{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 733,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0068212824010914054,
      "grad_norm": 1.9176777690560134,
      "learning_rate": 5.405405405405406e-06,
      "loss": 0.8544,
      "num_tokens": 3759778.0,
      "step": 5
    },
    {
      "epoch": 0.013642564802182811,
      "grad_norm": 1.6627094198294425,
      "learning_rate": 1.2162162162162164e-05,
      "loss": 0.8061,
      "num_tokens": 7670069.0,
      "step": 10
    },
    {
      "epoch": 0.020463847203274217,
      "grad_norm": 0.692154825441687,
      "learning_rate": 1.891891891891892e-05,
      "loss": 0.7189,
      "num_tokens": 11370767.0,
      "step": 15
    },
    {
      "epoch": 0.027285129604365622,
      "grad_norm": 0.568018891918523,
      "learning_rate": 2.5675675675675675e-05,
      "loss": 0.677,
      "num_tokens": 15120586.0,
      "step": 20
    },
    {
      "epoch": 0.034106412005457026,
      "grad_norm": 0.5004537886554214,
      "learning_rate": 3.2432432432432436e-05,
      "loss": 0.6543,
      "num_tokens": 18909989.0,
      "step": 25
    },
    {
      "epoch": 0.040927694406548434,
      "grad_norm": 0.44197300768134357,
      "learning_rate": 3.918918918918919e-05,
      "loss": 0.6319,
      "num_tokens": 22645536.0,
      "step": 30
    },
    {
      "epoch": 0.047748976807639835,
      "grad_norm": 0.4067554292845183,
      "learning_rate": 4.594594594594595e-05,
      "loss": 0.6276,
      "num_tokens": 26641035.0,
      "step": 35
    },
    {
      "epoch": 0.054570259208731244,
      "grad_norm": 0.4786866704049721,
      "learning_rate": 4.999908316574644e-05,
      "loss": 0.5971,
      "num_tokens": 30423007.0,
      "step": 40
    },
    {
      "epoch": 0.061391541609822645,
      "grad_norm": 0.3979368350874685,
      "learning_rate": 4.998876963847189e-05,
      "loss": 0.6001,
      "num_tokens": 34237001.0,
      "step": 45
    },
    {
      "epoch": 0.06821282401091405,
      "grad_norm": 0.4654195837280465,
      "learning_rate": 4.996700181165029e-05,
      "loss": 0.6001,
      "num_tokens": 37967720.0,
      "step": 50
    },
    {
      "epoch": 0.07503410641200546,
      "grad_norm": 0.4516488084979742,
      "learning_rate": 4.993379077238036e-05,
      "loss": 0.6008,
      "num_tokens": 41833783.0,
      "step": 55
    },
    {
      "epoch": 0.08185538881309687,
      "grad_norm": 0.49984152066023985,
      "learning_rate": 4.9889153436180295e-05,
      "loss": 0.5885,
      "num_tokens": 45550959.0,
      "step": 60
    },
    {
      "epoch": 0.08867667121418826,
      "grad_norm": 0.4879085584789433,
      "learning_rate": 4.983311253837213e-05,
      "loss": 0.5925,
      "num_tokens": 49377672.0,
      "step": 65
    },
    {
      "epoch": 0.09549795361527967,
      "grad_norm": 0.44782023094976653,
      "learning_rate": 4.9765696622501846e-05,
      "loss": 0.582,
      "num_tokens": 53019691.0,
      "step": 70
    },
    {
      "epoch": 0.10231923601637108,
      "grad_norm": 0.42799052136318366,
      "learning_rate": 4.968694002580118e-05,
      "loss": 0.5841,
      "num_tokens": 56919336.0,
      "step": 75
    },
    {
      "epoch": 0.10914051841746249,
      "grad_norm": 0.5198378195145373,
      "learning_rate": 4.959688286169851e-05,
      "loss": 0.5675,
      "num_tokens": 60660649.0,
      "step": 80
    },
    {
      "epoch": 0.11596180081855388,
      "grad_norm": 0.4009186218197995,
      "learning_rate": 4.9495570999387685e-05,
      "loss": 0.561,
      "num_tokens": 64575366.0,
      "step": 85
    },
    {
      "epoch": 0.12278308321964529,
      "grad_norm": 0.43972341232334455,
      "learning_rate": 4.9383056040465276e-05,
      "loss": 0.5792,
      "num_tokens": 68438217.0,
      "step": 90
    },
    {
      "epoch": 0.1296043656207367,
      "grad_norm": 0.5022875539848495,
      "learning_rate": 4.925939529264815e-05,
      "loss": 0.5751,
      "num_tokens": 72264786.0,
      "step": 95
    },
    {
      "epoch": 0.1364256480218281,
      "grad_norm": 0.4879174456867744,
      "learning_rate": 4.9124651740584684e-05,
      "loss": 0.5615,
      "num_tokens": 76173511.0,
      "step": 100
    },
    {
      "epoch": 0.1432469304229195,
      "grad_norm": 0.55054422042514,
      "learning_rate": 4.897889401377447e-05,
      "loss": 0.5634,
      "num_tokens": 80166162.0,
      "step": 105
    },
    {
      "epoch": 0.15006821282401092,
      "grad_norm": 0.3656687810097067,
      "learning_rate": 4.882219635161306e-05,
      "loss": 0.567,
      "num_tokens": 83915537.0,
      "step": 110
    },
    {
      "epoch": 0.15688949522510232,
      "grad_norm": 0.4628699637839558,
      "learning_rate": 4.865463856557922e-05,
      "loss": 0.5656,
      "num_tokens": 87705484.0,
      "step": 115
    },
    {
      "epoch": 0.16371077762619374,
      "grad_norm": 0.3979402679605098,
      "learning_rate": 4.847630599858426e-05,
      "loss": 0.5548,
      "num_tokens": 91557529.0,
      "step": 120
    },
    {
      "epoch": 0.17053206002728513,
      "grad_norm": 0.39083842800198015,
      "learning_rate": 4.8287289481503954e-05,
      "loss": 0.5615,
      "num_tokens": 95453893.0,
      "step": 125
    },
    {
      "epoch": 0.17735334242837653,
      "grad_norm": 0.41780246703962853,
      "learning_rate": 4.8087685286915276e-05,
      "loss": 0.5571,
      "num_tokens": 99400041.0,
      "step": 130
    },
    {
      "epoch": 0.18417462482946795,
      "grad_norm": 0.4031844524499151,
      "learning_rate": 4.787759508006147e-05,
      "loss": 0.5567,
      "num_tokens": 103240516.0,
      "step": 135
    },
    {
      "epoch": 0.19099590723055934,
      "grad_norm": 0.48919957010741905,
      "learning_rate": 4.765712586707048e-05,
      "loss": 0.5695,
      "num_tokens": 106919294.0,
      "step": 140
    },
    {
      "epoch": 0.19781718963165076,
      "grad_norm": 0.45120708271529725,
      "learning_rate": 4.7426389940453065e-05,
      "loss": 0.542,
      "num_tokens": 110858993.0,
      "step": 145
    },
    {
      "epoch": 0.20463847203274216,
      "grad_norm": 0.4566044280248485,
      "learning_rate": 4.718550482190837e-05,
      "loss": 0.5579,
      "num_tokens": 114540376.0,
      "step": 150
    },
    {
      "epoch": 0.21145975443383355,
      "grad_norm": 0.40666142897719454,
      "learning_rate": 4.6934593202466127e-05,
      "loss": 0.5427,
      "num_tokens": 118465258.0,
      "step": 155
    },
    {
      "epoch": 0.21828103683492497,
      "grad_norm": 0.4202500127067603,
      "learning_rate": 4.6673782879995896e-05,
      "loss": 0.5513,
      "num_tokens": 122331818.0,
      "step": 160
    },
    {
      "epoch": 0.22510231923601637,
      "grad_norm": 0.36280756345549253,
      "learning_rate": 4.640320669411526e-05,
      "loss": 0.5541,
      "num_tokens": 126115279.0,
      "step": 165
    },
    {
      "epoch": 0.23192360163710776,
      "grad_norm": 0.3777672205487623,
      "learning_rate": 4.612300245853004e-05,
      "loss": 0.5473,
      "num_tokens": 129992439.0,
      "step": 170
    },
    {
      "epoch": 0.23874488403819918,
      "grad_norm": 0.4906003744581939,
      "learning_rate": 4.5833312890841085e-05,
      "loss": 0.562,
      "num_tokens": 133787994.0,
      "step": 175
    },
    {
      "epoch": 0.24556616643929058,
      "grad_norm": 0.4239378854197107,
      "learning_rate": 4.553428553985329e-05,
      "loss": 0.5417,
      "num_tokens": 137544925.0,
      "step": 180
    },
    {
      "epoch": 0.252387448840382,
      "grad_norm": 0.42964921244752585,
      "learning_rate": 4.522607271042399e-05,
      "loss": 0.5366,
      "num_tokens": 141219364.0,
      "step": 185
    },
    {
      "epoch": 0.2592087312414734,
      "grad_norm": 0.5213144485813976,
      "learning_rate": 4.490883138588882e-05,
      "loss": 0.5481,
      "num_tokens": 145160611.0,
      "step": 190
    },
    {
      "epoch": 0.2660300136425648,
      "grad_norm": 0.35852284468504936,
      "learning_rate": 4.458272314810479e-05,
      "loss": 0.5359,
      "num_tokens": 148964661.0,
      "step": 195
    },
    {
      "epoch": 0.2728512960436562,
      "grad_norm": 0.33972968400294307,
      "learning_rate": 4.4247914095151086e-05,
      "loss": 0.5458,
      "num_tokens": 152834849.0,
      "step": 200
    },
    {
      "epoch": 0.27967257844474763,
      "grad_norm": 0.39019211196865305,
      "learning_rate": 4.390457475672966e-05,
      "loss": 0.5394,
      "num_tokens": 156709373.0,
      "step": 205
    },
    {
      "epoch": 0.286493860845839,
      "grad_norm": 0.37663298450178134,
      "learning_rate": 4.35528800073086e-05,
      "loss": 0.5409,
      "num_tokens": 160459756.0,
      "step": 210
    },
    {
      "epoch": 0.2933151432469304,
      "grad_norm": 0.36067884540954953,
      "learning_rate": 4.31930089770526e-05,
      "loss": 0.5442,
      "num_tokens": 164401378.0,
      "step": 215
    },
    {
      "epoch": 0.30013642564802184,
      "grad_norm": 0.3679116513152416,
      "learning_rate": 4.282514496058582e-05,
      "loss": 0.5238,
      "num_tokens": 168250987.0,
      "step": 220
    },
    {
      "epoch": 0.3069577080491132,
      "grad_norm": 0.3239388052959791,
      "learning_rate": 4.24494753236337e-05,
      "loss": 0.5365,
      "num_tokens": 172160313.0,
      "step": 225
    },
    {
      "epoch": 0.31377899045020463,
      "grad_norm": 0.31913103910612595,
      "learning_rate": 4.2066191407591125e-05,
      "loss": 0.5321,
      "num_tokens": 176115268.0,
      "step": 230
    },
    {
      "epoch": 0.32060027285129605,
      "grad_norm": 0.3331439854441546,
      "learning_rate": 4.1675488432065785e-05,
      "loss": 0.5243,
      "num_tokens": 179947208.0,
      "step": 235
    },
    {
      "epoch": 0.3274215552523875,
      "grad_norm": 0.3120568835661435,
      "learning_rate": 4.127756539544609e-05,
      "loss": 0.537,
      "num_tokens": 183776326.0,
      "step": 240
    },
    {
      "epoch": 0.33424283765347884,
      "grad_norm": 0.45645907440018146,
      "learning_rate": 4.087262497354452e-05,
      "loss": 0.5456,
      "num_tokens": 187730193.0,
      "step": 245
    },
    {
      "epoch": 0.34106412005457026,
      "grad_norm": 0.3653248040533821,
      "learning_rate": 4.046087341636789e-05,
      "loss": 0.528,
      "num_tokens": 191543591.0,
      "step": 250
    },
    {
      "epoch": 0.3478854024556617,
      "grad_norm": 0.32434915433858746,
      "learning_rate": 4.0042520443067176e-05,
      "loss": 0.5293,
      "num_tokens": 195399830.0,
      "step": 255
    },
    {
      "epoch": 0.35470668485675305,
      "grad_norm": 0.33553503337827456,
      "learning_rate": 3.961777913512035e-05,
      "loss": 0.5183,
      "num_tokens": 199248080.0,
      "step": 260
    },
    {
      "epoch": 0.3615279672578445,
      "grad_norm": 0.32517181298662434,
      "learning_rate": 3.9186865827802724e-05,
      "loss": 0.5379,
      "num_tokens": 202936389.0,
      "step": 265
    },
    {
      "epoch": 0.3683492496589359,
      "grad_norm": 0.3565608027112838,
      "learning_rate": 3.875e-05,
      "loss": 0.5266,
      "num_tokens": 206795183.0,
      "step": 270
    },
    {
      "epoch": 0.37517053206002726,
      "grad_norm": 0.3068560323748979,
      "learning_rate": 3.830740416242014e-05,
      "loss": 0.5225,
      "num_tokens": 210620229.0,
      "step": 275
    },
    {
      "epoch": 0.3819918144611187,
      "grad_norm": 0.3230438388020238,
      "learning_rate": 3.7859303744261064e-05,
      "loss": 0.5282,
      "num_tokens": 214296967.0,
      "step": 280
    },
    {
      "epoch": 0.3888130968622101,
      "grad_norm": 0.3364428624622822,
      "learning_rate": 3.740592697839185e-05,
      "loss": 0.5331,
      "num_tokens": 218179883.0,
      "step": 285
    },
    {
      "epoch": 0.3956343792633015,
      "grad_norm": 0.2962428407413462,
      "learning_rate": 3.694750478510596e-05,
      "loss": 0.5286,
      "num_tokens": 222093783.0,
      "step": 290
    },
    {
      "epoch": 0.4024556616643929,
      "grad_norm": 0.293071598457561,
      "learning_rate": 3.648427065450555e-05,
      "loss": 0.5198,
      "num_tokens": 225865685.0,
      "step": 295
    },
    {
      "epoch": 0.4092769440654843,
      "grad_norm": 0.31919217338705325,
      "learning_rate": 3.601646052757707e-05,
      "loss": 0.5191,
      "num_tokens": 229748231.0,
      "step": 300
    },
    {
      "epoch": 0.41609822646657574,
      "grad_norm": 0.35673159600574156,
      "learning_rate": 3.55443126760184e-05,
      "loss": 0.5345,
      "num_tokens": 233655900.0,
      "step": 305
    },
    {
      "epoch": 0.4229195088676671,
      "grad_norm": 0.4247633628818686,
      "learning_rate": 3.506806758087894e-05,
      "loss": 0.5321,
      "num_tokens": 237433481.0,
      "step": 310
    },
    {
      "epoch": 0.4297407912687585,
      "grad_norm": 0.37689245206385813,
      "learning_rate": 3.458796781007437e-05,
      "loss": 0.5268,
      "num_tokens": 241153898.0,
      "step": 315
    },
    {
      "epoch": 0.43656207366984995,
      "grad_norm": 0.3168940754783053,
      "learning_rate": 3.410425789483854e-05,
      "loss": 0.5271,
      "num_tokens": 245008254.0,
      "step": 320
    },
    {
      "epoch": 0.4433833560709413,
      "grad_norm": 0.31894359073537676,
      "learning_rate": 3.3617184205175304e-05,
      "loss": 0.5336,
      "num_tokens": 248791992.0,
      "step": 325
    },
    {
      "epoch": 0.45020463847203274,
      "grad_norm": 0.39040862996328446,
      "learning_rate": 3.312699482437392e-05,
      "loss": 0.5208,
      "num_tokens": 252648795.0,
      "step": 330
    },
    {
      "epoch": 0.45702592087312416,
      "grad_norm": 0.38056774885593053,
      "learning_rate": 3.263393942265168e-05,
      "loss": 0.5275,
      "num_tokens": 256460069.0,
      "step": 335
    },
    {
      "epoch": 0.4638472032742155,
      "grad_norm": 0.3072940523773231,
      "learning_rate": 3.213826912998838e-05,
      "loss": 0.52,
      "num_tokens": 260499223.0,
      "step": 340
    },
    {
      "epoch": 0.47066848567530695,
      "grad_norm": 0.31006481827047494,
      "learning_rate": 3.164023640821719e-05,
      "loss": 0.5132,
      "num_tokens": 264331327.0,
      "step": 345
    },
    {
      "epoch": 0.47748976807639837,
      "grad_norm": 0.29973194583053614,
      "learning_rate": 3.114009492243721e-05,
      "loss": 0.5216,
      "num_tokens": 268142845.0,
      "step": 350
    },
    {
      "epoch": 0.4843110504774898,
      "grad_norm": 0.313160334524215,
      "learning_rate": 3.063809941181321e-05,
      "loss": 0.5313,
      "num_tokens": 272018748.0,
      "step": 355
    },
    {
      "epoch": 0.49113233287858116,
      "grad_norm": 0.3520811242945727,
      "learning_rate": 3.0134505559828203e-05,
      "loss": 0.5351,
      "num_tokens": 275897357.0,
      "step": 360
    },
    {
      "epoch": 0.4979536152796726,
      "grad_norm": 0.31408461197771137,
      "learning_rate": 2.9629569864055125e-05,
      "loss": 0.5131,
      "num_tokens": 279550428.0,
      "step": 365
    },
    {
      "epoch": 0.504774897680764,
      "grad_norm": 0.30955677114945335,
      "learning_rate": 2.9123549505513868e-05,
      "loss": 0.5151,
      "num_tokens": 283508120.0,
      "step": 370
    },
    {
      "epoch": 0.5115961800818554,
      "grad_norm": 0.290796153781096,
      "learning_rate": 2.8616702217680134e-05,
      "loss": 0.523,
      "num_tokens": 287419124.0,
      "step": 375
    },
    {
      "epoch": 0.5184174624829468,
      "grad_norm": 0.30661975100626937,
      "learning_rate": 2.810928615521303e-05,
      "loss": 0.5096,
      "num_tokens": 291105578.0,
      "step": 380
    },
    {
      "epoch": 0.5252387448840382,
      "grad_norm": 0.3014228734854837,
      "learning_rate": 2.7601559762468022e-05,
      "loss": 0.5189,
      "num_tokens": 294930434.0,
      "step": 385
    },
    {
      "epoch": 0.5320600272851296,
      "grad_norm": 0.32239097194277977,
      "learning_rate": 2.7093781641862387e-05,
      "loss": 0.5214,
      "num_tokens": 298726998.0,
      "step": 390
    },
    {
      "epoch": 0.538881309686221,
      "grad_norm": 0.3081176434769925,
      "learning_rate": 2.658621042216021e-05,
      "loss": 0.5057,
      "num_tokens": 302437719.0,
      "step": 395
    },
    {
      "epoch": 0.5457025920873124,
      "grad_norm": 0.3051848872234331,
      "learning_rate": 2.6079104626743845e-05,
      "loss": 0.5275,
      "num_tokens": 306180961.0,
      "step": 400
    },
    {
      "epoch": 0.5525238744884038,
      "grad_norm": 0.3012019366236442,
      "learning_rate": 2.5572722541939113e-05,
      "loss": 0.525,
      "num_tokens": 310003006.0,
      "step": 405
    },
    {
      "epoch": 0.5593451568894953,
      "grad_norm": 0.3646744957708818,
      "learning_rate": 2.5067322085461315e-05,
      "loss": 0.5105,
      "num_tokens": 313777123.0,
      "step": 410
    },
    {
      "epoch": 0.5661664392905866,
      "grad_norm": 0.3717487439447167,
      "learning_rate": 2.4563160675048846e-05,
      "loss": 0.5155,
      "num_tokens": 317577413.0,
      "step": 415
    },
    {
      "epoch": 0.572987721691678,
      "grad_norm": 0.31243442111272723,
      "learning_rate": 2.406049509735156e-05,
      "loss": 0.5153,
      "num_tokens": 321462994.0,
      "step": 420
    },
    {
      "epoch": 0.5798090040927695,
      "grad_norm": 0.3138827497201017,
      "learning_rate": 2.355958137714056e-05,
      "loss": 0.5107,
      "num_tokens": 325155802.0,
      "step": 425
    },
    {
      "epoch": 0.5866302864938608,
      "grad_norm": 0.3183934334838428,
      "learning_rate": 2.3060674646906004e-05,
      "loss": 0.5155,
      "num_tokens": 328885222.0,
      "step": 430
    },
    {
      "epoch": 0.5934515688949522,
      "grad_norm": 0.347477351545875,
      "learning_rate": 2.2564029016909416e-05,
      "loss": 0.5049,
      "num_tokens": 332821824.0,
      "step": 435
    },
    {
      "epoch": 0.6002728512960437,
      "grad_norm": 0.28422582070555286,
      "learning_rate": 2.2069897445756627e-05,
      "loss": 0.5028,
      "num_tokens": 336651049.0,
      "step": 440
    },
    {
      "epoch": 0.607094133697135,
      "grad_norm": 0.3068506353486867,
      "learning_rate": 2.1578531611557322e-05,
      "loss": 0.5158,
      "num_tokens": 340414967.0,
      "step": 445
    },
    {
      "epoch": 0.6139154160982264,
      "grad_norm": 0.25681280618201996,
      "learning_rate": 2.109018178373675e-05,
      "loss": 0.5146,
      "num_tokens": 344295726.0,
      "step": 450
    },
    {
      "epoch": 0.6207366984993179,
      "grad_norm": 0.2885950963130501,
      "learning_rate": 2.0605096695564973e-05,
      "loss": 0.5181,
      "num_tokens": 348137882.0,
      "step": 455
    },
    {
      "epoch": 0.6275579809004093,
      "grad_norm": 0.263849619106182,
      "learning_rate": 2.0123523417468466e-05,
      "loss": 0.5115,
      "num_tokens": 351875623.0,
      "step": 460
    },
    {
      "epoch": 0.6343792633015006,
      "grad_norm": 0.2761424972238221,
      "learning_rate": 1.9645707231188742e-05,
      "loss": 0.5059,
      "num_tokens": 355697743.0,
      "step": 465
    },
    {
      "epoch": 0.6412005457025921,
      "grad_norm": 0.26542715591102417,
      "learning_rate": 1.9171891504851925e-05,
      "loss": 0.5245,
      "num_tokens": 359496768.0,
      "step": 470
    },
    {
      "epoch": 0.6480218281036835,
      "grad_norm": 0.28804005325228654,
      "learning_rate": 1.8702317569013094e-05,
      "loss": 0.5005,
      "num_tokens": 363249804.0,
      "step": 475
    },
    {
      "epoch": 0.654843110504775,
      "grad_norm": 0.2590332683110206,
      "learning_rate": 1.8237224593738327e-05,
      "loss": 0.5027,
      "num_tokens": 366923656.0,
      "step": 480
    },
    {
      "epoch": 0.6616643929058663,
      "grad_norm": 0.2479298846798914,
      "learning_rate": 1.7776849466787223e-05,
      "loss": 0.5171,
      "num_tokens": 370786942.0,
      "step": 485
    },
    {
      "epoch": 0.6684856753069577,
      "grad_norm": 0.248178044912107,
      "learning_rate": 1.7321426672957896e-05,
      "loss": 0.5025,
      "num_tokens": 374628223.0,
      "step": 490
    },
    {
      "epoch": 0.6753069577080492,
      "grad_norm": 0.28596505528159777,
      "learning_rate": 1.6871188174655787e-05,
      "loss": 0.4957,
      "num_tokens": 378392822.0,
      "step": 495
    },
    {
      "epoch": 0.6821282401091405,
      "grad_norm": 0.2547350816625695,
      "learning_rate": 1.6426363293747334e-05,
      "loss": 0.5001,
      "num_tokens": 382166430.0,
      "step": 500
    },
    {
      "epoch": 0.6889495225102319,
      "grad_norm": 0.26695292925007047,
      "learning_rate": 1.598717859475846e-05,
      "loss": 0.5087,
      "num_tokens": 385900887.0,
      "step": 505
    },
    {
      "epoch": 0.6957708049113234,
      "grad_norm": 0.26188575332170744,
      "learning_rate": 1.5553857769477553e-05,
      "loss": 0.5055,
      "num_tokens": 389650899.0,
      "step": 510
    },
    {
      "epoch": 0.7025920873124147,
      "grad_norm": 0.2479933595889963,
      "learning_rate": 1.5126621523021518e-05,
      "loss": 0.51,
      "num_tokens": 393493604.0,
      "step": 515
    },
    {
      "epoch": 0.7094133697135061,
      "grad_norm": 0.2909164355431986,
      "learning_rate": 1.4705687461423209e-05,
      "loss": 0.5221,
      "num_tokens": 397224172.0,
      "step": 520
    },
    {
      "epoch": 0.7162346521145976,
      "grad_norm": 0.2608112334029561,
      "learning_rate": 1.4291269980797139e-05,
      "loss": 0.5064,
      "num_tokens": 400990040.0,
      "step": 525
    },
    {
      "epoch": 0.723055934515689,
      "grad_norm": 0.2561771527326882,
      "learning_rate": 1.3883580158140291e-05,
      "loss": 0.5005,
      "num_tokens": 404752387.0,
      "step": 530
    },
    {
      "epoch": 0.7298772169167803,
      "grad_norm": 0.23857076212946476,
      "learning_rate": 1.3482825643823293e-05,
      "loss": 0.5061,
      "num_tokens": 408650161.0,
      "step": 535
    },
    {
      "epoch": 0.7366984993178718,
      "grad_norm": 0.2477873445432539,
      "learning_rate": 1.3089210555827086e-05,
      "loss": 0.5119,
      "num_tokens": 412454003.0,
      "step": 540
    },
    {
      "epoch": 0.7435197817189632,
      "grad_norm": 0.2359507356726729,
      "learning_rate": 1.270293537577855e-05,
      "loss": 0.5148,
      "num_tokens": 416440659.0,
      "step": 545
    },
    {
      "epoch": 0.7503410641200545,
      "grad_norm": 0.23189964569815721,
      "learning_rate": 1.232419684683844e-05,
      "loss": 0.4995,
      "num_tokens": 420121223.0,
      "step": 550
    },
    {
      "epoch": 0.757162346521146,
      "grad_norm": 0.24995560019248012,
      "learning_rate": 1.1953187873493303e-05,
      "loss": 0.5001,
      "num_tokens": 423755589.0,
      "step": 555
    },
    {
      "epoch": 0.7639836289222374,
      "grad_norm": 0.2509530321729054,
      "learning_rate": 1.1590097423302684e-05,
      "loss": 0.496,
      "num_tokens": 427476418.0,
      "step": 560
    },
    {
      "epoch": 0.7708049113233287,
      "grad_norm": 0.24017845417519926,
      "learning_rate": 1.1235110430651421e-05,
      "loss": 0.4968,
      "num_tokens": 431359520.0,
      "step": 565
    },
    {
      "epoch": 0.7776261937244202,
      "grad_norm": 0.2591754613816144,
      "learning_rate": 1.0888407702556284e-05,
      "loss": 0.4998,
      "num_tokens": 435149765.0,
      "step": 570
    },
    {
      "epoch": 0.7844474761255116,
      "grad_norm": 0.26496824313024014,
      "learning_rate": 1.0550165826574766e-05,
      "loss": 0.4997,
      "num_tokens": 439079263.0,
      "step": 575
    },
    {
      "epoch": 0.791268758526603,
      "grad_norm": 0.25738972167084373,
      "learning_rate": 1.0220557080862985e-05,
      "loss": 0.5149,
      "num_tokens": 443118717.0,
      "step": 580
    },
    {
      "epoch": 0.7980900409276944,
      "grad_norm": 0.23558612955473407,
      "learning_rate": 9.899749346428556e-06,
      "loss": 0.5021,
      "num_tokens": 446925677.0,
      "step": 585
    },
    {
      "epoch": 0.8049113233287858,
      "grad_norm": 0.22651636843721124,
      "learning_rate": 9.587906021623016e-06,
      "loss": 0.5161,
      "num_tokens": 450761568.0,
      "step": 590
    },
    {
      "epoch": 0.8117326057298773,
      "grad_norm": 0.2707894446795202,
      "learning_rate": 9.28518593891749e-06,
      "loss": 0.5013,
      "num_tokens": 454558803.0,
      "step": 595
    },
    {
      "epoch": 0.8185538881309686,
      "grad_norm": 0.24087744058717236,
      "learning_rate": 8.99174328400385e-06,
      "loss": 0.4997,
      "num_tokens": 458401399.0,
      "step": 600
    },
    {
      "epoch": 0.82537517053206,
      "grad_norm": 0.2510998466793629,
      "learning_rate": 8.707727517262697e-06,
      "loss": 0.5049,
      "num_tokens": 462132037.0,
      "step": 605
    },
    {
      "epoch": 0.8321964529331515,
      "grad_norm": 0.2289965692436888,
      "learning_rate": 8.433283297638053e-06,
      "loss": 0.4993,
      "num_tokens": 466050672.0,
      "step": 610
    },
    {
      "epoch": 0.8390177353342428,
      "grad_norm": 0.2456758738171012,
      "learning_rate": 8.168550408957632e-06,
      "loss": 0.497,
      "num_tokens": 469868923.0,
      "step": 615
    },
    {
      "epoch": 0.8458390177353342,
      "grad_norm": 0.2528320643035017,
      "learning_rate": 7.91366368873613e-06,
      "loss": 0.4944,
      "num_tokens": 473648633.0,
      "step": 620
    },
    {
      "epoch": 0.8526603001364257,
      "grad_norm": 0.24723521520394562,
      "learning_rate": 7.66875295949791e-06,
      "loss": 0.5105,
      "num_tokens": 477480036.0,
      "step": 625
    },
    {
      "epoch": 0.859481582537517,
      "grad_norm": 0.2170888923252067,
      "learning_rate": 7.4339429626539e-06,
      "loss": 0.5102,
      "num_tokens": 481428209.0,
      "step": 630
    },
    {
      "epoch": 0.8663028649386084,
      "grad_norm": 0.2577692978270651,
      "learning_rate": 7.2093532949665715e-06,
      "loss": 0.5081,
      "num_tokens": 485251856.0,
      "step": 635
    },
    {
      "epoch": 0.8731241473396999,
      "grad_norm": 0.2261578884095805,
      "learning_rate": 6.995098347635173e-06,
      "loss": 0.4934,
      "num_tokens": 489140124.0,
      "step": 640
    },
    {
      "epoch": 0.8799454297407913,
      "grad_norm": 0.23083645917591586,
      "learning_rate": 6.791287248032431e-06,
      "loss": 0.4966,
      "num_tokens": 492974236.0,
      "step": 645
    },
    {
      "epoch": 0.8867667121418826,
      "grad_norm": 0.24093733414141585,
      "learning_rate": 6.598023804122194e-06,
      "loss": 0.5038,
      "num_tokens": 496987511.0,
      "step": 650
    },
    {
      "epoch": 0.8935879945429741,
      "grad_norm": 0.2201860691663983,
      "learning_rate": 6.415406451586528e-06,
      "loss": 0.4994,
      "num_tokens": 500947010.0,
      "step": 655
    },
    {
      "epoch": 0.9004092769440655,
      "grad_norm": 0.21802735707137785,
      "learning_rate": 6.243528203689025e-06,
      "loss": 0.5032,
      "num_tokens": 504893463.0,
      "step": 660
    },
    {
      "epoch": 0.9072305593451568,
      "grad_norm": 0.23585791427445513,
      "learning_rate": 6.0824766039e-06,
      "loss": 0.4994,
      "num_tokens": 508690960.0,
      "step": 665
    },
    {
      "epoch": 0.9140518417462483,
      "grad_norm": 0.22116956778559868,
      "learning_rate": 5.932333681307571e-06,
      "loss": 0.5062,
      "num_tokens": 512558436.0,
      "step": 670
    },
    {
      "epoch": 0.9208731241473397,
      "grad_norm": 0.21955037974420277,
      "learning_rate": 5.793175908837471e-06,
      "loss": 0.4967,
      "num_tokens": 516301089.0,
      "step": 675
    },
    {
      "epoch": 0.927694406548431,
      "grad_norm": 0.22229974278530354,
      "learning_rate": 5.665074164302742e-06,
      "loss": 0.5064,
      "num_tokens": 520051962.0,
      "step": 680
    },
    {
      "epoch": 0.9345156889495225,
      "grad_norm": 0.20772743825814724,
      "learning_rate": 5.548093694303275e-06,
      "loss": 0.4919,
      "num_tokens": 523880084.0,
      "step": 685
    },
    {
      "epoch": 0.9413369713506139,
      "grad_norm": 0.2341093114448413,
      "learning_rate": 5.442294080993446e-06,
      "loss": 0.5059,
      "num_tokens": 527753736.0,
      "step": 690
    },
    {
      "epoch": 0.9481582537517054,
      "grad_norm": 0.2148921450505356,
      "learning_rate": 5.347729211734919e-06,
      "loss": 0.5034,
      "num_tokens": 531553862.0,
      "step": 695
    },
    {
      "epoch": 0.9549795361527967,
      "grad_norm": 0.2135752772739678,
      "learning_rate": 5.264447251649954e-06,
      "loss": 0.5057,
      "num_tokens": 535341346.0,
      "step": 700
    },
    {
      "epoch": 0.9618008185538881,
      "grad_norm": 0.22252189723093127,
      "learning_rate": 5.192490619089267e-06,
      "loss": 0.49,
      "num_tokens": 539226192.0,
      "step": 705
    },
    {
      "epoch": 0.9686221009549796,
      "grad_norm": 0.2167158306045372,
      "learning_rate": 5.1318959640269095e-06,
      "loss": 0.5005,
      "num_tokens": 542913487.0,
      "step": 710
    },
    {
      "epoch": 0.975443383356071,
      "grad_norm": 0.22839928449641014,
      "learning_rate": 5.082694149393189e-06,
      "loss": 0.5114,
      "num_tokens": 546668069.0,
      "step": 715
    },
    {
      "epoch": 0.9822646657571623,
      "grad_norm": 0.21859854387502445,
      "learning_rate": 5.044910235355121e-06,
      "loss": 0.4974,
      "num_tokens": 550468465.0,
      "step": 720
    },
    {
      "epoch": 0.9890859481582538,
      "grad_norm": 0.2040301960420221,
      "learning_rate": 5.0185634665524255e-06,
      "loss": 0.4934,
      "num_tokens": 554315309.0,
      "step": 725
    },
    {
      "epoch": 0.9959072305593452,
      "grad_norm": 0.20234396390669804,
      "learning_rate": 5.003667262295572e-06,
      "loss": 0.5015,
      "num_tokens": 558093284.0,
      "step": 730
    },
    {
      "epoch": 1.0,
      "num_tokens": 560403569.0,
      "step": 733,
      "total_flos": 1147535452274688.0,
      "train_loss": 0.534508459239507,
      "train_runtime": 12072.7525,
      "train_samples_per_second": 7.764,
      "train_steps_per_second": 0.061
    }
  ],
  "logging_steps": 5,
  "max_steps": 733,
  "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": 1147535452274688.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}