{
  "best_metric": 814.6945937054502,
  "best_model_checkpoint": "./whisper-small-vi/checkpoint-2000",
  "epoch": 4.322766570605188,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 5.7877,
      "step": 25
    },
    {
      "epoch": 0.07,
      "learning_rate": 9.000000000000001e-07,
      "loss": 4.8451,
      "step": 50
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.4000000000000001e-06,
      "loss": 3.5507,
      "step": 75
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9000000000000002e-06,
      "loss": 2.6175,
      "step": 100
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.6457,
      "step": 125
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.9e-06,
      "loss": 1.3034,
      "step": 150
    },
    {
      "epoch": 0.25,
      "learning_rate": 3.4000000000000005e-06,
      "loss": 1.1651,
      "step": 175
    },
    {
      "epoch": 0.29,
      "learning_rate": 3.900000000000001e-06,
      "loss": 1.109,
      "step": 200
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.38e-06,
      "loss": 0.9525,
      "step": 225
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.880000000000001e-06,
      "loss": 0.7372,
      "step": 250
    },
    {
      "epoch": 0.4,
      "learning_rate": 5.380000000000001e-06,
      "loss": 0.3308,
      "step": 275
    },
    {
      "epoch": 0.43,
      "learning_rate": 5.8800000000000005e-06,
      "loss": 0.288,
      "step": 300
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.380000000000001e-06,
      "loss": 0.3499,
      "step": 325
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.88e-06,
      "loss": 0.3132,
      "step": 350
    },
    {
      "epoch": 0.54,
      "learning_rate": 7.3800000000000005e-06,
      "loss": 0.2645,
      "step": 375
    },
    {
      "epoch": 0.58,
      "learning_rate": 7.88e-06,
      "loss": 0.2469,
      "step": 400
    },
    {
      "epoch": 0.61,
      "learning_rate": 8.380000000000001e-06,
      "loss": 0.2709,
      "step": 425
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.880000000000001e-06,
      "loss": 0.2689,
      "step": 450
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.38e-06,
      "loss": 0.2883,
      "step": 475
    },
    {
      "epoch": 0.72,
      "learning_rate": 9.88e-06,
      "loss": 0.2295,
      "step": 500
    },
    {
      "epoch": 0.76,
      "learning_rate": 9.945714285714286e-06,
      "loss": 0.2475,
      "step": 525
    },
    {
      "epoch": 0.79,
      "learning_rate": 9.874285714285715e-06,
      "loss": 0.2683,
      "step": 550
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.802857142857144e-06,
      "loss": 0.274,
      "step": 575
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.731428571428573e-06,
      "loss": 0.2694,
      "step": 600
    },
    {
      "epoch": 0.9,
      "learning_rate": 9.66e-06,
      "loss": 0.2621,
      "step": 625
    },
    {
      "epoch": 0.94,
      "learning_rate": 9.58857142857143e-06,
      "loss": 0.2185,
      "step": 650
    },
    {
      "epoch": 0.97,
      "learning_rate": 9.517142857142859e-06,
      "loss": 0.2433,
      "step": 675
    },
    {
      "epoch": 1.01,
      "learning_rate": 9.445714285714288e-06,
      "loss": 0.2303,
      "step": 700
    },
    {
      "epoch": 1.04,
      "learning_rate": 9.374285714285715e-06,
      "loss": 0.148,
      "step": 725
    },
    {
      "epoch": 1.08,
      "learning_rate": 9.302857142857144e-06,
      "loss": 0.1834,
      "step": 750
    },
    {
      "epoch": 1.12,
      "learning_rate": 9.231428571428573e-06,
      "loss": 0.1685,
      "step": 775
    },
    {
      "epoch": 1.15,
      "learning_rate": 9.16e-06,
      "loss": 0.1677,
      "step": 800
    },
    {
      "epoch": 1.19,
      "learning_rate": 9.08857142857143e-06,
      "loss": 0.1705,
      "step": 825
    },
    {
      "epoch": 1.22,
      "learning_rate": 9.017142857142858e-06,
      "loss": 0.1427,
      "step": 850
    },
    {
      "epoch": 1.26,
      "learning_rate": 8.945714285714286e-06,
      "loss": 0.1716,
      "step": 875
    },
    {
      "epoch": 1.3,
      "learning_rate": 8.874285714285715e-06,
      "loss": 0.1328,
      "step": 900
    },
    {
      "epoch": 1.33,
      "learning_rate": 8.802857142857144e-06,
      "loss": 0.1332,
      "step": 925
    },
    {
      "epoch": 1.37,
      "learning_rate": 8.731428571428571e-06,
      "loss": 0.1705,
      "step": 950
    },
    {
      "epoch": 1.4,
      "learning_rate": 8.66e-06,
      "loss": 0.1306,
      "step": 975
    },
    {
      "epoch": 1.44,
      "learning_rate": 8.588571428571429e-06,
      "loss": 0.1627,
      "step": 1000
    },
    {
      "epoch": 1.44,
      "eval_loss": 0.5460556149482727,
      "eval_runtime": 1644.0015,
      "eval_samples_per_second": 0.752,
      "eval_steps_per_second": 0.094,
      "eval_wer": 1004.5290053733962,
      "step": 1000
    },
    {
      "epoch": 1.48,
      "learning_rate": 8.517142857142858e-06,
      "loss": 0.1499,
      "step": 1025
    },
    {
      "epoch": 1.51,
      "learning_rate": 8.445714285714285e-06,
      "loss": 0.1401,
      "step": 1050
    },
    {
      "epoch": 1.55,
      "learning_rate": 8.374285714285714e-06,
      "loss": 0.1588,
      "step": 1075
    },
    {
      "epoch": 1.59,
      "learning_rate": 8.302857142857143e-06,
      "loss": 0.1326,
      "step": 1100
    },
    {
      "epoch": 1.62,
      "learning_rate": 8.231428571428572e-06,
      "loss": 0.1421,
      "step": 1125
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.16e-06,
      "loss": 0.1479,
      "step": 1150
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.088571428571429e-06,
      "loss": 0.1357,
      "step": 1175
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.017142857142858e-06,
      "loss": 0.1458,
      "step": 1200
    },
    {
      "epoch": 1.77,
      "learning_rate": 7.945714285714287e-06,
      "loss": 0.1338,
      "step": 1225
    },
    {
      "epoch": 1.8,
      "learning_rate": 7.874285714285716e-06,
      "loss": 0.1448,
      "step": 1250
    },
    {
      "epoch": 1.84,
      "learning_rate": 7.802857142857143e-06,
      "loss": 0.1507,
      "step": 1275
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.731428571428572e-06,
      "loss": 0.124,
      "step": 1300
    },
    {
      "epoch": 1.91,
      "learning_rate": 7.660000000000001e-06,
      "loss": 0.1538,
      "step": 1325
    },
    {
      "epoch": 1.95,
      "learning_rate": 7.588571428571429e-06,
      "loss": 0.1425,
      "step": 1350
    },
    {
      "epoch": 1.98,
      "learning_rate": 7.5171428571428575e-06,
      "loss": 0.1787,
      "step": 1375
    },
    {
      "epoch": 2.02,
      "learning_rate": 7.445714285714286e-06,
      "loss": 0.111,
      "step": 1400
    },
    {
      "epoch": 2.05,
      "learning_rate": 7.374285714285715e-06,
      "loss": 0.062,
      "step": 1425
    },
    {
      "epoch": 2.09,
      "learning_rate": 7.302857142857144e-06,
      "loss": 0.0726,
      "step": 1450
    },
    {
      "epoch": 2.13,
      "learning_rate": 7.231428571428573e-06,
      "loss": 0.0712,
      "step": 1475
    },
    {
      "epoch": 2.16,
      "learning_rate": 7.16e-06,
      "loss": 0.07,
      "step": 1500
    },
    {
      "epoch": 2.2,
      "learning_rate": 7.088571428571429e-06,
      "loss": 0.0758,
      "step": 1525
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.017142857142858e-06,
      "loss": 0.0817,
      "step": 1550
    },
    {
      "epoch": 2.27,
      "learning_rate": 6.945714285714287e-06,
      "loss": 0.068,
      "step": 1575
    },
    {
      "epoch": 2.31,
      "learning_rate": 6.874285714285714e-06,
      "loss": 0.0598,
      "step": 1600
    },
    {
      "epoch": 2.34,
      "learning_rate": 6.8028571428571434e-06,
      "loss": 0.0582,
      "step": 1625
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.7314285714285724e-06,
      "loss": 0.0921,
      "step": 1650
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.660000000000001e-06,
      "loss": 0.0647,
      "step": 1675
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.588571428571429e-06,
      "loss": 0.0769,
      "step": 1700
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.517142857142858e-06,
      "loss": 0.0833,
      "step": 1725
    },
    {
      "epoch": 2.52,
      "learning_rate": 6.445714285714286e-06,
      "loss": 0.0806,
      "step": 1750
    },
    {
      "epoch": 2.56,
      "learning_rate": 6.374285714285715e-06,
      "loss": 0.073,
      "step": 1775
    },
    {
      "epoch": 2.59,
      "learning_rate": 6.302857142857144e-06,
      "loss": 0.0691,
      "step": 1800
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.231428571428571e-06,
      "loss": 0.0553,
      "step": 1825
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.16e-06,
      "loss": 0.0937,
      "step": 1850
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.088571428571429e-06,
      "loss": 0.0746,
      "step": 1875
    },
    {
      "epoch": 2.74,
      "learning_rate": 6.017142857142858e-06,
      "loss": 0.0642,
      "step": 1900
    },
    {
      "epoch": 2.77,
      "learning_rate": 5.945714285714286e-06,
      "loss": 0.0809,
      "step": 1925
    },
    {
      "epoch": 2.81,
      "learning_rate": 5.874285714285715e-06,
      "loss": 0.062,
      "step": 1950
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.802857142857144e-06,
      "loss": 0.0652,
      "step": 1975
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.731428571428572e-06,
      "loss": 0.07,
      "step": 2000
    },
    {
      "epoch": 2.88,
      "eval_loss": 0.5394195318222046,
      "eval_runtime": 1644.6856,
      "eval_samples_per_second": 0.752,
      "eval_steps_per_second": 0.094,
      "eval_wer": 814.6945937054502,
      "step": 2000
    },
    {
      "epoch": 2.92,
      "learning_rate": 5.66e-06,
      "loss": 0.0761,
      "step": 2025
    },
    {
      "epoch": 2.95,
      "learning_rate": 5.588571428571429e-06,
      "loss": 0.0641,
      "step": 2050
    },
    {
      "epoch": 2.99,
      "learning_rate": 5.517142857142857e-06,
      "loss": 0.0699,
      "step": 2075
    },
    {
      "epoch": 3.03,
      "learning_rate": 5.445714285714286e-06,
      "loss": 0.0506,
      "step": 2100
    },
    {
      "epoch": 3.06,
      "learning_rate": 5.374285714285715e-06,
      "loss": 0.0347,
      "step": 2125
    },
    {
      "epoch": 3.1,
      "learning_rate": 5.3028571428571425e-06,
      "loss": 0.0264,
      "step": 2150
    },
    {
      "epoch": 3.13,
      "learning_rate": 5.2314285714285716e-06,
      "loss": 0.0262,
      "step": 2175
    },
    {
      "epoch": 3.17,
      "learning_rate": 5.1600000000000006e-06,
      "loss": 0.0321,
      "step": 2200
    },
    {
      "epoch": 3.21,
      "learning_rate": 5.08857142857143e-06,
      "loss": 0.0363,
      "step": 2225
    },
    {
      "epoch": 3.24,
      "learning_rate": 5.017142857142857e-06,
      "loss": 0.0313,
      "step": 2250
    },
    {
      "epoch": 3.28,
      "learning_rate": 4.945714285714286e-06,
      "loss": 0.0365,
      "step": 2275
    },
    {
      "epoch": 3.31,
      "learning_rate": 4.874285714285715e-06,
      "loss": 0.0285,
      "step": 2300
    },
    {
      "epoch": 3.35,
      "learning_rate": 4.802857142857143e-06,
      "loss": 0.0355,
      "step": 2325
    },
    {
      "epoch": 3.39,
      "learning_rate": 4.731428571428572e-06,
      "loss": 0.0346,
      "step": 2350
    },
    {
      "epoch": 3.42,
      "learning_rate": 4.66e-06,
      "loss": 0.0268,
      "step": 2375
    },
    {
      "epoch": 3.46,
      "learning_rate": 4.588571428571429e-06,
      "loss": 0.0304,
      "step": 2400
    },
    {
      "epoch": 3.49,
      "learning_rate": 4.5171428571428575e-06,
      "loss": 0.0297,
      "step": 2425
    },
    {
      "epoch": 3.53,
      "learning_rate": 4.445714285714286e-06,
      "loss": 0.0359,
      "step": 2450
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.374285714285715e-06,
      "loss": 0.0349,
      "step": 2475
    },
    {
      "epoch": 3.6,
      "learning_rate": 4.302857142857143e-06,
      "loss": 0.0278,
      "step": 2500
    },
    {
      "epoch": 3.64,
      "learning_rate": 4.231428571428572e-06,
      "loss": 0.0433,
      "step": 2525
    },
    {
      "epoch": 3.67,
      "learning_rate": 4.16e-06,
      "loss": 0.0322,
      "step": 2550
    },
    {
      "epoch": 3.71,
      "learning_rate": 4.088571428571429e-06,
      "loss": 0.0398,
      "step": 2575
    },
    {
      "epoch": 3.75,
      "learning_rate": 4.017142857142857e-06,
      "loss": 0.0316,
      "step": 2600
    },
    {
      "epoch": 3.78,
      "learning_rate": 3.945714285714286e-06,
      "loss": 0.0397,
      "step": 2625
    },
    {
      "epoch": 3.82,
      "learning_rate": 3.874285714285715e-06,
      "loss": 0.0318,
      "step": 2650
    },
    {
      "epoch": 3.85,
      "learning_rate": 3.802857142857143e-06,
      "loss": 0.0247,
      "step": 2675
    },
    {
      "epoch": 3.89,
      "learning_rate": 3.731428571428572e-06,
      "loss": 0.0285,
      "step": 2700
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.66e-06,
      "loss": 0.0321,
      "step": 2725
    },
    {
      "epoch": 3.96,
      "learning_rate": 3.588571428571429e-06,
      "loss": 0.028,
      "step": 2750
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.5171428571428573e-06,
      "loss": 0.0251,
      "step": 2775
    },
    {
      "epoch": 4.03,
      "learning_rate": 3.4457142857142863e-06,
      "loss": 0.0125,
      "step": 2800
    },
    {
      "epoch": 4.07,
      "learning_rate": 3.3742857142857145e-06,
      "loss": 0.0152,
      "step": 2825
    },
    {
      "epoch": 4.11,
      "learning_rate": 3.302857142857143e-06,
      "loss": 0.0131,
      "step": 2850
    },
    {
      "epoch": 4.14,
      "learning_rate": 3.2314285714285716e-06,
      "loss": 0.0117,
      "step": 2875
    },
    {
      "epoch": 4.18,
      "learning_rate": 3.1600000000000002e-06,
      "loss": 0.0156,
      "step": 2900
    },
    {
      "epoch": 4.21,
      "learning_rate": 3.0885714285714284e-06,
      "loss": 0.009,
      "step": 2925
    },
    {
      "epoch": 4.25,
      "learning_rate": 3.0171428571428574e-06,
      "loss": 0.0101,
      "step": 2950
    },
    {
      "epoch": 4.29,
      "learning_rate": 2.945714285714286e-06,
      "loss": 0.0143,
      "step": 2975
    },
    {
      "epoch": 4.32,
      "learning_rate": 2.8742857142857146e-06,
      "loss": 0.0151,
      "step": 3000
    },
    {
      "epoch": 4.32,
      "eval_loss": 0.5925257205963135,
      "eval_runtime": 1647.3609,
      "eval_samples_per_second": 0.751,
      "eval_steps_per_second": 0.094,
      "eval_wer": 1116.8110538436233,
      "step": 3000
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 6,
  "total_flos": 3.45956177903616e+18,
  "trial_name": null,
  "trial_params": null
}