{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 15.0,
  "global_step": 1035,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 8e-06,
      "loss": 11.3437,
      "step": 10
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 12.2411,
      "step": 20
    },
    {
      "epoch": 0.43,
      "learning_rate": 2.7e-05,
      "loss": 11.8912,
      "step": 30
    },
    {
      "epoch": 0.58,
      "learning_rate": 3.7e-05,
      "loss": 10.4142,
      "step": 40
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.7542,
      "step": 50
    },
    {
      "epoch": 0.87,
      "learning_rate": 5.7e-05,
      "loss": 4.6198,
      "step": 60
    },
    {
      "epoch": 1.01,
      "learning_rate": 6.7e-05,
      "loss": 3.8278,
      "step": 70
    },
    {
      "epoch": 1.16,
      "learning_rate": 7.7e-05,
      "loss": 3.2769,
      "step": 80
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.7e-05,
      "loss": 3.2017,
      "step": 90
    },
    {
      "epoch": 1.45,
      "learning_rate": 9.7e-05,
      "loss": 3.0602,
      "step": 100
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.000107,
      "loss": 2.963,
      "step": 110
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011700000000000001,
      "loss": 3.0003,
      "step": 120
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.000127,
      "loss": 2.9574,
      "step": 130
    },
    {
      "epoch": 2.03,
      "learning_rate": 0.00013700000000000002,
      "loss": 2.9113,
      "step": 140
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.000147,
      "loss": 2.9198,
      "step": 150
    },
    {
      "epoch": 2.32,
      "learning_rate": 0.000157,
      "loss": 2.9141,
      "step": 160
    },
    {
      "epoch": 2.46,
      "learning_rate": 0.00016700000000000002,
      "loss": 2.9118,
      "step": 170
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.000177,
      "loss": 2.8995,
      "step": 180
    },
    {
      "epoch": 2.75,
      "learning_rate": 0.000187,
      "loss": 2.9353,
      "step": 190
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.00019700000000000002,
      "loss": 2.8971,
      "step": 200
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.000207,
      "loss": 2.8813,
      "step": 210
    },
    {
      "epoch": 3.19,
      "learning_rate": 0.00021700000000000002,
      "loss": 2.903,
      "step": 220
    },
    {
      "epoch": 3.33,
      "learning_rate": 0.00022700000000000002,
      "loss": 2.8678,
      "step": 230
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.000237,
      "loss": 2.9094,
      "step": 240
    },
    {
      "epoch": 3.62,
      "learning_rate": 0.000247,
      "loss": 2.887,
      "step": 250
    },
    {
      "epoch": 3.77,
      "learning_rate": 0.000257,
      "loss": 2.9137,
      "step": 260
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00026700000000000004,
      "loss": 2.8704,
      "step": 270
    },
    {
      "epoch": 4.06,
      "learning_rate": 0.000277,
      "loss": 2.67,
      "step": 280
    },
    {
      "epoch": 4.2,
      "learning_rate": 0.000287,
      "loss": 2.3059,
      "step": 290
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.000297,
      "loss": 1.834,
      "step": 300
    },
    {
      "epoch": 4.49,
      "learning_rate": 0.000307,
      "loss": 1.0162,
      "step": 310
    },
    {
      "epoch": 4.64,
      "learning_rate": 0.000317,
      "loss": 0.6458,
      "step": 320
    },
    {
      "epoch": 4.78,
      "learning_rate": 0.00032700000000000003,
      "loss": 0.5226,
      "step": 330
    },
    {
      "epoch": 4.93,
      "learning_rate": 0.000337,
      "loss": 0.5121,
      "step": 340
    },
    {
      "epoch": 5.07,
      "learning_rate": 0.000347,
      "loss": 0.5253,
      "step": 350
    },
    {
      "epoch": 5.22,
      "learning_rate": 0.000357,
      "loss": 0.3705,
      "step": 360
    },
    {
      "epoch": 5.36,
      "learning_rate": 0.000367,
      "loss": 0.296,
      "step": 370
    },
    {
      "epoch": 5.51,
      "learning_rate": 0.000377,
      "loss": 0.29,
      "step": 380
    },
    {
      "epoch": 5.65,
      "learning_rate": 0.00038700000000000003,
      "loss": 0.289,
      "step": 390
    },
    {
      "epoch": 5.8,
      "learning_rate": 0.00039700000000000005,
      "loss": 0.2989,
      "step": 400
    },
    {
      "epoch": 5.94,
      "learning_rate": 0.00040699999999999997,
      "loss": 0.2531,
      "step": 410
    },
    {
      "epoch": 6.09,
      "learning_rate": 0.000417,
      "loss": 0.2184,
      "step": 420
    },
    {
      "epoch": 6.23,
      "learning_rate": 0.000427,
      "loss": 0.1819,
      "step": 430
    },
    {
      "epoch": 6.38,
      "learning_rate": 0.000437,
      "loss": 0.1702,
      "step": 440
    },
    {
      "epoch": 6.52,
      "learning_rate": 0.000447,
      "loss": 0.189,
      "step": 450
    },
    {
      "epoch": 6.67,
      "learning_rate": 0.00045700000000000005,
      "loss": 0.2173,
      "step": 460
    },
    {
      "epoch": 6.81,
      "learning_rate": 0.000467,
      "loss": 0.1703,
      "step": 470
    },
    {
      "epoch": 6.96,
      "learning_rate": 0.000477,
      "loss": 0.2397,
      "step": 480
    },
    {
      "epoch": 7.1,
      "learning_rate": 0.000487,
      "loss": 0.1703,
      "step": 490
    },
    {
      "epoch": 7.25,
      "learning_rate": 0.000497,
      "loss": 0.1612,
      "step": 500
    },
    {
      "epoch": 7.25,
      "eval_loss": 0.3108726739883423,
      "eval_runtime": 608.0436,
      "eval_samples_per_second": 5.582,
      "eval_steps_per_second": 0.699,
      "eval_wer": 0.2548134999876735,
      "step": 500
    },
    {
      "epoch": 7.39,
      "learning_rate": 0.0004934579439252336,
      "loss": 0.1538,
      "step": 510
    },
    {
      "epoch": 7.54,
      "learning_rate": 0.00048411214953271025,
      "loss": 0.1432,
      "step": 520
    },
    {
      "epoch": 7.68,
      "learning_rate": 0.0004747663551401869,
      "loss": 0.1354,
      "step": 530
    },
    {
      "epoch": 7.83,
      "learning_rate": 0.00046542056074766355,
      "loss": 0.1766,
      "step": 540
    },
    {
      "epoch": 7.97,
      "learning_rate": 0.00045607476635514015,
      "loss": 0.1654,
      "step": 550
    },
    {
      "epoch": 8.12,
      "learning_rate": 0.00044672897196261686,
      "loss": 0.1267,
      "step": 560
    },
    {
      "epoch": 8.26,
      "learning_rate": 0.0004373831775700935,
      "loss": 0.1313,
      "step": 570
    },
    {
      "epoch": 8.41,
      "learning_rate": 0.0004280373831775701,
      "loss": 0.1213,
      "step": 580
    },
    {
      "epoch": 8.55,
      "learning_rate": 0.00041869158878504677,
      "loss": 0.1221,
      "step": 590
    },
    {
      "epoch": 8.7,
      "learning_rate": 0.0004093457943925234,
      "loss": 0.119,
      "step": 600
    },
    {
      "epoch": 8.84,
      "learning_rate": 0.0004,
      "loss": 0.1085,
      "step": 610
    },
    {
      "epoch": 8.99,
      "learning_rate": 0.00039065420560747667,
      "loss": 0.1185,
      "step": 620
    },
    {
      "epoch": 9.13,
      "learning_rate": 0.00038130841121495327,
      "loss": 0.1115,
      "step": 630
    },
    {
      "epoch": 9.28,
      "learning_rate": 0.0003719626168224299,
      "loss": 0.0799,
      "step": 640
    },
    {
      "epoch": 9.42,
      "learning_rate": 0.0003626168224299066,
      "loss": 0.1101,
      "step": 650
    },
    {
      "epoch": 9.57,
      "learning_rate": 0.0003532710280373832,
      "loss": 0.0795,
      "step": 660
    },
    {
      "epoch": 9.71,
      "learning_rate": 0.00034392523364485983,
      "loss": 0.1179,
      "step": 670
    },
    {
      "epoch": 9.86,
      "learning_rate": 0.0003345794392523365,
      "loss": 0.1134,
      "step": 680
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0003252336448598131,
      "loss": 0.0883,
      "step": 690
    },
    {
      "epoch": 10.14,
      "learning_rate": 0.00031588785046728974,
      "loss": 0.0795,
      "step": 700
    },
    {
      "epoch": 10.29,
      "learning_rate": 0.0003065420560747664,
      "loss": 0.1018,
      "step": 710
    },
    {
      "epoch": 10.43,
      "learning_rate": 0.000297196261682243,
      "loss": 0.1054,
      "step": 720
    },
    {
      "epoch": 10.58,
      "learning_rate": 0.00028785046728971965,
      "loss": 0.0795,
      "step": 730
    },
    {
      "epoch": 10.72,
      "learning_rate": 0.0002785046728971963,
      "loss": 0.0796,
      "step": 740
    },
    {
      "epoch": 10.87,
      "learning_rate": 0.0002691588785046729,
      "loss": 0.0668,
      "step": 750
    },
    {
      "epoch": 11.01,
      "learning_rate": 0.00025981308411214955,
      "loss": 0.0732,
      "step": 760
    },
    {
      "epoch": 11.16,
      "learning_rate": 0.0002504672897196262,
      "loss": 0.061,
      "step": 770
    },
    {
      "epoch": 11.3,
      "learning_rate": 0.0002411214953271028,
      "loss": 0.0578,
      "step": 780
    },
    {
      "epoch": 11.45,
      "learning_rate": 0.00023177570093457946,
      "loss": 0.0555,
      "step": 790
    },
    {
      "epoch": 11.59,
      "learning_rate": 0.00022242990654205608,
      "loss": 0.0588,
      "step": 800
    },
    {
      "epoch": 11.74,
      "learning_rate": 0.0002130841121495327,
      "loss": 0.0699,
      "step": 810
    },
    {
      "epoch": 11.88,
      "learning_rate": 0.00020373831775700934,
      "loss": 0.0676,
      "step": 820
    },
    {
      "epoch": 12.03,
      "learning_rate": 0.000194392523364486,
      "loss": 0.0555,
      "step": 830
    },
    {
      "epoch": 12.17,
      "learning_rate": 0.00018504672897196262,
      "loss": 0.0612,
      "step": 840
    },
    {
      "epoch": 12.32,
      "learning_rate": 0.00017570093457943924,
      "loss": 0.0393,
      "step": 850
    },
    {
      "epoch": 12.46,
      "learning_rate": 0.0001663551401869159,
      "loss": 0.0508,
      "step": 860
    },
    {
      "epoch": 12.61,
      "learning_rate": 0.00015700934579439252,
      "loss": 0.0494,
      "step": 870
    },
    {
      "epoch": 12.75,
      "learning_rate": 0.00014766355140186915,
      "loss": 0.0445,
      "step": 880
    },
    {
      "epoch": 12.9,
      "learning_rate": 0.0001383177570093458,
      "loss": 0.0538,
      "step": 890
    },
    {
      "epoch": 13.04,
      "learning_rate": 0.00012897196261682243,
      "loss": 0.0509,
      "step": 900
    },
    {
      "epoch": 13.19,
      "learning_rate": 0.00011962616822429907,
      "loss": 0.0552,
      "step": 910
    },
    {
      "epoch": 13.33,
      "learning_rate": 0.00011028037383177571,
      "loss": 0.0363,
      "step": 920
    },
    {
      "epoch": 13.48,
      "learning_rate": 0.00010093457943925234,
      "loss": 0.0332,
      "step": 930
    },
    {
      "epoch": 13.62,
      "learning_rate": 9.158878504672898e-05,
      "loss": 0.0394,
      "step": 940
    },
    {
      "epoch": 13.77,
      "learning_rate": 8.224299065420562e-05,
      "loss": 0.0385,
      "step": 950
    },
    {
      "epoch": 13.91,
      "learning_rate": 7.289719626168224e-05,
      "loss": 0.042,
      "step": 960
    },
    {
      "epoch": 14.06,
      "learning_rate": 6.355140186915888e-05,
      "loss": 0.0427,
      "step": 970
    },
    {
      "epoch": 14.2,
      "learning_rate": 5.420560747663551e-05,
      "loss": 0.0352,
      "step": 980
    },
    {
      "epoch": 14.35,
      "learning_rate": 4.485981308411215e-05,
      "loss": 0.0448,
      "step": 990
    },
    {
      "epoch": 14.49,
      "learning_rate": 3.551401869158879e-05,
      "loss": 0.0433,
      "step": 1000
    },
    {
      "epoch": 14.49,
      "eval_loss": 0.2945314943790436,
      "eval_runtime": 593.7298,
      "eval_samples_per_second": 5.716,
      "eval_steps_per_second": 0.716,
      "eval_wer": 0.18705881386156514,
      "step": 1000
    },
    {
      "epoch": 14.64,
      "learning_rate": 2.6168224299065423e-05,
      "loss": 0.0409,
      "step": 1010
    },
    {
      "epoch": 14.78,
      "learning_rate": 1.6822429906542056e-05,
      "loss": 0.0354,
      "step": 1020
    },
    {
      "epoch": 14.93,
      "learning_rate": 7.476635514018692e-06,
      "loss": 0.0321,
      "step": 1030
    },
    {
      "epoch": 15.0,
      "step": 1035,
      "total_flos": 1.5036330849795359e+19,
      "train_loss": 1.3424202077342693,
      "train_runtime": 8416.5784,
      "train_samples_per_second": 3.91,
      "train_steps_per_second": 0.123
    }
  ],
  "max_steps": 1035,
  "num_train_epochs": 15,
  "total_flos": 1.5036330849795359e+19,
  "trial_name": null,
  "trial_params": null
}