{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 939,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 4.9730574359401585,
      "learning_rate": 5e-06,
      "loss": 0.9178,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 2.414364422517365,
      "learning_rate": 5e-06,
      "loss": 0.8185,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 2.355627570386528,
      "learning_rate": 5e-06,
      "loss": 0.7955,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 0.9368274495065052,
      "learning_rate": 5e-06,
      "loss": 0.777,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 1.0303864547714687,
      "learning_rate": 5e-06,
      "loss": 0.7544,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 0.8799746481470822,
      "learning_rate": 5e-06,
      "loss": 0.7414,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 2.571145704302205,
      "learning_rate": 5e-06,
      "loss": 0.7335,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 1.4093624036341388,
      "learning_rate": 5e-06,
      "loss": 0.7174,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 0.8615478628829397,
      "learning_rate": 5e-06,
      "loss": 0.7318,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.5831162653054314,
      "learning_rate": 5e-06,
      "loss": 0.72,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.6173922126615862,
      "learning_rate": 5e-06,
      "loss": 0.716,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.8888214325930446,
      "learning_rate": 5e-06,
      "loss": 0.7134,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.6979078907766634,
      "learning_rate": 5e-06,
      "loss": 0.7021,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.6491577783946951,
      "learning_rate": 5e-06,
      "loss": 0.6991,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 0.5395727376776203,
      "learning_rate": 5e-06,
      "loss": 0.7068,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.689753166346745,
      "learning_rate": 5e-06,
      "loss": 0.6935,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.5521859868964065,
      "learning_rate": 5e-06,
      "loss": 0.705,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.5708494730680989,
      "learning_rate": 5e-06,
      "loss": 0.6961,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.7742403437422439,
      "learning_rate": 5e-06,
      "loss": 0.7045,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.6022796802365044,
      "learning_rate": 5e-06,
      "loss": 0.6911,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.5364796054217524,
      "learning_rate": 5e-06,
      "loss": 0.695,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.47930189735330775,
      "learning_rate": 5e-06,
      "loss": 0.6961,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 0.5376571771457028,
      "learning_rate": 5e-06,
      "loss": 0.7035,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.7570921090383397,
      "learning_rate": 5e-06,
      "loss": 0.6801,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.5375423682771385,
      "learning_rate": 5e-06,
      "loss": 0.6809,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.5836884654889642,
      "learning_rate": 5e-06,
      "loss": 0.6864,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.598608895789362,
      "learning_rate": 5e-06,
      "loss": 0.6902,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.5331459973786111,
      "learning_rate": 5e-06,
      "loss": 0.6879,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.5067341091409512,
      "learning_rate": 5e-06,
      "loss": 0.6881,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.5948122722251207,
      "learning_rate": 5e-06,
      "loss": 0.6868,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.6883144495229085,
      "learning_rate": 5e-06,
      "loss": 0.6818,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.680176854133606,
      "eval_runtime": 30.2696,
      "eval_samples_per_second": 277.968,
      "eval_steps_per_second": 1.09,
      "step": 313
    },
    {
      "epoch": 1.0223642172523961,
      "grad_norm": 0.7468894293006159,
      "learning_rate": 5e-06,
      "loss": 0.656,
      "step": 320
    },
    {
      "epoch": 1.0543130990415335,
      "grad_norm": 0.6942190077761362,
      "learning_rate": 5e-06,
      "loss": 0.6387,
      "step": 330
    },
    {
      "epoch": 1.0862619808306708,
      "grad_norm": 0.6242193449068775,
      "learning_rate": 5e-06,
      "loss": 0.6407,
      "step": 340
    },
    {
      "epoch": 1.1182108626198084,
      "grad_norm": 0.6329861406612104,
      "learning_rate": 5e-06,
      "loss": 0.6428,
      "step": 350
    },
    {
      "epoch": 1.1501597444089458,
      "grad_norm": 0.6142460964851904,
      "learning_rate": 5e-06,
      "loss": 0.6356,
      "step": 360
    },
    {
      "epoch": 1.182108626198083,
      "grad_norm": 0.6850789950403501,
      "learning_rate": 5e-06,
      "loss": 0.636,
      "step": 370
    },
    {
      "epoch": 1.2140575079872205,
      "grad_norm": 0.5355828508870474,
      "learning_rate": 5e-06,
      "loss": 0.6382,
      "step": 380
    },
    {
      "epoch": 1.2460063897763578,
      "grad_norm": 0.5629126693045496,
      "learning_rate": 5e-06,
      "loss": 0.6344,
      "step": 390
    },
    {
      "epoch": 1.2779552715654952,
      "grad_norm": 0.5204488937445614,
      "learning_rate": 5e-06,
      "loss": 0.6401,
      "step": 400
    },
    {
      "epoch": 1.3099041533546325,
      "grad_norm": 0.5396176523426488,
      "learning_rate": 5e-06,
      "loss": 0.6382,
      "step": 410
    },
    {
      "epoch": 1.34185303514377,
      "grad_norm": 0.6023672235090743,
      "learning_rate": 5e-06,
      "loss": 0.6417,
      "step": 420
    },
    {
      "epoch": 1.3738019169329074,
      "grad_norm": 0.5459880912136352,
      "learning_rate": 5e-06,
      "loss": 0.6434,
      "step": 430
    },
    {
      "epoch": 1.4057507987220448,
      "grad_norm": 0.5092923506095731,
      "learning_rate": 5e-06,
      "loss": 0.6366,
      "step": 440
    },
    {
      "epoch": 1.4376996805111821,
      "grad_norm": 0.6382479698360073,
      "learning_rate": 5e-06,
      "loss": 0.6353,
      "step": 450
    },
    {
      "epoch": 1.4696485623003195,
      "grad_norm": 0.7533985947305522,
      "learning_rate": 5e-06,
      "loss": 0.6392,
      "step": 460
    },
    {
      "epoch": 1.5015974440894568,
      "grad_norm": 0.7229864015275371,
      "learning_rate": 5e-06,
      "loss": 0.6381,
      "step": 470
    },
    {
      "epoch": 1.5335463258785942,
      "grad_norm": 0.5758239404648058,
      "learning_rate": 5e-06,
      "loss": 0.636,
      "step": 480
    },
    {
      "epoch": 1.5654952076677318,
      "grad_norm": 0.5422584017870065,
      "learning_rate": 5e-06,
      "loss": 0.6393,
      "step": 490
    },
    {
      "epoch": 1.5974440894568689,
      "grad_norm": 0.500259651199132,
      "learning_rate": 5e-06,
      "loss": 0.6441,
      "step": 500
    },
    {
      "epoch": 1.6293929712460065,
      "grad_norm": 0.8813431746686585,
      "learning_rate": 5e-06,
      "loss": 0.6308,
      "step": 510
    },
    {
      "epoch": 1.6613418530351438,
      "grad_norm": 0.49923586083798,
      "learning_rate": 5e-06,
      "loss": 0.6365,
      "step": 520
    },
    {
      "epoch": 1.6932907348242812,
      "grad_norm": 0.46329712814558677,
      "learning_rate": 5e-06,
      "loss": 0.6279,
      "step": 530
    },
    {
      "epoch": 1.7252396166134185,
      "grad_norm": 0.7606638693711257,
      "learning_rate": 5e-06,
      "loss": 0.6341,
      "step": 540
    },
    {
      "epoch": 1.7571884984025559,
      "grad_norm": 0.5490152416615772,
      "learning_rate": 5e-06,
      "loss": 0.6378,
      "step": 550
    },
    {
      "epoch": 1.7891373801916934,
      "grad_norm": 0.5845892096604395,
      "learning_rate": 5e-06,
      "loss": 0.6395,
      "step": 560
    },
    {
      "epoch": 1.8210862619808306,
      "grad_norm": 0.5147171055711276,
      "learning_rate": 5e-06,
      "loss": 0.6297,
      "step": 570
    },
    {
      "epoch": 1.8530351437699681,
      "grad_norm": 0.5052763678219258,
      "learning_rate": 5e-06,
      "loss": 0.6342,
      "step": 580
    },
    {
      "epoch": 1.8849840255591053,
      "grad_norm": 0.5390734384968956,
      "learning_rate": 5e-06,
      "loss": 0.6369,
      "step": 590
    },
    {
      "epoch": 1.9169329073482428,
      "grad_norm": 0.6413516519865096,
      "learning_rate": 5e-06,
      "loss": 0.6418,
      "step": 600
    },
    {
      "epoch": 1.9488817891373802,
      "grad_norm": 0.5239174298332788,
      "learning_rate": 5e-06,
      "loss": 0.6274,
      "step": 610
    },
    {
      "epoch": 1.9808306709265175,
      "grad_norm": 0.7039020492664947,
      "learning_rate": 5e-06,
      "loss": 0.6385,
      "step": 620
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6707362532615662,
      "eval_runtime": 29.7735,
      "eval_samples_per_second": 282.6,
      "eval_steps_per_second": 1.108,
      "step": 626
    },
    {
      "epoch": 2.012779552715655,
      "grad_norm": 0.9555964669975626,
      "learning_rate": 5e-06,
      "loss": 0.6152,
      "step": 630
    },
    {
      "epoch": 2.0447284345047922,
      "grad_norm": 0.6929923210576585,
      "learning_rate": 5e-06,
      "loss": 0.5881,
      "step": 640
    },
    {
      "epoch": 2.07667731629393,
      "grad_norm": 0.5186537877634542,
      "learning_rate": 5e-06,
      "loss": 0.5871,
      "step": 650
    },
    {
      "epoch": 2.108626198083067,
      "grad_norm": 0.5532963474572669,
      "learning_rate": 5e-06,
      "loss": 0.5896,
      "step": 660
    },
    {
      "epoch": 2.1405750798722045,
      "grad_norm": 0.5667781850304423,
      "learning_rate": 5e-06,
      "loss": 0.589,
      "step": 670
    },
    {
      "epoch": 2.1725239616613417,
      "grad_norm": 0.5285098402420302,
      "learning_rate": 5e-06,
      "loss": 0.5901,
      "step": 680
    },
    {
      "epoch": 2.2044728434504792,
      "grad_norm": 0.536646702155888,
      "learning_rate": 5e-06,
      "loss": 0.5863,
      "step": 690
    },
    {
      "epoch": 2.236421725239617,
      "grad_norm": 0.5015237529663761,
      "learning_rate": 5e-06,
      "loss": 0.5846,
      "step": 700
    },
    {
      "epoch": 2.268370607028754,
      "grad_norm": 0.6152245702781334,
      "learning_rate": 5e-06,
      "loss": 0.5885,
      "step": 710
    },
    {
      "epoch": 2.3003194888178915,
      "grad_norm": 0.6897591535184826,
      "learning_rate": 5e-06,
      "loss": 0.584,
      "step": 720
    },
    {
      "epoch": 2.3322683706070286,
      "grad_norm": 0.48574756268499075,
      "learning_rate": 5e-06,
      "loss": 0.5888,
      "step": 730
    },
    {
      "epoch": 2.364217252396166,
      "grad_norm": 0.7776088479205505,
      "learning_rate": 5e-06,
      "loss": 0.591,
      "step": 740
    },
    {
      "epoch": 2.3961661341853033,
      "grad_norm": 0.5643419759040877,
      "learning_rate": 5e-06,
      "loss": 0.5975,
      "step": 750
    },
    {
      "epoch": 2.428115015974441,
      "grad_norm": 0.7618894474745995,
      "learning_rate": 5e-06,
      "loss": 0.5907,
      "step": 760
    },
    {
      "epoch": 2.460063897763578,
      "grad_norm": 0.566521654030305,
      "learning_rate": 5e-06,
      "loss": 0.5902,
      "step": 770
    },
    {
      "epoch": 2.4920127795527156,
      "grad_norm": 0.539754953993636,
      "learning_rate": 5e-06,
      "loss": 0.5884,
      "step": 780
    },
    {
      "epoch": 2.523961661341853,
      "grad_norm": 0.5625428853361434,
      "learning_rate": 5e-06,
      "loss": 0.5818,
      "step": 790
    },
    {
      "epoch": 2.5559105431309903,
      "grad_norm": 0.6279591943173818,
      "learning_rate": 5e-06,
      "loss": 0.588,
      "step": 800
    },
    {
      "epoch": 2.587859424920128,
      "grad_norm": 0.6224175939518917,
      "learning_rate": 5e-06,
      "loss": 0.5934,
      "step": 810
    },
    {
      "epoch": 2.619808306709265,
      "grad_norm": 0.5642613523825399,
      "learning_rate": 5e-06,
      "loss": 0.5865,
      "step": 820
    },
    {
      "epoch": 2.6517571884984026,
      "grad_norm": 0.6865432310936253,
      "learning_rate": 5e-06,
      "loss": 0.5926,
      "step": 830
    },
    {
      "epoch": 2.68370607028754,
      "grad_norm": 0.5373522424367686,
      "learning_rate": 5e-06,
      "loss": 0.5921,
      "step": 840
    },
    {
      "epoch": 2.7156549520766773,
      "grad_norm": 0.567457215981538,
      "learning_rate": 5e-06,
      "loss": 0.5954,
      "step": 850
    },
    {
      "epoch": 2.747603833865815,
      "grad_norm": 0.6426061724785515,
      "learning_rate": 5e-06,
      "loss": 0.5988,
      "step": 860
    },
    {
      "epoch": 2.779552715654952,
      "grad_norm": 0.59442339830324,
      "learning_rate": 5e-06,
      "loss": 0.6008,
      "step": 870
    },
    {
      "epoch": 2.8115015974440896,
      "grad_norm": 0.5592579452365836,
      "learning_rate": 5e-06,
      "loss": 0.6022,
      "step": 880
    },
    {
      "epoch": 2.8434504792332267,
      "grad_norm": 0.5094979594863669,
      "learning_rate": 5e-06,
      "loss": 0.592,
      "step": 890
    },
    {
      "epoch": 2.8753993610223643,
      "grad_norm": 0.5091387458726695,
      "learning_rate": 5e-06,
      "loss": 0.6,
      "step": 900
    },
    {
      "epoch": 2.9073482428115014,
      "grad_norm": 0.5094298110498509,
      "learning_rate": 5e-06,
      "loss": 0.5938,
      "step": 910
    },
    {
      "epoch": 2.939297124600639,
      "grad_norm": 0.5130578982623665,
      "learning_rate": 5e-06,
      "loss": 0.586,
      "step": 920
    },
    {
      "epoch": 2.9712460063897765,
      "grad_norm": 0.4928009047818368,
      "learning_rate": 5e-06,
      "loss": 0.5911,
      "step": 930
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6763173341751099,
      "eval_runtime": 29.5519,
      "eval_samples_per_second": 284.719,
      "eval_steps_per_second": 1.117,
      "step": 939
    },
    {
      "epoch": 3.0,
      "step": 939,
      "total_flos": 1572859973468160.0,
      "train_loss": 0.648996304907103,
      "train_runtime": 5978.5297,
      "train_samples_per_second": 80.212,
      "train_steps_per_second": 0.157
    }
  ],
  "logging_steps": 10,
  "max_steps": 939,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1572859973468160.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}