{
  "best_global_step": 600,
  "best_metric": 0.5363820680333409,
  "best_model_checkpoint": "./whisper-small-sw/checkpoint-600",
  "epoch": 2.0,
  "eval_steps": 100,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 4503124.5,
      "learning_rate": 4.800000000000001e-07,
      "loss": 4.1558,
      "step": 25
    },
    {
      "epoch": 0.1,
      "grad_norm": 1575086.625,
      "learning_rate": 9.800000000000001e-07,
      "loss": 3.3557,
      "step": 50
    },
    {
      "epoch": 0.15,
      "grad_norm": 1165318.0,
      "learning_rate": 1.48e-06,
      "loss": 2.4313,
      "step": 75
    },
    {
      "epoch": 0.2,
      "grad_norm": 1055806.75,
      "learning_rate": 1.98e-06,
      "loss": 1.9578,
      "step": 100
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.8956851959228516,
      "eval_runtime": 969.757,
      "eval_samples_per_second": 2.062,
      "eval_steps_per_second": 0.172,
      "eval_wer": 0.8697341743635955,
      "step": 100
    },
    {
      "epoch": 0.25,
      "grad_norm": 1095035.375,
      "learning_rate": 2.4800000000000004e-06,
      "loss": 1.8414,
      "step": 125
    },
    {
      "epoch": 0.3,
      "grad_norm": 1035096.4375,
      "learning_rate": 2.9800000000000003e-06,
      "loss": 1.6974,
      "step": 150
    },
    {
      "epoch": 0.35,
      "grad_norm": 1070272.75,
      "learning_rate": 3.48e-06,
      "loss": 1.621,
      "step": 175
    },
    {
      "epoch": 0.4,
      "grad_norm": 1132629.5,
      "learning_rate": 3.980000000000001e-06,
      "loss": 1.4927,
      "step": 200
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.4295355081558228,
      "eval_runtime": 924.4171,
      "eval_samples_per_second": 2.164,
      "eval_steps_per_second": 0.181,
      "eval_wer": 0.6874296012615454,
      "step": 200
    },
    {
      "epoch": 0.45,
      "grad_norm": 960689.375,
      "learning_rate": 4.48e-06,
      "loss": 1.3545,
      "step": 225
    },
    {
      "epoch": 0.5,
      "grad_norm": 920681.5,
      "learning_rate": 4.980000000000001e-06,
      "loss": 1.2992,
      "step": 250
    },
    {
      "epoch": 0.55,
      "grad_norm": 765348.0625,
      "learning_rate": 5.480000000000001e-06,
      "loss": 1.2236,
      "step": 275
    },
    {
      "epoch": 0.6,
      "grad_norm": 810740.3125,
      "learning_rate": 5.98e-06,
      "loss": 1.1184,
      "step": 300
    },
    {
      "epoch": 0.6,
      "eval_loss": 1.0826019048690796,
      "eval_runtime": 1007.0686,
      "eval_samples_per_second": 1.986,
      "eval_steps_per_second": 0.166,
      "eval_wer": 0.6485131786438387,
      "step": 300
    },
    {
      "epoch": 0.65,
      "grad_norm": 945555.75,
      "learning_rate": 6.480000000000001e-06,
      "loss": 1.0226,
      "step": 325
    },
    {
      "epoch": 0.7,
      "grad_norm": 629258.5625,
      "learning_rate": 6.98e-06,
      "loss": 0.9098,
      "step": 350
    },
    {
      "epoch": 0.75,
      "grad_norm": 720113.9375,
      "learning_rate": 7.48e-06,
      "loss": 0.848,
      "step": 375
    },
    {
      "epoch": 0.8,
      "grad_norm": 727505.875,
      "learning_rate": 7.980000000000002e-06,
      "loss": 0.7788,
      "step": 400
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.7836961150169373,
      "eval_runtime": 1040.1837,
      "eval_samples_per_second": 1.923,
      "eval_steps_per_second": 0.161,
      "eval_wer": 0.6084703762108583,
      "step": 400
    },
    {
      "epoch": 0.85,
      "grad_norm": 745182.125,
      "learning_rate": 8.48e-06,
      "loss": 0.7841,
      "step": 425
    },
    {
      "epoch": 0.9,
      "grad_norm": 624584.875,
      "learning_rate": 8.98e-06,
      "loss": 0.7609,
      "step": 450
    },
    {
      "epoch": 0.95,
      "grad_norm": 604703.5625,
      "learning_rate": 9.48e-06,
      "loss": 0.7249,
      "step": 475
    },
    {
      "epoch": 1.0,
      "grad_norm": 658865.0,
      "learning_rate": 9.980000000000001e-06,
      "loss": 0.6966,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7045486569404602,
      "eval_runtime": 1097.0835,
      "eval_samples_per_second": 1.823,
      "eval_steps_per_second": 0.152,
      "eval_wer": 0.688387024104528,
      "step": 500
    },
    {
      "epoch": 1.05,
      "grad_norm": 449478.71875,
      "learning_rate": 9.52e-06,
      "loss": 0.5287,
      "step": 525
    },
    {
      "epoch": 1.1,
      "grad_norm": 490002.71875,
      "learning_rate": 9.020000000000002e-06,
      "loss": 0.5736,
      "step": 550
    },
    {
      "epoch": 1.15,
      "grad_norm": 611781.625,
      "learning_rate": 8.52e-06,
      "loss": 0.5479,
      "step": 575
    },
    {
      "epoch": 1.2,
      "grad_norm": 440316.34375,
      "learning_rate": 8.020000000000001e-06,
      "loss": 0.5242,
      "step": 600
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.6567466855049133,
      "eval_runtime": 1001.8554,
      "eval_samples_per_second": 1.996,
      "eval_steps_per_second": 0.167,
      "eval_wer": 0.5363820680333409,
      "step": 600
    },
    {
      "epoch": 1.25,
      "grad_norm": 633716.0,
      "learning_rate": 7.520000000000001e-06,
      "loss": 0.5501,
      "step": 625
    },
    {
      "epoch": 1.3,
      "grad_norm": 482948.25,
      "learning_rate": 7.0200000000000006e-06,
      "loss": 0.5189,
      "step": 650
    },
    {
      "epoch": 1.35,
      "grad_norm": 623572.875,
      "learning_rate": 6.520000000000001e-06,
      "loss": 0.5392,
      "step": 675
    },
    {
      "epoch": 1.4,
      "grad_norm": 651482.125,
      "learning_rate": 6.02e-06,
      "loss": 0.506,
      "step": 700
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.6255258321762085,
      "eval_runtime": 1163.0764,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 0.144,
      "eval_wer": 0.6922730344672223,
      "step": 700
    },
    {
      "epoch": 1.45,
      "grad_norm": 462773.625,
      "learning_rate": 5.5200000000000005e-06,
      "loss": 0.4503,
      "step": 725
    },
    {
      "epoch": 1.5,
      "grad_norm": 498840.125,
      "learning_rate": 5.02e-06,
      "loss": 0.4676,
      "step": 750
    },
    {
      "epoch": 1.55,
      "grad_norm": 528649.8125,
      "learning_rate": 4.520000000000001e-06,
      "loss": 0.4561,
      "step": 775
    },
    {
      "epoch": 1.6,
      "grad_norm": 418120.5625,
      "learning_rate": 4.0200000000000005e-06,
      "loss": 0.4064,
      "step": 800
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.6007899641990662,
      "eval_runtime": 1235.1962,
      "eval_samples_per_second": 1.619,
      "eval_steps_per_second": 0.135,
      "eval_wer": 0.7322031989186754,
      "step": 800
    },
    {
      "epoch": 1.65,
      "grad_norm": 597861.5,
      "learning_rate": 3.52e-06,
      "loss": 0.4127,
      "step": 825
    },
    {
      "epoch": 1.7,
      "grad_norm": 377506.28125,
      "learning_rate": 3.0200000000000003e-06,
      "loss": 0.3911,
      "step": 850
    },
    {
      "epoch": 1.75,
      "grad_norm": 406077.25,
      "learning_rate": 2.52e-06,
      "loss": 0.3828,
      "step": 875
    },
    {
      "epoch": 1.8,
      "grad_norm": 495271.09375,
      "learning_rate": 2.02e-06,
      "loss": 0.3368,
      "step": 900
    },
    {
      "epoch": 1.8,
      "eval_loss": 0.5886157155036926,
      "eval_runtime": 1612.2956,
      "eval_samples_per_second": 1.24,
      "eval_steps_per_second": 0.104,
      "eval_wer": 0.9467785537283172,
      "step": 900
    },
    {
      "epoch": 1.85,
      "grad_norm": 552985.125,
      "learning_rate": 1.52e-06,
      "loss": 0.3455,
      "step": 925
    },
    {
      "epoch": 1.9,
      "grad_norm": 366518.8125,
      "learning_rate": 1.02e-06,
      "loss": 0.3354,
      "step": 950
    },
    {
      "epoch": 1.95,
      "grad_norm": 405846.40625,
      "learning_rate": 5.2e-07,
      "loss": 0.3129,
      "step": 975
    },
    {
      "epoch": 2.0,
      "grad_norm": 414410.40625,
      "learning_rate": 2e-08,
      "loss": 0.2974,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5830898880958557,
      "eval_runtime": 1396.4972,
      "eval_samples_per_second": 1.432,
      "eval_steps_per_second": 0.12,
      "eval_wer": 0.7279229556206352,
      "step": 1000
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.61736640512e+18,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}