{
  "best_metric": 38.85025498377376,
  "best_model_checkpoint": "./whisper-small-tonga\\checkpoint-4000",
  "epoch": 0.9006980409817609,
  "eval_steps": 1000,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.9972,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.7796,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.5e-06,
      "loss": 1.4897,
      "step": 75
    },
    {
      "epoch": 0.02,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.4191,
      "step": 100
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.5e-06,
      "loss": 1.3257,
      "step": 125
    },
    {
      "epoch": 0.03,
      "learning_rate": 3e-06,
      "loss": 1.3542,
      "step": 150
    },
    {
      "epoch": 0.04,
      "learning_rate": 3.5e-06,
      "loss": 1.1868,
      "step": 175
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.049,
      "step": 200
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.5e-06,
      "loss": 0.8937,
      "step": 225
    },
    {
      "epoch": 0.06,
      "learning_rate": 5e-06,
      "loss": 1.2414,
      "step": 250
    },
    {
      "epoch": 0.06,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.8938,
      "step": 275
    },
    {
      "epoch": 0.07,
      "learning_rate": 6e-06,
      "loss": 0.8904,
      "step": 300
    },
    {
      "epoch": 0.07,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.7698,
      "step": 325
    },
    {
      "epoch": 0.08,
      "learning_rate": 7e-06,
      "loss": 0.8931,
      "step": 350
    },
    {
      "epoch": 0.08,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.7658,
      "step": 375
    },
    {
      "epoch": 0.09,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.7351,
      "step": 400
    },
    {
      "epoch": 0.1,
      "learning_rate": 8.5e-06,
      "loss": 0.7168,
      "step": 425
    },
    {
      "epoch": 0.1,
      "learning_rate": 9e-06,
      "loss": 0.725,
      "step": 450
    },
    {
      "epoch": 0.11,
      "learning_rate": 9.5e-06,
      "loss": 0.7358,
      "step": 475
    },
    {
      "epoch": 0.11,
      "learning_rate": 1e-05,
      "loss": 0.7799,
      "step": 500
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.92857142857143e-06,
      "loss": 0.633,
      "step": 525
    },
    {
      "epoch": 0.12,
      "learning_rate": 9.857142857142859e-06,
      "loss": 0.6995,
      "step": 550
    },
    {
      "epoch": 0.13,
      "learning_rate": 9.785714285714286e-06,
      "loss": 0.7882,
      "step": 575
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.714285714285715e-06,
      "loss": 0.7329,
      "step": 600
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.642857142857144e-06,
      "loss": 0.6841,
      "step": 625
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.571428571428573e-06,
      "loss": 0.5597,
      "step": 650
    },
    {
      "epoch": 0.15,
      "learning_rate": 9.5e-06,
      "loss": 0.7231,
      "step": 675
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.42857142857143e-06,
      "loss": 0.6297,
      "step": 700
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.357142857142859e-06,
      "loss": 0.7377,
      "step": 725
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.285714285714288e-06,
      "loss": 0.6607,
      "step": 750
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.214285714285715e-06,
      "loss": 0.7616,
      "step": 775
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.142857142857144e-06,
      "loss": 0.5766,
      "step": 800
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.071428571428573e-06,
      "loss": 0.5233,
      "step": 825
    },
    {
      "epoch": 0.19,
      "learning_rate": 9e-06,
      "loss": 0.6328,
      "step": 850
    },
    {
      "epoch": 0.2,
      "learning_rate": 8.92857142857143e-06,
      "loss": 0.5359,
      "step": 875
    },
    {
      "epoch": 0.2,
      "learning_rate": 8.857142857142858e-06,
      "loss": 0.6946,
      "step": 900
    },
    {
      "epoch": 0.21,
      "learning_rate": 8.785714285714286e-06,
      "loss": 0.5996,
      "step": 925
    },
    {
      "epoch": 0.21,
      "learning_rate": 8.714285714285715e-06,
      "loss": 0.6294,
      "step": 950
    },
    {
      "epoch": 0.22,
      "learning_rate": 8.642857142857144e-06,
      "loss": 0.5598,
      "step": 975
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.571428571428571e-06,
      "loss": 0.6529,
      "step": 1000
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.6057160496711731,
      "eval_runtime": 2992.1646,
      "eval_samples_per_second": 0.155,
      "eval_steps_per_second": 0.019,
      "eval_wer": 46.43022716736208,
      "step": 1000
    },
    {
      "epoch": 0.23,
      "learning_rate": 8.5e-06,
      "loss": 0.5449,
      "step": 1025
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.428571428571429e-06,
      "loss": 0.6225,
      "step": 1050
    },
    {
      "epoch": 0.24,
      "learning_rate": 8.357142857142858e-06,
      "loss": 0.5594,
      "step": 1075
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.285714285714287e-06,
      "loss": 0.5715,
      "step": 1100
    },
    {
      "epoch": 0.25,
      "learning_rate": 8.214285714285714e-06,
      "loss": 0.5489,
      "step": 1125
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.142857142857143e-06,
      "loss": 0.7007,
      "step": 1150
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.071428571428572e-06,
      "loss": 0.5195,
      "step": 1175
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5413,
      "step": 1200
    },
    {
      "epoch": 0.28,
      "learning_rate": 7.928571428571429e-06,
      "loss": 0.4541,
      "step": 1225
    },
    {
      "epoch": 0.28,
      "learning_rate": 7.857142857142858e-06,
      "loss": 0.4984,
      "step": 1250
    },
    {
      "epoch": 0.29,
      "learning_rate": 7.785714285714287e-06,
      "loss": 0.5392,
      "step": 1275
    },
    {
      "epoch": 0.29,
      "learning_rate": 7.714285714285716e-06,
      "loss": 0.6058,
      "step": 1300
    },
    {
      "epoch": 0.3,
      "learning_rate": 7.642857142857143e-06,
      "loss": 0.5291,
      "step": 1325
    },
    {
      "epoch": 0.3,
      "learning_rate": 7.571428571428572e-06,
      "loss": 0.688,
      "step": 1350
    },
    {
      "epoch": 0.31,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.4688,
      "step": 1375
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.428571428571429e-06,
      "loss": 0.5756,
      "step": 1400
    },
    {
      "epoch": 0.32,
      "learning_rate": 7.357142857142858e-06,
      "loss": 0.5569,
      "step": 1425
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.285714285714286e-06,
      "loss": 0.4964,
      "step": 1450
    },
    {
      "epoch": 0.33,
      "learning_rate": 7.2142857142857145e-06,
      "loss": 0.4675,
      "step": 1475
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.4959,
      "step": 1500
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.0714285714285726e-06,
      "loss": 0.4597,
      "step": 1525
    },
    {
      "epoch": 0.35,
      "learning_rate": 7e-06,
      "loss": 0.5622,
      "step": 1550
    },
    {
      "epoch": 0.35,
      "learning_rate": 6.928571428571429e-06,
      "loss": 0.473,
      "step": 1575
    },
    {
      "epoch": 0.36,
      "learning_rate": 6.857142857142858e-06,
      "loss": 0.478,
      "step": 1600
    },
    {
      "epoch": 0.37,
      "learning_rate": 6.785714285714287e-06,
      "loss": 0.5204,
      "step": 1625
    },
    {
      "epoch": 0.37,
      "learning_rate": 6.714285714285714e-06,
      "loss": 0.5745,
      "step": 1650
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.642857142857143e-06,
      "loss": 0.5921,
      "step": 1675
    },
    {
      "epoch": 0.38,
      "learning_rate": 6.571428571428572e-06,
      "loss": 0.6358,
      "step": 1700
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 0.435,
      "step": 1725
    },
    {
      "epoch": 0.39,
      "learning_rate": 6.4285714285714295e-06,
      "loss": 0.4823,
      "step": 1750
    },
    {
      "epoch": 0.4,
      "learning_rate": 6.357142857142858e-06,
      "loss": 0.4833,
      "step": 1775
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.285714285714286e-06,
      "loss": 0.5223,
      "step": 1800
    },
    {
      "epoch": 0.41,
      "learning_rate": 6.214285714285715e-06,
      "loss": 0.5634,
      "step": 1825
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.142857142857144e-06,
      "loss": 0.5881,
      "step": 1850
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.071428571428571e-06,
      "loss": 0.5054,
      "step": 1875
    },
    {
      "epoch": 0.43,
      "learning_rate": 6e-06,
      "loss": 0.3668,
      "step": 1900
    },
    {
      "epoch": 0.43,
      "learning_rate": 5.928571428571429e-06,
      "loss": 0.4781,
      "step": 1925
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.857142857142858e-06,
      "loss": 0.4599,
      "step": 1950
    },
    {
      "epoch": 0.44,
      "learning_rate": 5.785714285714286e-06,
      "loss": 0.4176,
      "step": 1975
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.4709,
      "step": 2000
    },
    {
      "epoch": 0.45,
      "eval_loss": 0.5212363004684448,
      "eval_runtime": 3278.467,
      "eval_samples_per_second": 0.141,
      "eval_steps_per_second": 0.018,
      "eval_wer": 39.84700973574409,
      "step": 2000
    },
    {
      "epoch": 0.46,
      "learning_rate": 5.6428571428571435e-06,
      "loss": 0.541,
      "step": 2025
    },
    {
      "epoch": 0.46,
      "learning_rate": 5.571428571428572e-06,
      "loss": 0.4594,
      "step": 2050
    },
    {
      "epoch": 0.47,
      "learning_rate": 5.500000000000001e-06,
      "loss": 0.7156,
      "step": 2075
    },
    {
      "epoch": 0.47,
      "learning_rate": 5.428571428571429e-06,
      "loss": 0.5059,
      "step": 2100
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.357142857142857e-06,
      "loss": 0.3725,
      "step": 2125
    },
    {
      "epoch": 0.48,
      "learning_rate": 5.285714285714286e-06,
      "loss": 0.5438,
      "step": 2150
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.214285714285715e-06,
      "loss": 0.5069,
      "step": 2175
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.142857142857142e-06,
      "loss": 0.567,
      "step": 2200
    },
    {
      "epoch": 0.5,
      "learning_rate": 5.071428571428571e-06,
      "loss": 0.4593,
      "step": 2225
    },
    {
      "epoch": 0.51,
      "learning_rate": 5e-06,
      "loss": 0.5366,
      "step": 2250
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.928571428571429e-06,
      "loss": 0.3691,
      "step": 2275
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.857142857142858e-06,
      "loss": 0.3866,
      "step": 2300
    },
    {
      "epoch": 0.52,
      "learning_rate": 4.785714285714287e-06,
      "loss": 0.5432,
      "step": 2325
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.714285714285715e-06,
      "loss": 0.4391,
      "step": 2350
    },
    {
      "epoch": 0.53,
      "learning_rate": 4.642857142857144e-06,
      "loss": 0.4629,
      "step": 2375
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.571428571428572e-06,
      "loss": 0.4919,
      "step": 2400
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.5e-06,
      "loss": 0.5618,
      "step": 2425
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.428571428571429e-06,
      "loss": 0.3765,
      "step": 2450
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.357142857142857e-06,
      "loss": 0.4851,
      "step": 2475
    },
    {
      "epoch": 0.56,
      "learning_rate": 4.2857142857142855e-06,
      "loss": 0.4779,
      "step": 2500
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.2142857142857145e-06,
      "loss": 0.5239,
      "step": 2525
    },
    {
      "epoch": 0.57,
      "learning_rate": 4.1428571428571435e-06,
      "loss": 0.552,
      "step": 2550
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.071428571428572e-06,
      "loss": 0.4018,
      "step": 2575
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.4134,
      "step": 2600
    },
    {
      "epoch": 0.59,
      "learning_rate": 3.928571428571429e-06,
      "loss": 0.4725,
      "step": 2625
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.857142857142858e-06,
      "loss": 0.5316,
      "step": 2650
    },
    {
      "epoch": 0.6,
      "learning_rate": 3.785714285714286e-06,
      "loss": 0.3966,
      "step": 2675
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.7142857142857146e-06,
      "loss": 0.4601,
      "step": 2700
    },
    {
      "epoch": 0.61,
      "learning_rate": 3.642857142857143e-06,
      "loss": 0.5977,
      "step": 2725
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.5714285714285718e-06,
      "loss": 0.5199,
      "step": 2750
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.5e-06,
      "loss": 0.5612,
      "step": 2775
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.428571428571429e-06,
      "loss": 0.4065,
      "step": 2800
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.357142857142857e-06,
      "loss": 0.5322,
      "step": 2825
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.285714285714286e-06,
      "loss": 0.4151,
      "step": 2850
    },
    {
      "epoch": 0.65,
      "learning_rate": 3.2142857142857147e-06,
      "loss": 0.3623,
      "step": 2875
    },
    {
      "epoch": 0.65,
      "learning_rate": 3.142857142857143e-06,
      "loss": 0.488,
      "step": 2900
    },
    {
      "epoch": 0.66,
      "learning_rate": 3.071428571428572e-06,
      "loss": 0.3809,
      "step": 2925
    },
    {
      "epoch": 0.66,
      "learning_rate": 3e-06,
      "loss": 0.4215,
      "step": 2950
    },
    {
      "epoch": 0.67,
      "learning_rate": 2.928571428571429e-06,
      "loss": 0.3998,
      "step": 2975
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.3379,
      "step": 3000
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.47441861033439636,
      "eval_runtime": 5182.276,
      "eval_samples_per_second": 0.089,
      "eval_steps_per_second": 0.011,
      "eval_wer": 45.24802967083913,
      "step": 3000
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.785714285714286e-06,
      "loss": 0.3727,
      "step": 3025
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.7142857142857144e-06,
      "loss": 0.51,
      "step": 3050
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.642857142857143e-06,
      "loss": 0.4227,
      "step": 3075
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.571428571428571e-06,
      "loss": 0.4547,
      "step": 3100
    },
    {
      "epoch": 0.7,
      "learning_rate": 2.5e-06,
      "loss": 0.4407,
      "step": 3125
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.428571428571429e-06,
      "loss": 0.3879,
      "step": 3150
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.3571428571428574e-06,
      "loss": 0.4217,
      "step": 3175
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.285714285714286e-06,
      "loss": 0.4735,
      "step": 3200
    },
    {
      "epoch": 0.73,
      "learning_rate": 2.2142857142857146e-06,
      "loss": 0.4663,
      "step": 3225
    },
    {
      "epoch": 0.73,
      "learning_rate": 2.1428571428571427e-06,
      "loss": 0.4745,
      "step": 3250
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.0714285714285717e-06,
      "loss": 0.3319,
      "step": 3275
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.3633,
      "step": 3300
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.928571428571429e-06,
      "loss": 0.3954,
      "step": 3325
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.8571428571428573e-06,
      "loss": 0.4413,
      "step": 3350
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.7857142857142859e-06,
      "loss": 0.4784,
      "step": 3375
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.7142857142857145e-06,
      "loss": 0.4328,
      "step": 3400
    },
    {
      "epoch": 0.77,
      "learning_rate": 1.642857142857143e-06,
      "loss": 0.4514,
      "step": 3425
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5714285714285714e-06,
      "loss": 0.3688,
      "step": 3450
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5e-06,
      "loss": 0.3416,
      "step": 3475
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.4105,
      "step": 3500
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.3571428571428572e-06,
      "loss": 0.3837,
      "step": 3525
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.2857142857142856e-06,
      "loss": 0.2904,
      "step": 3550
    },
    {
      "epoch": 0.8,
      "learning_rate": 1.2142857142857144e-06,
      "loss": 0.5686,
      "step": 3575
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.142857142857143e-06,
      "loss": 0.5217,
      "step": 3600
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0714285714285714e-06,
      "loss": 0.4856,
      "step": 3625
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.3977,
      "step": 3650
    },
    {
      "epoch": 0.83,
      "learning_rate": 9.285714285714287e-07,
      "loss": 0.3874,
      "step": 3675
    },
    {
      "epoch": 0.83,
      "learning_rate": 8.571428571428572e-07,
      "loss": 0.4817,
      "step": 3700
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.857142857142857e-07,
      "loss": 0.4066,
      "step": 3725
    },
    {
      "epoch": 0.84,
      "learning_rate": 7.142857142857143e-07,
      "loss": 0.3815,
      "step": 3750
    },
    {
      "epoch": 0.85,
      "learning_rate": 6.428571428571428e-07,
      "loss": 0.4581,
      "step": 3775
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.714285714285715e-07,
      "loss": 0.5065,
      "step": 3800
    },
    {
      "epoch": 0.86,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4271,
      "step": 3825
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.285714285714286e-07,
      "loss": 0.392,
      "step": 3850
    },
    {
      "epoch": 0.87,
      "learning_rate": 3.5714285714285716e-07,
      "loss": 0.3524,
      "step": 3875
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.8571428571428575e-07,
      "loss": 0.3474,
      "step": 3900
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.142857142857143e-07,
      "loss": 0.4546,
      "step": 3925
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.4285714285714287e-07,
      "loss": 0.352,
      "step": 3950
    },
    {
      "epoch": 0.9,
      "learning_rate": 7.142857142857144e-08,
      "loss": 0.4057,
      "step": 3975
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0,
      "loss": 0.3411,
      "step": 4000
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.45225948095321655,
      "eval_runtime": 3736.5714,
      "eval_samples_per_second": 0.124,
      "eval_steps_per_second": 0.016,
      "eval_wer": 38.85025498377376,
      "step": 4000
    }
  ],
  "logging_steps": 25,
  "max_steps": 4000,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "total_flos": 2.30868320256e+18,
  "trial_name": null,
  "trial_params": null
}