{
  "best_metric": 15.736901620806634,
  "best_model_checkpoint": "./checkpoint-2000",
  "epoch": 76.012,
  "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": 4.6000000000000004e-07,
      "loss": 1.2023,
      "step": 25
    },
    {
      "epoch": 0.01,
      "learning_rate": 9.600000000000001e-07,
      "loss": 0.8225,
      "step": 50
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.46e-06,
      "loss": 0.4219,
      "step": 75
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.9600000000000003e-06,
      "loss": 0.2197,
      "step": 100
    },
    {
      "epoch": 2.01,
      "learning_rate": 2.46e-06,
      "loss": 0.2024,
      "step": 125
    },
    {
      "epoch": 2.01,
      "learning_rate": 2.96e-06,
      "loss": 0.1199,
      "step": 150
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.46e-06,
      "loss": 0.1207,
      "step": 175
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.96e-06,
      "loss": 0.0724,
      "step": 200
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.4600000000000005e-06,
      "loss": 0.062,
      "step": 225
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.960000000000001e-06,
      "loss": 0.0386,
      "step": 250
    },
    {
      "epoch": 5.0,
      "learning_rate": 5.460000000000001e-06,
      "loss": 0.0326,
      "step": 275
    },
    {
      "epoch": 5.01,
      "learning_rate": 5.9600000000000005e-06,
      "loss": 0.0206,
      "step": 300
    },
    {
      "epoch": 6.0,
      "learning_rate": 6.460000000000001e-06,
      "loss": 0.0183,
      "step": 325
    },
    {
      "epoch": 6.01,
      "learning_rate": 6.96e-06,
      "loss": 0.0136,
      "step": 350
    },
    {
      "epoch": 7.0,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.0121,
      "step": 375
    },
    {
      "epoch": 7.01,
      "learning_rate": 7.960000000000002e-06,
      "loss": 0.0107,
      "step": 400
    },
    {
      "epoch": 8.0,
      "learning_rate": 8.46e-06,
      "loss": 0.0105,
      "step": 425
    },
    {
      "epoch": 8.01,
      "learning_rate": 8.96e-06,
      "loss": 0.0101,
      "step": 450
    },
    {
      "epoch": 9.0,
      "learning_rate": 9.460000000000001e-06,
      "loss": 0.0116,
      "step": 475
    },
    {
      "epoch": 9.01,
      "learning_rate": 9.960000000000001e-06,
      "loss": 0.011,
      "step": 500
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.934285714285715e-06,
      "loss": 0.0129,
      "step": 525
    },
    {
      "epoch": 10.01,
      "learning_rate": 9.862857142857144e-06,
      "loss": 0.0108,
      "step": 550
    },
    {
      "epoch": 11.0,
      "learning_rate": 9.791428571428571e-06,
      "loss": 0.0088,
      "step": 575
    },
    {
      "epoch": 11.01,
      "learning_rate": 9.72e-06,
      "loss": 0.0066,
      "step": 600
    },
    {
      "epoch": 12.0,
      "learning_rate": 9.648571428571429e-06,
      "loss": 0.0056,
      "step": 625
    },
    {
      "epoch": 12.01,
      "learning_rate": 9.577142857142858e-06,
      "loss": 0.0052,
      "step": 650
    },
    {
      "epoch": 12.01,
      "learning_rate": 9.505714285714287e-06,
      "loss": 0.0043,
      "step": 675
    },
    {
      "epoch": 13.01,
      "learning_rate": 9.434285714285714e-06,
      "loss": 0.0051,
      "step": 700
    },
    {
      "epoch": 13.01,
      "learning_rate": 9.362857142857143e-06,
      "loss": 0.0032,
      "step": 725
    },
    {
      "epoch": 14.01,
      "learning_rate": 9.291428571428572e-06,
      "loss": 0.0037,
      "step": 750
    },
    {
      "epoch": 14.01,
      "learning_rate": 9.220000000000002e-06,
      "loss": 0.004,
      "step": 775
    },
    {
      "epoch": 15.01,
      "learning_rate": 9.148571428571429e-06,
      "loss": 0.004,
      "step": 800
    },
    {
      "epoch": 15.01,
      "learning_rate": 9.077142857142858e-06,
      "loss": 0.0043,
      "step": 825
    },
    {
      "epoch": 16.0,
      "learning_rate": 9.005714285714287e-06,
      "loss": 0.0037,
      "step": 850
    },
    {
      "epoch": 16.01,
      "learning_rate": 8.934285714285716e-06,
      "loss": 0.0025,
      "step": 875
    },
    {
      "epoch": 17.0,
      "learning_rate": 8.862857142857143e-06,
      "loss": 0.0036,
      "step": 900
    },
    {
      "epoch": 17.01,
      "learning_rate": 8.791428571428572e-06,
      "loss": 0.0016,
      "step": 925
    },
    {
      "epoch": 18.0,
      "learning_rate": 8.720000000000001e-06,
      "loss": 0.0024,
      "step": 950
    },
    {
      "epoch": 18.01,
      "learning_rate": 8.64857142857143e-06,
      "loss": 0.0012,
      "step": 975
    },
    {
      "epoch": 19.0,
      "learning_rate": 8.577142857142858e-06,
      "loss": 0.0014,
      "step": 1000
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.30293965339660645,
      "eval_runtime": 1421.7325,
      "eval_samples_per_second": 1.199,
      "eval_steps_per_second": 0.038,
      "eval_wer": 16.311722578213345,
      "step": 1000
    },
    {
      "epoch": 19.01,
      "learning_rate": 8.505714285714287e-06,
      "loss": 0.0012,
      "step": 1025
    },
    {
      "epoch": 20.0,
      "learning_rate": 8.434285714285716e-06,
      "loss": 0.0011,
      "step": 1050
    },
    {
      "epoch": 20.01,
      "learning_rate": 8.362857142857143e-06,
      "loss": 0.0009,
      "step": 1075
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.291428571428572e-06,
      "loss": 0.0011,
      "step": 1100
    },
    {
      "epoch": 21.01,
      "learning_rate": 8.220000000000001e-06,
      "loss": 0.0008,
      "step": 1125
    },
    {
      "epoch": 22.0,
      "learning_rate": 8.148571428571428e-06,
      "loss": 0.0015,
      "step": 1150
    },
    {
      "epoch": 22.01,
      "learning_rate": 8.077142857142857e-06,
      "loss": 0.0006,
      "step": 1175
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.005714285714286e-06,
      "loss": 0.0006,
      "step": 1200
    },
    {
      "epoch": 23.01,
      "learning_rate": 7.934285714285715e-06,
      "loss": 0.0005,
      "step": 1225
    },
    {
      "epoch": 24.0,
      "learning_rate": 7.862857142857143e-06,
      "loss": 0.0004,
      "step": 1250
    },
    {
      "epoch": 24.01,
      "learning_rate": 7.791428571428572e-06,
      "loss": 0.0004,
      "step": 1275
    },
    {
      "epoch": 24.01,
      "learning_rate": 7.72e-06,
      "loss": 0.0004,
      "step": 1300
    },
    {
      "epoch": 25.01,
      "learning_rate": 7.64857142857143e-06,
      "loss": 0.0004,
      "step": 1325
    },
    {
      "epoch": 25.01,
      "learning_rate": 7.577142857142857e-06,
      "loss": 0.0004,
      "step": 1350
    },
    {
      "epoch": 26.01,
      "learning_rate": 7.505714285714286e-06,
      "loss": 0.0004,
      "step": 1375
    },
    {
      "epoch": 26.01,
      "learning_rate": 7.434285714285715e-06,
      "loss": 0.0003,
      "step": 1400
    },
    {
      "epoch": 27.01,
      "learning_rate": 7.362857142857144e-06,
      "loss": 0.0004,
      "step": 1425
    },
    {
      "epoch": 27.01,
      "learning_rate": 7.291428571428571e-06,
      "loss": 0.0003,
      "step": 1450
    },
    {
      "epoch": 28.0,
      "learning_rate": 7.22e-06,
      "loss": 0.0003,
      "step": 1475
    },
    {
      "epoch": 28.01,
      "learning_rate": 7.148571428571429e-06,
      "loss": 0.0003,
      "step": 1500
    },
    {
      "epoch": 29.0,
      "learning_rate": 7.077142857142858e-06,
      "loss": 0.0003,
      "step": 1525
    },
    {
      "epoch": 29.01,
      "learning_rate": 7.0057142857142865e-06,
      "loss": 0.0003,
      "step": 1550
    },
    {
      "epoch": 30.0,
      "learning_rate": 6.934285714285715e-06,
      "loss": 0.0003,
      "step": 1575
    },
    {
      "epoch": 30.01,
      "learning_rate": 6.862857142857144e-06,
      "loss": 0.0003,
      "step": 1600
    },
    {
      "epoch": 31.0,
      "learning_rate": 6.791428571428572e-06,
      "loss": 0.0003,
      "step": 1625
    },
    {
      "epoch": 31.01,
      "learning_rate": 6.720000000000001e-06,
      "loss": 0.0003,
      "step": 1650
    },
    {
      "epoch": 32.0,
      "learning_rate": 6.648571428571429e-06,
      "loss": 0.0003,
      "step": 1675
    },
    {
      "epoch": 32.01,
      "learning_rate": 6.577142857142857e-06,
      "loss": 0.0003,
      "step": 1700
    },
    {
      "epoch": 33.0,
      "learning_rate": 6.505714285714286e-06,
      "loss": 0.0003,
      "step": 1725
    },
    {
      "epoch": 33.01,
      "learning_rate": 6.434285714285715e-06,
      "loss": 0.0003,
      "step": 1750
    },
    {
      "epoch": 34.0,
      "learning_rate": 6.3628571428571426e-06,
      "loss": 0.0002,
      "step": 1775
    },
    {
      "epoch": 34.01,
      "learning_rate": 6.2914285714285716e-06,
      "loss": 0.0002,
      "step": 1800
    },
    {
      "epoch": 35.0,
      "learning_rate": 6.220000000000001e-06,
      "loss": 0.0002,
      "step": 1825
    },
    {
      "epoch": 35.01,
      "learning_rate": 6.14857142857143e-06,
      "loss": 0.0002,
      "step": 1850
    },
    {
      "epoch": 36.0,
      "learning_rate": 6.077142857142858e-06,
      "loss": 0.0002,
      "step": 1875
    },
    {
      "epoch": 36.01,
      "learning_rate": 6.005714285714286e-06,
      "loss": 0.0002,
      "step": 1900
    },
    {
      "epoch": 37.0,
      "learning_rate": 5.934285714285715e-06,
      "loss": 0.0002,
      "step": 1925
    },
    {
      "epoch": 37.01,
      "learning_rate": 5.862857142857143e-06,
      "loss": 0.0002,
      "step": 1950
    },
    {
      "epoch": 37.01,
      "learning_rate": 5.791428571428572e-06,
      "loss": 0.0002,
      "step": 1975
    },
    {
      "epoch": 38.01,
      "learning_rate": 5.72e-06,
      "loss": 0.0002,
      "step": 2000
    },
    {
      "epoch": 38.01,
      "eval_loss": 0.3411926031112671,
      "eval_runtime": 1459.4397,
      "eval_samples_per_second": 1.168,
      "eval_steps_per_second": 0.037,
      "eval_wer": 15.736901620806634,
      "step": 2000
    },
    {
      "epoch": 38.01,
      "learning_rate": 5.6485714285714285e-06,
      "loss": 0.0002,
      "step": 2025
    },
    {
      "epoch": 39.01,
      "learning_rate": 5.5771428571428575e-06,
      "loss": 0.0002,
      "step": 2050
    },
    {
      "epoch": 39.01,
      "learning_rate": 5.5057142857142865e-06,
      "loss": 0.0002,
      "step": 2075
    },
    {
      "epoch": 40.01,
      "learning_rate": 5.4342857142857155e-06,
      "loss": 0.0002,
      "step": 2100
    },
    {
      "epoch": 40.01,
      "learning_rate": 5.362857142857143e-06,
      "loss": 0.0002,
      "step": 2125
    },
    {
      "epoch": 41.0,
      "learning_rate": 5.291428571428572e-06,
      "loss": 0.0002,
      "step": 2150
    },
    {
      "epoch": 41.01,
      "learning_rate": 5.220000000000001e-06,
      "loss": 0.0002,
      "step": 2175
    },
    {
      "epoch": 42.0,
      "learning_rate": 5.14857142857143e-06,
      "loss": 0.0002,
      "step": 2200
    },
    {
      "epoch": 42.01,
      "learning_rate": 5.077142857142857e-06,
      "loss": 0.0002,
      "step": 2225
    },
    {
      "epoch": 43.0,
      "learning_rate": 5.005714285714286e-06,
      "loss": 0.0002,
      "step": 2250
    },
    {
      "epoch": 43.01,
      "learning_rate": 4.934285714285715e-06,
      "loss": 0.0002,
      "step": 2275
    },
    {
      "epoch": 44.0,
      "learning_rate": 4.862857142857143e-06,
      "loss": 0.0002,
      "step": 2300
    },
    {
      "epoch": 44.01,
      "learning_rate": 4.7914285714285715e-06,
      "loss": 0.0002,
      "step": 2325
    },
    {
      "epoch": 45.0,
      "learning_rate": 4.7200000000000005e-06,
      "loss": 0.0002,
      "step": 2350
    },
    {
      "epoch": 45.01,
      "learning_rate": 4.648571428571429e-06,
      "loss": 0.0002,
      "step": 2375
    },
    {
      "epoch": 46.0,
      "learning_rate": 4.577142857142858e-06,
      "loss": 0.0002,
      "step": 2400
    },
    {
      "epoch": 46.01,
      "learning_rate": 4.505714285714286e-06,
      "loss": 0.0002,
      "step": 2425
    },
    {
      "epoch": 47.0,
      "learning_rate": 4.434285714285715e-06,
      "loss": 0.0002,
      "step": 2450
    },
    {
      "epoch": 47.01,
      "learning_rate": 4.362857142857143e-06,
      "loss": 0.0002,
      "step": 2475
    },
    {
      "epoch": 48.0,
      "learning_rate": 4.291428571428572e-06,
      "loss": 0.0001,
      "step": 2500
    },
    {
      "epoch": 48.01,
      "learning_rate": 4.22e-06,
      "loss": 0.0002,
      "step": 2525
    },
    {
      "epoch": 49.0,
      "learning_rate": 4.148571428571429e-06,
      "loss": 0.0001,
      "step": 2550
    },
    {
      "epoch": 49.01,
      "learning_rate": 4.0771428571428574e-06,
      "loss": 0.0001,
      "step": 2575
    },
    {
      "epoch": 49.01,
      "learning_rate": 4.0057142857142864e-06,
      "loss": 0.0001,
      "step": 2600
    },
    {
      "epoch": 50.01,
      "learning_rate": 3.934285714285715e-06,
      "loss": 0.0001,
      "step": 2625
    },
    {
      "epoch": 50.01,
      "learning_rate": 3.862857142857143e-06,
      "loss": 0.0001,
      "step": 2650
    },
    {
      "epoch": 51.01,
      "learning_rate": 3.7914285714285722e-06,
      "loss": 0.0001,
      "step": 2675
    },
    {
      "epoch": 51.01,
      "learning_rate": 3.7200000000000004e-06,
      "loss": 0.0001,
      "step": 2700
    },
    {
      "epoch": 52.01,
      "learning_rate": 3.648571428571429e-06,
      "loss": 0.0001,
      "step": 2725
    },
    {
      "epoch": 52.01,
      "learning_rate": 3.5771428571428576e-06,
      "loss": 0.0001,
      "step": 2750
    },
    {
      "epoch": 53.0,
      "learning_rate": 3.505714285714286e-06,
      "loss": 0.0001,
      "step": 2775
    },
    {
      "epoch": 53.01,
      "learning_rate": 3.4342857142857143e-06,
      "loss": 0.0001,
      "step": 2800
    },
    {
      "epoch": 54.0,
      "learning_rate": 3.3628571428571433e-06,
      "loss": 0.0001,
      "step": 2825
    },
    {
      "epoch": 54.01,
      "learning_rate": 3.2914285714285715e-06,
      "loss": 0.0001,
      "step": 2850
    },
    {
      "epoch": 55.0,
      "learning_rate": 3.2200000000000005e-06,
      "loss": 0.0001,
      "step": 2875
    },
    {
      "epoch": 55.01,
      "learning_rate": 3.1485714285714287e-06,
      "loss": 0.0001,
      "step": 2900
    },
    {
      "epoch": 56.0,
      "learning_rate": 3.0771428571428573e-06,
      "loss": 0.0001,
      "step": 2925
    },
    {
      "epoch": 56.01,
      "learning_rate": 3.005714285714286e-06,
      "loss": 0.0001,
      "step": 2950
    },
    {
      "epoch": 57.0,
      "learning_rate": 2.9342857142857144e-06,
      "loss": 0.0001,
      "step": 2975
    },
    {
      "epoch": 57.01,
      "learning_rate": 2.8628571428571435e-06,
      "loss": 0.0001,
      "step": 3000
    },
    {
      "epoch": 57.01,
      "eval_loss": 0.35916897654533386,
      "eval_runtime": 1464.8018,
      "eval_samples_per_second": 1.163,
      "eval_steps_per_second": 0.037,
      "eval_wer": 15.878251036562382,
      "step": 3000
    },
    {
      "epoch": 58.0,
      "learning_rate": 2.7914285714285716e-06,
      "loss": 0.0001,
      "step": 3025
    },
    {
      "epoch": 58.01,
      "learning_rate": 2.7200000000000002e-06,
      "loss": 0.0001,
      "step": 3050
    },
    {
      "epoch": 59.0,
      "learning_rate": 2.648571428571429e-06,
      "loss": 0.0001,
      "step": 3075
    },
    {
      "epoch": 59.01,
      "learning_rate": 2.5771428571428574e-06,
      "loss": 0.0001,
      "step": 3100
    },
    {
      "epoch": 60.0,
      "learning_rate": 2.5057142857142856e-06,
      "loss": 0.0001,
      "step": 3125
    },
    {
      "epoch": 60.01,
      "learning_rate": 2.4342857142857146e-06,
      "loss": 0.0001,
      "step": 3150
    },
    {
      "epoch": 61.0,
      "learning_rate": 2.362857142857143e-06,
      "loss": 0.0001,
      "step": 3175
    },
    {
      "epoch": 61.01,
      "learning_rate": 2.2914285714285718e-06,
      "loss": 0.0001,
      "step": 3200
    },
    {
      "epoch": 62.0,
      "learning_rate": 2.2200000000000003e-06,
      "loss": 0.0001,
      "step": 3225
    },
    {
      "epoch": 62.01,
      "learning_rate": 2.148571428571429e-06,
      "loss": 0.0001,
      "step": 3250
    },
    {
      "epoch": 62.01,
      "learning_rate": 2.077142857142857e-06,
      "loss": 0.0001,
      "step": 3275
    },
    {
      "epoch": 63.01,
      "learning_rate": 2.0057142857142857e-06,
      "loss": 0.0001,
      "step": 3300
    },
    {
      "epoch": 63.01,
      "learning_rate": 1.9342857142857143e-06,
      "loss": 0.0001,
      "step": 3325
    },
    {
      "epoch": 64.01,
      "learning_rate": 1.8628571428571429e-06,
      "loss": 0.0001,
      "step": 3350
    },
    {
      "epoch": 64.01,
      "learning_rate": 1.7914285714285715e-06,
      "loss": 0.0001,
      "step": 3375
    },
    {
      "epoch": 65.0,
      "learning_rate": 1.72e-06,
      "loss": 0.0001,
      "step": 3400
    },
    {
      "epoch": 65.01,
      "learning_rate": 1.6485714285714289e-06,
      "loss": 0.0001,
      "step": 3425
    },
    {
      "epoch": 66.0,
      "learning_rate": 1.5771428571428574e-06,
      "loss": 0.0001,
      "step": 3450
    },
    {
      "epoch": 66.01,
      "learning_rate": 1.5057142857142858e-06,
      "loss": 0.0001,
      "step": 3475
    },
    {
      "epoch": 67.0,
      "learning_rate": 1.4342857142857144e-06,
      "loss": 0.0001,
      "step": 3500
    },
    {
      "epoch": 67.01,
      "learning_rate": 1.362857142857143e-06,
      "loss": 0.0001,
      "step": 3525
    },
    {
      "epoch": 68.0,
      "learning_rate": 1.2914285714285716e-06,
      "loss": 0.0001,
      "step": 3550
    },
    {
      "epoch": 68.01,
      "learning_rate": 1.2200000000000002e-06,
      "loss": 0.0001,
      "step": 3575
    },
    {
      "epoch": 69.0,
      "learning_rate": 1.1485714285714286e-06,
      "loss": 0.0001,
      "step": 3600
    },
    {
      "epoch": 69.01,
      "learning_rate": 1.0771428571428574e-06,
      "loss": 0.0001,
      "step": 3625
    },
    {
      "epoch": 70.0,
      "learning_rate": 1.0057142857142857e-06,
      "loss": 0.0001,
      "step": 3650
    },
    {
      "epoch": 70.01,
      "learning_rate": 9.342857142857144e-07,
      "loss": 0.0001,
      "step": 3675
    },
    {
      "epoch": 71.0,
      "learning_rate": 8.628571428571429e-07,
      "loss": 0.0001,
      "step": 3700
    },
    {
      "epoch": 71.01,
      "learning_rate": 7.914285714285715e-07,
      "loss": 0.0001,
      "step": 3725
    },
    {
      "epoch": 72.0,
      "learning_rate": 7.2e-07,
      "loss": 0.0001,
      "step": 3750
    },
    {
      "epoch": 72.01,
      "learning_rate": 6.485714285714287e-07,
      "loss": 0.0001,
      "step": 3775
    },
    {
      "epoch": 73.0,
      "learning_rate": 5.771428571428572e-07,
      "loss": 0.0001,
      "step": 3800
    },
    {
      "epoch": 73.01,
      "learning_rate": 5.057142857142858e-07,
      "loss": 0.0001,
      "step": 3825
    },
    {
      "epoch": 74.0,
      "learning_rate": 4.342857142857143e-07,
      "loss": 0.0001,
      "step": 3850
    },
    {
      "epoch": 74.01,
      "learning_rate": 3.6285714285714283e-07,
      "loss": 0.0001,
      "step": 3875
    },
    {
      "epoch": 74.01,
      "learning_rate": 2.914285714285715e-07,
      "loss": 0.0001,
      "step": 3900
    },
    {
      "epoch": 75.01,
      "learning_rate": 2.2e-07,
      "loss": 0.0001,
      "step": 3925
    },
    {
      "epoch": 75.01,
      "learning_rate": 1.4857142857142857e-07,
      "loss": 0.0001,
      "step": 3950
    },
    {
      "epoch": 76.01,
      "learning_rate": 7.714285714285715e-08,
      "loss": 0.0001,
      "step": 3975
    },
    {
      "epoch": 76.01,
      "learning_rate": 5.714285714285715e-09,
      "loss": 0.0001,
      "step": 4000
    },
    {
      "epoch": 76.01,
      "eval_loss": 0.3655169606208801,
      "eval_runtime": 1406.1855,
      "eval_samples_per_second": 1.212,
      "eval_steps_per_second": 0.038,
      "eval_wer": 15.859404447794951,
      "step": 4000
    },
    {
      "epoch": 76.01,
      "step": 4000,
      "total_flos": 2.5910300908191744e+20,
      "train_loss": 0.022229147672012915,
      "train_runtime": 69257.2877,
      "train_samples_per_second": 3.696,
      "train_steps_per_second": 0.058
    }
  ],
  "max_steps": 4000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 2.5910300908191744e+20,
  "trial_name": null,
  "trial_params": null
}