{
  "epoch": 1.185633176993335,
  "global_step": 600,
  "max_steps": 4048,
  "logging_steps": 5,
  "eval_steps": 100,
  "save_steps": 100,
  "train_batch_size": 16,
  "num_train_epochs": 8,
  "num_input_tokens_seen": 0,
  "total_flos": 2.946671339427594e+17,
  "log_history": [
    {
      "loss": 10.7217,
      "grad_norm": 35.45301055908203,
      "learning_rate": 3.75e-05,
      "epoch": 0.009874105159219946,
      "step": 5
    },
    {
      "loss": 4.3632,
      "grad_norm": 12.563997268676758,
      "learning_rate": 7.5e-05,
      "epoch": 0.01974821031843989,
      "step": 10
    },
    {
      "loss": 1.6282,
      "grad_norm": 10.4967041015625,
      "learning_rate": 0.0001125,
      "epoch": 0.02962231547765984,
      "step": 15
    },
    {
      "loss": 1.1701,
      "grad_norm": 2.9293179512023926,
      "learning_rate": 0.00015,
      "epoch": 0.03949642063687978,
      "step": 20
    },
    {
      "loss": 0.9779,
      "grad_norm": 1.5904552936553955,
      "learning_rate": 0.00018749999999999998,
      "epoch": 0.049370525796099726,
      "step": 25
    },
    {
      "loss": 0.8266,
      "grad_norm": 1.494022011756897,
      "learning_rate": 0.000225,
      "epoch": 0.05924463095531968,
      "step": 30
    },
    {
      "loss": 0.7704,
      "grad_norm": 0.42715126276016235,
      "learning_rate": 0.0002625,
      "epoch": 0.06911873611453961,
      "step": 35
    },
    {
      "loss": 0.735,
      "grad_norm": 0.3201207220554352,
      "learning_rate": 0.0003,
      "epoch": 0.07899284127375956,
      "step": 40
    },
    {
      "loss": 0.7419,
      "grad_norm": 2.504087209701538,
      "learning_rate": 0.00029999884801974356,
      "epoch": 0.08886694643297952,
      "step": 45
    },
    {
      "loss": 0.8097,
      "grad_norm": 0.5647741556167603,
      "learning_rate": 0.00029999539209666856,
      "epoch": 0.09874105159219945,
      "step": 50
    },
    {
      "loss": 0.7196,
      "grad_norm": 0.5168487429618835,
      "learning_rate": 0.00029998963228385685,
      "epoch": 0.1086151567514194,
      "step": 55
    },
    {
      "loss": 0.707,
      "grad_norm": 0.6618458032608032,
      "learning_rate": 0.0002999815686697778,
      "epoch": 0.11848926191063935,
      "step": 60
    },
    {
      "loss": 0.7013,
      "grad_norm": 4.516430854797363,
      "learning_rate": 0.0002999712013782864,
      "epoch": 0.1283633670698593,
      "step": 65
    },
    {
      "loss": 0.8418,
      "grad_norm": 0.38112834095954895,
      "learning_rate": 0.0002999585305686214,
      "epoch": 0.13823747222907923,
      "step": 70
    },
    {
      "loss": 0.6765,
      "grad_norm": 0.4867613613605499,
      "learning_rate": 0.00029994355643540323,
      "epoch": 0.1481115773882992,
      "step": 75
    },
    {
      "loss": 0.675,
      "grad_norm": 0.4741807281970978,
      "learning_rate": 0.00029992627920863056,
      "epoch": 0.15798568254751913,
      "step": 80
    },
    {
      "loss": 0.6441,
      "grad_norm": 0.38224297761917114,
      "learning_rate": 0.000299906699153677,
      "epoch": 0.16785978770673907,
      "step": 85
    },
    {
      "loss": 0.64,
      "grad_norm": 0.40791240334510803,
      "learning_rate": 0.00029988481657128714,
      "epoch": 0.17773389286595903,
      "step": 90
    },
    {
      "loss": 0.6528,
      "grad_norm": 0.2506788969039917,
      "learning_rate": 0.00029986063179757166,
      "epoch": 0.18760799802517897,
      "step": 95
    },
    {
      "loss": 0.6278,
      "grad_norm": 0.2861074209213257,
      "learning_rate": 0.0002998341452040023,
      "epoch": 0.1974821031843989,
      "step": 100
    },
    {
      "eval_loss": 0.6399099230766296,
      "eval_runtime": 46.6556,
      "eval_samples_per_second": 154.301,
      "eval_steps_per_second": 9.645,
      "epoch": 0.1974821031843989,
      "step": 100
    },
    {
      "loss": 0.6385,
      "grad_norm": 0.2758176624774933,
      "learning_rate": 0.00029980535719740614,
      "epoch": 0.20735620834361887,
      "step": 105
    },
    {
      "loss": 0.6354,
      "grad_norm": 0.3389268219470978,
      "learning_rate": 0.0002997742682199594,
      "epoch": 0.2172303135028388,
      "step": 110
    },
    {
      "loss": 0.6263,
      "grad_norm": 0.35775822401046753,
      "learning_rate": 0.0002997408787491806,
      "epoch": 0.22710441866205874,
      "step": 115
    },
    {
      "loss": 0.6245,
      "grad_norm": 0.3307900130748749,
      "learning_rate": 0.0002997051892979233,
      "epoch": 0.2369785238212787,
      "step": 120
    },
    {
      "loss": 0.589,
      "grad_norm": 0.49249130487442017,
      "learning_rate": 0.0002996672004143679,
      "epoch": 0.24685262898049865,
      "step": 125
    },
    {
      "loss": 0.6172,
      "grad_norm": 0.48136135935783386,
      "learning_rate": 0.00029962691268201383,
      "epoch": 0.2567267341397186,
      "step": 130
    },
    {
      "loss": 0.5992,
      "grad_norm": 0.3200276494026184,
      "learning_rate": 0.0002995843267196699,
      "epoch": 0.2666008392989385,
      "step": 135
    },
    {
      "loss": 0.5685,
      "grad_norm": 0.39301344752311707,
      "learning_rate": 0.00029953944318144535,
      "epoch": 0.27647494445815846,
      "step": 140
    },
    {
      "loss": 0.5749,
      "grad_norm": 0.27214545011520386,
      "learning_rate": 0.00029949226275673954,
      "epoch": 0.28634904961737845,
      "step": 145
    },
    {
      "loss": 0.5586,
      "grad_norm": 0.43858101963996887,
      "learning_rate": 0.00029944278617023137,
      "epoch": 0.2962231547765984,
      "step": 150
    },
    {
      "loss": 0.5515,
      "grad_norm": 0.541606605052948,
      "learning_rate": 0.0002993910141818681,
      "epoch": 0.3060972599358183,
      "step": 155
    },
    {
      "loss": 0.556,
      "grad_norm": 0.48988083004951477,
      "learning_rate": 0.00029933694758685396,
      "epoch": 0.31597136509503826,
      "step": 160
    },
    {
      "loss": 0.5383,
      "grad_norm": 0.3207962214946747,
      "learning_rate": 0.0002992805872156376,
      "epoch": 0.3258454702542582,
      "step": 165
    },
    {
      "loss": 0.5315,
      "grad_norm": 0.3307562470436096,
      "learning_rate": 0.00029922193393389936,
      "epoch": 0.33571957541347813,
      "step": 170
    },
    {
      "loss": 0.5041,
      "grad_norm": 0.4709561765193939,
      "learning_rate": 0.0002991609886425383,
      "epoch": 0.3455936805726981,
      "step": 175
    },
    {
      "loss": 0.4966,
      "grad_norm": 0.39687275886535645,
      "learning_rate": 0.00029909775227765804,
      "epoch": 0.35546778573191806,
      "step": 180
    },
    {
      "loss": 0.5063,
      "grad_norm": 0.5469186305999756,
      "learning_rate": 0.0002990322258105525,
      "epoch": 0.365341890891138,
      "step": 185
    },
    {
      "loss": 0.5162,
      "grad_norm": 0.46143925189971924,
      "learning_rate": 0.000298964410247691,
      "epoch": 0.37521599605035794,
      "step": 190
    },
    {
      "loss": 0.5,
      "grad_norm": 0.454875111579895,
      "learning_rate": 0.00029889430663070267,
      "epoch": 0.3850901012095779,
      "step": 195
    },
    {
      "loss": 0.4925,
      "grad_norm": 0.5251834392547607,
      "learning_rate": 0.0002988219160363607,
      "epoch": 0.3949642063687978,
      "step": 200
    },
    {
      "eval_loss": 0.525215208530426,
      "eval_runtime": 46.6605,
      "eval_samples_per_second": 154.285,
      "eval_steps_per_second": 9.644,
      "epoch": 0.3949642063687978,
      "step": 200
    },
    {
      "loss": 0.4747,
      "grad_norm": 0.4320356249809265,
      "learning_rate": 0.00029874723957656543,
      "epoch": 0.40483831152801775,
      "step": 205
    },
    {
      "loss": 0.4638,
      "grad_norm": 0.43524378538131714,
      "learning_rate": 0.00029867027839832773,
      "epoch": 0.41471241668723774,
      "step": 210
    },
    {
      "loss": 0.4736,
      "grad_norm": 0.4714348018169403,
      "learning_rate": 0.000298591033683751,
      "epoch": 0.4245865218464577,
      "step": 215
    },
    {
      "loss": 0.4653,
      "grad_norm": 0.3850748836994171,
      "learning_rate": 0.0002985095066500132,
      "epoch": 0.4344606270056776,
      "step": 220
    },
    {
      "loss": 0.4525,
      "grad_norm": 0.41485950350761414,
      "learning_rate": 0.0002984256985493481,
      "epoch": 0.44433473216489755,
      "step": 225
    },
    {
      "loss": 0.4559,
      "grad_norm": 0.4150899052619934,
      "learning_rate": 0.0002983396106690261,
      "epoch": 0.4542088373241175,
      "step": 230
    },
    {
      "loss": 0.4523,
      "grad_norm": 0.5892351269721985,
      "learning_rate": 0.00029825124433133433,
      "epoch": 0.4640829424833374,
      "step": 235
    },
    {
      "loss": 0.4423,
      "grad_norm": 0.5627350807189941,
      "learning_rate": 0.0002981606008935565,
      "epoch": 0.4739570476425574,
      "step": 240
    },
    {
      "loss": 0.4596,
      "grad_norm": 0.5459627509117126,
      "learning_rate": 0.00029806768174795193,
      "epoch": 0.48383115280177735,
      "step": 245
    },
    {
      "loss": 0.4426,
      "grad_norm": 0.5007842183113098,
      "learning_rate": 0.00029797248832173426,
      "epoch": 0.4937052579609973,
      "step": 250
    },
    {
      "loss": 0.4299,
      "grad_norm": 0.5660277009010315,
      "learning_rate": 0.0002978750220770494,
      "epoch": 0.5035793631202172,
      "step": 255
    },
    {
      "loss": 0.4043,
      "grad_norm": 0.5617759227752686,
      "learning_rate": 0.0002977752845109534,
      "epoch": 0.5134534682794372,
      "step": 260
    },
    {
      "loss": 0.4117,
      "grad_norm": 10.37293529510498,
      "learning_rate": 0.00029767327715538873,
      "epoch": 0.5233275734386571,
      "step": 265
    },
    {
      "loss": 0.4237,
      "grad_norm": 0.5515385270118713,
      "learning_rate": 0.00029756900157716176,
      "epoch": 0.533201678597877,
      "step": 270
    },
    {
      "loss": 0.4268,
      "grad_norm": 0.5526891946792603,
      "learning_rate": 0.00029746245937791783,
      "epoch": 0.543075783757097,
      "step": 275
    },
    {
      "loss": 0.4088,
      "grad_norm": 0.6997721195220947,
      "learning_rate": 0.00029735365219411706,
      "epoch": 0.5529498889163169,
      "step": 280
    },
    {
      "loss": 0.4173,
      "grad_norm": 1.0989447832107544,
      "learning_rate": 0.0002972425816970091,
      "epoch": 0.562823994075537,
      "step": 285
    },
    {
      "loss": 0.3924,
      "grad_norm": 0.45413172245025635,
      "learning_rate": 0.0002971292495926077,
      "epoch": 0.5726980992347569,
      "step": 290
    },
    {
      "loss": 0.3877,
      "grad_norm": 0.53785640001297,
      "learning_rate": 0.000297013657621664,
      "epoch": 0.5825722043939768,
      "step": 295
    },
    {
      "loss": 0.3714,
      "grad_norm": 0.390790194272995,
      "learning_rate": 0.0002968958075596403,
      "epoch": 0.5924463095531968,
      "step": 300
    },
    {
      "eval_loss": 0.4794607162475586,
      "eval_runtime": 47.1461,
      "eval_samples_per_second": 152.696,
      "eval_steps_per_second": 9.545,
      "epoch": 0.5924463095531968,
      "step": 300
    },
    {
      "loss": 0.381,
      "grad_norm": 8.648950576782227,
      "learning_rate": 0.00029677570121668253,
      "epoch": 0.6023204147124167,
      "step": 305
    },
    {
      "loss": 0.3731,
      "grad_norm": 0.8738763332366943,
      "learning_rate": 0.0002966533404375925,
      "epoch": 0.6121945198716366,
      "step": 310
    },
    {
      "loss": 0.4041,
      "grad_norm": 1.4086726903915405,
      "learning_rate": 0.00029652872710179967,
      "epoch": 0.6220686250308566,
      "step": 315
    },
    {
      "loss": 0.3719,
      "grad_norm": 0.5402569770812988,
      "learning_rate": 0.0002964018631233319,
      "epoch": 0.6319427301900765,
      "step": 320
    },
    {
      "loss": 0.3624,
      "grad_norm": 0.6380510926246643,
      "learning_rate": 0.00029627275045078657,
      "epoch": 0.6418168353492965,
      "step": 325
    },
    {
      "loss": 0.3707,
      "grad_norm": 0.9881039261817932,
      "learning_rate": 0.0002961413910673003,
      "epoch": 0.6516909405085164,
      "step": 330
    },
    {
      "loss": 0.3544,
      "grad_norm": 0.5078769326210022,
      "learning_rate": 0.00029600778699051875,
      "epoch": 0.6615650456677363,
      "step": 335
    },
    {
      "loss": 0.3498,
      "grad_norm": 0.5671535134315491,
      "learning_rate": 0.0002958719402725653,
      "epoch": 0.6714391508269563,
      "step": 340
    },
    {
      "loss": 0.337,
      "grad_norm": 0.6489645838737488,
      "learning_rate": 0.0002957338530000097,
      "epoch": 0.6813132559861762,
      "step": 345
    },
    {
      "loss": 0.3578,
      "grad_norm": 0.8148907423019409,
      "learning_rate": 0.0002955935272938363,
      "epoch": 0.6911873611453963,
      "step": 350
    },
    {
      "loss": 0.3598,
      "grad_norm": 0.4891752004623413,
      "learning_rate": 0.0002954509653094108,
      "epoch": 0.7010614663046162,
      "step": 355
    },
    {
      "loss": 0.3502,
      "grad_norm": 6.3048176765441895,
      "learning_rate": 0.0002953061692364479,
      "epoch": 0.7109355714638361,
      "step": 360
    },
    {
      "loss": 0.3308,
      "grad_norm": 0.5603845715522766,
      "learning_rate": 0.0002951591412989771,
      "epoch": 0.7208096766230561,
      "step": 365
    },
    {
      "loss": 0.3173,
      "grad_norm": 0.5107300877571106,
      "learning_rate": 0.0002950098837553088,
      "epoch": 0.730683781782276,
      "step": 370
    },
    {
      "loss": 0.3348,
      "grad_norm": 0.6624521613121033,
      "learning_rate": 0.0002948583988979996,
      "epoch": 0.7405578869414959,
      "step": 375
    },
    {
      "loss": 0.3463,
      "grad_norm": 1.0949567556381226,
      "learning_rate": 0.00029470468905381706,
      "epoch": 0.7504319921007159,
      "step": 380
    },
    {
      "loss": 0.3428,
      "grad_norm": 0.6831812262535095,
      "learning_rate": 0.00029454875658370384,
      "epoch": 0.7603060972599358,
      "step": 385
    },
    {
      "loss": 0.3245,
      "grad_norm": 0.8317500352859497,
      "learning_rate": 0.0002943906038827417,
      "epoch": 0.7701802024191557,
      "step": 390
    },
    {
      "loss": 0.3336,
      "grad_norm": 0.5925719141960144,
      "learning_rate": 0.00029423023338011453,
      "epoch": 0.7800543075783757,
      "step": 395
    },
    {
      "loss": 0.3138,
      "grad_norm": 0.5986014604568481,
      "learning_rate": 0.00029406764753907094,
      "epoch": 0.7899284127375956,
      "step": 400
    },
    {
      "eval_loss": 0.48380041122436523,
      "eval_runtime": 46.653,
      "eval_samples_per_second": 154.31,
      "eval_steps_per_second": 9.646,
      "epoch": 0.7899284127375956,
      "step": 400
    },
    {
      "loss": 0.3153,
      "grad_norm": 1.081153392791748,
      "learning_rate": 0.0002939028488568867,
      "epoch": 0.7998025178968156,
      "step": 405
    },
    {
      "loss": 0.3295,
      "grad_norm": 1.3051981925964355,
      "learning_rate": 0.00029373583986482617,
      "epoch": 0.8096766230560355,
      "step": 410
    },
    {
      "loss": 0.3226,
      "grad_norm": 0.7439870834350586,
      "learning_rate": 0.0002935666231281035,
      "epoch": 0.8195507282152555,
      "step": 415
    },
    {
      "loss": 0.2997,
      "grad_norm": 0.5314565896987915,
      "learning_rate": 0.00029339520124584324,
      "epoch": 0.8294248333744755,
      "step": 420
    },
    {
      "loss": 0.2957,
      "grad_norm": 0.6831716299057007,
      "learning_rate": 0.0002932215768510404,
      "epoch": 0.8392989385336954,
      "step": 425
    },
    {
      "loss": 0.2846,
      "grad_norm": 0.49296361207962036,
      "learning_rate": 0.0002930457526105199,
      "epoch": 0.8491730436929154,
      "step": 430
    },
    {
      "loss": 0.285,
      "grad_norm": 0.5440292954444885,
      "learning_rate": 0.00029286773122489583,
      "epoch": 0.8590471488521353,
      "step": 435
    },
    {
      "loss": 0.2861,
      "grad_norm": 0.777549147605896,
      "learning_rate": 0.0002926875154285298,
      "epoch": 0.8689212540113552,
      "step": 440
    },
    {
      "loss": 0.2683,
      "grad_norm": 0.5027139186859131,
      "learning_rate": 0.0002925051079894891,
      "epoch": 0.8787953591705752,
      "step": 445
    },
    {
      "loss": 0.2897,
      "grad_norm": 0.833881139755249,
      "learning_rate": 0.0002923205117095037,
      "epoch": 0.8886694643297951,
      "step": 450
    },
    {
      "loss": 0.2792,
      "grad_norm": 0.7060643434524536,
      "learning_rate": 0.0002921337294239242,
      "epoch": 0.898543569489015,
      "step": 455
    },
    {
      "loss": 0.2663,
      "grad_norm": 0.6060872673988342,
      "learning_rate": 0.00029194476400167717,
      "epoch": 0.908417674648235,
      "step": 460
    },
    {
      "loss": 0.2677,
      "grad_norm": 0.7497969269752502,
      "learning_rate": 0.0002917536183452217,
      "epoch": 0.9182917798074549,
      "step": 465
    },
    {
      "loss": 0.2585,
      "grad_norm": 1.225437879562378,
      "learning_rate": 0.00029156029539050487,
      "epoch": 0.9281658849666748,
      "step": 470
    },
    {
      "loss": 0.2614,
      "grad_norm": 0.48100578784942627,
      "learning_rate": 0.0002913647981069164,
      "epoch": 0.9380399901258949,
      "step": 475
    },
    {
      "loss": 0.2488,
      "grad_norm": 4.289666175842285,
      "learning_rate": 0.0002911671294972429,
      "epoch": 0.9479140952851148,
      "step": 480
    },
    {
      "loss": 0.2548,
      "grad_norm": 0.7958076596260071,
      "learning_rate": 0.0002909672925976224,
      "epoch": 0.9577882004443348,
      "step": 485
    },
    {
      "loss": 0.2583,
      "grad_norm": 0.5692360997200012,
      "learning_rate": 0.00029076529047749694,
      "epoch": 0.9676623056035547,
      "step": 490
    },
    {
      "loss": 0.2643,
      "grad_norm": 0.6238569021224976,
      "learning_rate": 0.00029056112623956595,
      "epoch": 0.9775364107627746,
      "step": 495
    },
    {
      "loss": 0.2548,
      "grad_norm": 1.2753865718841553,
      "learning_rate": 0.00029035480301973837,
      "epoch": 0.9874105159219946,
      "step": 500
    },
    {
      "eval_loss": 0.49340417981147766,
      "eval_runtime": 46.5345,
      "eval_samples_per_second": 154.702,
      "eval_steps_per_second": 9.67,
      "epoch": 0.9874105159219946,
      "step": 500
    },
    {
      "loss": 0.259,
      "grad_norm": 0.5439327359199524,
      "learning_rate": 0.0002901463239870846,
      "epoch": 0.9972846210812145,
      "step": 505
    },
    {
      "loss": 0.3047,
      "grad_norm": 0.7135459184646606,
      "learning_rate": 0.00028993569234378756,
      "epoch": 1.007899284127376,
      "step": 510
    },
    {
      "loss": 0.2138,
      "grad_norm": 0.6349644064903259,
      "learning_rate": 0.00028972291132509396,
      "epoch": 1.0177733892865959,
      "step": 515
    },
    {
      "loss": 0.2317,
      "grad_norm": 0.6580847501754761,
      "learning_rate": 0.0002895079841992642,
      "epoch": 1.0276474944458158,
      "step": 520
    },
    {
      "loss": 0.2267,
      "grad_norm": 0.6198315024375916,
      "learning_rate": 0.00028929091426752236,
      "epoch": 1.0375215996050358,
      "step": 525
    },
    {
      "loss": 0.2394,
      "grad_norm": 0.4667389690876007,
      "learning_rate": 0.0002890717048640055,
      "epoch": 1.0473957047642557,
      "step": 530
    },
    {
      "loss": 0.2017,
      "grad_norm": 0.6087087988853455,
      "learning_rate": 0.00028885035935571223,
      "epoch": 1.0572698099234756,
      "step": 535
    },
    {
      "loss": 0.2006,
      "grad_norm": 0.6860356330871582,
      "learning_rate": 0.00028862688114245137,
      "epoch": 1.0671439150826956,
      "step": 540
    },
    {
      "loss": 0.2155,
      "grad_norm": 0.6059162020683289,
      "learning_rate": 0.00028840127365678943,
      "epoch": 1.0770180202419155,
      "step": 545
    },
    {
      "loss": 0.1982,
      "grad_norm": 0.5954364538192749,
      "learning_rate": 0.000288173540363998,
      "epoch": 1.0868921254011354,
      "step": 550
    },
    {
      "loss": 0.2235,
      "grad_norm": 0.5217412710189819,
      "learning_rate": 0.0002879436847620005,
      "epoch": 1.0967662305603554,
      "step": 555
    },
    {
      "loss": 0.2331,
      "grad_norm": 0.7081472277641296,
      "learning_rate": 0.0002877117103813185,
      "epoch": 1.1066403357195753,
      "step": 560
    },
    {
      "loss": 0.2262,
      "grad_norm": 0.6892974376678467,
      "learning_rate": 0.00028747762078501737,
      "epoch": 1.1165144408787953,
      "step": 565
    },
    {
      "loss": 0.2107,
      "grad_norm": 0.688247561454773,
      "learning_rate": 0.0002872414195686517,
      "epoch": 1.1263885460380152,
      "step": 570
    },
    {
      "loss": 0.2076,
      "grad_norm": 0.6474739909172058,
      "learning_rate": 0.00028700311036021004,
      "epoch": 1.1362626511972351,
      "step": 575
    },
    {
      "loss": 0.204,
      "grad_norm": 0.5220997333526611,
      "learning_rate": 0.00028676269682005904,
      "epoch": 1.1461367563564553,
      "step": 580
    },
    {
      "loss": 0.2003,
      "grad_norm": 0.5570688843727112,
      "learning_rate": 0.0002865201826408874,
      "epoch": 1.1560108615156752,
      "step": 585
    },
    {
      "loss": 0.2028,
      "grad_norm": 0.5454896092414856,
      "learning_rate": 0.0002862755715476491,
      "epoch": 1.1658849666748952,
      "step": 590
    },
    {
      "loss": 0.2011,
      "grad_norm": 0.5824912190437317,
      "learning_rate": 0.00028602886729750613,
      "epoch": 1.175759071834115,
      "step": 595
    },
    {
      "loss": 0.1875,
      "grad_norm": 0.612477719783783,
      "learning_rate": 0.00028578007367977074,
      "epoch": 1.185633176993335,
      "step": 600
    },
    {
      "eval_loss": 0.5274343490600586,
      "eval_runtime": 46.5304,
      "eval_samples_per_second": 154.716,
      "eval_steps_per_second": 9.671,
      "epoch": 1.185633176993335,
      "step": 600
    },
    {
      "train_runtime": 1998.6484,
      "train_samples_per_second": 259.379,
      "train_steps_per_second": 2.025,
      "total_flos": 2.946671339427594e+17,
      "train_loss": 0.5542280136545499,
      "epoch": 1.185633176993335,
      "step": 600
    }
  ],
  "best_metric": 0.4794607162475586,
  "best_model_checkpoint": "saves/adpr-llama/checkpoint-300",
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "is_hyper_param_search": false,
  "trial_name": null,
  "trial_params": null,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 3
      }
    },
    "TrainerControl": {
      "args": {
        "should_training_stop": true,
        "should_epoch_stop": false,
        "should_save": true,
        "should_evaluate": false,
        "should_log": false
      },
      "attributes": {}
    }
  },
  "model": "PeftModelForCausalLM(\n  (base_model): LoraModel(\n    (model): LlamaForCausalLM(\n      (model): LlamaModel(\n        (embed_tokens): Embedding(32000, 4096, padding_idx=0)\n        (layers): ModuleList(\n          (0-31): 32 x LlamaDecoderLayer(\n            (self_attn): LlamaAttention(\n              (q_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=4096, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (k_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=4096, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (v_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=4096, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (o_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=4096, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=4096, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n            )\n            (mlp): LlamaMLP(\n              (gate_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=11008, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=11008, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (up_proj): lora.Linear(\n                (base_layer): Linear(in_features=4096, out_features=11008, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=4096, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=11008, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (down_proj): lora.Linear(\n                (base_layer): Linear(in_features=11008, out_features=4096, bias=False)\n                (lora_dropout): ModuleDict(\n                  (default): Dropout(p=0.05, inplace=False)\n                )\n                (lora_A): ModuleDict(\n                  (default): Linear(in_features=11008, out_features=64, bias=False)\n                )\n                (lora_B): ModuleDict(\n                  (default): Linear(in_features=64, out_features=4096, bias=False)\n                )\n                (lora_embedding_A): ParameterDict()\n                (lora_embedding_B): ParameterDict()\n                (lora_magnitude_vector): ModuleDict()\n              )\n              (act_fn): SiLU()\n            )\n            (input_layernorm): LlamaRMSNorm((4096,), eps=1e-05)\n            (post_attention_layernorm): LlamaRMSNorm((4096,), eps=1e-05)\n          )\n        )\n        (norm): LlamaRMSNorm((4096,), eps=1e-05)\n        (rotary_emb): LlamaRotaryEmbedding()\n      )\n      (lm_head): Linear(in_features=4096, out_features=32000, bias=False)\n    )\n  )\n)",
  "optimizer": "AcceleratedOptimizer (\nParameter Group 0\n    amsgrad: False\n    betas: (0.9, 0.999)\n    capturable: False\n    decoupled_weight_decay: True\n    differentiable: False\n    eps: 1e-08\n    foreach: None\n    fused: None\n    initial_lr: 0.0003\n    lr: 0.00028578007367977074\n    maximize: False\n    weight_decay: 0.0\n\nParameter Group 1\n    amsgrad: False\n    betas: (0.9, 0.999)\n    capturable: False\n    decoupled_weight_decay: True\n    differentiable: False\n    eps: 1e-08\n    foreach: None\n    fused: None\n    initial_lr: 0.0003\n    lr: 0.00028578007367977074\n    maximize: False\n    weight_decay: 0.0\n)",
  "lr_scheduler": "<torch.optim.lr_scheduler.LambdaLR object at 0x7de720bc1580>",
  "train_dataloader": "<accelerate.data_loader.DataLoaderShard object at 0x7de619edf170>"
}