{
  "best_metric": 60.21685813863431,
  "best_model_checkpoint": "./whisper-small-taiwanese/checkpoint-1000",
  "epoch": 0.6451612903225806,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016129032258064516,
      "grad_norm": 241.39755249023438,
      "learning_rate": 5.376344086021506e-07,
      "loss": 8.0646,
      "step": 25
    },
    {
      "epoch": 0.03225806451612903,
      "grad_norm": 52.91600799560547,
      "learning_rate": 1.0752688172043011e-06,
      "loss": 5.6903,
      "step": 50
    },
    {
      "epoch": 0.04838709677419355,
      "grad_norm": 32.09747314453125,
      "learning_rate": 1.6129032258064516e-06,
      "loss": 3.6353,
      "step": 75
    },
    {
      "epoch": 0.06451612903225806,
      "grad_norm": 31.451000213623047,
      "learning_rate": 2.1505376344086023e-06,
      "loss": 2.6364,
      "step": 100
    },
    {
      "epoch": 0.08064516129032258,
      "grad_norm": 29.471986770629883,
      "learning_rate": 2.688172043010753e-06,
      "loss": 2.3125,
      "step": 125
    },
    {
      "epoch": 0.0967741935483871,
      "grad_norm": 28.64345932006836,
      "learning_rate": 3.225806451612903e-06,
      "loss": 2.1281,
      "step": 150
    },
    {
      "epoch": 0.11290322580645161,
      "grad_norm": 28.750173568725586,
      "learning_rate": 3.763440860215054e-06,
      "loss": 1.9073,
      "step": 175
    },
    {
      "epoch": 0.12903225806451613,
      "grad_norm": 23.051420211791992,
      "learning_rate": 4.3010752688172045e-06,
      "loss": 1.5977,
      "step": 200
    },
    {
      "epoch": 0.14516129032258066,
      "grad_norm": 18.67135238647461,
      "learning_rate": 4.838709677419355e-06,
      "loss": 1.5081,
      "step": 225
    },
    {
      "epoch": 0.16129032258064516,
      "grad_norm": 15.335652351379395,
      "learning_rate": 5.376344086021506e-06,
      "loss": 1.4169,
      "step": 250
    },
    {
      "epoch": 0.1774193548387097,
      "grad_norm": 16.2917537689209,
      "learning_rate": 5.9139784946236566e-06,
      "loss": 1.3469,
      "step": 275
    },
    {
      "epoch": 0.1935483870967742,
      "grad_norm": 15.212031364440918,
      "learning_rate": 6.451612903225806e-06,
      "loss": 1.4059,
      "step": 300
    },
    {
      "epoch": 0.20967741935483872,
      "grad_norm": 15.661399841308594,
      "learning_rate": 6.989247311827958e-06,
      "loss": 1.333,
      "step": 325
    },
    {
      "epoch": 0.22580645161290322,
      "grad_norm": 16.841798782348633,
      "learning_rate": 7.526881720430108e-06,
      "loss": 1.2252,
      "step": 350
    },
    {
      "epoch": 0.24193548387096775,
      "grad_norm": 17.468032836914062,
      "learning_rate": 8.064516129032258e-06,
      "loss": 1.2996,
      "step": 375
    },
    {
      "epoch": 0.25806451612903225,
      "grad_norm": 16.684844970703125,
      "learning_rate": 8.602150537634409e-06,
      "loss": 1.2653,
      "step": 400
    },
    {
      "epoch": 0.27419354838709675,
      "grad_norm": 14.749136924743652,
      "learning_rate": 9.13978494623656e-06,
      "loss": 1.1967,
      "step": 425
    },
    {
      "epoch": 0.2903225806451613,
      "grad_norm": 13.751141548156738,
      "learning_rate": 9.67741935483871e-06,
      "loss": 1.1865,
      "step": 450
    },
    {
      "epoch": 0.3064516129032258,
      "grad_norm": 16.48873519897461,
      "learning_rate": 9.97610513739546e-06,
      "loss": 1.1636,
      "step": 475
    },
    {
      "epoch": 0.3225806451612903,
      "grad_norm": 14.694608688354492,
      "learning_rate": 9.916367980884111e-06,
      "loss": 1.1796,
      "step": 500
    },
    {
      "epoch": 0.3387096774193548,
      "grad_norm": 15.619414329528809,
      "learning_rate": 9.856630824372761e-06,
      "loss": 1.1655,
      "step": 525
    },
    {
      "epoch": 0.3548387096774194,
      "grad_norm": 13.177242279052734,
      "learning_rate": 9.79689366786141e-06,
      "loss": 1.143,
      "step": 550
    },
    {
      "epoch": 0.3709677419354839,
      "grad_norm": 15.957605361938477,
      "learning_rate": 9.737156511350062e-06,
      "loss": 1.1414,
      "step": 575
    },
    {
      "epoch": 0.3870967741935484,
      "grad_norm": 12.467620849609375,
      "learning_rate": 9.67741935483871e-06,
      "loss": 1.0964,
      "step": 600
    },
    {
      "epoch": 0.4032258064516129,
      "grad_norm": 15.435978889465332,
      "learning_rate": 9.61768219832736e-06,
      "loss": 1.1512,
      "step": 625
    },
    {
      "epoch": 0.41935483870967744,
      "grad_norm": 13.087624549865723,
      "learning_rate": 9.557945041816011e-06,
      "loss": 1.1338,
      "step": 650
    },
    {
      "epoch": 0.43548387096774194,
      "grad_norm": 15.716456413269043,
      "learning_rate": 9.49820788530466e-06,
      "loss": 1.0783,
      "step": 675
    },
    {
      "epoch": 0.45161290322580644,
      "grad_norm": 14.517507553100586,
      "learning_rate": 9.43847072879331e-06,
      "loss": 1.0728,
      "step": 700
    },
    {
      "epoch": 0.46774193548387094,
      "grad_norm": 17.37009620666504,
      "learning_rate": 9.37873357228196e-06,
      "loss": 1.0317,
      "step": 725
    },
    {
      "epoch": 0.4838709677419355,
      "grad_norm": 14.03701400756836,
      "learning_rate": 9.31899641577061e-06,
      "loss": 1.0347,
      "step": 750
    },
    {
      "epoch": 0.5,
      "grad_norm": 12.431659698486328,
      "learning_rate": 9.25925925925926e-06,
      "loss": 1.0524,
      "step": 775
    },
    {
      "epoch": 0.5161290322580645,
      "grad_norm": 12.746413230895996,
      "learning_rate": 9.19952210274791e-06,
      "loss": 1.0826,
      "step": 800
    },
    {
      "epoch": 0.532258064516129,
      "grad_norm": 15.521408081054688,
      "learning_rate": 9.13978494623656e-06,
      "loss": 1.0377,
      "step": 825
    },
    {
      "epoch": 0.5483870967741935,
      "grad_norm": 15.342901229858398,
      "learning_rate": 9.08004778972521e-06,
      "loss": 0.9762,
      "step": 850
    },
    {
      "epoch": 0.5645161290322581,
      "grad_norm": 16.137371063232422,
      "learning_rate": 9.02031063321386e-06,
      "loss": 1.0725,
      "step": 875
    },
    {
      "epoch": 0.5806451612903226,
      "grad_norm": 14.61146068572998,
      "learning_rate": 8.96057347670251e-06,
      "loss": 0.9554,
      "step": 900
    },
    {
      "epoch": 0.5967741935483871,
      "grad_norm": 13.561723709106445,
      "learning_rate": 8.90083632019116e-06,
      "loss": 1.0127,
      "step": 925
    },
    {
      "epoch": 0.6129032258064516,
      "grad_norm": 16.037729263305664,
      "learning_rate": 8.84109916367981e-06,
      "loss": 0.9621,
      "step": 950
    },
    {
      "epoch": 0.6290322580645161,
      "grad_norm": 13.945268630981445,
      "learning_rate": 8.78136200716846e-06,
      "loss": 0.9479,
      "step": 975
    },
    {
      "epoch": 0.6451612903225806,
      "grad_norm": 15.826567649841309,
      "learning_rate": 8.72162485065711e-06,
      "loss": 0.9789,
      "step": 1000
    },
    {
      "epoch": 0.6451612903225806,
      "eval_cer": 60.21685813863431,
      "eval_loss": 0.9020848870277405,
      "eval_runtime": 953.7359,
      "eval_samples_per_second": 2.392,
      "eval_steps_per_second": 0.3,
      "step": 1000
    }
  ],
  "logging_steps": 25,
  "max_steps": 4650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 1000,
  "total_flos": 4.61736640512e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}