{
  "best_global_step": 1500,
  "best_metric": 63.903743315508024,
  "best_model_checkpoint": "./whisper-tiny-maithili/checkpoint-1500",
  "epoch": 13.69982847341338,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17152658662092624,
      "grad_norm": 18.268054962158203,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 9.723193969726562,
      "step": 25
    },
    {
      "epoch": 0.34305317324185247,
      "grad_norm": 13.877284049987793,
      "learning_rate": 2.4500000000000003e-06,
      "loss": 8.318059692382812,
      "step": 50
    },
    {
      "epoch": 0.5145797598627787,
      "grad_norm": 11.479126930236816,
      "learning_rate": 3.7e-06,
      "loss": 6.970803833007812,
      "step": 75
    },
    {
      "epoch": 0.6861063464837049,
      "grad_norm": 10.104531288146973,
      "learning_rate": 4.95e-06,
      "loss": 5.796900634765625,
      "step": 100
    },
    {
      "epoch": 0.8576329331046312,
      "grad_norm": 10.247244834899902,
      "learning_rate": 6.200000000000001e-06,
      "loss": 4.848861389160156,
      "step": 125
    },
    {
      "epoch": 1.0274442538593482,
      "grad_norm": 7.673226356506348,
      "learning_rate": 7.450000000000001e-06,
      "loss": 3.985304870605469,
      "step": 150
    },
    {
      "epoch": 1.1989708404802744,
      "grad_norm": 7.689826488494873,
      "learning_rate": 8.700000000000001e-06,
      "loss": 3.524297180175781,
      "step": 175
    },
    {
      "epoch": 1.3704974271012007,
      "grad_norm": 8.599395751953125,
      "learning_rate": 9.950000000000001e-06,
      "loss": 3.220665283203125,
      "step": 200
    },
    {
      "epoch": 1.542024013722127,
      "grad_norm": 8.532909393310547,
      "learning_rate": 9.866666666666668e-06,
      "loss": 3.0051885986328126,
      "step": 225
    },
    {
      "epoch": 1.7135506003430532,
      "grad_norm": 8.584299087524414,
      "learning_rate": 9.727777777777777e-06,
      "loss": 2.7648504638671874,
      "step": 250
    },
    {
      "epoch": 1.8850771869639793,
      "grad_norm": 8.186919212341309,
      "learning_rate": 9.58888888888889e-06,
      "loss": 2.6434637451171876,
      "step": 275
    },
    {
      "epoch": 2.0548885077186965,
      "grad_norm": 7.06594181060791,
      "learning_rate": 9.450000000000001e-06,
      "loss": 2.485320739746094,
      "step": 300
    },
    {
      "epoch": 2.2264150943396226,
      "grad_norm": 7.2849907875061035,
      "learning_rate": 9.311111111111112e-06,
      "loss": 2.2979925537109374,
      "step": 325
    },
    {
      "epoch": 2.3979416809605487,
      "grad_norm": 7.386153221130371,
      "learning_rate": 9.172222222222223e-06,
      "loss": 2.249031982421875,
      "step": 350
    },
    {
      "epoch": 2.5694682675814753,
      "grad_norm": 8.004015922546387,
      "learning_rate": 9.033333333333334e-06,
      "loss": 2.181380310058594,
      "step": 375
    },
    {
      "epoch": 2.7409948542024014,
      "grad_norm": 6.979090213775635,
      "learning_rate": 8.894444444444445e-06,
      "loss": 2.1656195068359376,
      "step": 400
    },
    {
      "epoch": 2.9125214408233275,
      "grad_norm": 6.589767932891846,
      "learning_rate": 8.755555555555556e-06,
      "loss": 2.2088014221191408,
      "step": 425
    },
    {
      "epoch": 3.0823327615780447,
      "grad_norm": 8.118329048156738,
      "learning_rate": 8.616666666666668e-06,
      "loss": 2.0147331237792967,
      "step": 450
    },
    {
      "epoch": 3.253859348198971,
      "grad_norm": 7.051534652709961,
      "learning_rate": 8.477777777777778e-06,
      "loss": 1.896328887939453,
      "step": 475
    },
    {
      "epoch": 3.425385934819897,
      "grad_norm": 6.275884628295898,
      "learning_rate": 8.33888888888889e-06,
      "loss": 1.886772918701172,
      "step": 500
    },
    {
      "epoch": 3.425385934819897,
      "eval_loss": 0.5728768110275269,
      "eval_runtime": 92.9426,
      "eval_samples_per_second": 2.797,
      "eval_steps_per_second": 0.699,
      "eval_wer": 68.77599524658348,
      "step": 500
    },
    {
      "epoch": 3.5969125214408235,
      "grad_norm": 7.222710609436035,
      "learning_rate": 8.2e-06,
      "loss": 1.8691773986816407,
      "step": 525
    },
    {
      "epoch": 3.7684391080617496,
      "grad_norm": 6.063198566436768,
      "learning_rate": 8.061111111111112e-06,
      "loss": 1.8207814025878906,
      "step": 550
    },
    {
      "epoch": 3.9399656946826758,
      "grad_norm": 6.641396522521973,
      "learning_rate": 7.922222222222223e-06,
      "loss": 1.7857606506347656,
      "step": 575
    },
    {
      "epoch": 4.109777015437393,
      "grad_norm": 6.6834397315979,
      "learning_rate": 7.783333333333334e-06,
      "loss": 1.713679656982422,
      "step": 600
    },
    {
      "epoch": 4.281303602058319,
      "grad_norm": 6.873757362365723,
      "learning_rate": 7.644444444444445e-06,
      "loss": 1.6278848266601562,
      "step": 625
    },
    {
      "epoch": 4.452830188679245,
      "grad_norm": 6.53254508972168,
      "learning_rate": 7.505555555555556e-06,
      "loss": 1.5813905334472655,
      "step": 650
    },
    {
      "epoch": 4.624356775300171,
      "grad_norm": 7.650671482086182,
      "learning_rate": 7.3666666666666676e-06,
      "loss": 1.6567597961425782,
      "step": 675
    },
    {
      "epoch": 4.795883361921097,
      "grad_norm": 6.784526348114014,
      "learning_rate": 7.227777777777778e-06,
      "loss": 1.6283486938476563,
      "step": 700
    },
    {
      "epoch": 4.967409948542024,
      "grad_norm": 7.425609588623047,
      "learning_rate": 7.0888888888888894e-06,
      "loss": 1.580086669921875,
      "step": 725
    },
    {
      "epoch": 5.137221269296741,
      "grad_norm": 7.560032367706299,
      "learning_rate": 6.95e-06,
      "loss": 1.4196165466308595,
      "step": 750
    },
    {
      "epoch": 5.308747855917667,
      "grad_norm": 7.114623069763184,
      "learning_rate": 6.811111111111111e-06,
      "loss": 1.4393228149414063,
      "step": 775
    },
    {
      "epoch": 5.480274442538594,
      "grad_norm": 6.174019813537598,
      "learning_rate": 6.672222222222223e-06,
      "loss": 1.4180364990234375,
      "step": 800
    },
    {
      "epoch": 5.65180102915952,
      "grad_norm": 6.444286346435547,
      "learning_rate": 6.533333333333334e-06,
      "loss": 1.4367037963867189,
      "step": 825
    },
    {
      "epoch": 5.823327615780446,
      "grad_norm": 6.457458019256592,
      "learning_rate": 6.394444444444445e-06,
      "loss": 1.4541001892089844,
      "step": 850
    },
    {
      "epoch": 5.994854202401372,
      "grad_norm": 6.65748929977417,
      "learning_rate": 6.255555555555556e-06,
      "loss": 1.4775286865234376,
      "step": 875
    },
    {
      "epoch": 6.164665523156089,
      "grad_norm": 6.2816057205200195,
      "learning_rate": 6.116666666666668e-06,
      "loss": 1.2730085754394531,
      "step": 900
    },
    {
      "epoch": 6.3361921097770155,
      "grad_norm": 6.40878963470459,
      "learning_rate": 5.977777777777778e-06,
      "loss": 1.2837075805664062,
      "step": 925
    },
    {
      "epoch": 6.507718696397942,
      "grad_norm": 6.427629470825195,
      "learning_rate": 5.8388888888888895e-06,
      "loss": 1.2479456329345704,
      "step": 950
    },
    {
      "epoch": 6.679245283018868,
      "grad_norm": 5.938961982727051,
      "learning_rate": 5.7e-06,
      "loss": 1.3641961669921876,
      "step": 975
    },
    {
      "epoch": 6.850771869639794,
      "grad_norm": 6.981799125671387,
      "learning_rate": 5.561111111111111e-06,
      "loss": 1.2829299926757813,
      "step": 1000
    },
    {
      "epoch": 6.850771869639794,
      "eval_loss": 0.5365099906921387,
      "eval_runtime": 92.6416,
      "eval_samples_per_second": 2.807,
      "eval_steps_per_second": 0.702,
      "eval_wer": 64.67617349970291,
      "step": 1000
    },
    {
      "epoch": 7.020583190394511,
      "grad_norm": 6.462595462799072,
      "learning_rate": 5.422222222222223e-06,
      "loss": 1.266632080078125,
      "step": 1025
    },
    {
      "epoch": 7.192109777015437,
      "grad_norm": 6.442847728729248,
      "learning_rate": 5.283333333333333e-06,
      "loss": 1.1616410827636718,
      "step": 1050
    },
    {
      "epoch": 7.363636363636363,
      "grad_norm": 5.910035133361816,
      "learning_rate": 5.144444444444445e-06,
      "loss": 1.1423403167724608,
      "step": 1075
    },
    {
      "epoch": 7.535162950257289,
      "grad_norm": 6.2111334800720215,
      "learning_rate": 5.005555555555556e-06,
      "loss": 1.1994317626953126,
      "step": 1100
    },
    {
      "epoch": 7.7066895368782165,
      "grad_norm": 5.6760783195495605,
      "learning_rate": 4.866666666666667e-06,
      "loss": 1.1873680877685546,
      "step": 1125
    },
    {
      "epoch": 7.878216123499143,
      "grad_norm": 6.996194839477539,
      "learning_rate": 4.727777777777779e-06,
      "loss": 1.1682706451416016,
      "step": 1150
    },
    {
      "epoch": 8.04802744425386,
      "grad_norm": 5.388121604919434,
      "learning_rate": 4.58888888888889e-06,
      "loss": 1.1501982879638672,
      "step": 1175
    },
    {
      "epoch": 8.219554030874786,
      "grad_norm": 6.324208736419678,
      "learning_rate": 4.450000000000001e-06,
      "loss": 1.0414784240722657,
      "step": 1200
    },
    {
      "epoch": 8.391080617495712,
      "grad_norm": 6.224495887756348,
      "learning_rate": 4.3111111111111115e-06,
      "loss": 1.0576978302001954,
      "step": 1225
    },
    {
      "epoch": 8.562607204116638,
      "grad_norm": 5.9399094581604,
      "learning_rate": 4.1722222222222225e-06,
      "loss": 1.0483750915527343,
      "step": 1250
    },
    {
      "epoch": 8.734133790737564,
      "grad_norm": 6.331912994384766,
      "learning_rate": 4.033333333333333e-06,
      "loss": 1.0921112060546876,
      "step": 1275
    },
    {
      "epoch": 8.90566037735849,
      "grad_norm": 7.319565296173096,
      "learning_rate": 3.894444444444444e-06,
      "loss": 1.0816537475585937,
      "step": 1300
    },
    {
      "epoch": 9.075471698113208,
      "grad_norm": 6.406201362609863,
      "learning_rate": 3.7555555555555557e-06,
      "loss": 1.0307456207275392,
      "step": 1325
    },
    {
      "epoch": 9.246998284734135,
      "grad_norm": 6.348610877990723,
      "learning_rate": 3.616666666666667e-06,
      "loss": 0.9632396697998047,
      "step": 1350
    },
    {
      "epoch": 9.41852487135506,
      "grad_norm": 6.033849239349365,
      "learning_rate": 3.4777777777777784e-06,
      "loss": 0.9932768249511719,
      "step": 1375
    },
    {
      "epoch": 9.590051457975987,
      "grad_norm": 6.423510551452637,
      "learning_rate": 3.3388888888888893e-06,
      "loss": 0.978936538696289,
      "step": 1400
    },
    {
      "epoch": 9.761578044596913,
      "grad_norm": 5.581418514251709,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.9529872894287109,
      "step": 1425
    },
    {
      "epoch": 9.933104631217839,
      "grad_norm": 7.128633499145508,
      "learning_rate": 3.0611111111111112e-06,
      "loss": 1.00614990234375,
      "step": 1450
    },
    {
      "epoch": 10.102915951972555,
      "grad_norm": 6.873051643371582,
      "learning_rate": 2.9222222222222226e-06,
      "loss": 0.9351605224609375,
      "step": 1475
    },
    {
      "epoch": 10.274442538593481,
      "grad_norm": 6.594143390655518,
      "learning_rate": 2.7833333333333335e-06,
      "loss": 0.8834098815917969,
      "step": 1500
    },
    {
      "epoch": 10.274442538593481,
      "eval_loss": 0.5569483637809753,
      "eval_runtime": 93.3641,
      "eval_samples_per_second": 2.785,
      "eval_steps_per_second": 0.696,
      "eval_wer": 63.903743315508024,
      "step": 1500
    },
    {
      "epoch": 10.445969125214408,
      "grad_norm": 6.543798446655273,
      "learning_rate": 2.6444444444444444e-06,
      "loss": 0.9357952880859375,
      "step": 1525
    },
    {
      "epoch": 10.617495711835334,
      "grad_norm": 5.903841018676758,
      "learning_rate": 2.5055555555555554e-06,
      "loss": 0.8977721405029296,
      "step": 1550
    },
    {
      "epoch": 10.78902229845626,
      "grad_norm": 5.874545574188232,
      "learning_rate": 2.3666666666666667e-06,
      "loss": 0.8883360290527343,
      "step": 1575
    },
    {
      "epoch": 10.960548885077188,
      "grad_norm": 6.661474227905273,
      "learning_rate": 2.2277777777777777e-06,
      "loss": 0.920865249633789,
      "step": 1600
    },
    {
      "epoch": 11.130360205831904,
      "grad_norm": 5.809503555297852,
      "learning_rate": 2.088888888888889e-06,
      "loss": 0.8493796539306641,
      "step": 1625
    },
    {
      "epoch": 11.30188679245283,
      "grad_norm": 5.753544807434082,
      "learning_rate": 1.9500000000000004e-06,
      "loss": 0.825905532836914,
      "step": 1650
    },
    {
      "epoch": 11.473413379073756,
      "grad_norm": 6.0916972160339355,
      "learning_rate": 1.8111111111111113e-06,
      "loss": 0.8476572418212891,
      "step": 1675
    },
    {
      "epoch": 11.644939965694682,
      "grad_norm": 6.63301420211792,
      "learning_rate": 1.6722222222222223e-06,
      "loss": 0.8341015625,
      "step": 1700
    },
    {
      "epoch": 11.816466552315608,
      "grad_norm": 5.515976428985596,
      "learning_rate": 1.5333333333333334e-06,
      "loss": 0.8632970428466797,
      "step": 1725
    },
    {
      "epoch": 11.987993138936535,
      "grad_norm": 5.419062614440918,
      "learning_rate": 1.3944444444444446e-06,
      "loss": 0.8705248260498046,
      "step": 1750
    },
    {
      "epoch": 12.157804459691253,
      "grad_norm": 5.655745506286621,
      "learning_rate": 1.2555555555555557e-06,
      "loss": 0.7931866455078125,
      "step": 1775
    },
    {
      "epoch": 12.329331046312179,
      "grad_norm": 5.437541961669922,
      "learning_rate": 1.1166666666666666e-06,
      "loss": 0.813487548828125,
      "step": 1800
    },
    {
      "epoch": 12.500857632933105,
      "grad_norm": 5.816484451293945,
      "learning_rate": 9.77777777777778e-07,
      "loss": 0.8068718719482422,
      "step": 1825
    },
    {
      "epoch": 12.672384219554031,
      "grad_norm": 5.428255081176758,
      "learning_rate": 8.388888888888889e-07,
      "loss": 0.7925386810302735,
      "step": 1850
    },
    {
      "epoch": 12.843910806174957,
      "grad_norm": 5.579376220703125,
      "learning_rate": 7.000000000000001e-07,
      "loss": 0.8237639617919922,
      "step": 1875
    },
    {
      "epoch": 13.013722126929673,
      "grad_norm": 5.45307731628418,
      "learning_rate": 5.611111111111111e-07,
      "loss": 0.7975894165039062,
      "step": 1900
    },
    {
      "epoch": 13.1852487135506,
      "grad_norm": 5.217875003814697,
      "learning_rate": 4.2222222222222226e-07,
      "loss": 0.8227002716064453,
      "step": 1925
    },
    {
      "epoch": 13.356775300171527,
      "grad_norm": 5.0090155601501465,
      "learning_rate": 2.8333333333333336e-07,
      "loss": 0.7616297149658203,
      "step": 1950
    },
    {
      "epoch": 13.528301886792454,
      "grad_norm": 5.426841735839844,
      "learning_rate": 1.4444444444444445e-07,
      "loss": 0.7497710418701172,
      "step": 1975
    },
    {
      "epoch": 13.69982847341338,
      "grad_norm": 5.744300365447998,
      "learning_rate": 5.555555555555556e-09,
      "loss": 0.8125200653076172,
      "step": 2000
    },
    {
      "epoch": 13.69982847341338,
      "eval_loss": 0.5726562738418579,
      "eval_runtime": 93.1097,
      "eval_samples_per_second": 2.792,
      "eval_steps_per_second": 0.698,
      "eval_wer": 64.20083184789067,
      "step": 2000
    }
  ],
  "logging_steps": 25,
  "max_steps": 2000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 14,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.8652414181376e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}