
      "learning_rate": 5.933664535396346e-05,
      "loss": 0.178,
      "mean_token_accuracy": 0.8876104235649109,
      "num_tokens": 7677276.0,
      "step": 230
    },
    {
      "epoch": 0.4906054279749478,
      "grad_norm": 0.05092785879969597,
      "learning_rate": 5.822848669861723e-05,
      "loss": 0.1786,
      "mean_token_accuracy": 0.8900844812393188,
      "num_tokens": 7841210.0,
      "step": 235
    },
    {
      "epoch": 0.5010438413361169,
      "grad_norm": 0.042054884135723114,
      "learning_rate": 5.7113871499416016e-05,
      "loss": 0.1702,
      "mean_token_accuracy": 0.8916158556938172,
      "num_tokens": 8011338.0,
      "step": 240
    },
    {
      "epoch": 0.511482254697286,
      "grad_norm": 0.08291042596101761,
      "learning_rate": 5.5994194689770917e-05,
      "loss": 0.1692,
      "mean_token_accuracy": 0.8925593614578247,
      "num_tokens": 8177977.0,
      "step": 245
    },
    {
      "epoch": 0.5219206680584552,
      "grad_norm": 0.07634980976581573,
      "learning_rate": 5.487085753766434e-05,
      "loss": 0.177,
      "mean_token_accuracy": 0.8844507575035095,
      "num_tokens": 8345865.0,
      "step": 250
    },
    {
      "epoch": 0.5323590814196242,
      "grad_norm": 0.07123459130525589,
      "learning_rate": 5.374526589197267e-05,
      "loss": 0.1752,
      "mean_token_accuracy": 0.888279914855957,
      "num_tokens": 8511359.0,
      "step": 255
    },
    {
      "epoch": 0.5427974947807933,
      "grad_norm": 0.09457564353942871,
      "learning_rate": 5.261882842305581e-05,
      "loss": 0.1765,
      "mean_token_accuracy": 0.8875098109245301,
      "num_tokens": 8677713.0,
      "step": 260
    },
    {
      "epoch": 0.5532359081419624,
      "grad_norm": 0.05131667107343674,
      "learning_rate": 5.1492954859815675e-05,
      "loss": 0.1741,
      "mean_token_accuracy": 0.8929031252861023,
      "num_tokens": 8844996.0,
      "step": 265
    },
    {
      "epoch": 0.5636743215031316,
      "grad_norm": 0.0476628802716732,
      "learning_rate": 5.0369054225430114e-05,
      "loss": 0.1641,
      "mean_token_accuracy": 0.8998504161834717,
      "num_tokens": 9014325.0,
      "step": 270
    },
    {
      "epoch": 0.5741127348643006,
      "grad_norm": 0.04344228282570839,
      "learning_rate": 4.924853307396972e-05,
      "loss": 0.1728,
      "mean_token_accuracy": 0.8919055461883545,
      "num_tokens": 9184322.0,
      "step": 275
    },
    {
      "epoch": 0.5845511482254697,
      "grad_norm": 0.054395176470279694,
      "learning_rate": 4.8132793730105065e-05,
      "loss": 0.1682,
      "mean_token_accuracy": 0.894045627117157,
      "num_tokens": 9352926.0,
      "step": 280
    },
    {
      "epoch": 0.5949895615866388,
      "grad_norm": 0.0685492604970932,
      "learning_rate": 4.702323253410666e-05,
      "loss": 0.1732,
      "mean_token_accuracy": 0.89416663646698,
      "num_tokens": 9519005.0,
      "step": 285
    },
    {
      "epoch": 0.605427974947808,
      "grad_norm": 0.04440397769212723,
      "learning_rate": 4.5921238094334656e-05,
      "loss": 0.1775,
      "mean_token_accuracy": 0.8938532590866088,
      "num_tokens": 9684137.0,
      "step": 290
    },
    {
      "epoch": 0.615866388308977,
      "grad_norm": 0.08006265759468079,
      "learning_rate": 4.482818954940473e-05,
      "loss": 0.1751,
      "mean_token_accuracy": 0.8890296816825867,
      "num_tokens": 9848337.0,
      "step": 295
    },
    {
      "epoch": 0.6263048016701461,
      "grad_norm": 0.06337990611791611,
      "learning_rate": 4.374545484220546e-05,
      "loss": 0.172,
      "mean_token_accuracy": 0.8964735388755798,
      "num_tokens": 10013375.0,
      "step": 300
    },
    {
      "epoch": 0.6367432150313153,
      "grad_norm": 0.06805319339036942,
      "learning_rate": 4.2674389007926984e-05,
      "loss": 0.1743,
      "mean_token_accuracy": 0.891275691986084,
      "num_tokens": 10178365.0,
      "step": 305
    },
    {
      "epoch": 0.6471816283924844,
      "grad_norm": 0.08113426715135574,
      "learning_rate": 4.1616332478243656e-05,
      "loss": 0.1705,
      "mean_token_accuracy": 0.890809166431427,
      "num_tokens": 10344289.0,
      "step": 310
    },
    {
      "epoch": 0.6576200417536534,
      "grad_norm": 0.049595121294260025,
      "learning_rate": 4.057260940377274e-05,
      "loss": 0.1718,
      "mean_token_accuracy": 0.8930620074272155,
      "num_tokens": 10511152.0,
      "step": 315
    },
    {
      "epoch": 0.6680584551148225,
      "grad_norm": 0.0932910367846489,
      "learning_rate": 3.9544525996908924e-05,
      "loss": 0.1695,
      "mean_token_accuracy": 0.8938508629798889,
      "num_tokens": 10678699.0,
      "step": 320
    },
    {
      "epoch": 0.6784968684759917,
      "grad_norm": 0.06598310172557831,
      "learning_rate": 3.853336889710825e-05,
      "loss": 0.175,
      "mean_token_accuracy": 0.893925416469574,
      "num_tokens": 10846204.0,
      "step": 325
    },
    {
      "epoch": 0.6889352818371608,
      "grad_norm": 0.03508803993463516,
      "learning_rate": 3.7540403560667546e-05,
      "loss": 0.1693,
      "mean_token_accuracy": 0.8978321552276611,
      "num_tokens": 11012783.0,
      "step": 330
    },
    {
      "epoch": 0.6993736951983298,
      "grad_norm": 0.07559879869222641,
      "learning_rate": 3.656687267701434e-05,
      "loss": 0.1731,
      "mean_token_accuracy": 0.8905814051628113,
      "num_tokens": 11180472.0,
      "step": 335
    },
    {
      "epoch": 0.7098121085594989,
      "grad_norm": 0.04261172190308571,
      "learning_rate": 3.56139946134895e-05,
      "loss": 0.1732,
      "mean_token_accuracy": 0.890303087234497,
      "num_tokens": 11346372.0,
      "step": 340
    },
    {
      "epoch": 0.7202505219206681,
      "grad_norm": 0.0686853900551796,
      "learning_rate": 3.468296189056864e-05,
      "loss": 0.1732,
      "mean_token_accuracy": 0.8982097268104553,
      "num_tokens": 11513828.0,
      "step": 345
    },
    {
      "epoch": 0.7306889352818372,
      "grad_norm": 0.03751448541879654,
      "learning_rate": 3.377493968943082e-05,
      "loss": 0.1655,
      "mean_token_accuracy": 0.8962634086608887,
      "num_tokens": 11681244.0,
      "step": 350
    },
    {
      "epoch": 0.7411273486430062,
      "grad_norm": 0.09118922799825668,
      "learning_rate": 3.289106439374217e-05,
      "loss": 0.1757,
      "mean_token_accuracy": 0.8874221444129944,
      "num_tokens": 11845089.0,
      "step": 355
    },
    {
      "epoch": 0.7515657620041754,
      "grad_norm": 0.05042296275496483,
      "learning_rate": 3.20324421674793e-05,
      "loss": 0.1729,
      "mean_token_accuracy": 0.8906779885292053,
      "num_tokens": 12012657.0,
      "step": 360
    },
    {
      "epoch": 0.7620041753653445,
      "grad_norm": 0.08142732828855515,
      "learning_rate": 3.120014757057265e-05,
      "loss": 0.1759,
      "mean_token_accuracy": 0.8841803073883057,
      "num_tokens": 12178933.0,
      "step": 365
    },
    {
      "epoch": 0.7724425887265136,
      "grad_norm": 0.054538823664188385,
      "learning_rate": 3.039522221410191e-05,
      "loss": 0.1671,
      "mean_token_accuracy": 0.9001896619796753,
      "num_tokens": 12346020.0,
      "step": 370
    },
    {
      "epoch": 0.7828810020876826,
      "grad_norm": 0.0754442811012268,
      "learning_rate": 2.9618673456726793e-05,
      "loss": 0.1697,
      "mean_token_accuracy": 0.8957448601722717,
      "num_tokens": 12513307.0,
      "step": 375
    },
    {
      "epoch": 0.7933194154488518,
      "grad_norm": 0.05027084797620773,
      "learning_rate": 2.887147314398446e-05,
      "loss": 0.171,
      "mean_token_accuracy": 0.9047865271568298,
      "num_tokens": 12683268.0,
      "step": 380
    },
    {
      "epoch": 0.8037578288100209,
      "grad_norm": 0.037379130721092224,
      "learning_rate": 2.815455639203135e-05,
      "loss": 0.1673,
      "mean_token_accuracy": 0.8971027612686158,
      "num_tokens": 12849541.0,
      "step": 385
    },
    {
      "epoch": 0.81419624217119,
      "grad_norm": 0.06629011780023575,
      "learning_rate": 2.746882041735168e-05,
      "loss": 0.1708,
      "mean_token_accuracy": 0.9001821756362915,
      "num_tokens": 13018604.0,
      "step": 390
    },
    {
      "epoch": 0.824634655532359,
      "grad_norm": 0.03887145593762398,
      "learning_rate": 2.6815123413896988e-05,
      "loss": 0.178,
      "mean_token_accuracy": 0.890343713760376,
      "num_tokens": 13183578.0,
      "step": 395
    },
    {
      "epoch": 0.8350730688935282,
      "grad_norm": 0.05481924116611481,
      "learning_rate": 2.6194283479062222e-05,
      "loss": 0.1708,
      "mean_token_accuracy": 0.8925650119781494,
      "num_tokens": 13351564.0,
      "step": 400
    },
    {
      "epoch": 0.8455114822546973,
      "grad_norm": 0.06296750158071518,
      "learning_rate": 2.5607077589842365e-05,
      "loss": 0.1675,
      "mean_token_accuracy": 0.8997923254966735,
      "num_tokens": 13517204.0,
      "step": 405
    },
    {
      "epoch": 0.8559498956158664,
      "grad_norm": 0.037203434854745865,
      "learning_rate": 2.5054240630450926e-05,
      "loss": 0.1714,
      "mean_token_accuracy": 0.893981146812439,
      "num_tokens": 13682958.0,
      "step": 410
    },
    {
      "epoch": 0.8663883089770354,
      "grad_norm": 0.056684475392103195,
      "learning_rate": 2.4536464472617357e-05,
      "loss": 0.1749,
      "mean_token_accuracy": 0.89893057346344,
      "num_tokens": 13846601.0,
      "step": 415
    },
    {
      "epoch": 0.8768267223382046,
      "grad_norm": 0.044169504195451736,
      "learning_rate": 2.4054397109714206e-05,
      "loss": 0.172,
      "mean_token_accuracy": 0.8983666896820068,
      "num_tokens": 14010731.0,
      "step": 420
    },
    {
      "epoch": 0.8872651356993737,
      "grad_norm": 0.04433104768395424,
      "learning_rate": 2.3608641845797782e-05,
      "loss": 0.1774,
      "mean_token_accuracy": 0.8897690415382385,
      "num_tokens": 14179072.0,
      "step": 425
    },
    {
      "epoch": 0.8977035490605428,
      "grad_norm": 0.051063422113657,
      "learning_rate": 2.3199756540577187e-05,
      "loss": 0.1741,
      "mean_token_accuracy": 0.8910220861434937,
      "num_tokens": 14345417.0,
      "step": 430
    },
    {
      "epoch": 0.9081419624217119,
      "grad_norm": 0.06690821051597595,
      "learning_rate": 2.2828252911256678e-05,
      "loss": 0.1765,
      "mean_token_accuracy": 0.8882366180419922,
      "num_tokens": 14510966.0,
      "step": 435
    },
    {
      "epoch": 0.918580375782881,
      "grad_norm": 0.06864584237337112,
      "learning_rate": 2.2494595892125003e-05,
      "loss": 0.1761,
      "mean_token_accuracy": 0.8908858180046082,
      "num_tokens": 14677608.0,
      "step": 440
    },
    {
      "epoch": 0.9290187891440501,
      "grad_norm": 0.05033930391073227,
      "learning_rate": 2.21992030526933e-05,
      "loss": 0.1728,
      "mean_token_accuracy": 0.887229859828949,
      "num_tokens": 14845430.0,
      "step": 445
    },
    {
      "epoch": 0.9394572025052192,
      "grad_norm": 0.041298288851976395,
      "learning_rate": 2.1942444075109632e-05,
      "loss": 0.1768,
      "mean_token_accuracy": 0.8911224722862243,
      "num_tokens": 15010392.0,
      "step": 450
    },
    {
      "epoch": 0.9498956158663883,
      "grad_norm": 0.043606266379356384,
      "learning_rate": 2.1724640291504296e-05,
      "loss": 0.1821,
      "mean_token_accuracy": 0.8849847674369812,
      "num_tokens": 15173861.0,
      "step": 455
    },
    {
      "epoch": 0.9603340292275574,
      "grad_norm": 0.05044990032911301,
      "learning_rate": 2.1546064281844835e-05,
      "loss": 0.1755,
      "mean_token_accuracy": 0.895173990726471,
      "num_tokens": 15340411.0,
      "step": 460
    },
    {
      "epoch": 0.9707724425887265,
      "grad_norm": 0.05344529077410698,
      "learning_rate": 2.1406939532804015e-05,
      "loss": 0.1759,
      "mean_token_accuracy": 0.8931485772132873,
      "num_tokens": 15505974.0,
      "step": 465
    },
    {
      "epoch": 0.9812108559498957,
      "grad_norm": 0.05700814723968506,
      "learning_rate": 2.130744015806778e-05,
      "loss": 0.1748,
      "mean_token_accuracy": 0.8841968178749084,
      "num_tokens": 15673298.0,
      "step": 470
    },
    {
      "epoch": 0.9916492693110647,
      "grad_norm": 0.041500721126794815,
      "learning_rate": 2.1247690680433146e-05,
      "loss": 0.1709,
      "mean_token_accuracy": 0.8919184803962708,
      "num_tokens": 15839500.0,
      "step": 475
    }
  ],
  "logging_steps": 5,
  "max_steps": 479,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.625292560472474e+17,
  "train_batch_size": 28,
  "trial_name": null,
  "trial_params": null
}