{
  "best_metric": 27.6637932833796,
  "best_model_checkpoint": "./checkpoint-1000",
  "epoch": 65.78947368421052,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33,
      "learning_rate": 5e-08,
      "loss": 0.2946,
      "step": 25
    },
    {
      "epoch": 0.66,
      "learning_rate": 1e-07,
      "loss": 0.2857,
      "step": 50
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.5e-07,
      "loss": 0.2658,
      "step": 75
    },
    {
      "epoch": 1.32,
      "learning_rate": 2e-07,
      "loss": 0.2487,
      "step": 100
    },
    {
      "epoch": 1.64,
      "learning_rate": 2.5e-07,
      "loss": 0.2276,
      "step": 125
    },
    {
      "epoch": 1.97,
      "learning_rate": 3e-07,
      "loss": 0.2196,
      "step": 150
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.5e-07,
      "loss": 0.2082,
      "step": 175
    },
    {
      "epoch": 2.63,
      "learning_rate": 4e-07,
      "loss": 0.2007,
      "step": 200
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.5e-07,
      "loss": 0.2016,
      "step": 225
    },
    {
      "epoch": 3.29,
      "learning_rate": 5e-07,
      "loss": 0.195,
      "step": 250
    },
    {
      "epoch": 3.62,
      "learning_rate": 5.5e-07,
      "loss": 0.1884,
      "step": 275
    },
    {
      "epoch": 3.95,
      "learning_rate": 6e-07,
      "loss": 0.1869,
      "step": 300
    },
    {
      "epoch": 4.28,
      "learning_rate": 6.5e-07,
      "loss": 0.1776,
      "step": 325
    },
    {
      "epoch": 4.61,
      "learning_rate": 7e-07,
      "loss": 0.1815,
      "step": 350
    },
    {
      "epoch": 4.93,
      "learning_rate": 7.5e-07,
      "loss": 0.1791,
      "step": 375
    },
    {
      "epoch": 5.26,
      "learning_rate": 8e-07,
      "loss": 0.1765,
      "step": 400
    },
    {
      "epoch": 5.59,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.1707,
      "step": 425
    },
    {
      "epoch": 5.92,
      "learning_rate": 9e-07,
      "loss": 0.1673,
      "step": 450
    },
    {
      "epoch": 6.25,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.1665,
      "step": 475
    },
    {
      "epoch": 6.58,
      "learning_rate": 1e-06,
      "loss": 0.1639,
      "step": 500
    },
    {
      "epoch": 6.91,
      "learning_rate": 9.944444444444444e-07,
      "loss": 0.1611,
      "step": 525
    },
    {
      "epoch": 7.24,
      "learning_rate": 9.88888888888889e-07,
      "loss": 0.1584,
      "step": 550
    },
    {
      "epoch": 7.57,
      "learning_rate": 9.833333333333332e-07,
      "loss": 0.1571,
      "step": 575
    },
    {
      "epoch": 7.89,
      "learning_rate": 9.777777777777778e-07,
      "loss": 0.1544,
      "step": 600
    },
    {
      "epoch": 8.22,
      "learning_rate": 9.722222222222222e-07,
      "loss": 0.1539,
      "step": 625
    },
    {
      "epoch": 8.55,
      "learning_rate": 9.666666666666666e-07,
      "loss": 0.1516,
      "step": 650
    },
    {
      "epoch": 8.88,
      "learning_rate": 9.61111111111111e-07,
      "loss": 0.149,
      "step": 675
    },
    {
      "epoch": 9.21,
      "learning_rate": 9.555555555555556e-07,
      "loss": 0.1451,
      "step": 700
    },
    {
      "epoch": 9.54,
      "learning_rate": 9.499999999999999e-07,
      "loss": 0.1425,
      "step": 725
    },
    {
      "epoch": 9.87,
      "learning_rate": 9.444444444444444e-07,
      "loss": 0.1421,
      "step": 750
    },
    {
      "epoch": 10.2,
      "learning_rate": 9.388888888888888e-07,
      "loss": 0.1416,
      "step": 775
    },
    {
      "epoch": 10.53,
      "learning_rate": 9.333333333333333e-07,
      "loss": 0.1394,
      "step": 800
    },
    {
      "epoch": 10.86,
      "learning_rate": 9.277777777777777e-07,
      "loss": 0.1412,
      "step": 825
    },
    {
      "epoch": 11.18,
      "learning_rate": 9.222222222222222e-07,
      "loss": 0.1359,
      "step": 850
    },
    {
      "epoch": 11.51,
      "learning_rate": 9.166666666666665e-07,
      "loss": 0.1326,
      "step": 875
    },
    {
      "epoch": 11.84,
      "learning_rate": 9.11111111111111e-07,
      "loss": 0.1361,
      "step": 900
    },
    {
      "epoch": 12.17,
      "learning_rate": 9.055555555555556e-07,
      "loss": 0.1328,
      "step": 925
    },
    {
      "epoch": 12.5,
      "learning_rate": 9e-07,
      "loss": 0.1302,
      "step": 950
    },
    {
      "epoch": 12.83,
      "learning_rate": 8.944444444444445e-07,
      "loss": 0.1306,
      "step": 975
    },
    {
      "epoch": 13.16,
      "learning_rate": 8.888888888888888e-07,
      "loss": 0.1251,
      "step": 1000
    },
    {
      "epoch": 13.16,
      "eval_loss": 0.46811801195144653,
      "eval_runtime": 303.5864,
      "eval_samples_per_second": 10.234,
      "eval_steps_per_second": 0.323,
      "eval_wer": 27.6637932833796,
      "step": 1000
    },
    {
      "epoch": 13.49,
      "learning_rate": 8.833333333333333e-07,
      "loss": 0.1253,
      "step": 1025
    },
    {
      "epoch": 13.82,
      "learning_rate": 8.777777777777777e-07,
      "loss": 0.1247,
      "step": 1050
    },
    {
      "epoch": 14.14,
      "learning_rate": 8.722222222222222e-07,
      "loss": 0.1244,
      "step": 1075
    },
    {
      "epoch": 14.47,
      "learning_rate": 8.666666666666667e-07,
      "loss": 0.1217,
      "step": 1100
    },
    {
      "epoch": 14.8,
      "learning_rate": 8.611111111111111e-07,
      "loss": 0.12,
      "step": 1125
    },
    {
      "epoch": 15.13,
      "learning_rate": 8.555555555555555e-07,
      "loss": 0.1184,
      "step": 1150
    },
    {
      "epoch": 15.46,
      "learning_rate": 8.499999999999999e-07,
      "loss": 0.1177,
      "step": 1175
    },
    {
      "epoch": 15.79,
      "learning_rate": 8.444444444444444e-07,
      "loss": 0.1193,
      "step": 1200
    },
    {
      "epoch": 16.12,
      "learning_rate": 8.388888888888888e-07,
      "loss": 0.1131,
      "step": 1225
    },
    {
      "epoch": 16.45,
      "learning_rate": 8.333333333333333e-07,
      "loss": 0.1156,
      "step": 1250
    },
    {
      "epoch": 16.78,
      "learning_rate": 8.277777777777777e-07,
      "loss": 0.1135,
      "step": 1275
    },
    {
      "epoch": 17.11,
      "learning_rate": 8.222222222222221e-07,
      "loss": 0.1136,
      "step": 1300
    },
    {
      "epoch": 17.43,
      "learning_rate": 8.166666666666666e-07,
      "loss": 0.1122,
      "step": 1325
    },
    {
      "epoch": 17.76,
      "learning_rate": 8.11111111111111e-07,
      "loss": 0.1068,
      "step": 1350
    },
    {
      "epoch": 18.09,
      "learning_rate": 8.055555555555556e-07,
      "loss": 0.1069,
      "step": 1375
    },
    {
      "epoch": 18.42,
      "learning_rate": 8e-07,
      "loss": 0.1078,
      "step": 1400
    },
    {
      "epoch": 18.75,
      "learning_rate": 7.944444444444444e-07,
      "loss": 0.1064,
      "step": 1425
    },
    {
      "epoch": 19.08,
      "learning_rate": 7.888888888888889e-07,
      "loss": 0.1034,
      "step": 1450
    },
    {
      "epoch": 19.41,
      "learning_rate": 7.833333333333333e-07,
      "loss": 0.1013,
      "step": 1475
    },
    {
      "epoch": 19.74,
      "learning_rate": 7.777777777777778e-07,
      "loss": 0.1044,
      "step": 1500
    },
    {
      "epoch": 20.07,
      "learning_rate": 7.722222222222222e-07,
      "loss": 0.1025,
      "step": 1525
    },
    {
      "epoch": 20.39,
      "learning_rate": 7.666666666666667e-07,
      "loss": 0.0967,
      "step": 1550
    },
    {
      "epoch": 20.72,
      "learning_rate": 7.61111111111111e-07,
      "loss": 0.0996,
      "step": 1575
    },
    {
      "epoch": 21.05,
      "learning_rate": 7.555555555555555e-07,
      "loss": 0.1016,
      "step": 1600
    },
    {
      "epoch": 21.38,
      "learning_rate": 7.5e-07,
      "loss": 0.0953,
      "step": 1625
    },
    {
      "epoch": 21.71,
      "learning_rate": 7.444444444444444e-07,
      "loss": 0.0959,
      "step": 1650
    },
    {
      "epoch": 22.04,
      "learning_rate": 7.388888888888889e-07,
      "loss": 0.0991,
      "step": 1675
    },
    {
      "epoch": 22.37,
      "learning_rate": 7.333333333333332e-07,
      "loss": 0.0926,
      "step": 1700
    },
    {
      "epoch": 22.7,
      "learning_rate": 7.277777777777777e-07,
      "loss": 0.0948,
      "step": 1725
    },
    {
      "epoch": 23.03,
      "learning_rate": 7.222222222222221e-07,
      "loss": 0.0921,
      "step": 1750
    },
    {
      "epoch": 23.36,
      "learning_rate": 7.166666666666667e-07,
      "loss": 0.0922,
      "step": 1775
    },
    {
      "epoch": 23.68,
      "learning_rate": 7.111111111111111e-07,
      "loss": 0.0885,
      "step": 1800
    },
    {
      "epoch": 24.01,
      "learning_rate": 7.055555555555556e-07,
      "loss": 0.0928,
      "step": 1825
    },
    {
      "epoch": 24.34,
      "learning_rate": 7e-07,
      "loss": 0.0886,
      "step": 1850
    },
    {
      "epoch": 24.67,
      "learning_rate": 6.944444444444444e-07,
      "loss": 0.0888,
      "step": 1875
    },
    {
      "epoch": 25.0,
      "learning_rate": 6.888888888888889e-07,
      "loss": 0.0861,
      "step": 1900
    },
    {
      "epoch": 25.33,
      "learning_rate": 6.833333333333333e-07,
      "loss": 0.0876,
      "step": 1925
    },
    {
      "epoch": 25.66,
      "learning_rate": 6.777777777777778e-07,
      "loss": 0.084,
      "step": 1950
    },
    {
      "epoch": 25.99,
      "learning_rate": 6.722222222222222e-07,
      "loss": 0.0846,
      "step": 1975
    },
    {
      "epoch": 26.32,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.0812,
      "step": 2000
    },
    {
      "epoch": 26.32,
      "eval_loss": 0.5046022534370422,
      "eval_runtime": 304.5123,
      "eval_samples_per_second": 10.203,
      "eval_steps_per_second": 0.322,
      "eval_wer": 28.20646615599758,
      "step": 2000
    },
    {
      "epoch": 26.64,
      "learning_rate": 6.611111111111111e-07,
      "loss": 0.0822,
      "step": 2025
    },
    {
      "epoch": 26.97,
      "learning_rate": 6.555555555555555e-07,
      "loss": 0.0843,
      "step": 2050
    },
    {
      "epoch": 27.3,
      "learning_rate": 6.5e-07,
      "loss": 0.0795,
      "step": 2075
    },
    {
      "epoch": 27.63,
      "learning_rate": 6.444444444444444e-07,
      "loss": 0.079,
      "step": 2100
    },
    {
      "epoch": 27.96,
      "learning_rate": 6.388888888888888e-07,
      "loss": 0.0824,
      "step": 2125
    },
    {
      "epoch": 28.29,
      "learning_rate": 6.333333333333332e-07,
      "loss": 0.0756,
      "step": 2150
    },
    {
      "epoch": 28.62,
      "learning_rate": 6.277777777777777e-07,
      "loss": 0.0781,
      "step": 2175
    },
    {
      "epoch": 28.95,
      "learning_rate": 6.222222222222223e-07,
      "loss": 0.0817,
      "step": 2200
    },
    {
      "epoch": 29.28,
      "learning_rate": 6.166666666666667e-07,
      "loss": 0.0769,
      "step": 2225
    },
    {
      "epoch": 29.61,
      "learning_rate": 6.111111111111112e-07,
      "loss": 0.0742,
      "step": 2250
    },
    {
      "epoch": 29.93,
      "learning_rate": 6.055555555555555e-07,
      "loss": 0.0758,
      "step": 2275
    },
    {
      "epoch": 30.26,
      "learning_rate": 6e-07,
      "loss": 0.0718,
      "step": 2300
    },
    {
      "epoch": 30.59,
      "learning_rate": 5.944444444444444e-07,
      "loss": 0.0738,
      "step": 2325
    },
    {
      "epoch": 30.92,
      "learning_rate": 5.888888888888889e-07,
      "loss": 0.0741,
      "step": 2350
    },
    {
      "epoch": 31.25,
      "learning_rate": 5.833333333333334e-07,
      "loss": 0.072,
      "step": 2375
    },
    {
      "epoch": 31.58,
      "learning_rate": 5.777777777777777e-07,
      "loss": 0.0725,
      "step": 2400
    },
    {
      "epoch": 31.91,
      "learning_rate": 5.722222222222222e-07,
      "loss": 0.0715,
      "step": 2425
    },
    {
      "epoch": 32.24,
      "learning_rate": 5.666666666666666e-07,
      "loss": 0.0675,
      "step": 2450
    },
    {
      "epoch": 32.57,
      "learning_rate": 5.611111111111111e-07,
      "loss": 0.0731,
      "step": 2475
    },
    {
      "epoch": 32.89,
      "learning_rate": 5.555555555555555e-07,
      "loss": 0.0689,
      "step": 2500
    },
    {
      "epoch": 33.22,
      "learning_rate": 5.5e-07,
      "loss": 0.0675,
      "step": 2525
    },
    {
      "epoch": 33.55,
      "learning_rate": 5.444444444444443e-07,
      "loss": 0.0684,
      "step": 2550
    },
    {
      "epoch": 33.88,
      "learning_rate": 5.388888888888888e-07,
      "loss": 0.0684,
      "step": 2575
    },
    {
      "epoch": 34.21,
      "learning_rate": 5.333333333333333e-07,
      "loss": 0.0658,
      "step": 2600
    },
    {
      "epoch": 34.54,
      "learning_rate": 5.277777777777777e-07,
      "loss": 0.0663,
      "step": 2625
    },
    {
      "epoch": 34.87,
      "learning_rate": 5.222222222222223e-07,
      "loss": 0.0653,
      "step": 2650
    },
    {
      "epoch": 35.2,
      "learning_rate": 5.166666666666667e-07,
      "loss": 0.0651,
      "step": 2675
    },
    {
      "epoch": 35.53,
      "learning_rate": 5.111111111111111e-07,
      "loss": 0.0621,
      "step": 2700
    },
    {
      "epoch": 35.86,
      "learning_rate": 5.055555555555555e-07,
      "loss": 0.0639,
      "step": 2725
    },
    {
      "epoch": 36.18,
      "learning_rate": 5e-07,
      "loss": 0.0628,
      "step": 2750
    },
    {
      "epoch": 36.51,
      "learning_rate": 4.944444444444445e-07,
      "loss": 0.0622,
      "step": 2775
    },
    {
      "epoch": 36.84,
      "learning_rate": 4.888888888888889e-07,
      "loss": 0.0625,
      "step": 2800
    },
    {
      "epoch": 37.17,
      "learning_rate": 4.833333333333333e-07,
      "loss": 0.0617,
      "step": 2825
    },
    {
      "epoch": 37.5,
      "learning_rate": 4.777777777777778e-07,
      "loss": 0.0606,
      "step": 2850
    },
    {
      "epoch": 37.83,
      "learning_rate": 4.722222222222222e-07,
      "loss": 0.0618,
      "step": 2875
    },
    {
      "epoch": 38.16,
      "learning_rate": 4.6666666666666666e-07,
      "loss": 0.0599,
      "step": 2900
    },
    {
      "epoch": 38.49,
      "learning_rate": 4.611111111111111e-07,
      "loss": 0.0585,
      "step": 2925
    },
    {
      "epoch": 38.82,
      "learning_rate": 4.555555555555555e-07,
      "loss": 0.0603,
      "step": 2950
    },
    {
      "epoch": 39.14,
      "learning_rate": 4.5e-07,
      "loss": 0.0579,
      "step": 2975
    },
    {
      "epoch": 39.47,
      "learning_rate": 4.444444444444444e-07,
      "loss": 0.0584,
      "step": 3000
    },
    {
      "epoch": 39.47,
      "eval_loss": 0.5393078327178955,
      "eval_runtime": 305.5977,
      "eval_samples_per_second": 10.167,
      "eval_steps_per_second": 0.321,
      "eval_wer": 28.30456471374006,
      "step": 3000
    },
    {
      "epoch": 39.8,
      "learning_rate": 4.3888888888888884e-07,
      "loss": 0.0582,
      "step": 3025
    },
    {
      "epoch": 40.13,
      "learning_rate": 4.3333333333333335e-07,
      "loss": 0.0556,
      "step": 3050
    },
    {
      "epoch": 40.46,
      "learning_rate": 4.2777777777777775e-07,
      "loss": 0.0556,
      "step": 3075
    },
    {
      "epoch": 40.79,
      "learning_rate": 4.222222222222222e-07,
      "loss": 0.0566,
      "step": 3100
    },
    {
      "epoch": 41.12,
      "learning_rate": 4.1666666666666667e-07,
      "loss": 0.0561,
      "step": 3125
    },
    {
      "epoch": 41.45,
      "learning_rate": 4.1111111111111107e-07,
      "loss": 0.0561,
      "step": 3150
    },
    {
      "epoch": 41.78,
      "learning_rate": 4.055555555555555e-07,
      "loss": 0.0548,
      "step": 3175
    },
    {
      "epoch": 42.11,
      "learning_rate": 4e-07,
      "loss": 0.0545,
      "step": 3200
    },
    {
      "epoch": 42.43,
      "learning_rate": 3.9444444444444444e-07,
      "loss": 0.0547,
      "step": 3225
    },
    {
      "epoch": 42.76,
      "learning_rate": 3.888888888888889e-07,
      "loss": 0.0542,
      "step": 3250
    },
    {
      "epoch": 43.09,
      "learning_rate": 3.8333333333333335e-07,
      "loss": 0.0519,
      "step": 3275
    },
    {
      "epoch": 43.42,
      "learning_rate": 3.7777777777777775e-07,
      "loss": 0.0523,
      "step": 3300
    },
    {
      "epoch": 43.75,
      "learning_rate": 3.722222222222222e-07,
      "loss": 0.053,
      "step": 3325
    },
    {
      "epoch": 44.08,
      "learning_rate": 3.666666666666666e-07,
      "loss": 0.052,
      "step": 3350
    },
    {
      "epoch": 44.41,
      "learning_rate": 3.6111111111111107e-07,
      "loss": 0.0525,
      "step": 3375
    },
    {
      "epoch": 44.74,
      "learning_rate": 3.5555555555555553e-07,
      "loss": 0.0525,
      "step": 3400
    },
    {
      "epoch": 45.07,
      "learning_rate": 3.5e-07,
      "loss": 0.0496,
      "step": 3425
    },
    {
      "epoch": 45.39,
      "learning_rate": 3.4444444444444444e-07,
      "loss": 0.0495,
      "step": 3450
    },
    {
      "epoch": 45.72,
      "learning_rate": 3.388888888888889e-07,
      "loss": 0.0511,
      "step": 3475
    },
    {
      "epoch": 46.05,
      "learning_rate": 3.333333333333333e-07,
      "loss": 0.0505,
      "step": 3500
    },
    {
      "epoch": 46.38,
      "learning_rate": 3.2777777777777776e-07,
      "loss": 0.0491,
      "step": 3525
    },
    {
      "epoch": 46.71,
      "learning_rate": 3.222222222222222e-07,
      "loss": 0.0504,
      "step": 3550
    },
    {
      "epoch": 47.04,
      "learning_rate": 3.166666666666666e-07,
      "loss": 0.0487,
      "step": 3575
    },
    {
      "epoch": 47.37,
      "learning_rate": 3.111111111111111e-07,
      "loss": 0.0474,
      "step": 3600
    },
    {
      "epoch": 47.7,
      "learning_rate": 3.055555555555556e-07,
      "loss": 0.0485,
      "step": 3625
    },
    {
      "epoch": 48.03,
      "learning_rate": 3e-07,
      "loss": 0.0484,
      "step": 3650
    },
    {
      "epoch": 48.36,
      "learning_rate": 2.9444444444444444e-07,
      "loss": 0.0465,
      "step": 3675
    },
    {
      "epoch": 48.68,
      "learning_rate": 2.8888888888888885e-07,
      "loss": 0.0463,
      "step": 3700
    },
    {
      "epoch": 49.01,
      "learning_rate": 2.833333333333333e-07,
      "loss": 0.0503,
      "step": 3725
    },
    {
      "epoch": 49.34,
      "learning_rate": 2.7777777777777776e-07,
      "loss": 0.0484,
      "step": 3750
    },
    {
      "epoch": 49.67,
      "learning_rate": 2.7222222222222216e-07,
      "loss": 0.0451,
      "step": 3775
    },
    {
      "epoch": 50.0,
      "learning_rate": 2.6666666666666667e-07,
      "loss": 0.0465,
      "step": 3800
    },
    {
      "epoch": 50.33,
      "learning_rate": 2.6111111111111113e-07,
      "loss": 0.0467,
      "step": 3825
    },
    {
      "epoch": 50.66,
      "learning_rate": 2.5555555555555553e-07,
      "loss": 0.0452,
      "step": 3850
    },
    {
      "epoch": 50.99,
      "learning_rate": 2.5e-07,
      "loss": 0.0453,
      "step": 3875
    },
    {
      "epoch": 51.32,
      "learning_rate": 2.4444444444444445e-07,
      "loss": 0.0468,
      "step": 3900
    },
    {
      "epoch": 51.64,
      "learning_rate": 2.388888888888889e-07,
      "loss": 0.0453,
      "step": 3925
    },
    {
      "epoch": 51.97,
      "learning_rate": 2.3333333333333333e-07,
      "loss": 0.0449,
      "step": 3950
    },
    {
      "epoch": 52.3,
      "learning_rate": 2.2777777777777776e-07,
      "loss": 0.0427,
      "step": 3975
    },
    {
      "epoch": 52.63,
      "learning_rate": 2.222222222222222e-07,
      "loss": 0.0441,
      "step": 4000
    },
    {
      "epoch": 52.63,
      "eval_loss": 0.5639278888702393,
      "eval_runtime": 305.7525,
      "eval_samples_per_second": 10.162,
      "eval_steps_per_second": 0.321,
      "eval_wer": 28.49241301580013,
      "step": 4000
    },
    {
      "epoch": 52.96,
      "learning_rate": 2.1666666666666667e-07,
      "loss": 0.0465,
      "step": 4025
    },
    {
      "epoch": 53.29,
      "learning_rate": 2.111111111111111e-07,
      "loss": 0.0431,
      "step": 4050
    },
    {
      "epoch": 53.62,
      "learning_rate": 2.0555555555555553e-07,
      "loss": 0.0445,
      "step": 4075
    },
    {
      "epoch": 53.95,
      "learning_rate": 2e-07,
      "loss": 0.0436,
      "step": 4100
    },
    {
      "epoch": 54.28,
      "learning_rate": 1.9444444444444445e-07,
      "loss": 0.0438,
      "step": 4125
    },
    {
      "epoch": 54.61,
      "learning_rate": 1.8888888888888888e-07,
      "loss": 0.0437,
      "step": 4150
    },
    {
      "epoch": 54.93,
      "learning_rate": 1.833333333333333e-07,
      "loss": 0.0421,
      "step": 4175
    },
    {
      "epoch": 55.26,
      "learning_rate": 1.7777777777777776e-07,
      "loss": 0.0442,
      "step": 4200
    },
    {
      "epoch": 55.59,
      "learning_rate": 1.7222222222222222e-07,
      "loss": 0.0418,
      "step": 4225
    },
    {
      "epoch": 55.92,
      "learning_rate": 1.6666666666666665e-07,
      "loss": 0.043,
      "step": 4250
    },
    {
      "epoch": 56.25,
      "learning_rate": 1.611111111111111e-07,
      "loss": 0.0419,
      "step": 4275
    },
    {
      "epoch": 56.58,
      "learning_rate": 1.5555555555555556e-07,
      "loss": 0.0427,
      "step": 4300
    },
    {
      "epoch": 56.91,
      "learning_rate": 1.5e-07,
      "loss": 0.0423,
      "step": 4325
    },
    {
      "epoch": 57.24,
      "learning_rate": 1.4444444444444442e-07,
      "loss": 0.0416,
      "step": 4350
    },
    {
      "epoch": 57.57,
      "learning_rate": 1.3888888888888888e-07,
      "loss": 0.043,
      "step": 4375
    },
    {
      "epoch": 57.89,
      "learning_rate": 1.3333333333333334e-07,
      "loss": 0.0419,
      "step": 4400
    },
    {
      "epoch": 58.22,
      "learning_rate": 1.2777777777777777e-07,
      "loss": 0.0418,
      "step": 4425
    },
    {
      "epoch": 58.55,
      "learning_rate": 1.2222222222222222e-07,
      "loss": 0.0409,
      "step": 4450
    },
    {
      "epoch": 58.88,
      "learning_rate": 1.1666666666666667e-07,
      "loss": 0.042,
      "step": 4475
    },
    {
      "epoch": 59.21,
      "learning_rate": 1.111111111111111e-07,
      "loss": 0.041,
      "step": 4500
    },
    {
      "epoch": 59.54,
      "learning_rate": 1.0555555555555555e-07,
      "loss": 0.0411,
      "step": 4525
    },
    {
      "epoch": 59.87,
      "learning_rate": 1e-07,
      "loss": 0.0415,
      "step": 4550
    },
    {
      "epoch": 60.2,
      "learning_rate": 9.444444444444444e-08,
      "loss": 0.0414,
      "step": 4575
    },
    {
      "epoch": 60.53,
      "learning_rate": 8.888888888888888e-08,
      "loss": 0.0406,
      "step": 4600
    },
    {
      "epoch": 60.86,
      "learning_rate": 8.333333333333333e-08,
      "loss": 0.0409,
      "step": 4625
    },
    {
      "epoch": 61.18,
      "learning_rate": 7.777777777777778e-08,
      "loss": 0.0394,
      "step": 4650
    },
    {
      "epoch": 61.51,
      "learning_rate": 7.222222222222221e-08,
      "loss": 0.0422,
      "step": 4675
    },
    {
      "epoch": 61.84,
      "learning_rate": 6.666666666666667e-08,
      "loss": 0.0405,
      "step": 4700
    },
    {
      "epoch": 62.17,
      "learning_rate": 6.111111111111111e-08,
      "loss": 0.0405,
      "step": 4725
    },
    {
      "epoch": 62.5,
      "learning_rate": 5.555555555555555e-08,
      "loss": 0.0402,
      "step": 4750
    },
    {
      "epoch": 62.83,
      "learning_rate": 5e-08,
      "loss": 0.041,
      "step": 4775
    },
    {
      "epoch": 63.16,
      "learning_rate": 4.444444444444444e-08,
      "loss": 0.0387,
      "step": 4800
    },
    {
      "epoch": 63.49,
      "learning_rate": 3.888888888888889e-08,
      "loss": 0.0405,
      "step": 4825
    },
    {
      "epoch": 63.82,
      "learning_rate": 3.3333333333333334e-08,
      "loss": 0.0393,
      "step": 4850
    },
    {
      "epoch": 64.14,
      "learning_rate": 2.7777777777777774e-08,
      "loss": 0.0419,
      "step": 4875
    },
    {
      "epoch": 64.47,
      "learning_rate": 2.222222222222222e-08,
      "loss": 0.0408,
      "step": 4900
    },
    {
      "epoch": 64.8,
      "learning_rate": 1.6666666666666667e-08,
      "loss": 0.0402,
      "step": 4925
    },
    {
      "epoch": 65.13,
      "learning_rate": 1.111111111111111e-08,
      "loss": 0.0393,
      "step": 4950
    },
    {
      "epoch": 65.46,
      "learning_rate": 5.555555555555555e-09,
      "loss": 0.0414,
      "step": 4975
    },
    {
      "epoch": 65.79,
      "learning_rate": 0.0,
      "loss": 0.0392,
      "step": 5000
    },
    {
      "epoch": 65.79,
      "eval_loss": 0.57337486743927,
      "eval_runtime": 305.9735,
      "eval_samples_per_second": 10.154,
      "eval_steps_per_second": 0.32,
      "eval_wer": 28.586337166830166,
      "step": 5000
    },
    {
      "epoch": 65.79,
      "step": 5000,
      "total_flos": 2.06793125756928e+19,
      "train_loss": 0.08767790142297745,
      "train_runtime": 14504.947,
      "train_samples_per_second": 22.061,
      "train_steps_per_second": 0.345
    }
  ],
  "logging_steps": 25,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 66,
  "save_steps": 1000,
  "total_flos": 2.06793125756928e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}