{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8479067302596714,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010598834128245893,
      "grad_norm": 24.94797134399414,
      "learning_rate": 2.162496240601504e-06,
      "loss": 0.3988,
      "mean_token_accuracy": 0.8821502715349198,
      "num_tokens": 758934.0,
      "step": 5
    },
    {
      "epoch": 0.021197668256491786,
      "grad_norm": 3.5783743858337402,
      "learning_rate": 4.865616541353383e-06,
      "loss": 0.1951,
      "mean_token_accuracy": 0.9503676801919937,
      "num_tokens": 1515405.0,
      "step": 10
    },
    {
      "epoch": 0.03179650238473768,
      "grad_norm": 1.2595021724700928,
      "learning_rate": 7.568736842105265e-06,
      "loss": 0.1331,
      "mean_token_accuracy": 0.9593322277069092,
      "num_tokens": 2274569.0,
      "step": 15
    },
    {
      "epoch": 0.04239533651298357,
      "grad_norm": 0.8977835178375244,
      "learning_rate": 1.0271857142857143e-05,
      "loss": 0.117,
      "mean_token_accuracy": 0.9656193614006042,
      "num_tokens": 3025921.0,
      "step": 20
    },
    {
      "epoch": 0.052994170641229466,
      "grad_norm": 0.962152361869812,
      "learning_rate": 1.2974977443609024e-05,
      "loss": 0.1346,
      "mean_token_accuracy": 0.9613782346248627,
      "num_tokens": 3773329.0,
      "step": 25
    },
    {
      "epoch": 0.06359300476947535,
      "grad_norm": 1.1860769987106323,
      "learning_rate": 1.5678097744360906e-05,
      "loss": 0.1274,
      "mean_token_accuracy": 0.9628078043460846,
      "num_tokens": 4531567.0,
      "step": 30
    },
    {
      "epoch": 0.07419183889772125,
      "grad_norm": 0.9026021361351013,
      "learning_rate": 1.8381218045112784e-05,
      "loss": 0.1157,
      "mean_token_accuracy": 0.9642972320318222,
      "num_tokens": 5290076.0,
      "step": 35
    },
    {
      "epoch": 0.08479067302596714,
      "grad_norm": 0.8118496537208557,
      "learning_rate": 1.891889509687966e-05,
      "loss": 0.0866,
      "mean_token_accuracy": 0.9727037996053696,
      "num_tokens": 6035596.0,
      "step": 40
    },
    {
      "epoch": 0.09538950715421304,
      "grad_norm": 0.8422428965568542,
      "learning_rate": 1.89069270706161e-05,
      "loss": 0.1146,
      "mean_token_accuracy": 0.9646131128072739,
      "num_tokens": 6780221.0,
      "step": 45
    },
    {
      "epoch": 0.10598834128245893,
      "grad_norm": 0.7775574326515198,
      "learning_rate": 1.888576935902918e-05,
      "loss": 0.0965,
      "mean_token_accuracy": 0.9704228937625885,
      "num_tokens": 7535930.0,
      "step": 50
    },
    {
      "epoch": 0.11658717541070482,
      "grad_norm": 0.9478778839111328,
      "learning_rate": 1.88554494213576e-05,
      "loss": 0.1195,
      "mean_token_accuracy": 0.962310528755188,
      "num_tokens": 8281705.0,
      "step": 55
    },
    {
      "epoch": 0.1271860095389507,
      "grad_norm": 0.7263641357421875,
      "learning_rate": 1.8816006607906446e-05,
      "loss": 0.099,
      "mean_token_accuracy": 0.9673406720161438,
      "num_tokens": 9031160.0,
      "step": 60
    },
    {
      "epoch": 0.13778484366719662,
      "grad_norm": 1.1775802373886108,
      "learning_rate": 1.8767492108976963e-05,
      "loss": 0.1018,
      "mean_token_accuracy": 0.9681402653455734,
      "num_tokens": 9779765.0,
      "step": 65
    },
    {
      "epoch": 0.1483836777954425,
      "grad_norm": 0.5858120322227478,
      "learning_rate": 1.8709968888429915e-05,
      "loss": 0.0876,
      "mean_token_accuracy": 0.9693532943725586,
      "num_tokens": 10524281.0,
      "step": 70
    },
    {
      "epoch": 0.1589825119236884,
      "grad_norm": 0.6502286791801453,
      "learning_rate": 1.8643511601968855e-05,
      "loss": 0.1153,
      "mean_token_accuracy": 0.9650196969509125,
      "num_tokens": 11266684.0,
      "step": 75
    },
    {
      "epoch": 0.16958134605193428,
      "grad_norm": 0.4491387605667114,
      "learning_rate": 1.8568206500249235e-05,
      "loss": 0.1245,
      "mean_token_accuracy": 0.9640865415334702,
      "num_tokens": 12021422.0,
      "step": 80
    },
    {
      "epoch": 0.18018018018018017,
      "grad_norm": 0.7483686208724976,
      "learning_rate": 1.8484151316939235e-05,
      "loss": 0.1068,
      "mean_token_accuracy": 0.9680378794670105,
      "num_tokens": 12768213.0,
      "step": 85
    },
    {
      "epoch": 0.1907790143084261,
      "grad_norm": 0.534408450126648,
      "learning_rate": 1.8391455141877475e-05,
      "loss": 0.0942,
      "mean_token_accuracy": 0.968615373969078,
      "num_tokens": 13508156.0,
      "step": 90
    },
    {
      "epoch": 0.20137784843667197,
      "grad_norm": 0.4876060485839844,
      "learning_rate": 1.8290238279492314e-05,
      "loss": 0.0893,
      "mean_token_accuracy": 0.9708556532859802,
      "num_tokens": 14256282.0,
      "step": 95
    },
    {
      "epoch": 0.21197668256491786,
      "grad_norm": 0.45095881819725037,
      "learning_rate": 1.8180632092666422e-05,
      "loss": 0.0893,
      "mean_token_accuracy": 0.9683458656072617,
      "num_tokens": 15010215.0,
      "step": 100
    },
    {
      "epoch": 0.22257551669316375,
      "grad_norm": 0.5163909792900085,
      "learning_rate": 1.806277883224933e-05,
      "loss": 0.1052,
      "mean_token_accuracy": 0.967122933268547,
      "num_tokens": 15761644.0,
      "step": 105
    },
    {
      "epoch": 0.23317435082140964,
      "grad_norm": 0.5106090307235718,
      "learning_rate": 1.7936831452439144e-05,
      "loss": 0.0949,
      "mean_token_accuracy": 0.9696859061717987,
      "num_tokens": 16508299.0,
      "step": 110
    },
    {
      "epoch": 0.24377318494965552,
      "grad_norm": 0.46762996912002563,
      "learning_rate": 1.78029534122731e-05,
      "loss": 0.0779,
      "mean_token_accuracy": 0.9739178985357284,
      "num_tokens": 17258670.0,
      "step": 115
    },
    {
      "epoch": 0.2543720190779014,
      "grad_norm": 0.7346382737159729,
      "learning_rate": 1.7661318463484542e-05,
      "loss": 0.1008,
      "mean_token_accuracy": 0.9689333975315094,
      "num_tokens": 18009652.0,
      "step": 120
    },
    {
      "epoch": 0.2649708532061473,
      "grad_norm": 0.7018872499465942,
      "learning_rate": 1.7512110425001687e-05,
      "loss": 0.0777,
      "mean_token_accuracy": 0.9726225912570954,
      "num_tokens": 18763805.0,
      "step": 125
    },
    {
      "epoch": 0.27556968733439324,
      "grad_norm": 0.5918297171592712,
      "learning_rate": 1.735552294438079e-05,
      "loss": 0.0793,
      "mean_token_accuracy": 0.9710846275091172,
      "num_tokens": 19519084.0,
      "step": 130
    },
    {
      "epoch": 0.2861685214626391,
      "grad_norm": 0.48590484261512756,
      "learning_rate": 1.719175924648341e-05,
      "loss": 0.095,
      "mean_token_accuracy": 0.9694568663835526,
      "num_tokens": 20260881.0,
      "step": 135
    },
    {
      "epoch": 0.296767355590885,
      "grad_norm": 0.6686930060386658,
      "learning_rate": 1.702103186972387e-05,
      "loss": 0.1036,
      "mean_token_accuracy": 0.9665713518857956,
      "num_tokens": 21009526.0,
      "step": 140
    },
    {
      "epoch": 0.3073661897191309,
      "grad_norm": 0.544148862361908,
      "learning_rate": 1.6843562390229243e-05,
      "loss": 0.0945,
      "mean_token_accuracy": 0.9712516963481903,
      "num_tokens": 21757997.0,
      "step": 145
    },
    {
      "epoch": 0.3179650238473768,
      "grad_norm": 0.30081477761268616,
      "learning_rate": 1.665958113426993e-05,
      "loss": 0.0695,
      "mean_token_accuracy": 0.9746049761772155,
      "num_tokens": 22509571.0,
      "step": 150
    },
    {
      "epoch": 0.3285638579756227,
      "grad_norm": 0.5244784355163574,
      "learning_rate": 1.6469326879333894e-05,
      "loss": 0.0827,
      "mean_token_accuracy": 0.9729004025459289,
      "num_tokens": 23258193.0,
      "step": 155
    },
    {
      "epoch": 0.33916269210386857,
      "grad_norm": 0.5562810897827148,
      "learning_rate": 1.6273046544232703e-05,
      "loss": 0.0786,
      "mean_token_accuracy": 0.9723935961723328,
      "num_tokens": 24017620.0,
      "step": 160
    },
    {
      "epoch": 0.3497615262321145,
      "grad_norm": 0.3627864122390747,
      "learning_rate": 1.607099486864138e-05,
      "loss": 0.0892,
      "mean_token_accuracy": 0.9698468178510666,
      "num_tokens": 24767985.0,
      "step": 165
    },
    {
      "epoch": 0.36036036036036034,
      "grad_norm": 1.233094573020935,
      "learning_rate": 1.5863434082488125e-05,
      "loss": 0.0848,
      "mean_token_accuracy": 0.9692422747612,
      "num_tokens": 25516396.0,
      "step": 170
    },
    {
      "epoch": 0.37095919448860626,
      "grad_norm": 0.7834928631782532,
      "learning_rate": 1.5650633565622858e-05,
      "loss": 0.0866,
      "mean_token_accuracy": 0.9709006309509277,
      "num_tokens": 26268109.0,
      "step": 175
    },
    {
      "epoch": 0.3815580286168522,
      "grad_norm": 0.4577222764492035,
      "learning_rate": 1.5432869498206403e-05,
      "loss": 0.0842,
      "mean_token_accuracy": 0.9687849968671799,
      "num_tokens": 27011216.0,
      "step": 180
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 0.8089472651481628,
      "learning_rate": 1.5210424502273902e-05,
      "loss": 0.0969,
      "mean_token_accuracy": 0.9705709159374237,
      "num_tokens": 27748732.0,
      "step": 185
    },
    {
      "epoch": 0.40275569687334395,
      "grad_norm": 0.6289703845977783,
      "learning_rate": 1.4983587274937822e-05,
      "loss": 0.0974,
      "mean_token_accuracy": 0.9689016461372375,
      "num_tokens": 28499187.0,
      "step": 190
    },
    {
      "epoch": 0.4133545310015898,
      "grad_norm": 0.5271142721176147,
      "learning_rate": 1.4752652213706427e-05,
      "loss": 0.083,
      "mean_token_accuracy": 0.9725337207317353,
      "num_tokens": 29244962.0,
      "step": 195
    },
    {
      "epoch": 0.4239533651298357,
      "grad_norm": 0.5489286184310913,
      "learning_rate": 1.4517919034404131e-05,
      "loss": 0.1041,
      "mean_token_accuracy": 0.9659698814153671,
      "num_tokens": 29991318.0,
      "step": 200
    },
    {
      "epoch": 0.4345521992580816,
      "grad_norm": 0.5454921722412109,
      "learning_rate": 1.4279692382189522e-05,
      "loss": 0.0951,
      "mean_token_accuracy": 0.968726572394371,
      "num_tokens": 30737470.0,
      "step": 205
    },
    {
      "epoch": 0.4451510333863275,
      "grad_norm": 0.5300664901733398,
      "learning_rate": 1.4038281436175922e-05,
      "loss": 0.0878,
      "mean_token_accuracy": 0.9712866872549057,
      "num_tokens": 31490009.0,
      "step": 210
    },
    {
      "epoch": 0.4557498675145734,
      "grad_norm": 0.40223416686058044,
      "learning_rate": 1.3793999508167609e-05,
      "loss": 0.0699,
      "mean_token_accuracy": 0.9755781441926956,
      "num_tokens": 32246084.0,
      "step": 215
    },
    {
      "epoch": 0.4663487016428193,
      "grad_norm": 0.7377273440361023,
      "learning_rate": 1.3547163636032515e-05,
      "loss": 0.0881,
      "mean_token_accuracy": 0.9704002678394318,
      "num_tokens": 32988513.0,
      "step": 220
    },
    {
      "epoch": 0.4769475357710652,
      "grad_norm": 0.48401543498039246,
      "learning_rate": 1.3298094172239069e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9755650013685226,
      "num_tokens": 33740877.0,
      "step": 225
    },
    {
      "epoch": 0.48754636989931105,
      "grad_norm": 0.48549309372901917,
      "learning_rate": 1.304711436809124e-05,
      "loss": 0.0745,
      "mean_token_accuracy": 0.9718391954898834,
      "num_tokens": 34491295.0,
      "step": 230
    },
    {
      "epoch": 0.49814520402755696,
      "grad_norm": 0.593839704990387,
      "learning_rate": 1.2794549954201399e-05,
      "loss": 0.0856,
      "mean_token_accuracy": 0.9719352841377258,
      "num_tokens": 35243143.0,
      "step": 235
    },
    {
      "epoch": 0.5087440381558028,
      "grad_norm": 0.5269816517829895,
      "learning_rate": 1.2540728717745402e-05,
      "loss": 0.0852,
      "mean_token_accuracy": 0.9718370169401169,
      "num_tokens": 35994968.0,
      "step": 240
    },
    {
      "epoch": 0.5193428722840487,
      "grad_norm": 0.7908631563186646,
      "learning_rate": 1.2285980077048606e-05,
      "loss": 0.1059,
      "mean_token_accuracy": 0.9660630851984024,
      "num_tokens": 36746487.0,
      "step": 245
    },
    {
      "epoch": 0.5299417064122947,
      "grad_norm": 0.7140752077102661,
      "learning_rate": 1.2030634654054938e-05,
      "loss": 0.0968,
      "mean_token_accuracy": 0.9680245071649551,
      "num_tokens": 37493043.0,
      "step": 250
    },
    {
      "epoch": 0.5405405405405406,
      "grad_norm": 0.579908013343811,
      "learning_rate": 1.1775023845233844e-05,
      "loss": 0.1078,
      "mean_token_accuracy": 0.9634087413549424,
      "num_tokens": 38247808.0,
      "step": 255
    },
    {
      "epoch": 0.5511393746687865,
      "grad_norm": 0.5308011770248413,
      "learning_rate": 1.151947939148203e-05,
      "loss": 0.0683,
      "mean_token_accuracy": 0.9758787631988526,
      "num_tokens": 38995447.0,
      "step": 260
    },
    {
      "epoch": 0.5617382087970323,
      "grad_norm": 0.48115476965904236,
      "learning_rate": 1.1264332947578215e-05,
      "loss": 0.0895,
      "mean_token_accuracy": 0.9713383793830872,
      "num_tokens": 39737011.0,
      "step": 265
    },
    {
      "epoch": 0.5723370429252782,
      "grad_norm": 0.4850735366344452,
      "learning_rate": 1.1009915651749629e-05,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9750222563743591,
      "num_tokens": 40490523.0,
      "step": 270
    },
    {
      "epoch": 0.5829358770535241,
      "grad_norm": 0.4546583890914917,
      "learning_rate": 1.0756557695908914e-05,
      "loss": 0.0731,
      "mean_token_accuracy": 0.9746813982725143,
      "num_tokens": 41234962.0,
      "step": 275
    },
    {
      "epoch": 0.59353471118177,
      "grad_norm": 0.37385496497154236,
      "learning_rate": 1.0504587897119192e-05,
      "loss": 0.0684,
      "mean_token_accuracy": 0.9749377816915512,
      "num_tokens": 41985881.0,
      "step": 280
    },
    {
      "epoch": 0.604133545310016,
      "grad_norm": 0.5346734523773193,
      "learning_rate": 1.0254333270843464e-05,
      "loss": 0.0763,
      "mean_token_accuracy": 0.9734444499015809,
      "num_tokens": 42732592.0,
      "step": 285
    },
    {
      "epoch": 0.6147323794382618,
      "grad_norm": 0.4579368531703949,
      "learning_rate": 1.0006118606532189e-05,
      "loss": 0.0958,
      "mean_token_accuracy": 0.9702183693647385,
      "num_tokens": 43476258.0,
      "step": 290
    },
    {
      "epoch": 0.6253312135665077,
      "grad_norm": 0.5687949657440186,
      "learning_rate": 9.760266046099837e-06,
      "loss": 0.0962,
      "mean_token_accuracy": 0.9674538969993591,
      "num_tokens": 44231340.0,
      "step": 295
    },
    {
      "epoch": 0.6359300476947536,
      "grad_norm": 0.47385069727897644,
      "learning_rate": 9.51709466583757e-06,
      "loss": 0.0728,
      "mean_token_accuracy": 0.975405541062355,
      "num_tokens": 44984728.0,
      "step": 300
    },
    {
      "epoch": 0.6465288818229995,
      "grad_norm": 0.5263253450393677,
      "learning_rate": 9.276920062304517e-06,
      "loss": 0.0843,
      "mean_token_accuracy": 0.972364267706871,
      "num_tokens": 45738975.0,
      "step": 305
    },
    {
      "epoch": 0.6571277159512454,
      "grad_norm": 0.43684136867523193,
      "learning_rate": 9.040053942735253e-06,
      "loss": 0.0986,
      "mean_token_accuracy": 0.9670877307653427,
      "num_tokens": 46483508.0,
      "step": 310
    },
    {
      "epoch": 0.6677265500794912,
      "grad_norm": 0.8255428671836853,
      "learning_rate": 8.806803720494912e-06,
      "loss": 0.0903,
      "mean_token_accuracy": 0.9708788752555847,
      "num_tokens": 47225297.0,
      "step": 315
    },
    {
      "epoch": 0.6783253842077371,
      "grad_norm": 0.4157679080963135,
      "learning_rate": 8.577472116107104e-06,
      "loss": 0.0767,
      "mean_token_accuracy": 0.9740481734275818,
      "num_tokens": 47975826.0,
      "step": 320
    },
    {
      "epoch": 0.688924218335983,
      "grad_norm": 0.5178333520889282,
      "learning_rate": 8.3523567643723e-06,
      "loss": 0.0784,
      "mean_token_accuracy": 0.9755507856607437,
      "num_tokens": 48726601.0,
      "step": 325
    },
    {
      "epoch": 0.699523052464229,
      "grad_norm": 0.6876512765884399,
      "learning_rate": 8.131749828086715e-06,
      "loss": 0.1076,
      "mean_token_accuracy": 0.9637717425823211,
      "num_tokens": 49481248.0,
      "step": 330
    },
    {
      "epoch": 0.7101218865924749,
      "grad_norm": 0.6507133841514587,
      "learning_rate": 7.915937618862925e-06,
      "loss": 0.1007,
      "mean_token_accuracy": 0.9670225590467453,
      "num_tokens": 50235517.0,
      "step": 335
    },
    {
      "epoch": 0.7207207207207207,
      "grad_norm": 0.48040467500686646,
      "learning_rate": 7.705200225544349e-06,
      "loss": 0.0742,
      "mean_token_accuracy": 0.9743717759847641,
      "num_tokens": 50988614.0,
      "step": 340
    },
    {
      "epoch": 0.7313195548489666,
      "grad_norm": 0.6268354058265686,
      "learning_rate": 7.499811150695894e-06,
      "loss": 0.0805,
      "mean_token_accuracy": 0.9720939308404922,
      "num_tokens": 51744083.0,
      "step": 345
    },
    {
      "epoch": 0.7419183889772125,
      "grad_norm": 0.5063605904579163,
      "learning_rate": 7.300036955642486e-06,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9762273490428924,
      "num_tokens": 52498859.0,
      "step": 350
    },
    {
      "epoch": 0.7525172231054584,
      "grad_norm": 0.6706928014755249,
      "learning_rate": 7.106136914516202e-06,
      "loss": 0.079,
      "mean_token_accuracy": 0.9714215576648713,
      "num_tokens": 53241304.0,
      "step": 355
    },
    {
      "epoch": 0.7631160572337043,
      "grad_norm": 0.6424875259399414,
      "learning_rate": 6.91836267776098e-06,
      "loss": 0.0619,
      "mean_token_accuracy": 0.9780495315790176,
      "num_tokens": 53980433.0,
      "step": 360
    },
    {
      "epoch": 0.7737148913619502,
      "grad_norm": 0.6865656971931458,
      "learning_rate": 6.736957945531631e-06,
      "loss": 0.0787,
      "mean_token_accuracy": 0.973663604259491,
      "num_tokens": 54727968.0,
      "step": 365
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.6949437856674194,
      "learning_rate": 6.562158151411012e-06,
      "loss": 0.074,
      "mean_token_accuracy": 0.9744630962610245,
      "num_tokens": 55478691.0,
      "step": 370
    },
    {
      "epoch": 0.794912559618442,
      "grad_norm": 0.5605303645133972,
      "learning_rate": 6.394190156855855e-06,
      "loss": 0.0751,
      "mean_token_accuracy": 0.9737562090158463,
      "num_tokens": 56231131.0,
      "step": 375
    },
    {
      "epoch": 0.8055113937466879,
      "grad_norm": 0.5303736925125122,
      "learning_rate": 6.23327195676782e-06,
      "loss": 0.0814,
      "mean_token_accuracy": 0.972227942943573,
      "num_tokens": 56983848.0,
      "step": 380
    },
    {
      "epoch": 0.8161102278749337,
      "grad_norm": 0.5973182320594788,
      "learning_rate": 6.079612396571878e-06,
      "loss": 0.0819,
      "mean_token_accuracy": 0.9728082656860352,
      "num_tokens": 57738513.0,
      "step": 385
    },
    {
      "epoch": 0.8267090620031796,
      "grad_norm": 0.4760076403617859,
      "learning_rate": 5.933410901169198e-06,
      "loss": 0.0887,
      "mean_token_accuracy": 0.97094267308712,
      "num_tokens": 58482083.0,
      "step": 390
    },
    {
      "epoch": 0.8373078961314255,
      "grad_norm": 0.48814713954925537,
      "learning_rate": 5.794857216116354e-06,
      "loss": 0.0925,
      "mean_token_accuracy": 0.970235088467598,
      "num_tokens": 59236014.0,
      "step": 395
    },
    {
      "epoch": 0.8479067302596714,
      "grad_norm": 0.5259151458740234,
      "learning_rate": 5.66413116136671e-06,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9760579377412796,
      "num_tokens": 59973002.0,
      "step": 400
    }
  ],
  "logging_steps": 5,
  "max_steps": 471,
  "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": 3.5718615033823887e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}