{
  "best_metric": 0.9917032698877501,
  "best_model_checkpoint": "./checkpoint/checkpoint-905",
  "epoch": 4.997245179063361,
  "global_step": 905,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06,
      "learning_rate": 3.6764705882352942e-06,
      "loss": 1.0502,
      "step": 10
    },
    {
      "epoch": 0.11,
      "learning_rate": 7.3529411764705884e-06,
      "loss": 1.0183,
      "step": 20
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.1029411764705883e-05,
      "loss": 0.9565,
      "step": 30
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.4705882352941177e-05,
      "loss": 0.8426,
      "step": 40
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.8382352941176472e-05,
      "loss": 0.6867,
      "step": 50
    },
    {
      "epoch": 0.33,
      "learning_rate": 2.2058823529411766e-05,
      "loss": 0.5298,
      "step": 60
    },
    {
      "epoch": 0.39,
      "learning_rate": 2.5735294117647057e-05,
      "loss": 0.3856,
      "step": 70
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.9411764705882354e-05,
      "loss": 0.2925,
      "step": 80
    },
    {
      "epoch": 0.5,
      "learning_rate": 3.308823529411765e-05,
      "loss": 0.234,
      "step": 90
    },
    {
      "epoch": 0.55,
      "learning_rate": 3.6764705882352945e-05,
      "loss": 0.1777,
      "step": 100
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.044117647058824e-05,
      "loss": 0.1539,
      "step": 110
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.411764705882353e-05,
      "loss": 0.114,
      "step": 120
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.7794117647058826e-05,
      "loss": 0.123,
      "step": 130
    },
    {
      "epoch": 0.77,
      "learning_rate": 5.147058823529411e-05,
      "loss": 0.1118,
      "step": 140
    },
    {
      "epoch": 0.83,
      "learning_rate": 5.514705882352942e-05,
      "loss": 0.1032,
      "step": 150
    },
    {
      "epoch": 0.88,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.1201,
      "step": 160
    },
    {
      "epoch": 0.94,
      "learning_rate": 6.25e-05,
      "loss": 0.0914,
      "step": 170
    },
    {
      "epoch": 0.99,
      "learning_rate": 6.61764705882353e-05,
      "loss": 0.0825,
      "step": 180
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9858467545143973,
      "eval_loss": 0.07265167683362961,
      "eval_runtime": 59.8085,
      "eval_samples_per_second": 68.519,
      "eval_steps_per_second": 2.157,
      "step": 181
    },
    {
      "epoch": 1.05,
      "learning_rate": 6.985294117647059e-05,
      "loss": 0.0836,
      "step": 190
    },
    {
      "epoch": 1.1,
      "learning_rate": 7.352941176470589e-05,
      "loss": 0.0691,
      "step": 200
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.720588235294119e-05,
      "loss": 0.0834,
      "step": 210
    },
    {
      "epoch": 1.21,
      "learning_rate": 8.088235294117648e-05,
      "loss": 0.0876,
      "step": 220
    },
    {
      "epoch": 1.27,
      "learning_rate": 8.455882352941176e-05,
      "loss": 0.077,
      "step": 230
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.823529411764706e-05,
      "loss": 0.0671,
      "step": 240
    },
    {
      "epoch": 1.38,
      "learning_rate": 9.191176470588235e-05,
      "loss": 0.0623,
      "step": 250
    },
    {
      "epoch": 1.44,
      "learning_rate": 9.558823529411765e-05,
      "loss": 0.0539,
      "step": 260
    },
    {
      "epoch": 1.49,
      "learning_rate": 9.926470588235295e-05,
      "loss": 0.0645,
      "step": 270
    },
    {
      "epoch": 1.55,
      "learning_rate": 9.873617693522908e-05,
      "loss": 0.061,
      "step": 280
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.715639810426542e-05,
      "loss": 0.0519,
      "step": 290
    },
    {
      "epoch": 1.66,
      "learning_rate": 9.557661927330174e-05,
      "loss": 0.0697,
      "step": 300
    },
    {
      "epoch": 1.71,
      "learning_rate": 9.399684044233808e-05,
      "loss": 0.062,
      "step": 310
    },
    {
      "epoch": 1.77,
      "learning_rate": 9.241706161137442e-05,
      "loss": 0.0681,
      "step": 320
    },
    {
      "epoch": 1.82,
      "learning_rate": 9.083728278041075e-05,
      "loss": 0.0698,
      "step": 330
    },
    {
      "epoch": 1.88,
      "learning_rate": 8.925750394944709e-05,
      "loss": 0.067,
      "step": 340
    },
    {
      "epoch": 1.93,
      "learning_rate": 8.767772511848341e-05,
      "loss": 0.0728,
      "step": 350
    },
    {
      "epoch": 1.99,
      "learning_rate": 8.609794628751975e-05,
      "loss": 0.06,
      "step": 360
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9812103465104929,
      "eval_loss": 0.06039927154779434,
      "eval_runtime": 59.4929,
      "eval_samples_per_second": 68.882,
      "eval_steps_per_second": 2.168,
      "step": 362
    },
    {
      "epoch": 2.04,
      "learning_rate": 8.451816745655609e-05,
      "loss": 0.065,
      "step": 370
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.293838862559243e-05,
      "loss": 0.0635,
      "step": 380
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.135860979462876e-05,
      "loss": 0.0532,
      "step": 390
    },
    {
      "epoch": 2.21,
      "learning_rate": 7.977883096366509e-05,
      "loss": 0.0497,
      "step": 400
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.819905213270142e-05,
      "loss": 0.0665,
      "step": 410
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.661927330173776e-05,
      "loss": 0.0525,
      "step": 420
    },
    {
      "epoch": 2.37,
      "learning_rate": 7.50394944707741e-05,
      "loss": 0.0514,
      "step": 430
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.345971563981043e-05,
      "loss": 0.0649,
      "step": 440
    },
    {
      "epoch": 2.48,
      "learning_rate": 7.187993680884676e-05,
      "loss": 0.064,
      "step": 450
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.03001579778831e-05,
      "loss": 0.0488,
      "step": 460
    },
    {
      "epoch": 2.6,
      "learning_rate": 6.872037914691943e-05,
      "loss": 0.0405,
      "step": 470
    },
    {
      "epoch": 2.65,
      "learning_rate": 6.714060031595577e-05,
      "loss": 0.0357,
      "step": 480
    },
    {
      "epoch": 2.71,
      "learning_rate": 6.556082148499211e-05,
      "loss": 0.0439,
      "step": 490
    },
    {
      "epoch": 2.76,
      "learning_rate": 6.398104265402843e-05,
      "loss": 0.0469,
      "step": 500
    },
    {
      "epoch": 2.82,
      "learning_rate": 6.240126382306477e-05,
      "loss": 0.0289,
      "step": 510
    },
    {
      "epoch": 2.87,
      "learning_rate": 6.0821484992101105e-05,
      "loss": 0.0431,
      "step": 520
    },
    {
      "epoch": 2.93,
      "learning_rate": 5.924170616113744e-05,
      "loss": 0.0284,
      "step": 530
    },
    {
      "epoch": 2.98,
      "learning_rate": 5.766192733017378e-05,
      "loss": 0.0467,
      "step": 540
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9838945827232797,
      "eval_loss": 0.04717087373137474,
      "eval_runtime": 59.6298,
      "eval_samples_per_second": 68.724,
      "eval_steps_per_second": 2.163,
      "step": 543
    },
    {
      "epoch": 3.04,
      "learning_rate": 5.608214849921012e-05,
      "loss": 0.0503,
      "step": 550
    },
    {
      "epoch": 3.09,
      "learning_rate": 5.450236966824645e-05,
      "loss": 0.043,
      "step": 560
    },
    {
      "epoch": 3.15,
      "learning_rate": 5.2922590837282785e-05,
      "loss": 0.0381,
      "step": 570
    },
    {
      "epoch": 3.2,
      "learning_rate": 5.134281200631912e-05,
      "loss": 0.0295,
      "step": 580
    },
    {
      "epoch": 3.26,
      "learning_rate": 4.976303317535545e-05,
      "loss": 0.0411,
      "step": 590
    },
    {
      "epoch": 3.31,
      "learning_rate": 4.818325434439179e-05,
      "loss": 0.0242,
      "step": 600
    },
    {
      "epoch": 3.37,
      "learning_rate": 4.660347551342813e-05,
      "loss": 0.0365,
      "step": 610
    },
    {
      "epoch": 3.42,
      "learning_rate": 4.502369668246446e-05,
      "loss": 0.021,
      "step": 620
    },
    {
      "epoch": 3.48,
      "learning_rate": 4.3443917851500794e-05,
      "loss": 0.0492,
      "step": 630
    },
    {
      "epoch": 3.53,
      "learning_rate": 4.1864139020537125e-05,
      "loss": 0.0391,
      "step": 640
    },
    {
      "epoch": 3.59,
      "learning_rate": 4.028436018957346e-05,
      "loss": 0.0291,
      "step": 650
    },
    {
      "epoch": 3.64,
      "learning_rate": 3.87045813586098e-05,
      "loss": 0.0317,
      "step": 660
    },
    {
      "epoch": 3.7,
      "learning_rate": 3.712480252764613e-05,
      "loss": 0.0355,
      "step": 670
    },
    {
      "epoch": 3.75,
      "learning_rate": 3.554502369668247e-05,
      "loss": 0.0407,
      "step": 680
    },
    {
      "epoch": 3.81,
      "learning_rate": 3.39652448657188e-05,
      "loss": 0.0257,
      "step": 690
    },
    {
      "epoch": 3.87,
      "learning_rate": 3.2385466034755135e-05,
      "loss": 0.0382,
      "step": 700
    },
    {
      "epoch": 3.92,
      "learning_rate": 3.080568720379147e-05,
      "loss": 0.0323,
      "step": 710
    },
    {
      "epoch": 3.98,
      "learning_rate": 2.9225908372827802e-05,
      "loss": 0.0233,
      "step": 720
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9897510980966325,
      "eval_loss": 0.028671853244304657,
      "eval_runtime": 59.9206,
      "eval_samples_per_second": 68.391,
      "eval_steps_per_second": 2.153,
      "step": 724
    },
    {
      "epoch": 4.03,
      "learning_rate": 2.764612954186414e-05,
      "loss": 0.031,
      "step": 730
    },
    {
      "epoch": 4.09,
      "learning_rate": 2.6066350710900477e-05,
      "loss": 0.0272,
      "step": 740
    },
    {
      "epoch": 4.14,
      "learning_rate": 2.448657187993681e-05,
      "loss": 0.0283,
      "step": 750
    },
    {
      "epoch": 4.2,
      "learning_rate": 2.2906793048973144e-05,
      "loss": 0.0208,
      "step": 760
    },
    {
      "epoch": 4.25,
      "learning_rate": 2.132701421800948e-05,
      "loss": 0.0439,
      "step": 770
    },
    {
      "epoch": 4.31,
      "learning_rate": 1.9747235387045815e-05,
      "loss": 0.0339,
      "step": 780
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.816745655608215e-05,
      "loss": 0.0298,
      "step": 790
    },
    {
      "epoch": 4.42,
      "learning_rate": 1.6587677725118483e-05,
      "loss": 0.0279,
      "step": 800
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.500789889415482e-05,
      "loss": 0.0227,
      "step": 810
    },
    {
      "epoch": 4.53,
      "learning_rate": 1.3428120063191154e-05,
      "loss": 0.0215,
      "step": 820
    },
    {
      "epoch": 4.58,
      "learning_rate": 1.184834123222749e-05,
      "loss": 0.0276,
      "step": 830
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.0268562401263823e-05,
      "loss": 0.0356,
      "step": 840
    },
    {
      "epoch": 4.69,
      "learning_rate": 8.688783570300159e-06,
      "loss": 0.0168,
      "step": 850
    },
    {
      "epoch": 4.75,
      "learning_rate": 7.109004739336493e-06,
      "loss": 0.0316,
      "step": 860
    },
    {
      "epoch": 4.8,
      "learning_rate": 5.529225908372828e-06,
      "loss": 0.0237,
      "step": 870
    },
    {
      "epoch": 4.86,
      "learning_rate": 3.949447077409163e-06,
      "loss": 0.0207,
      "step": 880
    },
    {
      "epoch": 4.91,
      "learning_rate": 2.3696682464454976e-06,
      "loss": 0.0305,
      "step": 890
    },
    {
      "epoch": 4.97,
      "learning_rate": 7.898894154818326e-07,
      "loss": 0.0327,
      "step": 900
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9917032698877501,
      "eval_loss": 0.025840837508440018,
      "eval_runtime": 59.7361,
      "eval_samples_per_second": 68.602,
      "eval_steps_per_second": 2.159,
      "step": 905
    }
  ],
  "max_steps": 905,
  "num_train_epochs": 5,
  "total_flos": 8.992561039593578e+18,
  "trial_name": null,
  "trial_params": null
}