{
  "best_metric": 59.0146750524109,
  "best_model_checkpoint": "./whisper-small-bn/checkpoint-2000",
  "epoch": 4.282655246252677,
  "eval_steps": 1000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 27.802248001098633,
      "learning_rate": 3.2e-07,
      "loss": 2.2811,
      "step": 20
    },
    {
      "epoch": 0.13,
      "grad_norm": 18.253421783447266,
      "learning_rate": 7.2e-07,
      "loss": 2.1964,
      "step": 40
    },
    {
      "epoch": 0.2,
      "grad_norm": 13.585515975952148,
      "learning_rate": 1.12e-06,
      "loss": 1.8635,
      "step": 60
    },
    {
      "epoch": 0.27,
      "grad_norm": 22.400362014770508,
      "learning_rate": 1.52e-06,
      "loss": 1.703,
      "step": 80
    },
    {
      "epoch": 0.33,
      "grad_norm": 13.77535343170166,
      "learning_rate": 1.9200000000000003e-06,
      "loss": 1.5336,
      "step": 100
    },
    {
      "epoch": 0.4,
      "grad_norm": 14.989215850830078,
      "learning_rate": 2.3200000000000002e-06,
      "loss": 1.4044,
      "step": 120
    },
    {
      "epoch": 0.47,
      "grad_norm": 22.01858139038086,
      "learning_rate": 2.7200000000000002e-06,
      "loss": 1.2402,
      "step": 140
    },
    {
      "epoch": 0.53,
      "grad_norm": 16.482311248779297,
      "learning_rate": 3.12e-06,
      "loss": 1.0922,
      "step": 160
    },
    {
      "epoch": 0.6,
      "grad_norm": 17.529874801635742,
      "learning_rate": 3.52e-06,
      "loss": 0.9583,
      "step": 180
    },
    {
      "epoch": 0.67,
      "grad_norm": 18.252761840820312,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.8783,
      "step": 200
    },
    {
      "epoch": 0.73,
      "grad_norm": 16.48067283630371,
      "learning_rate": 4.32e-06,
      "loss": 0.7643,
      "step": 220
    },
    {
      "epoch": 0.8,
      "grad_norm": 15.062904357910156,
      "learning_rate": 4.7200000000000005e-06,
      "loss": 0.6991,
      "step": 240
    },
    {
      "epoch": 0.87,
      "grad_norm": 14.231403350830078,
      "learning_rate": 5.12e-06,
      "loss": 0.6594,
      "step": 260
    },
    {
      "epoch": 0.93,
      "grad_norm": 18.425230026245117,
      "learning_rate": 5.5200000000000005e-06,
      "loss": 0.5724,
      "step": 280
    },
    {
      "epoch": 1.0,
      "grad_norm": 11.786844253540039,
      "learning_rate": 5.92e-06,
      "loss": 0.5687,
      "step": 300
    },
    {
      "epoch": 1.07,
      "grad_norm": 15.346015930175781,
      "learning_rate": 6.3200000000000005e-06,
      "loss": 0.4696,
      "step": 320
    },
    {
      "epoch": 1.13,
      "grad_norm": 7.936849594116211,
      "learning_rate": 6.720000000000001e-06,
      "loss": 0.4569,
      "step": 340
    },
    {
      "epoch": 1.2,
      "grad_norm": 13.894779205322266,
      "learning_rate": 7.1200000000000004e-06,
      "loss": 0.4089,
      "step": 360
    },
    {
      "epoch": 1.27,
      "grad_norm": 11.665081977844238,
      "learning_rate": 7.520000000000001e-06,
      "loss": 0.4168,
      "step": 380
    },
    {
      "epoch": 1.33,
      "grad_norm": 16.7696475982666,
      "learning_rate": 7.92e-06,
      "loss": 0.4433,
      "step": 400
    },
    {
      "epoch": 1.4,
      "grad_norm": 15.60254192352295,
      "learning_rate": 8.32e-06,
      "loss": 0.4013,
      "step": 420
    },
    {
      "epoch": 1.47,
      "grad_norm": 13.72472858428955,
      "learning_rate": 8.720000000000001e-06,
      "loss": 0.3903,
      "step": 440
    },
    {
      "epoch": 1.53,
      "grad_norm": 10.506444931030273,
      "learning_rate": 9.12e-06,
      "loss": 0.3581,
      "step": 460
    },
    {
      "epoch": 1.6,
      "grad_norm": 13.643900871276855,
      "learning_rate": 9.52e-06,
      "loss": 0.3605,
      "step": 480
    },
    {
      "epoch": 1.67,
      "grad_norm": 11.669689178466797,
      "learning_rate": 9.920000000000002e-06,
      "loss": 0.3548,
      "step": 500
    },
    {
      "epoch": 1.73,
      "grad_norm": 9.907474517822266,
      "learning_rate": 9.936000000000001e-06,
      "loss": 0.3535,
      "step": 520
    },
    {
      "epoch": 1.8,
      "grad_norm": 11.366904258728027,
      "learning_rate": 9.856000000000002e-06,
      "loss": 0.2943,
      "step": 540
    },
    {
      "epoch": 1.87,
      "grad_norm": 7.5114240646362305,
      "learning_rate": 9.776000000000001e-06,
      "loss": 0.3057,
      "step": 560
    },
    {
      "epoch": 1.93,
      "grad_norm": 11.717443466186523,
      "learning_rate": 9.696000000000002e-06,
      "loss": 0.2919,
      "step": 580
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.275129318237305,
      "learning_rate": 9.616e-06,
      "loss": 0.3115,
      "step": 600
    },
    {
      "epoch": 2.07,
      "grad_norm": 9.388382911682129,
      "learning_rate": 9.536000000000002e-06,
      "loss": 0.1926,
      "step": 620
    },
    {
      "epoch": 2.13,
      "grad_norm": 8.574433326721191,
      "learning_rate": 9.456e-06,
      "loss": 0.2193,
      "step": 640
    },
    {
      "epoch": 2.2,
      "grad_norm": 5.512338638305664,
      "learning_rate": 9.376000000000001e-06,
      "loss": 0.18,
      "step": 660
    },
    {
      "epoch": 2.27,
      "grad_norm": 9.644317626953125,
      "learning_rate": 9.296e-06,
      "loss": 0.1916,
      "step": 680
    },
    {
      "epoch": 2.33,
      "grad_norm": 9.321474075317383,
      "learning_rate": 9.216000000000001e-06,
      "loss": 0.2055,
      "step": 700
    },
    {
      "epoch": 2.4,
      "grad_norm": 4.868476390838623,
      "learning_rate": 9.136e-06,
      "loss": 0.1843,
      "step": 720
    },
    {
      "epoch": 2.47,
      "grad_norm": 6.592445373535156,
      "learning_rate": 9.056000000000001e-06,
      "loss": 0.1713,
      "step": 740
    },
    {
      "epoch": 2.53,
      "grad_norm": 8.138116836547852,
      "learning_rate": 8.976e-06,
      "loss": 0.1792,
      "step": 760
    },
    {
      "epoch": 2.6,
      "grad_norm": 9.782510757446289,
      "learning_rate": 8.896000000000001e-06,
      "loss": 0.1796,
      "step": 780
    },
    {
      "epoch": 2.67,
      "grad_norm": 5.867671012878418,
      "learning_rate": 8.816000000000002e-06,
      "loss": 0.1652,
      "step": 800
    },
    {
      "epoch": 2.73,
      "grad_norm": 6.637588977813721,
      "learning_rate": 8.736e-06,
      "loss": 0.1735,
      "step": 820
    },
    {
      "epoch": 2.8,
      "grad_norm": 11.652266502380371,
      "learning_rate": 8.656000000000001e-06,
      "loss": 0.1734,
      "step": 840
    },
    {
      "epoch": 2.87,
      "grad_norm": 6.051203727722168,
      "learning_rate": 8.576e-06,
      "loss": 0.1557,
      "step": 860
    },
    {
      "epoch": 2.93,
      "grad_norm": 11.018319129943848,
      "learning_rate": 8.496000000000001e-06,
      "loss": 0.1668,
      "step": 880
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.878730297088623,
      "learning_rate": 8.416e-06,
      "loss": 0.1706,
      "step": 900
    },
    {
      "epoch": 3.07,
      "grad_norm": 6.223515510559082,
      "learning_rate": 8.336000000000001e-06,
      "loss": 0.0917,
      "step": 920
    },
    {
      "epoch": 3.13,
      "grad_norm": 7.449841022491455,
      "learning_rate": 8.256e-06,
      "loss": 0.0853,
      "step": 940
    },
    {
      "epoch": 3.2,
      "grad_norm": 3.824934244155884,
      "learning_rate": 8.176000000000001e-06,
      "loss": 0.1033,
      "step": 960
    },
    {
      "epoch": 3.27,
      "grad_norm": 6.369926929473877,
      "learning_rate": 8.096e-06,
      "loss": 0.0997,
      "step": 980
    },
    {
      "epoch": 3.33,
      "grad_norm": 6.938141822814941,
      "learning_rate": 8.016e-06,
      "loss": 0.095,
      "step": 1000
    },
    {
      "epoch": 3.33,
      "eval_loss": 0.27550819516181946,
      "eval_runtime": 345.1723,
      "eval_samples_per_second": 0.29,
      "eval_steps_per_second": 0.29,
      "eval_wer": 60.18099547511312,
      "step": 1000
    },
    {
      "epoch": 2.18,
      "grad_norm": 7.869150638580322,
      "learning_rate": 7.936e-06,
      "loss": 0.1443,
      "step": 1020
    },
    {
      "epoch": 2.23,
      "grad_norm": 3.6447396278381348,
      "learning_rate": 7.860000000000001e-06,
      "loss": 0.1974,
      "step": 1040
    },
    {
      "epoch": 2.27,
      "grad_norm": 9.588157653808594,
      "learning_rate": 7.78e-06,
      "loss": 0.1427,
      "step": 1060
    },
    {
      "epoch": 2.31,
      "grad_norm": 10.796669006347656,
      "learning_rate": 7.7e-06,
      "loss": 0.187,
      "step": 1080
    },
    {
      "epoch": 2.36,
      "grad_norm": 4.941339015960693,
      "learning_rate": 7.620000000000001e-06,
      "loss": 0.1568,
      "step": 1100
    },
    {
      "epoch": 2.4,
      "grad_norm": 11.114888191223145,
      "learning_rate": 7.540000000000001e-06,
      "loss": 0.1697,
      "step": 1120
    },
    {
      "epoch": 2.44,
      "grad_norm": 10.200749397277832,
      "learning_rate": 7.4600000000000006e-06,
      "loss": 0.1675,
      "step": 1140
    },
    {
      "epoch": 2.48,
      "grad_norm": 5.67927360534668,
      "learning_rate": 7.3800000000000005e-06,
      "loss": 0.1557,
      "step": 1160
    },
    {
      "epoch": 2.53,
      "grad_norm": 7.413832664489746,
      "learning_rate": 7.304000000000001e-06,
      "loss": 0.1629,
      "step": 1180
    },
    {
      "epoch": 2.57,
      "grad_norm": 6.933262348175049,
      "learning_rate": 7.224000000000001e-06,
      "loss": 0.17,
      "step": 1200
    },
    {
      "epoch": 2.61,
      "grad_norm": 4.847285747528076,
      "learning_rate": 7.1440000000000005e-06,
      "loss": 0.1389,
      "step": 1220
    },
    {
      "epoch": 2.66,
      "grad_norm": 19.55495834350586,
      "learning_rate": 7.0640000000000005e-06,
      "loss": 0.1338,
      "step": 1240
    },
    {
      "epoch": 2.7,
      "grad_norm": 7.148222923278809,
      "learning_rate": 6.984e-06,
      "loss": 0.1296,
      "step": 1260
    },
    {
      "epoch": 2.74,
      "grad_norm": 6.323164939880371,
      "learning_rate": 6.904e-06,
      "loss": 0.163,
      "step": 1280
    },
    {
      "epoch": 2.78,
      "grad_norm": 9.096046447753906,
      "learning_rate": 6.824e-06,
      "loss": 0.1942,
      "step": 1300
    },
    {
      "epoch": 2.83,
      "grad_norm": 8.462915420532227,
      "learning_rate": 6.744e-06,
      "loss": 0.1576,
      "step": 1320
    },
    {
      "epoch": 2.87,
      "grad_norm": 7.848359107971191,
      "learning_rate": 6.664e-06,
      "loss": 0.1219,
      "step": 1340
    },
    {
      "epoch": 2.91,
      "grad_norm": 6.30383825302124,
      "learning_rate": 6.584e-06,
      "loss": 0.1404,
      "step": 1360
    },
    {
      "epoch": 2.96,
      "grad_norm": 5.657530784606934,
      "learning_rate": 6.504e-06,
      "loss": 0.1449,
      "step": 1380
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.4172234535217285,
      "learning_rate": 6.424e-06,
      "loss": 0.1234,
      "step": 1400
    },
    {
      "epoch": 3.04,
      "grad_norm": 6.550841331481934,
      "learning_rate": 6.344e-06,
      "loss": 0.0885,
      "step": 1420
    },
    {
      "epoch": 3.08,
      "grad_norm": 6.212351322174072,
      "learning_rate": 6.264e-06,
      "loss": 0.0725,
      "step": 1440
    },
    {
      "epoch": 3.13,
      "grad_norm": 5.2562575340271,
      "learning_rate": 6.184e-06,
      "loss": 0.0886,
      "step": 1460
    },
    {
      "epoch": 3.17,
      "grad_norm": 7.921740531921387,
      "learning_rate": 6.104000000000001e-06,
      "loss": 0.0805,
      "step": 1480
    },
    {
      "epoch": 3.21,
      "grad_norm": 3.5878069400787354,
      "learning_rate": 6.024000000000001e-06,
      "loss": 0.0628,
      "step": 1500
    },
    {
      "epoch": 3.25,
      "grad_norm": 8.585208892822266,
      "learning_rate": 5.944000000000001e-06,
      "loss": 0.0812,
      "step": 1520
    },
    {
      "epoch": 3.3,
      "grad_norm": 4.111368656158447,
      "learning_rate": 5.868e-06,
      "loss": 0.0694,
      "step": 1540
    },
    {
      "epoch": 3.34,
      "grad_norm": 5.304944038391113,
      "learning_rate": 5.788e-06,
      "loss": 0.0739,
      "step": 1560
    },
    {
      "epoch": 3.38,
      "grad_norm": 6.168578147888184,
      "learning_rate": 5.708e-06,
      "loss": 0.0821,
      "step": 1580
    },
    {
      "epoch": 3.43,
      "grad_norm": 12.2472505569458,
      "learning_rate": 5.628e-06,
      "loss": 0.0885,
      "step": 1600
    },
    {
      "epoch": 3.47,
      "grad_norm": 9.349952697753906,
      "learning_rate": 5.548e-06,
      "loss": 0.0607,
      "step": 1620
    },
    {
      "epoch": 3.51,
      "grad_norm": 5.98253870010376,
      "learning_rate": 5.468e-06,
      "loss": 0.0619,
      "step": 1640
    },
    {
      "epoch": 3.55,
      "grad_norm": 7.106723785400391,
      "learning_rate": 5.388e-06,
      "loss": 0.0992,
      "step": 1660
    },
    {
      "epoch": 3.6,
      "grad_norm": 3.693020820617676,
      "learning_rate": 5.308000000000001e-06,
      "loss": 0.0855,
      "step": 1680
    },
    {
      "epoch": 3.64,
      "grad_norm": 7.3649678230285645,
      "learning_rate": 5.228000000000001e-06,
      "loss": 0.081,
      "step": 1700
    },
    {
      "epoch": 3.68,
      "grad_norm": 5.4454240798950195,
      "learning_rate": 5.1480000000000005e-06,
      "loss": 0.0872,
      "step": 1720
    },
    {
      "epoch": 3.73,
      "grad_norm": 8.854460716247559,
      "learning_rate": 5.0680000000000004e-06,
      "loss": 0.0869,
      "step": 1740
    },
    {
      "epoch": 3.77,
      "grad_norm": 3.8615269660949707,
      "learning_rate": 4.988e-06,
      "loss": 0.0821,
      "step": 1760
    },
    {
      "epoch": 3.81,
      "grad_norm": 3.7190287113189697,
      "learning_rate": 4.908e-06,
      "loss": 0.0636,
      "step": 1780
    },
    {
      "epoch": 3.85,
      "grad_norm": 7.683879375457764,
      "learning_rate": 4.828e-06,
      "loss": 0.0892,
      "step": 1800
    },
    {
      "epoch": 3.9,
      "grad_norm": 3.645141839981079,
      "learning_rate": 4.748e-06,
      "loss": 0.0803,
      "step": 1820
    },
    {
      "epoch": 3.94,
      "grad_norm": 3.8167316913604736,
      "learning_rate": 4.668e-06,
      "loss": 0.1012,
      "step": 1840
    },
    {
      "epoch": 3.98,
      "grad_norm": 7.247605323791504,
      "learning_rate": 4.588e-06,
      "loss": 0.0728,
      "step": 1860
    },
    {
      "epoch": 4.03,
      "grad_norm": 3.5495615005493164,
      "learning_rate": 4.508e-06,
      "loss": 0.0439,
      "step": 1880
    },
    {
      "epoch": 4.07,
      "grad_norm": 4.105050086975098,
      "learning_rate": 4.428000000000001e-06,
      "loss": 0.0357,
      "step": 1900
    },
    {
      "epoch": 4.11,
      "grad_norm": 9.399913787841797,
      "learning_rate": 4.3480000000000006e-06,
      "loss": 0.0363,
      "step": 1920
    },
    {
      "epoch": 4.15,
      "grad_norm": 3.0376510620117188,
      "learning_rate": 4.2680000000000005e-06,
      "loss": 0.0221,
      "step": 1940
    },
    {
      "epoch": 4.2,
      "grad_norm": 0.6908012628555298,
      "learning_rate": 4.188e-06,
      "loss": 0.0323,
      "step": 1960
    },
    {
      "epoch": 4.24,
      "grad_norm": 6.255786895751953,
      "learning_rate": 4.108e-06,
      "loss": 0.0327,
      "step": 1980
    },
    {
      "epoch": 4.28,
      "grad_norm": 7.1336669921875,
      "learning_rate": 4.028e-06,
      "loss": 0.0416,
      "step": 2000
    },
    {
      "epoch": 4.28,
      "eval_loss": 0.27948442101478577,
      "eval_runtime": 349.9899,
      "eval_samples_per_second": 0.286,
      "eval_steps_per_second": 0.286,
      "eval_wer": 59.0146750524109,
      "step": 2000
    }
  ],
  "logging_steps": 20,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 1000,
  "total_flos": 1.73093523111936e+18,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}