{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 100,
  "global_step": 915,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 0.00019781420765027324,
      "loss": 2.1723,
      "step": 10
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019562841530054644,
      "loss": 1.7534,
      "step": 20
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019344262295081967,
      "loss": 1.4381,
      "step": 30
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001912568306010929,
      "loss": 1.3181,
      "step": 40
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018907103825136615,
      "loss": 1.2203,
      "step": 50
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00018688524590163935,
      "loss": 1.0679,
      "step": 60
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00018469945355191258,
      "loss": 1.0329,
      "step": 70
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001825136612021858,
      "loss": 0.9817,
      "step": 80
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00018032786885245904,
      "loss": 0.866,
      "step": 90
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00017814207650273224,
      "loss": 0.8634,
      "step": 100
    },
    {
      "epoch": 0.55,
      "eval_accuracy": 0.6975,
      "eval_loss": 0.9265598654747009,
      "eval_runtime": 17.6341,
      "eval_samples_per_second": 68.05,
      "eval_steps_per_second": 8.506,
      "step": 100
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00017595628415300547,
      "loss": 0.6979,
      "step": 110
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001737704918032787,
      "loss": 0.7132,
      "step": 120
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00017158469945355192,
      "loss": 0.6692,
      "step": 130
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00016939890710382515,
      "loss": 0.6,
      "step": 140
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00016721311475409838,
      "loss": 0.5497,
      "step": 150
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0001650273224043716,
      "loss": 0.4835,
      "step": 160
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001628415300546448,
      "loss": 0.4699,
      "step": 170
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00016065573770491804,
      "loss": 0.4311,
      "step": 180
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00015846994535519127,
      "loss": 0.4363,
      "step": 190
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001562841530054645,
      "loss": 0.3225,
      "step": 200
    },
    {
      "epoch": 1.09,
      "eval_accuracy": 0.7325,
      "eval_loss": 0.8994238376617432,
      "eval_runtime": 16.2421,
      "eval_samples_per_second": 73.882,
      "eval_steps_per_second": 9.235,
      "step": 200
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001540983606557377,
      "loss": 0.3329,
      "step": 210
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015191256830601093,
      "loss": 0.3262,
      "step": 220
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00014972677595628418,
      "loss": 0.3394,
      "step": 230
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00014754098360655738,
      "loss": 0.3319,
      "step": 240
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001453551912568306,
      "loss": 0.2554,
      "step": 250
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014316939890710384,
      "loss": 0.2912,
      "step": 260
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00014098360655737707,
      "loss": 0.2801,
      "step": 270
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00013879781420765027,
      "loss": 0.2443,
      "step": 280
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0001366120218579235,
      "loss": 0.278,
      "step": 290
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00013442622950819673,
      "loss": 0.2353,
      "step": 300
    },
    {
      "epoch": 1.64,
      "eval_accuracy": 0.73,
      "eval_loss": 0.968270480632782,
      "eval_runtime": 17.23,
      "eval_samples_per_second": 69.646,
      "eval_steps_per_second": 8.706,
      "step": 300
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00013224043715846995,
      "loss": 0.2092,
      "step": 310
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00013005464480874316,
      "loss": 0.1959,
      "step": 320
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0001278688524590164,
      "loss": 0.1841,
      "step": 330
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00012568306010928964,
      "loss": 0.2678,
      "step": 340
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00012349726775956284,
      "loss": 0.2366,
      "step": 350
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00012131147540983607,
      "loss": 0.2198,
      "step": 360
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0001191256830601093,
      "loss": 0.1459,
      "step": 370
    },
    {
      "epoch": 2.08,
      "learning_rate": 0.00011693989071038251,
      "loss": 0.0904,
      "step": 380
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.00011475409836065574,
      "loss": 0.1093,
      "step": 390
    },
    {
      "epoch": 2.19,
      "learning_rate": 0.00011256830601092896,
      "loss": 0.1119,
      "step": 400
    },
    {
      "epoch": 2.19,
      "eval_accuracy": 0.7491666666666666,
      "eval_loss": 0.9247021675109863,
      "eval_runtime": 16.092,
      "eval_samples_per_second": 74.571,
      "eval_steps_per_second": 9.321,
      "step": 400
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00011038251366120218,
      "loss": 0.1183,
      "step": 410
    },
    {
      "epoch": 2.3,
      "learning_rate": 0.00010819672131147543,
      "loss": 0.1155,
      "step": 420
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.00010601092896174864,
      "loss": 0.138,
      "step": 430
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.00010382513661202187,
      "loss": 0.1124,
      "step": 440
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00010163934426229508,
      "loss": 0.1535,
      "step": 450
    },
    {
      "epoch": 2.51,
      "learning_rate": 9.945355191256831e-05,
      "loss": 0.1895,
      "step": 460
    },
    {
      "epoch": 2.57,
      "learning_rate": 9.726775956284153e-05,
      "loss": 0.128,
      "step": 470
    },
    {
      "epoch": 2.62,
      "learning_rate": 9.508196721311476e-05,
      "loss": 0.0633,
      "step": 480
    },
    {
      "epoch": 2.68,
      "learning_rate": 9.289617486338798e-05,
      "loss": 0.0788,
      "step": 490
    },
    {
      "epoch": 2.73,
      "learning_rate": 9.071038251366121e-05,
      "loss": 0.049,
      "step": 500
    },
    {
      "epoch": 2.73,
      "eval_accuracy": 0.7566666666666667,
      "eval_loss": 0.9662973880767822,
      "eval_runtime": 15.9004,
      "eval_samples_per_second": 75.47,
      "eval_steps_per_second": 9.434,
      "step": 500
    },
    {
      "epoch": 2.79,
      "learning_rate": 8.852459016393443e-05,
      "loss": 0.0829,
      "step": 510
    },
    {
      "epoch": 2.84,
      "learning_rate": 8.633879781420766e-05,
      "loss": 0.0821,
      "step": 520
    },
    {
      "epoch": 2.9,
      "learning_rate": 8.415300546448088e-05,
      "loss": 0.0554,
      "step": 530
    },
    {
      "epoch": 2.95,
      "learning_rate": 8.19672131147541e-05,
      "loss": 0.0756,
      "step": 540
    },
    {
      "epoch": 3.01,
      "learning_rate": 7.978142076502733e-05,
      "loss": 0.136,
      "step": 550
    },
    {
      "epoch": 3.06,
      "learning_rate": 7.759562841530054e-05,
      "loss": 0.0903,
      "step": 560
    },
    {
      "epoch": 3.11,
      "learning_rate": 7.540983606557377e-05,
      "loss": 0.0868,
      "step": 570
    },
    {
      "epoch": 3.17,
      "learning_rate": 7.3224043715847e-05,
      "loss": 0.0523,
      "step": 580
    },
    {
      "epoch": 3.22,
      "learning_rate": 7.103825136612023e-05,
      "loss": 0.0396,
      "step": 590
    },
    {
      "epoch": 3.28,
      "learning_rate": 6.885245901639344e-05,
      "loss": 0.0537,
      "step": 600
    },
    {
      "epoch": 3.28,
      "eval_accuracy": 0.7566666666666667,
      "eval_loss": 1.0557572841644287,
      "eval_runtime": 16.1696,
      "eval_samples_per_second": 74.213,
      "eval_steps_per_second": 9.277,
      "step": 600
    },
    {
      "epoch": 3.33,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.064,
      "step": 610
    },
    {
      "epoch": 3.39,
      "learning_rate": 6.44808743169399e-05,
      "loss": 0.0206,
      "step": 620
    },
    {
      "epoch": 3.44,
      "learning_rate": 6.229508196721313e-05,
      "loss": 0.028,
      "step": 630
    },
    {
      "epoch": 3.5,
      "learning_rate": 6.010928961748634e-05,
      "loss": 0.026,
      "step": 640
    },
    {
      "epoch": 3.55,
      "learning_rate": 5.792349726775956e-05,
      "loss": 0.0229,
      "step": 650
    },
    {
      "epoch": 3.61,
      "learning_rate": 5.5737704918032785e-05,
      "loss": 0.032,
      "step": 660
    },
    {
      "epoch": 3.66,
      "learning_rate": 5.355191256830602e-05,
      "loss": 0.023,
      "step": 670
    },
    {
      "epoch": 3.72,
      "learning_rate": 5.136612021857924e-05,
      "loss": 0.0269,
      "step": 680
    },
    {
      "epoch": 3.77,
      "learning_rate": 4.918032786885246e-05,
      "loss": 0.023,
      "step": 690
    },
    {
      "epoch": 3.83,
      "learning_rate": 4.6994535519125685e-05,
      "loss": 0.0274,
      "step": 700
    },
    {
      "epoch": 3.83,
      "eval_accuracy": 0.7691666666666667,
      "eval_loss": 1.0343540906906128,
      "eval_runtime": 16.2987,
      "eval_samples_per_second": 73.626,
      "eval_steps_per_second": 9.203,
      "step": 700
    },
    {
      "epoch": 3.88,
      "learning_rate": 4.4808743169398906e-05,
      "loss": 0.0197,
      "step": 710
    },
    {
      "epoch": 3.93,
      "learning_rate": 4.262295081967213e-05,
      "loss": 0.0195,
      "step": 720
    },
    {
      "epoch": 3.99,
      "learning_rate": 4.0437158469945356e-05,
      "loss": 0.0122,
      "step": 730
    },
    {
      "epoch": 4.04,
      "learning_rate": 3.825136612021858e-05,
      "loss": 0.0156,
      "step": 740
    },
    {
      "epoch": 4.1,
      "learning_rate": 3.6065573770491806e-05,
      "loss": 0.0115,
      "step": 750
    },
    {
      "epoch": 4.15,
      "learning_rate": 3.387978142076503e-05,
      "loss": 0.0118,
      "step": 760
    },
    {
      "epoch": 4.21,
      "learning_rate": 3.1693989071038256e-05,
      "loss": 0.0123,
      "step": 770
    },
    {
      "epoch": 4.26,
      "learning_rate": 2.9508196721311478e-05,
      "loss": 0.0108,
      "step": 780
    },
    {
      "epoch": 4.32,
      "learning_rate": 2.7322404371584703e-05,
      "loss": 0.0112,
      "step": 790
    },
    {
      "epoch": 4.37,
      "learning_rate": 2.5136612021857924e-05,
      "loss": 0.0102,
      "step": 800
    },
    {
      "epoch": 4.37,
      "eval_accuracy": 0.7941666666666667,
      "eval_loss": 0.9259100556373596,
      "eval_runtime": 16.5205,
      "eval_samples_per_second": 72.637,
      "eval_steps_per_second": 9.08,
      "step": 800
    },
    {
      "epoch": 4.43,
      "learning_rate": 2.295081967213115e-05,
      "loss": 0.0101,
      "step": 810
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.0765027322404374e-05,
      "loss": 0.0099,
      "step": 820
    },
    {
      "epoch": 4.54,
      "learning_rate": 1.85792349726776e-05,
      "loss": 0.0097,
      "step": 830
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.6393442622950818e-05,
      "loss": 0.011,
      "step": 840
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.4207650273224044e-05,
      "loss": 0.0094,
      "step": 850
    },
    {
      "epoch": 4.7,
      "learning_rate": 1.2021857923497268e-05,
      "loss": 0.01,
      "step": 860
    },
    {
      "epoch": 4.75,
      "learning_rate": 9.836065573770493e-06,
      "loss": 0.0096,
      "step": 870
    },
    {
      "epoch": 4.81,
      "learning_rate": 7.650273224043716e-06,
      "loss": 0.0098,
      "step": 880
    },
    {
      "epoch": 4.86,
      "learning_rate": 5.46448087431694e-06,
      "loss": 0.0163,
      "step": 890
    },
    {
      "epoch": 4.92,
      "learning_rate": 3.278688524590164e-06,
      "loss": 0.0095,
      "step": 900
    },
    {
      "epoch": 4.92,
      "eval_accuracy": 0.785,
      "eval_loss": 0.9604464769363403,
      "eval_runtime": 17.3811,
      "eval_samples_per_second": 69.04,
      "eval_steps_per_second": 8.63,
      "step": 900
    },
    {
      "epoch": 4.97,
      "learning_rate": 1.092896174863388e-06,
      "loss": 0.0092,
      "step": 910
    },
    {
      "epoch": 5.0,
      "step": 915,
      "total_flos": 3.4029172406502605e+18,
      "train_loss": 0.2756463099698551,
      "train_runtime": 974.1891,
      "train_samples_per_second": 45.073,
      "train_steps_per_second": 0.939
    }
  ],
  "logging_steps": 10,
  "max_steps": 915,
  "num_train_epochs": 5,
  "save_steps": 1000,
  "total_flos": 3.4029172406502605e+18,
  "trial_name": null,
  "trial_params": null
}