{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 625,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0016,
      "grad_norm": 0.6777933835983276,
      "learning_rate": 3.1746031746031746e-06,
      "loss": 1.4507,
      "mean_token_accuracy": 0.7135298848152161,
      "step": 1
    },
    {
      "epoch": 0.008,
      "grad_norm": 0.6817252039909363,
      "learning_rate": 1.5873015873015872e-05,
      "loss": 1.4573,
      "mean_token_accuracy": 0.7123897969722748,
      "step": 5
    },
    {
      "epoch": 0.016,
      "grad_norm": 0.6862361431121826,
      "learning_rate": 3.1746031746031745e-05,
      "loss": 1.4579,
      "mean_token_accuracy": 0.7112078011035919,
      "step": 10
    },
    {
      "epoch": 0.024,
      "grad_norm": 0.6690642237663269,
      "learning_rate": 4.761904761904762e-05,
      "loss": 1.3632,
      "mean_token_accuracy": 0.7240169644355774,
      "step": 15
    },
    {
      "epoch": 0.032,
      "grad_norm": 0.7569957375526428,
      "learning_rate": 6.349206349206349e-05,
      "loss": 1.185,
      "mean_token_accuracy": 0.741723358631134,
      "step": 20
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.6911544799804688,
      "learning_rate": 7.936507936507937e-05,
      "loss": 0.9249,
      "mean_token_accuracy": 0.7771691501140594,
      "step": 25
    },
    {
      "epoch": 0.048,
      "grad_norm": 0.8835947513580322,
      "learning_rate": 9.523809523809524e-05,
      "loss": 0.6314,
      "mean_token_accuracy": 0.8349982738494873,
      "step": 30
    },
    {
      "epoch": 0.056,
      "grad_norm": 0.5303900241851807,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.3246,
      "mean_token_accuracy": 0.9103669762611389,
      "step": 35
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.3050856292247772,
      "learning_rate": 0.00012698412698412698,
      "loss": 0.145,
      "mean_token_accuracy": 0.9529362678527832,
      "step": 40
    },
    {
      "epoch": 0.072,
      "grad_norm": 0.1506454050540924,
      "learning_rate": 0.00014285714285714287,
      "loss": 0.0887,
      "mean_token_accuracy": 0.9687770783901215,
      "step": 45
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.09974420815706253,
      "learning_rate": 0.00015873015873015873,
      "loss": 0.0805,
      "mean_token_accuracy": 0.9694386482238769,
      "step": 50
    },
    {
      "epoch": 0.088,
      "grad_norm": 0.12219829857349396,
      "learning_rate": 0.00017460317460317462,
      "loss": 0.0855,
      "mean_token_accuracy": 0.9688551783561706,
      "step": 55
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.08872757107019424,
      "learning_rate": 0.00019047619047619048,
      "loss": 0.077,
      "mean_token_accuracy": 0.9706803619861603,
      "step": 60
    },
    {
      "epoch": 0.104,
      "grad_norm": 0.07106012105941772,
      "learning_rate": 0.00019999375039475277,
      "loss": 0.0763,
      "mean_token_accuracy": 0.970524275302887,
      "step": 65
    },
    {
      "epoch": 0.112,
      "grad_norm": 0.07300037890672684,
      "learning_rate": 0.0001999234513064475,
      "loss": 0.0714,
      "mean_token_accuracy": 0.9725041687488556,
      "step": 70
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.08171433955430984,
      "learning_rate": 0.00019977509622105233,
      "loss": 0.0741,
      "mean_token_accuracy": 0.9726600170135498,
      "step": 75
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.05939149111509323,
      "learning_rate": 0.0001995488010273198,
      "loss": 0.072,
      "mean_token_accuracy": 0.9725209295749664,
      "step": 80
    },
    {
      "epoch": 0.136,
      "grad_norm": 0.06247780844569206,
      "learning_rate": 0.00019924474249753655,
      "loss": 0.0685,
      "mean_token_accuracy": 0.9738408207893372,
      "step": 85
    },
    {
      "epoch": 0.144,
      "grad_norm": 0.08245568722486496,
      "learning_rate": 0.00019886315814943647,
      "loss": 0.073,
      "mean_token_accuracy": 0.9711012184619904,
      "step": 90
    },
    {
      "epoch": 0.152,
      "grad_norm": 0.0997772142291069,
      "learning_rate": 0.0001984043460606618,
      "loss": 0.0736,
      "mean_token_accuracy": 0.9707770466804504,
      "step": 95
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.038036637008190155,
      "learning_rate": 0.0001978686646359173,
      "loss": 0.07,
      "mean_token_accuracy": 0.9724892556667328,
      "step": 100
    },
    {
      "epoch": 0.168,
      "grad_norm": 0.058324337005615234,
      "learning_rate": 0.0001972565323269996,
      "loss": 0.0693,
      "mean_token_accuracy": 0.9737250328063964,
      "step": 105
    },
    {
      "epoch": 0.176,
      "grad_norm": 0.07653159648180008,
      "learning_rate": 0.00019656842730592046,
      "loss": 0.0697,
      "mean_token_accuracy": 0.972350025177002,
      "step": 110
    },
    {
      "epoch": 0.184,
      "grad_norm": 0.05668872222304344,
      "learning_rate": 0.0001958048870913786,
      "loss": 0.0673,
      "mean_token_accuracy": 0.9740376353263855,
      "step": 115
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.07520148903131485,
      "learning_rate": 0.0001949665081288729,
      "loss": 0.0686,
      "mean_token_accuracy": 0.9735703825950622,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.06685350835323334,
      "learning_rate": 0.00019405394532478424,
      "loss": 0.0686,
      "mean_token_accuracy": 0.9737636029720307,
      "step": 125
    },
    {
      "epoch": 0.208,
      "grad_norm": 0.07289742678403854,
      "learning_rate": 0.00019306791153479006,
      "loss": 0.0698,
      "mean_token_accuracy": 0.9733847618103028,
      "step": 130
    },
    {
      "epoch": 0.216,
      "grad_norm": 0.05272844061255455,
      "learning_rate": 0.00019200917700701176,
      "loss": 0.0683,
      "mean_token_accuracy": 0.9743804275989533,
      "step": 135
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.04753327742218971,
      "learning_rate": 0.0001908785687803289,
      "loss": 0.0681,
      "mean_token_accuracy": 0.9739800930023194,
      "step": 140
    },
    {
      "epoch": 0.232,
      "grad_norm": 0.03577865660190582,
      "learning_rate": 0.00018967697003833157,
      "loss": 0.0665,
      "mean_token_accuracy": 0.9742783665657043,
      "step": 145
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.0762329250574112,
      "learning_rate": 0.0001884053194194142,
      "loss": 0.0701,
      "mean_token_accuracy": 0.9719095170497895,
      "step": 150
    },
    {
      "epoch": 0.248,
      "grad_norm": 0.06152822822332382,
      "learning_rate": 0.00018706461028355104,
      "loss": 0.0714,
      "mean_token_accuracy": 0.972782963514328,
      "step": 155
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.05897537246346474,
      "learning_rate": 0.00018565588993632487,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9727080404758454,
      "step": 160
    },
    {
      "epoch": 0.264,
      "grad_norm": 0.05083536356687546,
      "learning_rate": 0.0001841802588108161,
      "loss": 0.0699,
      "mean_token_accuracy": 0.9732326388359069,
      "step": 165
    },
    {
      "epoch": 0.272,
      "grad_norm": 0.10682022571563721,
      "learning_rate": 0.00018263886960799062,
      "loss": 0.0694,
      "mean_token_accuracy": 0.9737395167350769,
      "step": 170
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.04953533783555031,
      "learning_rate": 0.00018103292639625837,
      "loss": 0.0674,
      "mean_token_accuracy": 0.9735804319381713,
      "step": 175
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.09398102015256882,
      "learning_rate": 0.0001793636836709057,
      "loss": 0.0703,
      "mean_token_accuracy": 0.9727153956890107,
      "step": 180
    },
    {
      "epoch": 0.296,
      "grad_norm": 0.03411261737346649,
      "learning_rate": 0.0001776324453741365,
      "loss": 0.0697,
      "mean_token_accuracy": 0.9730988323688508,
      "step": 185
    },
    {
      "epoch": 0.304,
      "grad_norm": 0.04288816452026367,
      "learning_rate": 0.00017584056387648727,
      "loss": 0.0674,
      "mean_token_accuracy": 0.9743736147880554,
      "step": 190
    },
    {
      "epoch": 0.312,
      "grad_norm": 0.029879208654165268,
      "learning_rate": 0.0001739894389204122,
      "loss": 0.0702,
      "mean_token_accuracy": 0.9731830060482025,
      "step": 195
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.08851434290409088,
      "learning_rate": 0.00017208051652686335,
      "loss": 0.069,
      "mean_token_accuracy": 0.9728053629398346,
      "step": 200
    },
    {
      "epoch": 0.328,
      "grad_norm": 0.0552876852452755,
      "learning_rate": 0.00017011528786571969,
      "loss": 0.068,
      "mean_token_accuracy": 0.9748190581798554,
      "step": 205
    },
    {
      "epoch": 0.336,
      "grad_norm": 0.034572843462228775,
      "learning_rate": 0.00016809528809094807,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9749053597450257,
      "step": 210
    },
    {
      "epoch": 0.344,
      "grad_norm": 0.0606369711458683,
      "learning_rate": 0.0001660220951414055,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9748689591884613,
      "step": 215
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.04933500289916992,
      "learning_rate": 0.00016389732850821966,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9748404204845429,
      "step": 220
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.04625103250145912,
      "learning_rate": 0.0001617226479697105,
      "loss": 0.0681,
      "mean_token_accuracy": 0.9743536829948425,
      "step": 225
    },
    {
      "epoch": 0.368,
      "grad_norm": 0.04646700620651245,
      "learning_rate": 0.00015949975229484134,
      "loss": 0.0688,
      "mean_token_accuracy": 0.9732876718044281,
      "step": 230
    },
    {
      "epoch": 0.376,
      "grad_norm": 0.048528991639614105,
      "learning_rate": 0.00015723037791621193,
      "loss": 0.0689,
      "mean_token_accuracy": 0.973936015367508,
      "step": 235
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.03883543610572815,
      "learning_rate": 0.00015491629757363032,
      "loss": 0.0672,
      "mean_token_accuracy": 0.9743247866630554,
      "step": 240
    },
    {
      "epoch": 0.392,
      "grad_norm": 0.024075213819742203,
      "learning_rate": 0.00015255931892932333,
      "loss": 0.0675,
      "mean_token_accuracy": 0.974286425113678,
      "step": 245
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.04204133525490761,
      "learning_rate": 0.0001501612831558664,
      "loss": 0.0682,
      "mean_token_accuracy": 0.9739055633544922,
      "step": 250
    },
    {
      "epoch": 0.408,
      "grad_norm": 0.042601633816957474,
      "learning_rate": 0.00014772406349793744,
      "loss": 0.0676,
      "mean_token_accuracy": 0.9740329682826996,
      "step": 255
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.07800234854221344,
      "learning_rate": 0.0001452495638090167,
      "loss": 0.0682,
      "mean_token_accuracy": 0.9739810705184937,
      "step": 260
    },
    {
      "epoch": 0.424,
      "grad_norm": 0.08913061022758484,
      "learning_rate": 0.00014273971706417647,
      "loss": 0.0683,
      "mean_token_accuracy": 0.97349653840065,
      "step": 265
    },
    {
      "epoch": 0.432,
      "grad_norm": 0.041133053600788116,
      "learning_rate": 0.00014019648385012244,
      "loss": 0.0674,
      "mean_token_accuracy": 0.9733855128288269,
      "step": 270
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.045423462986946106,
      "learning_rate": 0.00013762185083366556,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9736481547355652,
      "step": 275
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.06275457888841629,
      "learning_rate": 0.00013501782920982184,
      "loss": 0.0674,
      "mean_token_accuracy": 0.9724659562110901,
      "step": 280
    },
    {
      "epoch": 0.456,
      "grad_norm": 0.05909360572695732,
      "learning_rate": 0.00013238645313075104,
      "loss": 0.0673,
      "mean_token_accuracy": 0.9730458855628967,
      "step": 285
    },
    {
      "epoch": 0.464,
      "grad_norm": 0.04676037281751633,
      "learning_rate": 0.00012972977811676287,
      "loss": 0.0662,
      "mean_token_accuracy": 0.974538516998291,
      "step": 290
    },
    {
      "epoch": 0.472,
      "grad_norm": 0.041126739233732224,
      "learning_rate": 0.00012704987945063068,
      "loss": 0.0711,
      "mean_token_accuracy": 0.9727458715438843,
      "step": 295
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.0714876651763916,
      "learning_rate": 0.00012434885055646823,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9741096377372742,
      "step": 300
    },
    {
      "epoch": 0.488,
      "grad_norm": 0.04500213637948036,
      "learning_rate": 0.00012162880136443447,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9740184247493744,
      "step": 305
    },
    {
      "epoch": 0.496,
      "grad_norm": 0.05178278684616089,
      "learning_rate": 0.00011889185666254506,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9743685066699982,
      "step": 310
    },
    {
      "epoch": 0.504,
      "grad_norm": 0.04735963046550751,
      "learning_rate": 0.00011614015443687722,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9740488171577454,
      "step": 315
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.034043699502944946,
      "learning_rate": 0.0001133758442014651,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9741357147693634,
      "step": 320
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.0381355844438076,
      "learning_rate": 0.00011060108531918971,
      "loss": 0.0664,
      "mean_token_accuracy": 0.9744893789291382,
      "step": 325
    },
    {
      "epoch": 0.528,
      "grad_norm": 0.03367630019783974,
      "learning_rate": 0.0001078180453149754,
      "loss": 0.0665,
      "mean_token_accuracy": 0.973786473274231,
      "step": 330
    },
    {
      "epoch": 0.536,
      "grad_norm": 0.049847234040498734,
      "learning_rate": 0.00010502889818261075,
      "loss": 0.0662,
      "mean_token_accuracy": 0.9747727394104004,
      "step": 335
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.04152824729681015,
      "learning_rate": 0.00010223582268651586,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9741198182106018,
      "step": 340
    },
    {
      "epoch": 0.552,
      "grad_norm": 0.031035475432872772,
      "learning_rate": 9.94410006597835e-05,
      "loss": 0.0666,
      "mean_token_accuracy": 0.9741639375686646,
      "step": 345
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.06370649486780167,
      "learning_rate": 9.66466152998226e-05,
      "loss": 0.0657,
      "mean_token_accuracy": 0.974870103597641,
      "step": 350
    },
    {
      "epoch": 0.568,
      "grad_norm": 0.05804061517119408,
      "learning_rate": 9.385484946293637e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9743902027606964,
      "step": 355
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.04537951946258545,
      "learning_rate": 9.106788395916678e-05,
      "loss": 0.0655,
      "mean_token_accuracy": 0.9747992336750031,
      "step": 360
    },
    {
      "epoch": 0.584,
      "grad_norm": 0.044710028916597366,
      "learning_rate": 8.828789584873754e-05,
      "loss": 0.0648,
      "mean_token_accuracy": 0.9747158527374268,
      "step": 365
    },
    {
      "epoch": 0.592,
      "grad_norm": 0.06087673082947731,
      "learning_rate": 8.551705674142617e-05,
      "loss": 0.0659,
      "mean_token_accuracy": 0.975097918510437,
      "step": 370
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.042375147342681885,
      "learning_rate": 8.275753110019367e-05,
      "loss": 0.0648,
      "mean_token_accuracy": 0.9747480094432831,
      "step": 375
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.10627017915248871,
      "learning_rate": 8.001147455039735e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9743880510330201,
      "step": 380
    },
    {
      "epoch": 0.616,
      "grad_norm": 0.042157676070928574,
      "learning_rate": 7.728103219590681e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9746121108531952,
      "step": 385
    },
    {
      "epoch": 0.624,
      "grad_norm": 0.03764188662171364,
      "learning_rate": 7.456833694343906e-05,
      "loss": 0.0668,
      "mean_token_accuracy": 0.9736917674541473,
      "step": 390
    },
    {
      "epoch": 0.632,
      "grad_norm": 0.03813442215323448,
      "learning_rate": 7.18755078364214e-05,
      "loss": 0.0655,
      "mean_token_accuracy": 0.9746114015579224,
      "step": 395
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.059106457978487015,
      "learning_rate": 6.920464839968405e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9745768249034882,
      "step": 400
    },
    {
      "epoch": 0.648,
      "grad_norm": 0.03773843124508858,
      "learning_rate": 6.65578449962749e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9745455861091614,
      "step": 405
    },
    {
      "epoch": 0.656,
      "grad_norm": 0.03554068133234978,
      "learning_rate": 6.393716519768047e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9740902602672576,
      "step": 410
    },
    {
      "epoch": 0.664,
      "grad_norm": 0.03690912574529648,
      "learning_rate": 6.134465616872598e-05,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9752342998981476,
      "step": 415
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.045288536697626114,
      "learning_rate": 5.878234306841637e-05,
      "loss": 0.0667,
      "mean_token_accuracy": 0.9734826624393463,
      "step": 420
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.0370134674012661,
      "learning_rate": 5.62522274679673e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9740849792957306,
      "step": 425
    },
    {
      "epoch": 0.688,
      "grad_norm": 0.028062770143151283,
      "learning_rate": 5.375628578726181e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9742703199386596,
      "step": 430
    },
    {
      "epoch": 0.696,
      "grad_norm": 0.04216380789875984,
      "learning_rate": 5.1296467750954314e-05,
      "loss": 0.0655,
      "mean_token_accuracy": 0.9746261894702911,
      "step": 435
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.02920725755393505,
      "learning_rate": 4.8874694865427676e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9742297470569611,
      "step": 440
    },
    {
      "epoch": 0.712,
      "grad_norm": 0.03160529211163521,
      "learning_rate": 4.649285891779327e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.975237387418747,
      "step": 445
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.03332898020744324,
      "learning_rate": 4.415282049810644e-05,
      "loss": 0.0653,
      "mean_token_accuracy": 0.974715632200241,
      "step": 450
    },
    {
      "epoch": 0.728,
      "grad_norm": 0.026405662298202515,
      "learning_rate": 4.1856407545951834e-05,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9741382241249085,
      "step": 455
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.030400417745113373,
      "learning_rate": 3.9605413922533874e-05,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9746271312236786,
      "step": 460
    },
    {
      "epoch": 0.744,
      "grad_norm": 0.03024221584200859,
      "learning_rate": 3.740159800938784e-05,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9744933009147644,
      "step": 465
    },
    {
      "epoch": 0.752,
      "grad_norm": 0.03216560184955597,
      "learning_rate": 3.5246681334806175e-05,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9744032263755799,
      "step": 470
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.026212945580482483,
      "learning_rate": 3.3142347229053015e-05,
      "loss": 0.0661,
      "mean_token_accuracy": 0.9738227069377899,
      "step": 475
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.034428227692842484,
      "learning_rate": 3.109023950941736e-05,
      "loss": 0.0646,
      "mean_token_accuracy": 0.9747149765491485,
      "step": 480
    },
    {
      "epoch": 0.776,
      "grad_norm": 0.034947581589221954,
      "learning_rate": 2.909196119613218e-05,
      "loss": 0.0645,
      "mean_token_accuracy": 0.9754405736923217,
      "step": 485
    },
    {
      "epoch": 0.784,
      "grad_norm": 0.07648847997188568,
      "learning_rate": 2.7149073260162416e-05,
      "loss": 0.0648,
      "mean_token_accuracy": 0.9742410600185394,
      "step": 490
    },
    {
      "epoch": 0.792,
      "grad_norm": 0.03137823939323425,
      "learning_rate": 2.5263093403840142e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9745282530784607,
      "step": 495
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.0399913415312767,
      "learning_rate": 2.3435494875299314e-05,
      "loss": 0.0641,
      "mean_token_accuracy": 0.9746397316455842,
      "step": 500
    },
    {
      "epoch": 0.808,
      "grad_norm": 0.04372737184166908,
      "learning_rate": 2.166770531763633e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9738189876079559,
      "step": 505
    },
    {
      "epoch": 0.816,
      "grad_norm": 0.04505913704633713,
      "learning_rate": 1.9961105653695266e-05,
      "loss": 0.0648,
      "mean_token_accuracy": 0.9754888594150544,
      "step": 510
    },
    {
      "epoch": 0.824,
      "grad_norm": 0.04182517156004906,
      "learning_rate": 1.8317029007349085e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9731900095939636,
      "step": 515
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.028708158060908318,
      "learning_rate": 1.6736759662119183e-05,
      "loss": 0.0652,
      "mean_token_accuracy": 0.9751022756099701,
      "step": 520
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.03188589587807655,
      "learning_rate": 1.5221532057947419e-05,
      "loss": 0.065,
      "mean_token_accuracy": 0.9747376561164856,
      "step": 525
    },
    {
      "epoch": 0.848,
      "grad_norm": 0.04484269767999649,
      "learning_rate": 1.3772529826903269e-05,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9743803679943085,
      "step": 530
    },
    {
      "epoch": 0.856,
      "grad_norm": 0.033399421721696854,
      "learning_rate": 1.23908848685804e-05,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9744038641452789,
      "step": 535
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.03684014827013016,
      "learning_rate": 1.1077676465904208e-05,
      "loss": 0.0658,
      "mean_token_accuracy": 0.9743221640586853,
      "step": 540
    },
    {
      "epoch": 0.872,
      "grad_norm": 0.038499992340803146,
      "learning_rate": 9.833930442041506e-06,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9748572528362274,
      "step": 545
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.04119903966784477,
      "learning_rate": 8.660618359070604e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9743854105472565,
      "step": 550
    },
    {
      "epoch": 0.888,
      "grad_norm": 0.0345255583524704,
      "learning_rate": 7.558656759037797e-06,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9747420430183411,
      "step": 555
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.04247881844639778,
      "learning_rate": 6.528906447993288e-06,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9744344353675842,
      "step": 560
    },
    {
      "epoch": 0.904,
      "grad_norm": 0.0425385944545269,
      "learning_rate": 5.572171823565797e-06,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9747009396553039,
      "step": 565
    },
    {
      "epoch": 0.912,
      "grad_norm": 0.038496922701597214,
      "learning_rate": 4.689200246600867e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9749464929103852,
      "step": 570
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.09051476418972015,
      "learning_rate": 3.880681457354118e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9746138632297516,
      "step": 575
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.03681914880871773,
      "learning_rate": 3.1472470366950334e-06,
      "loss": 0.065,
      "mean_token_accuracy": 0.9747094750404358,
      "step": 580
    },
    {
      "epoch": 0.936,
      "grad_norm": 0.027946218848228455,
      "learning_rate": 2.4894699127426367e-06,
      "loss": 0.0663,
      "mean_token_accuracy": 0.9741702735424042,
      "step": 585
    },
    {
      "epoch": 0.944,
      "grad_norm": 0.03258182853460312,
      "learning_rate": 1.907863913318153e-06,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9744637787342072,
      "step": 590
    },
    {
      "epoch": 0.952,
      "grad_norm": 0.048518866300582886,
      "learning_rate": 1.4028833645643113e-06,
      "loss": 0.0668,
      "mean_token_accuracy": 0.97352836728096,
      "step": 595
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.03360092639923096,
      "learning_rate": 9.749227360448143e-07,
      "loss": 0.0653,
      "mean_token_accuracy": 0.9744040012359619,
      "step": 600
    },
    {
      "epoch": 0.968,
      "grad_norm": 0.026586443185806274,
      "learning_rate": 6.243163326014267e-07,
      "loss": 0.0644,
      "mean_token_accuracy": 0.9743619084358215,
      "step": 605
    },
    {
      "epoch": 0.976,
      "grad_norm": 0.03370346128940582,
      "learning_rate": 3.5133803320896994e-07,
      "loss": 0.0654,
      "mean_token_accuracy": 0.9755459070205689,
      "step": 610
    },
    {
      "epoch": 0.984,
      "grad_norm": 0.030597997829318047,
      "learning_rate": 1.562010770326916e-07,
      "loss": 0.0646,
      "mean_token_accuracy": 0.9748627007007599,
      "step": 615
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.030849959701299667,
      "learning_rate": 3.905789685471062e-08,
      "loss": 0.0647,
      "mean_token_accuracy": 0.9752645015716552,
      "step": 620
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.046669185161590576,
      "learning_rate": 0.0,
      "loss": 0.0657,
      "mean_token_accuracy": 0.9741253137588501,
      "step": 625
    },
    {
      "epoch": 1.0,
      "step": 625,
      "total_flos": 1.7169334809919488e+16,
      "train_loss": 0.12309885902404785,
      "train_runtime": 558.0955,
      "train_samples_per_second": 8.959,
      "train_steps_per_second": 1.12
    }
  ],
  "logging_steps": 5,
  "max_steps": 625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.7169334809919488e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}