{
  "best_metric": 6.313937972585015,
  "best_model_checkpoint": "./checkpoint-3000",
  "epoch": 1.0,
  "global_step": 3000,
  "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": 0.6304,
      "step": 25
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.400000000000001e-07,
      "loss": 0.4908,
      "step": 50
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.44e-06,
      "loss": 0.2708,
      "step": 75
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.94e-06,
      "loss": 0.2194,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.4400000000000004e-06,
      "loss": 0.1984,
      "step": 125
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.9400000000000002e-06,
      "loss": 0.1976,
      "step": 150
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.44e-06,
      "loss": 0.1758,
      "step": 175
    },
    {
      "epoch": 0.07,
      "learning_rate": 3.94e-06,
      "loss": 0.1747,
      "step": 200
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.440000000000001e-06,
      "loss": 0.1578,
      "step": 225
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.94e-06,
      "loss": 0.1605,
      "step": 250
    },
    {
      "epoch": 0.09,
      "learning_rate": 5.4400000000000004e-06,
      "loss": 0.1654,
      "step": 275
    },
    {
      "epoch": 0.1,
      "learning_rate": 5.94e-06,
      "loss": 0.1633,
      "step": 300
    },
    {
      "epoch": 0.11,
      "learning_rate": 6.440000000000001e-06,
      "loss": 0.1522,
      "step": 325
    },
    {
      "epoch": 0.12,
      "learning_rate": 6.9400000000000005e-06,
      "loss": 0.1494,
      "step": 350
    },
    {
      "epoch": 0.12,
      "learning_rate": 7.440000000000001e-06,
      "loss": 0.1372,
      "step": 375
    },
    {
      "epoch": 0.13,
      "learning_rate": 7.94e-06,
      "loss": 0.1338,
      "step": 400
    },
    {
      "epoch": 0.14,
      "learning_rate": 8.44e-06,
      "loss": 0.1441,
      "step": 425
    },
    {
      "epoch": 0.15,
      "learning_rate": 8.94e-06,
      "loss": 0.1401,
      "step": 450
    },
    {
      "epoch": 0.16,
      "learning_rate": 9.440000000000001e-06,
      "loss": 0.1274,
      "step": 475
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.940000000000001e-06,
      "loss": 0.127,
      "step": 500
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.15694677829742432,
      "eval_runtime": 17256.1308,
      "eval_samples_per_second": 0.932,
      "eval_steps_per_second": 0.117,
      "eval_wer": 7.258771043090319,
      "step": 500
    },
    {
      "epoch": 0.17,
      "learning_rate": 9.912000000000001e-06,
      "loss": 0.1335,
      "step": 525
    },
    {
      "epoch": 0.18,
      "learning_rate": 9.812e-06,
      "loss": 0.1282,
      "step": 550
    },
    {
      "epoch": 0.19,
      "learning_rate": 9.716000000000002e-06,
      "loss": 0.1318,
      "step": 575
    },
    {
      "epoch": 0.2,
      "learning_rate": 9.616e-06,
      "loss": 0.1305,
      "step": 600
    },
    {
      "epoch": 0.21,
      "learning_rate": 9.516e-06,
      "loss": 0.1293,
      "step": 625
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.416000000000001e-06,
      "loss": 0.1148,
      "step": 650
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.316e-06,
      "loss": 0.1349,
      "step": 675
    },
    {
      "epoch": 0.23,
      "learning_rate": 9.216000000000001e-06,
      "loss": 0.1178,
      "step": 700
    },
    {
      "epoch": 0.24,
      "learning_rate": 9.116e-06,
      "loss": 0.1252,
      "step": 725
    },
    {
      "epoch": 0.25,
      "learning_rate": 9.016e-06,
      "loss": 0.1253,
      "step": 750
    },
    {
      "epoch": 0.26,
      "learning_rate": 8.916e-06,
      "loss": 0.1203,
      "step": 775
    },
    {
      "epoch": 0.27,
      "learning_rate": 8.816000000000002e-06,
      "loss": 0.1222,
      "step": 800
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.716000000000001e-06,
      "loss": 0.1207,
      "step": 825
    },
    {
      "epoch": 0.28,
      "learning_rate": 8.616000000000002e-06,
      "loss": 0.1133,
      "step": 850
    },
    {
      "epoch": 0.29,
      "learning_rate": 8.516000000000001e-06,
      "loss": 0.1179,
      "step": 875
    },
    {
      "epoch": 0.3,
      "learning_rate": 8.416e-06,
      "loss": 0.1164,
      "step": 900
    },
    {
      "epoch": 0.31,
      "learning_rate": 8.316000000000001e-06,
      "loss": 0.1176,
      "step": 925
    },
    {
      "epoch": 0.32,
      "learning_rate": 8.216e-06,
      "loss": 0.1151,
      "step": 950
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.116e-06,
      "loss": 0.1205,
      "step": 975
    },
    {
      "epoch": 0.33,
      "learning_rate": 8.016e-06,
      "loss": 0.1116,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.15050184726715088,
      "eval_runtime": 17277.2664,
      "eval_samples_per_second": 0.931,
      "eval_steps_per_second": 0.116,
      "eval_wer": 7.237205666302751,
      "step": 1000
    },
    {
      "epoch": 0.34,
      "learning_rate": 7.916e-06,
      "loss": 0.1148,
      "step": 1025
    },
    {
      "epoch": 0.35,
      "learning_rate": 7.816000000000001e-06,
      "loss": 0.111,
      "step": 1050
    },
    {
      "epoch": 0.36,
      "learning_rate": 7.716e-06,
      "loss": 0.1198,
      "step": 1075
    },
    {
      "epoch": 0.37,
      "learning_rate": 7.616000000000001e-06,
      "loss": 0.1192,
      "step": 1100
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.516000000000001e-06,
      "loss": 0.1093,
      "step": 1125
    },
    {
      "epoch": 0.38,
      "learning_rate": 7.416000000000001e-06,
      "loss": 0.1012,
      "step": 1150
    },
    {
      "epoch": 0.39,
      "learning_rate": 7.316000000000001e-06,
      "loss": 0.1219,
      "step": 1175
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.216000000000001e-06,
      "loss": 0.108,
      "step": 1200
    },
    {
      "epoch": 0.41,
      "learning_rate": 7.116000000000001e-06,
      "loss": 0.117,
      "step": 1225
    },
    {
      "epoch": 0.42,
      "learning_rate": 7.016e-06,
      "loss": 0.1139,
      "step": 1250
    },
    {
      "epoch": 0.42,
      "learning_rate": 6.916e-06,
      "loss": 0.1117,
      "step": 1275
    },
    {
      "epoch": 0.43,
      "learning_rate": 6.8160000000000005e-06,
      "loss": 0.1066,
      "step": 1300
    },
    {
      "epoch": 0.44,
      "learning_rate": 6.716000000000001e-06,
      "loss": 0.1188,
      "step": 1325
    },
    {
      "epoch": 0.45,
      "learning_rate": 6.616e-06,
      "loss": 0.1115,
      "step": 1350
    },
    {
      "epoch": 0.46,
      "learning_rate": 6.516e-06,
      "loss": 0.0991,
      "step": 1375
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.416e-06,
      "loss": 0.1097,
      "step": 1400
    },
    {
      "epoch": 0.47,
      "learning_rate": 6.316000000000001e-06,
      "loss": 0.1064,
      "step": 1425
    },
    {
      "epoch": 0.48,
      "learning_rate": 6.216000000000001e-06,
      "loss": 0.1214,
      "step": 1450
    },
    {
      "epoch": 0.49,
      "learning_rate": 6.116000000000001e-06,
      "loss": 0.1139,
      "step": 1475
    },
    {
      "epoch": 0.5,
      "learning_rate": 6.0160000000000005e-06,
      "loss": 0.1132,
      "step": 1500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.1435445249080658,
      "eval_runtime": 17236.1573,
      "eval_samples_per_second": 0.933,
      "eval_steps_per_second": 0.117,
      "eval_wer": 6.882050867332498,
      "step": 1500
    },
    {
      "epoch": 0.51,
      "learning_rate": 5.916000000000001e-06,
      "loss": 0.0995,
      "step": 1525
    },
    {
      "epoch": 0.52,
      "learning_rate": 5.816000000000001e-06,
      "loss": 0.1168,
      "step": 1550
    },
    {
      "epoch": 0.53,
      "learning_rate": 5.716000000000001e-06,
      "loss": 0.102,
      "step": 1575
    },
    {
      "epoch": 0.53,
      "learning_rate": 5.616e-06,
      "loss": 0.0999,
      "step": 1600
    },
    {
      "epoch": 0.54,
      "learning_rate": 5.516e-06,
      "loss": 0.1065,
      "step": 1625
    },
    {
      "epoch": 0.55,
      "learning_rate": 5.416e-06,
      "loss": 0.1092,
      "step": 1650
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.3160000000000004e-06,
      "loss": 0.1014,
      "step": 1675
    },
    {
      "epoch": 0.57,
      "learning_rate": 5.216e-06,
      "loss": 0.1016,
      "step": 1700
    },
    {
      "epoch": 0.57,
      "learning_rate": 5.116000000000001e-06,
      "loss": 0.096,
      "step": 1725
    },
    {
      "epoch": 0.58,
      "learning_rate": 5.016000000000001e-06,
      "loss": 0.0948,
      "step": 1750
    },
    {
      "epoch": 0.59,
      "learning_rate": 4.916e-06,
      "loss": 0.1017,
      "step": 1775
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.816e-06,
      "loss": 0.0969,
      "step": 1800
    },
    {
      "epoch": 0.61,
      "learning_rate": 4.716e-06,
      "loss": 0.1081,
      "step": 1825
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.616e-06,
      "loss": 0.1128,
      "step": 1850
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.5160000000000005e-06,
      "loss": 0.1017,
      "step": 1875
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.416000000000001e-06,
      "loss": 0.0963,
      "step": 1900
    },
    {
      "epoch": 0.64,
      "learning_rate": 4.316e-06,
      "loss": 0.1004,
      "step": 1925
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.216e-06,
      "loss": 0.1001,
      "step": 1950
    },
    {
      "epoch": 0.66,
      "learning_rate": 4.116000000000001e-06,
      "loss": 0.0999,
      "step": 1975
    },
    {
      "epoch": 0.67,
      "learning_rate": 4.016e-06,
      "loss": 0.0939,
      "step": 2000
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.13536924123764038,
      "eval_runtime": 17221.386,
      "eval_samples_per_second": 0.934,
      "eval_steps_per_second": 0.117,
      "eval_wer": 6.534309166632971,
      "step": 2000
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.916e-06,
      "loss": 0.0911,
      "step": 2025
    },
    {
      "epoch": 0.68,
      "learning_rate": 3.816e-06,
      "loss": 0.0985,
      "step": 2050
    },
    {
      "epoch": 0.69,
      "learning_rate": 3.716e-06,
      "loss": 0.0954,
      "step": 2075
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.616e-06,
      "loss": 0.0866,
      "step": 2100
    },
    {
      "epoch": 0.71,
      "learning_rate": 3.5160000000000007e-06,
      "loss": 0.0945,
      "step": 2125
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.4160000000000004e-06,
      "loss": 0.1051,
      "step": 2150
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.3160000000000005e-06,
      "loss": 0.0984,
      "step": 2175
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.216e-06,
      "loss": 0.1037,
      "step": 2200
    },
    {
      "epoch": 0.74,
      "learning_rate": 3.1160000000000003e-06,
      "loss": 0.0993,
      "step": 2225
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.016e-06,
      "loss": 0.0924,
      "step": 2250
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.9160000000000005e-06,
      "loss": 0.0972,
      "step": 2275
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.8160000000000002e-06,
      "loss": 0.0898,
      "step": 2300
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.7160000000000003e-06,
      "loss": 0.0928,
      "step": 2325
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.616e-06,
      "loss": 0.095,
      "step": 2350
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.516e-06,
      "loss": 0.0831,
      "step": 2375
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.4160000000000002e-06,
      "loss": 0.0853,
      "step": 2400
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.3160000000000004e-06,
      "loss": 0.0845,
      "step": 2425
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.216e-06,
      "loss": 0.08,
      "step": 2450
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.116e-06,
      "loss": 0.0873,
      "step": 2475
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.02e-06,
      "loss": 0.0819,
      "step": 2500
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.13387082517147064,
      "eval_runtime": 17283.6443,
      "eval_samples_per_second": 0.93,
      "eval_steps_per_second": 0.116,
      "eval_wer": 6.497917593303951,
      "step": 2500
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 0.0903,
      "step": 2525
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.8200000000000002e-06,
      "loss": 0.0923,
      "step": 2550
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.72e-06,
      "loss": 0.0855,
      "step": 2575
    },
    {
      "epoch": 0.87,
      "learning_rate": 1.6200000000000002e-06,
      "loss": 0.088,
      "step": 2600
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.52e-06,
      "loss": 0.0833,
      "step": 2625
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.42e-06,
      "loss": 0.0879,
      "step": 2650
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.32e-06,
      "loss": 0.0951,
      "step": 2675
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.2200000000000002e-06,
      "loss": 0.0945,
      "step": 2700
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.12e-06,
      "loss": 0.0828,
      "step": 2725
    },
    {
      "epoch": 0.92,
      "learning_rate": 1.02e-06,
      "loss": 0.0893,
      "step": 2750
    },
    {
      "epoch": 0.93,
      "learning_rate": 9.200000000000001e-07,
      "loss": 0.082,
      "step": 2775
    },
    {
      "epoch": 0.93,
      "learning_rate": 8.200000000000001e-07,
      "loss": 0.0915,
      "step": 2800
    },
    {
      "epoch": 0.94,
      "learning_rate": 7.2e-07,
      "loss": 0.095,
      "step": 2825
    },
    {
      "epoch": 0.95,
      "learning_rate": 6.200000000000001e-07,
      "loss": 0.0921,
      "step": 2850
    },
    {
      "epoch": 0.96,
      "learning_rate": 5.2e-07,
      "loss": 0.0839,
      "step": 2875
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.2000000000000006e-07,
      "loss": 0.0944,
      "step": 2900
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.2e-07,
      "loss": 0.0871,
      "step": 2925
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.2e-07,
      "loss": 0.0834,
      "step": 2950
    },
    {
      "epoch": 0.99,
      "learning_rate": 1.2000000000000002e-07,
      "loss": 0.089,
      "step": 2975
    },
    {
      "epoch": 1.0,
      "learning_rate": 2e-08,
      "loss": 0.0892,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.13065485656261444,
      "eval_runtime": 17232.2238,
      "eval_samples_per_second": 0.933,
      "eval_steps_per_second": 0.117,
      "eval_wer": 6.313937972585015,
      "step": 3000
    },
    {
      "epoch": 1.0,
      "step": 3000,
      "total_flos": 4.076488359936e+20,
      "train_loss": 0.12058185605208079,
      "train_runtime": 159323.6173,
      "train_samples_per_second": 1.205,
      "train_steps_per_second": 0.019
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 4.076488359936e+20,
  "trial_name": null,
  "trial_params": null
}