{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "global_step": 20000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 3.5624999999999998e-06,
      "loss": 137.0348,
      "step": 100
    },
    {
      "epoch": 1.0,
      "learning_rate": 7.275e-06,
      "loss": 106.075,
      "step": 200
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.1024999999999999e-05,
      "loss": 95.1437,
      "step": 300
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.4775e-05,
      "loss": 88.0386,
      "step": 400
    },
    {
      "epoch": 2.5,
      "learning_rate": 1.8525e-05,
      "loss": 82.628,
      "step": 500
    },
    {
      "epoch": 2.5,
      "eval_loss": 79.55873107910156,
      "eval_runtime": 24.8027,
      "eval_samples_per_second": 19.877,
      "eval_steps_per_second": 9.959,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.2274999999999996e-05,
      "loss": 73.1146,
      "step": 600
    },
    {
      "epoch": 3.5,
      "learning_rate": 2.6024999999999996e-05,
      "loss": 62.7571,
      "step": 700
    },
    {
      "epoch": 4.0,
      "learning_rate": 2.9775e-05,
      "loss": 48.9782,
      "step": 800
    },
    {
      "epoch": 4.5,
      "learning_rate": 3.3524999999999995e-05,
      "loss": 33.2033,
      "step": 900
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.7275e-05,
      "loss": 17.5635,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_loss": 11.592885971069336,
      "eval_runtime": 24.3352,
      "eval_samples_per_second": 20.259,
      "eval_steps_per_second": 10.15,
      "eval_wer": 1.0,
      "step": 1000
    },
    {
      "epoch": 5.5,
      "learning_rate": 4.1025e-05,
      "loss": 8.6246,
      "step": 1100
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4775e-05,
      "loss": 6.5245,
      "step": 1200
    },
    {
      "epoch": 6.5,
      "learning_rate": 4.8525e-05,
      "loss": 6.4322,
      "step": 1300
    },
    {
      "epoch": 7.0,
      "learning_rate": 5.227499999999999e-05,
      "loss": 6.4273,
      "step": 1400
    },
    {
      "epoch": 7.5,
      "learning_rate": 5.602499999999999e-05,
      "loss": 6.4288,
      "step": 1500
    },
    {
      "epoch": 7.5,
      "eval_loss": 6.447471618652344,
      "eval_runtime": 24.3824,
      "eval_samples_per_second": 20.219,
      "eval_steps_per_second": 10.13,
      "eval_wer": 1.0,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "learning_rate": 5.9774999999999996e-05,
      "loss": 6.4087,
      "step": 1600
    },
    {
      "epoch": 8.5,
      "learning_rate": 6.3525e-05,
      "loss": 6.4099,
      "step": 1700
    },
    {
      "epoch": 9.0,
      "learning_rate": 6.7275e-05,
      "loss": 6.4216,
      "step": 1800
    },
    {
      "epoch": 9.5,
      "learning_rate": 7.102499999999999e-05,
      "loss": 6.4159,
      "step": 1900
    },
    {
      "epoch": 10.0,
      "learning_rate": 7.477499999999999e-05,
      "loss": 6.4092,
      "step": 2000
    },
    {
      "epoch": 10.0,
      "eval_loss": 6.457908630371094,
      "eval_runtime": 24.1035,
      "eval_samples_per_second": 20.453,
      "eval_steps_per_second": 10.247,
      "eval_wer": 1.0,
      "step": 2000
    },
    {
      "epoch": 10.5,
      "learning_rate": 7.460833333333332e-05,
      "loss": 6.4008,
      "step": 2100
    },
    {
      "epoch": 11.0,
      "learning_rate": 7.419166666666666e-05,
      "loss": 6.4257,
      "step": 2200
    },
    {
      "epoch": 11.5,
      "learning_rate": 7.377499999999999e-05,
      "loss": 6.4109,
      "step": 2300
    },
    {
      "epoch": 12.0,
      "learning_rate": 7.335833333333333e-05,
      "loss": 6.4135,
      "step": 2400
    },
    {
      "epoch": 12.5,
      "learning_rate": 7.294166666666666e-05,
      "loss": 6.3982,
      "step": 2500
    },
    {
      "epoch": 12.5,
      "eval_loss": 6.466182708740234,
      "eval_runtime": 24.3477,
      "eval_samples_per_second": 20.248,
      "eval_steps_per_second": 10.145,
      "eval_wer": 1.0,
      "step": 2500
    },
    {
      "epoch": 13.0,
      "learning_rate": 7.252499999999999e-05,
      "loss": 6.4234,
      "step": 2600
    },
    {
      "epoch": 13.5,
      "learning_rate": 7.210833333333333e-05,
      "loss": 6.4333,
      "step": 2700
    },
    {
      "epoch": 14.0,
      "learning_rate": 7.169166666666666e-05,
      "loss": 6.3853,
      "step": 2800
    },
    {
      "epoch": 14.5,
      "learning_rate": 7.1275e-05,
      "loss": 6.4267,
      "step": 2900
    },
    {
      "epoch": 15.0,
      "learning_rate": 7.085833333333333e-05,
      "loss": 6.391,
      "step": 3000
    },
    {
      "epoch": 15.0,
      "eval_loss": 6.465456008911133,
      "eval_runtime": 24.1868,
      "eval_samples_per_second": 20.383,
      "eval_steps_per_second": 10.212,
      "eval_wer": 1.0,
      "step": 3000
    },
    {
      "epoch": 15.5,
      "learning_rate": 7.044166666666665e-05,
      "loss": 6.4101,
      "step": 3100
    },
    {
      "epoch": 16.0,
      "learning_rate": 7.0025e-05,
      "loss": 6.4082,
      "step": 3200
    },
    {
      "epoch": 16.5,
      "learning_rate": 6.960833333333332e-05,
      "loss": 6.4155,
      "step": 3300
    },
    {
      "epoch": 17.0,
      "learning_rate": 6.919166666666666e-05,
      "loss": 6.4019,
      "step": 3400
    },
    {
      "epoch": 17.5,
      "learning_rate": 6.877499999999999e-05,
      "loss": 6.4097,
      "step": 3500
    },
    {
      "epoch": 17.5,
      "eval_loss": 6.469109535217285,
      "eval_runtime": 24.7207,
      "eval_samples_per_second": 19.943,
      "eval_steps_per_second": 9.992,
      "eval_wer": 1.0,
      "step": 3500
    },
    {
      "epoch": 18.0,
      "learning_rate": 6.835833333333332e-05,
      "loss": 6.4073,
      "step": 3600
    },
    {
      "epoch": 18.5,
      "learning_rate": 6.794166666666666e-05,
      "loss": 6.4142,
      "step": 3700
    },
    {
      "epoch": 19.0,
      "learning_rate": 6.752499999999999e-05,
      "loss": 6.4013,
      "step": 3800
    },
    {
      "epoch": 19.5,
      "learning_rate": 6.710833333333333e-05,
      "loss": 6.4181,
      "step": 3900
    },
    {
      "epoch": 20.0,
      "learning_rate": 6.669166666666666e-05,
      "loss": 6.3986,
      "step": 4000
    },
    {
      "epoch": 20.0,
      "eval_loss": 6.470158100128174,
      "eval_runtime": 24.0989,
      "eval_samples_per_second": 20.457,
      "eval_steps_per_second": 10.249,
      "eval_wer": 1.0,
      "step": 4000
    },
    {
      "epoch": 20.5,
      "learning_rate": 6.6275e-05,
      "loss": 6.4044,
      "step": 4100
    },
    {
      "epoch": 21.0,
      "learning_rate": 6.585833333333333e-05,
      "loss": 6.4134,
      "step": 4200
    },
    {
      "epoch": 21.5,
      "learning_rate": 6.544166666666665e-05,
      "loss": 6.4066,
      "step": 4300
    },
    {
      "epoch": 22.0,
      "learning_rate": 6.5025e-05,
      "loss": 6.4098,
      "step": 4400
    },
    {
      "epoch": 22.5,
      "learning_rate": 6.460833333333332e-05,
      "loss": 6.4069,
      "step": 4500
    },
    {
      "epoch": 22.5,
      "eval_loss": 6.47609806060791,
      "eval_runtime": 24.5428,
      "eval_samples_per_second": 20.087,
      "eval_steps_per_second": 10.064,
      "eval_wer": 1.0,
      "step": 4500
    },
    {
      "epoch": 23.0,
      "learning_rate": 6.419166666666666e-05,
      "loss": 6.4075,
      "step": 4600
    },
    {
      "epoch": 23.5,
      "learning_rate": 6.377499999999999e-05,
      "loss": 6.4069,
      "step": 4700
    },
    {
      "epoch": 24.0,
      "learning_rate": 6.335833333333332e-05,
      "loss": 6.407,
      "step": 4800
    },
    {
      "epoch": 24.5,
      "learning_rate": 6.294166666666666e-05,
      "loss": 6.3966,
      "step": 4900
    },
    {
      "epoch": 25.0,
      "learning_rate": 6.252499999999999e-05,
      "loss": 6.4158,
      "step": 5000
    },
    {
      "epoch": 25.0,
      "eval_loss": 6.47500467300415,
      "eval_runtime": 24.1455,
      "eval_samples_per_second": 20.418,
      "eval_steps_per_second": 10.23,
      "eval_wer": 1.0,
      "step": 5000
    },
    {
      "epoch": 25.5,
      "learning_rate": 6.210833333333333e-05,
      "loss": 6.4081,
      "step": 5100
    },
    {
      "epoch": 26.0,
      "learning_rate": 6.169166666666667e-05,
      "loss": 6.4024,
      "step": 5200
    },
    {
      "epoch": 26.5,
      "learning_rate": 6.127499999999999e-05,
      "loss": 6.4122,
      "step": 5300
    },
    {
      "epoch": 27.0,
      "learning_rate": 6.0858333333333326e-05,
      "loss": 6.4011,
      "step": 5400
    },
    {
      "epoch": 27.5,
      "learning_rate": 6.044166666666666e-05,
      "loss": 6.4117,
      "step": 5500
    },
    {
      "epoch": 27.5,
      "eval_loss": 6.48163366317749,
      "eval_runtime": 24.2083,
      "eval_samples_per_second": 20.365,
      "eval_steps_per_second": 10.203,
      "eval_wer": 1.0,
      "step": 5500
    },
    {
      "epoch": 28.0,
      "learning_rate": 6.0024999999999995e-05,
      "loss": 6.4013,
      "step": 5600
    },
    {
      "epoch": 28.5,
      "learning_rate": 5.960833333333333e-05,
      "loss": 6.4158,
      "step": 5700
    },
    {
      "epoch": 29.0,
      "learning_rate": 5.9191666666666664e-05,
      "loss": 6.3951,
      "step": 5800
    },
    {
      "epoch": 29.5,
      "learning_rate": 5.877499999999999e-05,
      "loss": 6.4047,
      "step": 5900
    },
    {
      "epoch": 30.0,
      "learning_rate": 5.8358333333333327e-05,
      "loss": 6.4086,
      "step": 6000
    },
    {
      "epoch": 30.0,
      "eval_loss": 6.4805707931518555,
      "eval_runtime": 24.395,
      "eval_samples_per_second": 20.209,
      "eval_steps_per_second": 10.125,
      "eval_wer": 1.0,
      "step": 6000
    },
    {
      "epoch": 30.5,
      "learning_rate": 5.794166666666666e-05,
      "loss": 6.3924,
      "step": 6100
    },
    {
      "epoch": 31.0,
      "learning_rate": 5.7524999999999996e-05,
      "loss": 6.4216,
      "step": 6200
    },
    {
      "epoch": 31.5,
      "learning_rate": 5.710833333333333e-05,
      "loss": 6.3965,
      "step": 6300
    },
    {
      "epoch": 32.0,
      "learning_rate": 5.6691666666666665e-05,
      "loss": 6.4165,
      "step": 6400
    },
    {
      "epoch": 32.5,
      "learning_rate": 5.627499999999999e-05,
      "loss": 6.3992,
      "step": 6500
    },
    {
      "epoch": 32.5,
      "eval_loss": 6.487205505371094,
      "eval_runtime": 24.2919,
      "eval_samples_per_second": 20.295,
      "eval_steps_per_second": 10.168,
      "eval_wer": 1.0,
      "step": 6500
    },
    {
      "epoch": 33.0,
      "learning_rate": 5.585833333333333e-05,
      "loss": 6.414,
      "step": 6600
    },
    {
      "epoch": 33.5,
      "learning_rate": 5.544166666666666e-05,
      "loss": 6.4138,
      "step": 6700
    },
    {
      "epoch": 34.0,
      "learning_rate": 5.5024999999999996e-05,
      "loss": 6.3973,
      "step": 6800
    },
    {
      "epoch": 34.5,
      "learning_rate": 5.460833333333333e-05,
      "loss": 6.4154,
      "step": 6900
    },
    {
      "epoch": 35.0,
      "learning_rate": 5.419166666666666e-05,
      "loss": 6.3946,
      "step": 7000
    },
    {
      "epoch": 35.0,
      "eval_loss": 6.486586570739746,
      "eval_runtime": 24.3253,
      "eval_samples_per_second": 20.267,
      "eval_steps_per_second": 10.154,
      "eval_wer": 1.0,
      "step": 7000
    },
    {
      "epoch": 35.5,
      "learning_rate": 5.377499999999999e-05,
      "loss": 6.4203,
      "step": 7100
    },
    {
      "epoch": 36.0,
      "learning_rate": 5.335833333333333e-05,
      "loss": 6.3905,
      "step": 7200
    },
    {
      "epoch": 36.5,
      "learning_rate": 5.294166666666666e-05,
      "loss": 6.4104,
      "step": 7300
    },
    {
      "epoch": 37.0,
      "learning_rate": 5.2524999999999996e-05,
      "loss": 6.4002,
      "step": 7400
    },
    {
      "epoch": 37.5,
      "learning_rate": 5.210833333333333e-05,
      "loss": 6.4212,
      "step": 7500
    },
    {
      "epoch": 37.5,
      "eval_loss": 6.489477634429932,
      "eval_runtime": 24.1492,
      "eval_samples_per_second": 20.415,
      "eval_steps_per_second": 10.228,
      "eval_wer": 1.0,
      "step": 7500
    },
    {
      "epoch": 38.0,
      "learning_rate": 5.169166666666666e-05,
      "loss": 6.3871,
      "step": 7600
    },
    {
      "epoch": 38.5,
      "learning_rate": 5.127499999999999e-05,
      "loss": 6.3982,
      "step": 7700
    },
    {
      "epoch": 39.0,
      "learning_rate": 5.085833333333333e-05,
      "loss": 6.4097,
      "step": 7800
    },
    {
      "epoch": 39.5,
      "learning_rate": 5.044166666666666e-05,
      "loss": 6.4015,
      "step": 7900
    },
    {
      "epoch": 40.0,
      "learning_rate": 5.0024999999999996e-05,
      "loss": 6.4051,
      "step": 8000
    },
    {
      "epoch": 40.0,
      "eval_loss": 6.492602348327637,
      "eval_runtime": 24.1509,
      "eval_samples_per_second": 20.413,
      "eval_steps_per_second": 10.227,
      "eval_wer": 1.0,
      "step": 8000
    },
    {
      "epoch": 40.5,
      "learning_rate": 4.9608333333333324e-05,
      "loss": 6.3881,
      "step": 8100
    },
    {
      "epoch": 41.0,
      "learning_rate": 4.919166666666666e-05,
      "loss": 6.4196,
      "step": 8200
    },
    {
      "epoch": 41.5,
      "learning_rate": 4.877499999999999e-05,
      "loss": 6.4132,
      "step": 8300
    },
    {
      "epoch": 42.0,
      "learning_rate": 4.835833333333333e-05,
      "loss": 6.3967,
      "step": 8400
    },
    {
      "epoch": 42.5,
      "learning_rate": 4.794166666666666e-05,
      "loss": 6.398,
      "step": 8500
    },
    {
      "epoch": 42.5,
      "eval_loss": 6.50146484375,
      "eval_runtime": 24.2251,
      "eval_samples_per_second": 20.351,
      "eval_steps_per_second": 10.196,
      "eval_wer": 1.0,
      "step": 8500
    },
    {
      "epoch": 43.0,
      "learning_rate": 4.7524999999999996e-05,
      "loss": 6.4101,
      "step": 8600
    },
    {
      "epoch": 43.5,
      "learning_rate": 4.7108333333333324e-05,
      "loss": 6.3947,
      "step": 8700
    },
    {
      "epoch": 44.0,
      "learning_rate": 4.669166666666666e-05,
      "loss": 6.4115,
      "step": 8800
    },
    {
      "epoch": 44.5,
      "learning_rate": 4.627499999999999e-05,
      "loss": 6.4088,
      "step": 8900
    },
    {
      "epoch": 45.0,
      "learning_rate": 4.585833333333333e-05,
      "loss": 6.3967,
      "step": 9000
    },
    {
      "epoch": 45.0,
      "eval_loss": 6.496038913726807,
      "eval_runtime": 24.4306,
      "eval_samples_per_second": 20.18,
      "eval_steps_per_second": 10.11,
      "eval_wer": 1.0,
      "step": 9000
    },
    {
      "epoch": 45.5,
      "learning_rate": 4.544166666666666e-05,
      "loss": 6.401,
      "step": 9100
    },
    {
      "epoch": 46.0,
      "learning_rate": 4.5025e-05,
      "loss": 6.4063,
      "step": 9200
    },
    {
      "epoch": 46.5,
      "learning_rate": 4.4608333333333324e-05,
      "loss": 6.4101,
      "step": 9300
    },
    {
      "epoch": 47.0,
      "learning_rate": 4.419166666666666e-05,
      "loss": 6.3957,
      "step": 9400
    },
    {
      "epoch": 47.5,
      "learning_rate": 4.377499999999999e-05,
      "loss": 6.4096,
      "step": 9500
    },
    {
      "epoch": 47.5,
      "eval_loss": 6.500304222106934,
      "eval_runtime": 24.1983,
      "eval_samples_per_second": 20.373,
      "eval_steps_per_second": 10.207,
      "eval_wer": 1.0,
      "step": 9500
    },
    {
      "epoch": 48.0,
      "learning_rate": 4.335833333333333e-05,
      "loss": 6.3966,
      "step": 9600
    },
    {
      "epoch": 48.5,
      "learning_rate": 4.294166666666667e-05,
      "loss": 6.409,
      "step": 9700
    },
    {
      "epoch": 49.0,
      "learning_rate": 4.252499999999999e-05,
      "loss": 6.3979,
      "step": 9800
    },
    {
      "epoch": 49.5,
      "learning_rate": 4.2108333333333325e-05,
      "loss": 6.3985,
      "step": 9900
    },
    {
      "epoch": 50.0,
      "learning_rate": 4.169166666666666e-05,
      "loss": 6.4068,
      "step": 10000
    },
    {
      "epoch": 50.0,
      "eval_loss": 6.502621650695801,
      "eval_runtime": 24.1622,
      "eval_samples_per_second": 20.404,
      "eval_steps_per_second": 10.223,
      "eval_wer": 1.0,
      "step": 10000
    },
    {
      "epoch": 50.5,
      "learning_rate": 4.1274999999999994e-05,
      "loss": 6.4059,
      "step": 10100
    },
    {
      "epoch": 51.0,
      "learning_rate": 4.0858333333333335e-05,
      "loss": 6.4012,
      "step": 10200
    },
    {
      "epoch": 51.5,
      "learning_rate": 4.044166666666667e-05,
      "loss": 6.4118,
      "step": 10300
    },
    {
      "epoch": 52.0,
      "learning_rate": 4.002499999999999e-05,
      "loss": 6.3924,
      "step": 10400
    },
    {
      "epoch": 52.5,
      "learning_rate": 3.9608333333333325e-05,
      "loss": 6.4062,
      "step": 10500
    },
    {
      "epoch": 52.5,
      "eval_loss": 6.507063865661621,
      "eval_runtime": 24.2765,
      "eval_samples_per_second": 20.308,
      "eval_steps_per_second": 10.174,
      "eval_wer": 1.0,
      "step": 10500
    },
    {
      "epoch": 53.0,
      "learning_rate": 3.9191666666666666e-05,
      "loss": 6.3992,
      "step": 10600
    },
    {
      "epoch": 53.5,
      "learning_rate": 3.8775e-05,
      "loss": 6.4136,
      "step": 10700
    },
    {
      "epoch": 54.0,
      "learning_rate": 3.8358333333333335e-05,
      "loss": 6.3915,
      "step": 10800
    },
    {
      "epoch": 54.5,
      "learning_rate": 3.794166666666667e-05,
      "loss": 6.4096,
      "step": 10900
    },
    {
      "epoch": 55.0,
      "learning_rate": 3.7525e-05,
      "loss": 6.395,
      "step": 11000
    },
    {
      "epoch": 55.0,
      "eval_loss": 6.506567001342773,
      "eval_runtime": 24.2863,
      "eval_samples_per_second": 20.3,
      "eval_steps_per_second": 10.17,
      "eval_wer": 1.0,
      "step": 11000
    },
    {
      "epoch": 55.5,
      "learning_rate": 3.710833333333333e-05,
      "loss": 6.4042,
      "step": 11100
    },
    {
      "epoch": 56.0,
      "learning_rate": 3.6691666666666666e-05,
      "loss": 6.402,
      "step": 11200
    },
    {
      "epoch": 56.5,
      "learning_rate": 3.6275e-05,
      "loss": 6.4056,
      "step": 11300
    },
    {
      "epoch": 57.0,
      "learning_rate": 3.585833333333333e-05,
      "loss": 6.4005,
      "step": 11400
    },
    {
      "epoch": 57.5,
      "learning_rate": 3.544166666666666e-05,
      "loss": 6.4079,
      "step": 11500
    },
    {
      "epoch": 57.5,
      "eval_loss": 6.50927734375,
      "eval_runtime": 24.3197,
      "eval_samples_per_second": 20.272,
      "eval_steps_per_second": 10.156,
      "eval_wer": 1.0,
      "step": 11500
    },
    {
      "epoch": 58.0,
      "learning_rate": 3.5025e-05,
      "loss": 6.3957,
      "step": 11600
    },
    {
      "epoch": 58.5,
      "learning_rate": 3.460833333333333e-05,
      "loss": 6.4057,
      "step": 11700
    },
    {
      "epoch": 59.0,
      "learning_rate": 3.4191666666666667e-05,
      "loss": 6.3981,
      "step": 11800
    },
    {
      "epoch": 59.5,
      "learning_rate": 3.3774999999999994e-05,
      "loss": 6.394,
      "step": 11900
    },
    {
      "epoch": 60.0,
      "learning_rate": 3.335833333333333e-05,
      "loss": 6.411,
      "step": 12000
    },
    {
      "epoch": 60.0,
      "eval_loss": 6.510580539703369,
      "eval_runtime": 24.2221,
      "eval_samples_per_second": 20.353,
      "eval_steps_per_second": 10.197,
      "eval_wer": 1.0,
      "step": 12000
    },
    {
      "epoch": 60.5,
      "learning_rate": 3.294166666666666e-05,
      "loss": 6.3902,
      "step": 12100
    },
    {
      "epoch": 61.0,
      "learning_rate": 3.2525e-05,
      "loss": 6.413,
      "step": 12200
    },
    {
      "epoch": 61.5,
      "learning_rate": 3.210833333333333e-05,
      "loss": 6.3998,
      "step": 12300
    },
    {
      "epoch": 62.0,
      "learning_rate": 3.169166666666667e-05,
      "loss": 6.404,
      "step": 12400
    },
    {
      "epoch": 62.5,
      "learning_rate": 3.1274999999999994e-05,
      "loss": 6.4023,
      "step": 12500
    },
    {
      "epoch": 62.5,
      "eval_loss": 6.511165142059326,
      "eval_runtime": 24.2209,
      "eval_samples_per_second": 20.354,
      "eval_steps_per_second": 10.198,
      "eval_wer": 1.0,
      "step": 12500
    },
    {
      "epoch": 63.0,
      "learning_rate": 3.085833333333333e-05,
      "loss": 6.3992,
      "step": 12600
    },
    {
      "epoch": 63.5,
      "learning_rate": 3.0441666666666667e-05,
      "loss": 6.4093,
      "step": 12700
    },
    {
      "epoch": 64.0,
      "learning_rate": 3.0024999999999995e-05,
      "loss": 6.3919,
      "step": 12800
    },
    {
      "epoch": 64.5,
      "learning_rate": 2.9608333333333332e-05,
      "loss": 6.3972,
      "step": 12900
    },
    {
      "epoch": 65.0,
      "learning_rate": 2.9191666666666667e-05,
      "loss": 6.4053,
      "step": 13000
    },
    {
      "epoch": 65.0,
      "eval_loss": 6.51427698135376,
      "eval_runtime": 24.1929,
      "eval_samples_per_second": 20.378,
      "eval_steps_per_second": 10.21,
      "eval_wer": 1.0,
      "step": 13000
    },
    {
      "epoch": 65.5,
      "learning_rate": 2.8774999999999998e-05,
      "loss": 6.3993,
      "step": 13100
    },
    {
      "epoch": 66.0,
      "learning_rate": 2.8358333333333333e-05,
      "loss": 6.403,
      "step": 13200
    },
    {
      "epoch": 66.5,
      "learning_rate": 2.7941666666666664e-05,
      "loss": 6.3973,
      "step": 13300
    },
    {
      "epoch": 67.0,
      "learning_rate": 2.7524999999999998e-05,
      "loss": 6.405,
      "step": 13400
    },
    {
      "epoch": 67.5,
      "learning_rate": 2.7108333333333333e-05,
      "loss": 6.4103,
      "step": 13500
    },
    {
      "epoch": 67.5,
      "eval_loss": 6.517162322998047,
      "eval_runtime": 24.3081,
      "eval_samples_per_second": 20.281,
      "eval_steps_per_second": 10.161,
      "eval_wer": 1.0,
      "step": 13500
    },
    {
      "epoch": 68.0,
      "learning_rate": 2.6691666666666664e-05,
      "loss": 6.3907,
      "step": 13600
    },
    {
      "epoch": 68.5,
      "learning_rate": 2.6275e-05,
      "loss": 6.3932,
      "step": 13700
    },
    {
      "epoch": 69.0,
      "learning_rate": 2.5858333333333333e-05,
      "loss": 6.4082,
      "step": 13800
    },
    {
      "epoch": 69.5,
      "learning_rate": 2.5441666666666664e-05,
      "loss": 6.4119,
      "step": 13900
    },
    {
      "epoch": 70.0,
      "learning_rate": 2.5025e-05,
      "loss": 6.3899,
      "step": 14000
    },
    {
      "epoch": 70.0,
      "eval_loss": 6.518171310424805,
      "eval_runtime": 24.1998,
      "eval_samples_per_second": 20.372,
      "eval_steps_per_second": 10.207,
      "eval_wer": 1.0,
      "step": 14000
    },
    {
      "epoch": 70.5,
      "learning_rate": 2.4608333333333333e-05,
      "loss": 6.4104,
      "step": 14100
    },
    {
      "epoch": 71.0,
      "learning_rate": 2.4191666666666664e-05,
      "loss": 6.3909,
      "step": 14200
    },
    {
      "epoch": 71.5,
      "learning_rate": 2.3775e-05,
      "loss": 6.3933,
      "step": 14300
    },
    {
      "epoch": 72.0,
      "learning_rate": 2.3358333333333333e-05,
      "loss": 6.4065,
      "step": 14400
    },
    {
      "epoch": 72.5,
      "learning_rate": 2.2941666666666664e-05,
      "loss": 6.4054,
      "step": 14500
    },
    {
      "epoch": 72.5,
      "eval_loss": 6.519650936126709,
      "eval_runtime": 24.4435,
      "eval_samples_per_second": 20.169,
      "eval_steps_per_second": 10.105,
      "eval_wer": 1.0,
      "step": 14500
    },
    {
      "epoch": 73.0,
      "learning_rate": 2.2525e-05,
      "loss": 6.3945,
      "step": 14600
    },
    {
      "epoch": 73.5,
      "learning_rate": 2.2108333333333333e-05,
      "loss": 6.3972,
      "step": 14700
    },
    {
      "epoch": 74.0,
      "learning_rate": 2.1691666666666664e-05,
      "loss": 6.4027,
      "step": 14800
    },
    {
      "epoch": 74.5,
      "learning_rate": 2.1275e-05,
      "loss": 6.4082,
      "step": 14900
    },
    {
      "epoch": 75.0,
      "learning_rate": 2.085833333333333e-05,
      "loss": 6.391,
      "step": 15000
    },
    {
      "epoch": 75.0,
      "eval_loss": 6.51995849609375,
      "eval_runtime": 24.299,
      "eval_samples_per_second": 20.289,
      "eval_steps_per_second": 10.165,
      "eval_wer": 1.0,
      "step": 15000
    },
    {
      "epoch": 75.5,
      "learning_rate": 2.0441666666666664e-05,
      "loss": 6.4077,
      "step": 15100
    },
    {
      "epoch": 76.0,
      "learning_rate": 2.0025e-05,
      "loss": 6.3905,
      "step": 15200
    },
    {
      "epoch": 76.5,
      "learning_rate": 1.960833333333333e-05,
      "loss": 6.3987,
      "step": 15300
    },
    {
      "epoch": 77.0,
      "learning_rate": 1.9191666666666664e-05,
      "loss": 6.4004,
      "step": 15400
    },
    {
      "epoch": 77.5,
      "learning_rate": 1.8775e-05,
      "loss": 6.3988,
      "step": 15500
    },
    {
      "epoch": 77.5,
      "eval_loss": 6.522036075592041,
      "eval_runtime": 24.4796,
      "eval_samples_per_second": 20.139,
      "eval_steps_per_second": 10.09,
      "eval_wer": 1.0,
      "step": 15500
    },
    {
      "epoch": 78.0,
      "learning_rate": 1.8358333333333333e-05,
      "loss": 6.3996,
      "step": 15600
    },
    {
      "epoch": 78.5,
      "learning_rate": 1.7941666666666665e-05,
      "loss": 6.404,
      "step": 15700
    },
    {
      "epoch": 79.0,
      "learning_rate": 1.7525e-05,
      "loss": 6.3918,
      "step": 15800
    },
    {
      "epoch": 79.5,
      "learning_rate": 1.710833333333333e-05,
      "loss": 6.3915,
      "step": 15900
    },
    {
      "epoch": 80.0,
      "learning_rate": 1.6691666666666665e-05,
      "loss": 6.4059,
      "step": 16000
    },
    {
      "epoch": 80.0,
      "eval_loss": 6.522817611694336,
      "eval_runtime": 24.2749,
      "eval_samples_per_second": 20.309,
      "eval_steps_per_second": 10.175,
      "eval_wer": 1.0,
      "step": 16000
    },
    {
      "epoch": 80.5,
      "learning_rate": 1.6275e-05,
      "loss": 6.3983,
      "step": 16100
    },
    {
      "epoch": 81.0,
      "learning_rate": 1.585833333333333e-05,
      "loss": 6.3996,
      "step": 16200
    },
    {
      "epoch": 81.5,
      "learning_rate": 1.5445833333333334e-05,
      "loss": 6.4038,
      "step": 16300
    },
    {
      "epoch": 82.0,
      "learning_rate": 1.5029166666666665e-05,
      "loss": 6.3918,
      "step": 16400
    },
    {
      "epoch": 82.5,
      "learning_rate": 1.4612499999999998e-05,
      "loss": 6.392,
      "step": 16500
    },
    {
      "epoch": 82.5,
      "eval_loss": 6.523283004760742,
      "eval_runtime": 24.4538,
      "eval_samples_per_second": 20.16,
      "eval_steps_per_second": 10.101,
      "eval_wer": 1.0,
      "step": 16500
    },
    {
      "epoch": 83.0,
      "learning_rate": 1.419583333333333e-05,
      "loss": 6.4043,
      "step": 16600
    },
    {
      "epoch": 83.5,
      "learning_rate": 1.3779166666666667e-05,
      "loss": 6.3989,
      "step": 16700
    },
    {
      "epoch": 84.0,
      "learning_rate": 1.33625e-05,
      "loss": 6.3956,
      "step": 16800
    },
    {
      "epoch": 84.5,
      "learning_rate": 1.2945833333333332e-05,
      "loss": 6.403,
      "step": 16900
    },
    {
      "epoch": 85.0,
      "learning_rate": 1.2529166666666667e-05,
      "loss": 6.3947,
      "step": 17000
    },
    {
      "epoch": 85.0,
      "eval_loss": 6.525250434875488,
      "eval_runtime": 24.2494,
      "eval_samples_per_second": 20.33,
      "eval_steps_per_second": 10.186,
      "eval_wer": 1.0,
      "step": 17000
    },
    {
      "epoch": 85.5,
      "learning_rate": 1.21125e-05,
      "loss": 6.4083,
      "step": 17100
    },
    {
      "epoch": 86.0,
      "learning_rate": 1.1695833333333332e-05,
      "loss": 6.3874,
      "step": 17200
    },
    {
      "epoch": 86.5,
      "learning_rate": 1.1279166666666665e-05,
      "loss": 6.3944,
      "step": 17300
    },
    {
      "epoch": 87.0,
      "learning_rate": 1.08625e-05,
      "loss": 6.3974,
      "step": 17400
    },
    {
      "epoch": 87.5,
      "learning_rate": 1.0445833333333333e-05,
      "loss": 6.3966,
      "step": 17500
    },
    {
      "epoch": 87.5,
      "eval_loss": 6.525856971740723,
      "eval_runtime": 24.4696,
      "eval_samples_per_second": 20.147,
      "eval_steps_per_second": 10.094,
      "eval_wer": 1.0,
      "step": 17500
    },
    {
      "epoch": 88.0,
      "learning_rate": 1.0029166666666665e-05,
      "loss": 6.397,
      "step": 17600
    },
    {
      "epoch": 88.5,
      "learning_rate": 9.6125e-06,
      "loss": 6.4091,
      "step": 17700
    },
    {
      "epoch": 89.0,
      "learning_rate": 9.195833333333333e-06,
      "loss": 6.3844,
      "step": 17800
    },
    {
      "epoch": 89.5,
      "learning_rate": 8.779166666666665e-06,
      "loss": 6.4037,
      "step": 17900
    },
    {
      "epoch": 90.0,
      "learning_rate": 8.3625e-06,
      "loss": 6.3905,
      "step": 18000
    },
    {
      "epoch": 90.0,
      "eval_loss": 6.526439666748047,
      "eval_runtime": 24.3911,
      "eval_samples_per_second": 20.212,
      "eval_steps_per_second": 10.127,
      "eval_wer": 1.0,
      "step": 18000
    },
    {
      "epoch": 90.5,
      "learning_rate": 7.945833333333333e-06,
      "loss": 6.4008,
      "step": 18100
    },
    {
      "epoch": 91.0,
      "learning_rate": 7.529166666666666e-06,
      "loss": 6.3926,
      "step": 18200
    },
    {
      "epoch": 91.5,
      "learning_rate": 7.116666666666666e-06,
      "loss": 6.3982,
      "step": 18300
    },
    {
      "epoch": 92.0,
      "learning_rate": 6.699999999999999e-06,
      "loss": 6.3951,
      "step": 18400
    },
    {
      "epoch": 92.5,
      "learning_rate": 6.283333333333333e-06,
      "loss": 6.4003,
      "step": 18500
    },
    {
      "epoch": 92.5,
      "eval_loss": 6.527181148529053,
      "eval_runtime": 24.2624,
      "eval_samples_per_second": 20.32,
      "eval_steps_per_second": 10.18,
      "eval_wer": 1.0,
      "step": 18500
    },
    {
      "epoch": 93.0,
      "learning_rate": 5.866666666666666e-06,
      "loss": 6.3932,
      "step": 18600
    },
    {
      "epoch": 93.5,
      "learning_rate": 5.4499999999999995e-06,
      "loss": 6.4076,
      "step": 18700
    },
    {
      "epoch": 94.0,
      "learning_rate": 5.033333333333332e-06,
      "loss": 6.3857,
      "step": 18800
    },
    {
      "epoch": 94.5,
      "learning_rate": 4.616666666666667e-06,
      "loss": 6.4045,
      "step": 18900
    },
    {
      "epoch": 95.0,
      "learning_rate": 4.2e-06,
      "loss": 6.3877,
      "step": 19000
    },
    {
      "epoch": 95.0,
      "eval_loss": 6.527469635009766,
      "eval_runtime": 24.1993,
      "eval_samples_per_second": 20.372,
      "eval_steps_per_second": 10.207,
      "eval_wer": 1.0,
      "step": 19000
    },
    {
      "epoch": 95.5,
      "learning_rate": 3.783333333333333e-06,
      "loss": 6.4,
      "step": 19100
    },
    {
      "epoch": 96.0,
      "learning_rate": 3.3666666666666665e-06,
      "loss": 6.3918,
      "step": 19200
    },
    {
      "epoch": 96.5,
      "learning_rate": 2.9499999999999997e-06,
      "loss": 6.4001,
      "step": 19300
    },
    {
      "epoch": 97.0,
      "learning_rate": 2.533333333333333e-06,
      "loss": 6.3935,
      "step": 19400
    },
    {
      "epoch": 97.5,
      "learning_rate": 2.1166666666666666e-06,
      "loss": 6.3903,
      "step": 19500
    },
    {
      "epoch": 97.5,
      "eval_loss": 6.527697563171387,
      "eval_runtime": 24.3683,
      "eval_samples_per_second": 20.231,
      "eval_steps_per_second": 10.136,
      "eval_wer": 1.0,
      "step": 19500
    },
    {
      "epoch": 98.0,
      "learning_rate": 1.7e-06,
      "loss": 6.4019,
      "step": 19600
    },
    {
      "epoch": 98.5,
      "learning_rate": 1.2833333333333333e-06,
      "loss": 6.4,
      "step": 19700
    },
    {
      "epoch": 99.0,
      "learning_rate": 8.666666666666666e-07,
      "loss": 6.3896,
      "step": 19800
    },
    {
      "epoch": 99.5,
      "learning_rate": 4.5e-07,
      "loss": 6.3972,
      "step": 19900
    },
    {
      "epoch": 100.0,
      "learning_rate": 3.3333333333333334e-08,
      "loss": 6.3944,
      "step": 20000
    },
    {
      "epoch": 100.0,
      "eval_loss": 6.5278449058532715,
      "eval_runtime": 24.1079,
      "eval_samples_per_second": 20.45,
      "eval_steps_per_second": 10.246,
      "eval_wer": 1.0,
      "step": 20000
    },
    {
      "epoch": 100.0,
      "step": 20000,
      "total_flos": 2.051320516956903e+19,
      "train_loss": 9.81739658203125,
      "train_runtime": 23393.5837,
      "train_samples_per_second": 27.272,
      "train_steps_per_second": 0.855
    }
  ],
  "max_steps": 20000,
  "num_train_epochs": 100,
  "total_flos": 2.051320516956903e+19,
  "trial_name": null,
  "trial_params": null
}