{
  "best_metric": 4.192136287689209,
  "best_model_checkpoint": "./checkpoint-1000",
  "epoch": 12.658227848101266,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 7.500000000000001e-09,
      "loss": 22.5894,
      "step": 10
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.5000000000000002e-08,
      "loss": 20.4369,
      "step": 20
    },
    {
      "epoch": 0.38,
      "learning_rate": 2.25e-08,
      "loss": 22.056,
      "step": 30
    },
    {
      "epoch": 0.51,
      "learning_rate": 3.0000000000000004e-08,
      "loss": 19.983,
      "step": 40
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.7500000000000005e-08,
      "loss": 21.5341,
      "step": 50
    },
    {
      "epoch": 0.76,
      "learning_rate": 4.4249999999999996e-08,
      "loss": 20.8669,
      "step": 60
    },
    {
      "epoch": 0.89,
      "learning_rate": 5.1750000000000003e-08,
      "loss": 21.3542,
      "step": 70
    },
    {
      "epoch": 1.01,
      "learning_rate": 5.9250000000000004e-08,
      "loss": 20.587,
      "step": 80
    },
    {
      "epoch": 1.14,
      "learning_rate": 6.675e-08,
      "loss": 21.7896,
      "step": 90
    },
    {
      "epoch": 1.27,
      "learning_rate": 7.425e-08,
      "loss": 19.9558,
      "step": 100
    },
    {
      "epoch": 1.27,
      "eval_cer": 3.2660447130816737,
      "eval_loss": 20.919719696044922,
      "eval_runtime": 51.6953,
      "eval_samples_per_second": 9.305,
      "eval_steps_per_second": 2.341,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 1.39,
      "learning_rate": 8.175e-08,
      "loss": 21.8368,
      "step": 110
    },
    {
      "epoch": 1.52,
      "learning_rate": 8.925e-08,
      "loss": 20.2781,
      "step": 120
    },
    {
      "epoch": 1.65,
      "learning_rate": 9.675e-08,
      "loss": 20.5284,
      "step": 130
    },
    {
      "epoch": 1.77,
      "learning_rate": 1.0425000000000001e-07,
      "loss": 20.2507,
      "step": 140
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.1174999999999999e-07,
      "loss": 20.063,
      "step": 150
    },
    {
      "epoch": 2.03,
      "learning_rate": 1.1925e-07,
      "loss": 20.3082,
      "step": 160
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.2675000000000002e-07,
      "loss": 20.5946,
      "step": 170
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.3425e-07,
      "loss": 19.3427,
      "step": 180
    },
    {
      "epoch": 2.41,
      "learning_rate": 1.4175e-07,
      "loss": 19.4337,
      "step": 190
    },
    {
      "epoch": 2.53,
      "learning_rate": 1.4925000000000002e-07,
      "loss": 19.7186,
      "step": 200
    },
    {
      "epoch": 2.53,
      "eval_cer": 1.1691851406904061,
      "eval_loss": 19.244659423828125,
      "eval_runtime": 51.3422,
      "eval_samples_per_second": 9.369,
      "eval_steps_per_second": 2.357,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 2.66,
      "learning_rate": 1.56e-07,
      "loss": 19.3518,
      "step": 210
    },
    {
      "epoch": 2.78,
      "learning_rate": 1.635e-07,
      "loss": 18.8406,
      "step": 220
    },
    {
      "epoch": 2.91,
      "learning_rate": 1.71e-07,
      "loss": 18.1267,
      "step": 230
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.785e-07,
      "loss": 18.2315,
      "step": 240
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.86e-07,
      "loss": 17.8997,
      "step": 250
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.935e-07,
      "loss": 17.5011,
      "step": 260
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.01e-07,
      "loss": 16.8782,
      "step": 270
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.0850000000000002e-07,
      "loss": 16.628,
      "step": 280
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.16e-07,
      "loss": 15.2879,
      "step": 290
    },
    {
      "epoch": 3.8,
      "learning_rate": 2.2349999999999998e-07,
      "loss": 15.203,
      "step": 300
    },
    {
      "epoch": 3.8,
      "eval_cer": 0.9687083513224335,
      "eval_loss": 15.005261421203613,
      "eval_runtime": 54.4312,
      "eval_samples_per_second": 8.837,
      "eval_steps_per_second": 1.121,
      "eval_wer": 0.9998294679399727,
      "step": 300
    },
    {
      "epoch": 3.92,
      "learning_rate": 2.31e-07,
      "loss": 13.8484,
      "step": 310
    },
    {
      "epoch": 4.05,
      "learning_rate": 2.385e-07,
      "loss": 13.6877,
      "step": 320
    },
    {
      "epoch": 4.18,
      "learning_rate": 2.46e-07,
      "loss": 12.1026,
      "step": 330
    },
    {
      "epoch": 4.3,
      "learning_rate": 2.5350000000000004e-07,
      "loss": 11.1649,
      "step": 340
    },
    {
      "epoch": 4.43,
      "learning_rate": 2.6099999999999997e-07,
      "loss": 9.5398,
      "step": 350
    },
    {
      "epoch": 4.56,
      "learning_rate": 2.6775e-07,
      "loss": 8.7484,
      "step": 360
    },
    {
      "epoch": 4.68,
      "learning_rate": 2.7525e-07,
      "loss": 7.7766,
      "step": 370
    },
    {
      "epoch": 4.81,
      "learning_rate": 2.8275e-07,
      "loss": 7.2303,
      "step": 380
    },
    {
      "epoch": 4.94,
      "learning_rate": 2.9025e-07,
      "loss": 6.5927,
      "step": 390
    },
    {
      "epoch": 5.06,
      "learning_rate": 2.9775000000000003e-07,
      "loss": 6.4303,
      "step": 400
    },
    {
      "epoch": 5.06,
      "eval_cer": 0.9911212058641338,
      "eval_loss": 6.543659687042236,
      "eval_runtime": 55.4335,
      "eval_samples_per_second": 8.677,
      "eval_steps_per_second": 1.1,
      "eval_wer": 0.9631650750341064,
      "step": 400
    },
    {
      "epoch": 5.19,
      "learning_rate": 3.0525e-07,
      "loss": 5.9555,
      "step": 410
    },
    {
      "epoch": 5.32,
      "learning_rate": 3.1275e-07,
      "loss": 5.9212,
      "step": 420
    },
    {
      "epoch": 5.44,
      "learning_rate": 3.2025e-07,
      "loss": 5.4995,
      "step": 430
    },
    {
      "epoch": 5.57,
      "learning_rate": 3.2775e-07,
      "loss": 5.5903,
      "step": 440
    },
    {
      "epoch": 5.7,
      "learning_rate": 3.3525000000000004e-07,
      "loss": 5.1376,
      "step": 450
    },
    {
      "epoch": 5.82,
      "learning_rate": 3.4275e-07,
      "loss": 4.9402,
      "step": 460
    },
    {
      "epoch": 5.95,
      "learning_rate": 3.5025e-07,
      "loss": 4.9112,
      "step": 470
    },
    {
      "epoch": 6.08,
      "learning_rate": 3.5775e-07,
      "loss": 4.8001,
      "step": 480
    },
    {
      "epoch": 6.2,
      "learning_rate": 3.6525e-07,
      "loss": 4.482,
      "step": 490
    },
    {
      "epoch": 6.33,
      "learning_rate": 3.7275e-07,
      "loss": 4.5712,
      "step": 500
    },
    {
      "epoch": 6.33,
      "eval_cer": 0.9545736113978939,
      "eval_loss": 4.904029369354248,
      "eval_runtime": 64.6266,
      "eval_samples_per_second": 7.443,
      "eval_steps_per_second": 0.944,
      "eval_wer": 0.9322987721691678,
      "step": 500
    },
    {
      "epoch": 6.46,
      "learning_rate": 3.8025000000000003e-07,
      "loss": 4.3992,
      "step": 510
    },
    {
      "epoch": 6.58,
      "learning_rate": 3.8775e-07,
      "loss": 4.1942,
      "step": 520
    },
    {
      "epoch": 6.71,
      "learning_rate": 3.9525000000000005e-07,
      "loss": 4.339,
      "step": 530
    },
    {
      "epoch": 6.84,
      "learning_rate": 4.0275000000000003e-07,
      "loss": 4.2826,
      "step": 540
    },
    {
      "epoch": 6.96,
      "learning_rate": 4.1025000000000006e-07,
      "loss": 4.3752,
      "step": 550
    },
    {
      "epoch": 7.09,
      "learning_rate": 4.1775000000000004e-07,
      "loss": 3.9576,
      "step": 560
    },
    {
      "epoch": 7.22,
      "learning_rate": 4.2524999999999997e-07,
      "loss": 4.0814,
      "step": 570
    },
    {
      "epoch": 7.34,
      "learning_rate": 4.3274999999999995e-07,
      "loss": 4.3274,
      "step": 580
    },
    {
      "epoch": 7.47,
      "learning_rate": 4.4025e-07,
      "loss": 3.9749,
      "step": 590
    },
    {
      "epoch": 7.59,
      "learning_rate": 4.4774999999999997e-07,
      "loss": 3.8373,
      "step": 600
    },
    {
      "epoch": 7.72,
      "learning_rate": 4.5525e-07,
      "loss": 4.1076,
      "step": 610
    },
    {
      "epoch": 7.85,
      "learning_rate": 4.6275e-07,
      "loss": 3.8279,
      "step": 620
    },
    {
      "epoch": 7.97,
      "learning_rate": 4.7025e-07,
      "loss": 4.0263,
      "step": 630
    },
    {
      "epoch": 8.1,
      "learning_rate": 4.7775e-07,
      "loss": 3.8141,
      "step": 640
    },
    {
      "epoch": 8.23,
      "learning_rate": 4.8525e-07,
      "loss": 3.8791,
      "step": 650
    },
    {
      "epoch": 8.35,
      "learning_rate": 4.927500000000001e-07,
      "loss": 3.6906,
      "step": 660
    },
    {
      "epoch": 8.48,
      "learning_rate": 5.0025e-07,
      "loss": 4.0741,
      "step": 670
    },
    {
      "epoch": 8.61,
      "learning_rate": 5.0775e-07,
      "loss": 3.7322,
      "step": 680
    },
    {
      "epoch": 8.73,
      "learning_rate": 5.152500000000001e-07,
      "loss": 4.1834,
      "step": 690
    },
    {
      "epoch": 8.86,
      "learning_rate": 5.2275e-07,
      "loss": 3.8571,
      "step": 700
    },
    {
      "epoch": 8.99,
      "learning_rate": 5.3025e-07,
      "loss": 3.9223,
      "step": 710
    },
    {
      "epoch": 9.11,
      "learning_rate": 5.3775e-07,
      "loss": 3.6997,
      "step": 720
    },
    {
      "epoch": 9.24,
      "learning_rate": 5.4525e-07,
      "loss": 4.0024,
      "step": 730
    },
    {
      "epoch": 9.37,
      "learning_rate": 5.5275e-07,
      "loss": 3.7653,
      "step": 740
    },
    {
      "epoch": 9.49,
      "learning_rate": 5.602500000000001e-07,
      "loss": 3.873,
      "step": 750
    },
    {
      "epoch": 9.62,
      "learning_rate": 5.6775e-07,
      "loss": 3.597,
      "step": 760
    },
    {
      "epoch": 9.75,
      "learning_rate": 5.7525e-07,
      "loss": 3.7902,
      "step": 770
    },
    {
      "epoch": 9.87,
      "learning_rate": 5.827500000000001e-07,
      "loss": 3.8314,
      "step": 780
    },
    {
      "epoch": 10.0,
      "learning_rate": 5.902500000000001e-07,
      "loss": 3.956,
      "step": 790
    },
    {
      "epoch": 10.13,
      "learning_rate": 5.9775e-07,
      "loss": 3.6297,
      "step": 800
    },
    {
      "epoch": 10.25,
      "learning_rate": 6.0525e-07,
      "loss": 4.0449,
      "step": 810
    },
    {
      "epoch": 10.38,
      "learning_rate": 6.1275e-07,
      "loss": 3.5669,
      "step": 820
    },
    {
      "epoch": 10.51,
      "learning_rate": 6.2025e-07,
      "loss": 3.6994,
      "step": 830
    },
    {
      "epoch": 10.63,
      "learning_rate": 6.277499999999999e-07,
      "loss": 3.6146,
      "step": 840
    },
    {
      "epoch": 10.76,
      "learning_rate": 6.3525e-07,
      "loss": 3.8637,
      "step": 850
    },
    {
      "epoch": 10.89,
      "learning_rate": 6.4275e-07,
      "loss": 3.7,
      "step": 860
    },
    {
      "epoch": 11.01,
      "learning_rate": 6.5025e-07,
      "loss": 3.8526,
      "step": 870
    },
    {
      "epoch": 11.14,
      "learning_rate": 6.5775e-07,
      "loss": 3.7276,
      "step": 880
    },
    {
      "epoch": 11.27,
      "learning_rate": 6.6525e-07,
      "loss": 3.989,
      "step": 890
    },
    {
      "epoch": 11.39,
      "learning_rate": 6.7275e-07,
      "loss": 3.7838,
      "step": 900
    },
    {
      "epoch": 11.52,
      "learning_rate": 6.802500000000001e-07,
      "loss": 3.698,
      "step": 910
    },
    {
      "epoch": 11.65,
      "learning_rate": 6.8775e-07,
      "loss": 3.4391,
      "step": 920
    },
    {
      "epoch": 11.77,
      "learning_rate": 6.9525e-07,
      "loss": 3.7371,
      "step": 930
    },
    {
      "epoch": 11.9,
      "learning_rate": 7.027500000000001e-07,
      "loss": 3.4201,
      "step": 940
    },
    {
      "epoch": 12.03,
      "learning_rate": 7.1025e-07,
      "loss": 3.7721,
      "step": 950
    },
    {
      "epoch": 12.15,
      "learning_rate": 7.1775e-07,
      "loss": 3.7842,
      "step": 960
    },
    {
      "epoch": 12.28,
      "learning_rate": 7.2525e-07,
      "loss": 3.8125,
      "step": 970
    },
    {
      "epoch": 12.41,
      "learning_rate": 7.3275e-07,
      "loss": 3.7673,
      "step": 980
    },
    {
      "epoch": 12.53,
      "learning_rate": 7.395e-07,
      "loss": 3.7048,
      "step": 990
    },
    {
      "epoch": 12.66,
      "learning_rate": 7.47e-07,
      "loss": 3.3986,
      "step": 1000
    },
    {
      "epoch": 12.66,
      "eval_cer": 0.9608244326394233,
      "eval_loss": 4.192136287689209,
      "eval_runtime": 52.6187,
      "eval_samples_per_second": 9.141,
      "eval_steps_per_second": 1.159,
      "eval_wer": 0.9294849931787176,
      "step": 1000
    },
    {
      "epoch": 12.66,
      "step": 1000,
      "total_flos": 3.5131981783950033e+19,
      "train_loss": 1.9340453338623047,
      "train_runtime": 2875.5301,
      "train_samples_per_second": 11.128,
      "train_steps_per_second": 0.348
    }
  ],
  "max_steps": 1000,
  "num_train_epochs": 13,
  "total_flos": 3.5131981783950033e+19,
  "trial_name": null,
  "trial_params": null
}