{
  "best_metric": 44.35982865302237,
  "best_model_checkpoint": "whisper-sk-small-augmented/checkpoint-5000",
  "epoch": 71.42857142857143,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.71,
      "learning_rate": 4.6e-06,
      "loss": 3.0357,
      "step": 50
    },
    {
      "epoch": 1.43,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.2337,
      "step": 100
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.4599999999999999e-05,
      "loss": 0.7656,
      "step": 150
    },
    {
      "epoch": 2.86,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.4719,
      "step": 200
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.46e-05,
      "loss": 0.3019,
      "step": 250
    },
    {
      "epoch": 4.29,
      "learning_rate": 2.96e-05,
      "loss": 0.2215,
      "step": 300
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.46e-05,
      "loss": 0.1645,
      "step": 350
    },
    {
      "epoch": 5.71,
      "learning_rate": 3.960000000000001e-05,
      "loss": 0.0987,
      "step": 400
    },
    {
      "epoch": 6.43,
      "learning_rate": 4.46e-05,
      "loss": 0.0914,
      "step": 450
    },
    {
      "epoch": 7.14,
      "learning_rate": 4.96e-05,
      "loss": 0.0885,
      "step": 500
    },
    {
      "epoch": 7.86,
      "learning_rate": 4.948888888888889e-05,
      "loss": 0.0722,
      "step": 550
    },
    {
      "epoch": 8.57,
      "learning_rate": 4.8933333333333335e-05,
      "loss": 0.059,
      "step": 600
    },
    {
      "epoch": 9.29,
      "learning_rate": 4.837777777777778e-05,
      "loss": 0.0532,
      "step": 650
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.782222222222222e-05,
      "loss": 0.0464,
      "step": 700
    },
    {
      "epoch": 10.71,
      "learning_rate": 4.726666666666667e-05,
      "loss": 0.0373,
      "step": 750
    },
    {
      "epoch": 11.43,
      "learning_rate": 4.671111111111111e-05,
      "loss": 0.0333,
      "step": 800
    },
    {
      "epoch": 12.14,
      "learning_rate": 4.615555555555556e-05,
      "loss": 0.0278,
      "step": 850
    },
    {
      "epoch": 12.86,
      "learning_rate": 4.5600000000000004e-05,
      "loss": 0.0228,
      "step": 900
    },
    {
      "epoch": 13.57,
      "learning_rate": 4.504444444444445e-05,
      "loss": 0.0268,
      "step": 950
    },
    {
      "epoch": 14.29,
      "learning_rate": 4.448888888888889e-05,
      "loss": 0.0232,
      "step": 1000
    },
    {
      "epoch": 14.29,
      "eval_loss": 0.7425360679626465,
      "eval_runtime": 568.891,
      "eval_samples_per_second": 3.985,
      "eval_steps_per_second": 0.125,
      "eval_wer": 51.880057115659206,
      "step": 1000
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.3933333333333335e-05,
      "loss": 0.0208,
      "step": 1050
    },
    {
      "epoch": 15.71,
      "learning_rate": 4.337777777777778e-05,
      "loss": 0.0201,
      "step": 1100
    },
    {
      "epoch": 16.43,
      "learning_rate": 4.282222222222222e-05,
      "loss": 0.0173,
      "step": 1150
    },
    {
      "epoch": 17.14,
      "learning_rate": 4.226666666666667e-05,
      "loss": 0.0192,
      "step": 1200
    },
    {
      "epoch": 17.86,
      "learning_rate": 4.171111111111111e-05,
      "loss": 0.0181,
      "step": 1250
    },
    {
      "epoch": 18.57,
      "learning_rate": 4.115555555555556e-05,
      "loss": 0.0138,
      "step": 1300
    },
    {
      "epoch": 19.29,
      "learning_rate": 4.0600000000000004e-05,
      "loss": 0.014,
      "step": 1350
    },
    {
      "epoch": 20.0,
      "learning_rate": 4.004444444444445e-05,
      "loss": 0.0125,
      "step": 1400
    },
    {
      "epoch": 20.71,
      "learning_rate": 3.948888888888889e-05,
      "loss": 0.0097,
      "step": 1450
    },
    {
      "epoch": 21.43,
      "learning_rate": 3.8933333333333336e-05,
      "loss": 0.0122,
      "step": 1500
    },
    {
      "epoch": 22.14,
      "learning_rate": 3.837777777777778e-05,
      "loss": 0.0109,
      "step": 1550
    },
    {
      "epoch": 22.86,
      "learning_rate": 3.782222222222222e-05,
      "loss": 0.0094,
      "step": 1600
    },
    {
      "epoch": 23.57,
      "learning_rate": 3.726666666666667e-05,
      "loss": 0.009,
      "step": 1650
    },
    {
      "epoch": 24.29,
      "learning_rate": 3.671111111111111e-05,
      "loss": 0.007,
      "step": 1700
    },
    {
      "epoch": 25.0,
      "learning_rate": 3.615555555555556e-05,
      "loss": 0.008,
      "step": 1750
    },
    {
      "epoch": 25.71,
      "learning_rate": 3.56e-05,
      "loss": 0.008,
      "step": 1800
    },
    {
      "epoch": 26.43,
      "learning_rate": 3.504444444444445e-05,
      "loss": 0.0082,
      "step": 1850
    },
    {
      "epoch": 27.14,
      "learning_rate": 3.448888888888889e-05,
      "loss": 0.0077,
      "step": 1900
    },
    {
      "epoch": 27.86,
      "learning_rate": 3.3933333333333336e-05,
      "loss": 0.0064,
      "step": 1950
    },
    {
      "epoch": 28.57,
      "learning_rate": 3.337777777777778e-05,
      "loss": 0.0083,
      "step": 2000
    },
    {
      "epoch": 28.57,
      "eval_loss": 0.7697533369064331,
      "eval_runtime": 560.6254,
      "eval_samples_per_second": 4.044,
      "eval_steps_per_second": 0.127,
      "eval_wer": 48.48881485007139,
      "step": 2000
    },
    {
      "epoch": 29.29,
      "learning_rate": 3.2822222222222223e-05,
      "loss": 0.0057,
      "step": 2050
    },
    {
      "epoch": 30.0,
      "learning_rate": 3.226666666666667e-05,
      "loss": 0.0052,
      "step": 2100
    },
    {
      "epoch": 30.71,
      "learning_rate": 3.171111111111111e-05,
      "loss": 0.0055,
      "step": 2150
    },
    {
      "epoch": 31.43,
      "learning_rate": 3.1155555555555555e-05,
      "loss": 0.0028,
      "step": 2200
    },
    {
      "epoch": 32.14,
      "learning_rate": 3.06e-05,
      "loss": 0.003,
      "step": 2250
    },
    {
      "epoch": 32.86,
      "learning_rate": 3.004444444444445e-05,
      "loss": 0.0027,
      "step": 2300
    },
    {
      "epoch": 33.57,
      "learning_rate": 2.948888888888889e-05,
      "loss": 0.0023,
      "step": 2350
    },
    {
      "epoch": 34.29,
      "learning_rate": 2.8933333333333333e-05,
      "loss": 0.0024,
      "step": 2400
    },
    {
      "epoch": 35.0,
      "learning_rate": 2.837777777777778e-05,
      "loss": 0.0018,
      "step": 2450
    },
    {
      "epoch": 35.71,
      "learning_rate": 2.782222222222222e-05,
      "loss": 0.001,
      "step": 2500
    },
    {
      "epoch": 36.43,
      "learning_rate": 2.7266666666666668e-05,
      "loss": 0.0014,
      "step": 2550
    },
    {
      "epoch": 37.14,
      "learning_rate": 2.6711111111111115e-05,
      "loss": 0.0015,
      "step": 2600
    },
    {
      "epoch": 37.86,
      "learning_rate": 2.6155555555555555e-05,
      "loss": 0.001,
      "step": 2650
    },
    {
      "epoch": 38.57,
      "learning_rate": 2.5600000000000002e-05,
      "loss": 0.0009,
      "step": 2700
    },
    {
      "epoch": 39.29,
      "learning_rate": 2.504444444444445e-05,
      "loss": 0.0015,
      "step": 2750
    },
    {
      "epoch": 40.0,
      "learning_rate": 2.448888888888889e-05,
      "loss": 0.0013,
      "step": 2800
    },
    {
      "epoch": 40.71,
      "learning_rate": 2.3933333333333337e-05,
      "loss": 0.0007,
      "step": 2850
    },
    {
      "epoch": 41.43,
      "learning_rate": 2.337777777777778e-05,
      "loss": 0.0003,
      "step": 2900
    },
    {
      "epoch": 42.14,
      "learning_rate": 2.282222222222222e-05,
      "loss": 0.0004,
      "step": 2950
    },
    {
      "epoch": 42.86,
      "learning_rate": 2.2266666666666668e-05,
      "loss": 0.0006,
      "step": 3000
    },
    {
      "epoch": 42.86,
      "eval_loss": 0.7640286684036255,
      "eval_runtime": 528.4764,
      "eval_samples_per_second": 4.29,
      "eval_steps_per_second": 0.134,
      "eval_wer": 47.596382674916704,
      "step": 3000
    },
    {
      "epoch": 43.57,
      "learning_rate": 2.1711111111111112e-05,
      "loss": 0.0009,
      "step": 3050
    },
    {
      "epoch": 44.29,
      "learning_rate": 2.1155555555555556e-05,
      "loss": 0.0008,
      "step": 3100
    },
    {
      "epoch": 45.0,
      "learning_rate": 2.06e-05,
      "loss": 0.0012,
      "step": 3150
    },
    {
      "epoch": 45.71,
      "learning_rate": 2.0044444444444446e-05,
      "loss": 0.0007,
      "step": 3200
    },
    {
      "epoch": 46.43,
      "learning_rate": 1.948888888888889e-05,
      "loss": 0.0008,
      "step": 3250
    },
    {
      "epoch": 47.14,
      "learning_rate": 1.8933333333333334e-05,
      "loss": 0.0004,
      "step": 3300
    },
    {
      "epoch": 47.86,
      "learning_rate": 1.837777777777778e-05,
      "loss": 0.0008,
      "step": 3350
    },
    {
      "epoch": 48.57,
      "learning_rate": 1.7822222222222225e-05,
      "loss": 0.0003,
      "step": 3400
    },
    {
      "epoch": 49.29,
      "learning_rate": 1.726666666666667e-05,
      "loss": 0.0007,
      "step": 3450
    },
    {
      "epoch": 50.0,
      "learning_rate": 1.6711111111111112e-05,
      "loss": 0.001,
      "step": 3500
    },
    {
      "epoch": 50.71,
      "learning_rate": 1.6155555555555556e-05,
      "loss": 0.0004,
      "step": 3550
    },
    {
      "epoch": 51.43,
      "learning_rate": 1.56e-05,
      "loss": 0.0004,
      "step": 3600
    },
    {
      "epoch": 52.14,
      "learning_rate": 1.5044444444444445e-05,
      "loss": 0.0004,
      "step": 3650
    },
    {
      "epoch": 52.86,
      "learning_rate": 1.448888888888889e-05,
      "loss": 0.0002,
      "step": 3700
    },
    {
      "epoch": 53.57,
      "learning_rate": 1.3933333333333334e-05,
      "loss": 0.0007,
      "step": 3750
    },
    {
      "epoch": 54.29,
      "learning_rate": 1.3377777777777778e-05,
      "loss": 0.0006,
      "step": 3800
    },
    {
      "epoch": 55.0,
      "learning_rate": 1.2822222222222222e-05,
      "loss": 0.0005,
      "step": 3850
    },
    {
      "epoch": 55.71,
      "learning_rate": 1.2266666666666667e-05,
      "loss": 0.0002,
      "step": 3900
    },
    {
      "epoch": 56.43,
      "learning_rate": 1.1711111111111111e-05,
      "loss": 0.0002,
      "step": 3950
    },
    {
      "epoch": 57.14,
      "learning_rate": 1.1155555555555556e-05,
      "loss": 0.0005,
      "step": 4000
    },
    {
      "epoch": 57.14,
      "eval_loss": 0.7649062275886536,
      "eval_runtime": 488.6276,
      "eval_samples_per_second": 4.64,
      "eval_steps_per_second": 0.145,
      "eval_wer": 44.89528795811518,
      "step": 4000
    },
    {
      "epoch": 57.86,
      "learning_rate": 1.06e-05,
      "loss": 0.0002,
      "step": 4050
    },
    {
      "epoch": 58.57,
      "learning_rate": 1.0044444444444446e-05,
      "loss": 0.0003,
      "step": 4100
    },
    {
      "epoch": 59.29,
      "learning_rate": 9.48888888888889e-06,
      "loss": 0.0005,
      "step": 4150
    },
    {
      "epoch": 60.0,
      "learning_rate": 8.933333333333333e-06,
      "loss": 0.0005,
      "step": 4200
    },
    {
      "epoch": 60.71,
      "learning_rate": 8.377777777777779e-06,
      "loss": 0.0002,
      "step": 4250
    },
    {
      "epoch": 61.43,
      "learning_rate": 7.822222222222222e-06,
      "loss": 0.0003,
      "step": 4300
    },
    {
      "epoch": 62.14,
      "learning_rate": 7.266666666666668e-06,
      "loss": 0.001,
      "step": 4350
    },
    {
      "epoch": 62.86,
      "learning_rate": 6.711111111111111e-06,
      "loss": 0.0002,
      "step": 4400
    },
    {
      "epoch": 63.57,
      "learning_rate": 6.155555555555556e-06,
      "loss": 0.0003,
      "step": 4450
    },
    {
      "epoch": 64.29,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.0004,
      "step": 4500
    },
    {
      "epoch": 65.0,
      "learning_rate": 5.044444444444444e-06,
      "loss": 0.0003,
      "step": 4550
    },
    {
      "epoch": 65.71,
      "learning_rate": 4.488888888888889e-06,
      "loss": 0.0003,
      "step": 4600
    },
    {
      "epoch": 66.43,
      "learning_rate": 3.9333333333333335e-06,
      "loss": 0.0002,
      "step": 4650
    },
    {
      "epoch": 67.14,
      "learning_rate": 3.3777777777777777e-06,
      "loss": 0.0002,
      "step": 4700
    },
    {
      "epoch": 67.86,
      "learning_rate": 2.8222222222222223e-06,
      "loss": 0.0003,
      "step": 4750
    },
    {
      "epoch": 68.57,
      "learning_rate": 2.266666666666667e-06,
      "loss": 0.0002,
      "step": 4800
    },
    {
      "epoch": 69.29,
      "learning_rate": 1.7111111111111112e-06,
      "loss": 0.0003,
      "step": 4850
    },
    {
      "epoch": 70.0,
      "learning_rate": 1.1555555555555556e-06,
      "loss": 0.0004,
      "step": 4900
    },
    {
      "epoch": 70.71,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.0002,
      "step": 4950
    },
    {
      "epoch": 71.43,
      "learning_rate": 4.444444444444445e-08,
      "loss": 0.0002,
      "step": 5000
    },
    {
      "epoch": 71.43,
      "eval_loss": 0.7440354824066162,
      "eval_runtime": 495.5118,
      "eval_samples_per_second": 4.575,
      "eval_steps_per_second": 0.143,
      "eval_wer": 44.35982865302237,
      "step": 5000
    },
    {
      "epoch": 71.43,
      "step": 5000,
      "total_flos": 9.230634897555456e+19,
      "train_loss": 0.07175957694733516,
      "train_runtime": 15470.3261,
      "train_samples_per_second": 20.685,
      "train_steps_per_second": 0.323
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 72,
  "total_flos": 9.230634897555456e+19,
  "trial_name": null,
  "trial_params": null
}