checkpoint-600/trainer_state.json

{
  "best_metric": 0.5542770299609205,
  "best_model_checkpoint": "./whisper-large-v3-quantized/checkpoint-200",
  "epoch": 0.36877688998156116,
  "eval_steps": 200,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015365703749231715,
      "grad_norm": 0.47920289635658264,
      "learning_rate": 5.000000000000001e-07,
      "loss": 1.5655,
      "step": 25
    },
    {
      "epoch": 0.03073140749846343,
      "grad_norm": 0.15030227601528168,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.0487,
      "step": 50
    },
    {
      "epoch": 0.046097111247695145,
      "grad_norm": 0.5764341354370117,
      "learning_rate": 1.5e-06,
      "loss": 1.6787,
      "step": 75
    },
    {
      "epoch": 0.06146281499692686,
      "grad_norm": 0.6326006650924683,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.3895,
      "step": 100
    },
    {
      "epoch": 0.07682851874615858,
      "grad_norm": 0.5107858180999756,
      "learning_rate": 2.5e-06,
      "loss": 1.5308,
      "step": 125
    },
    {
      "epoch": 0.09219422249539029,
      "grad_norm": 0.4186854362487793,
      "learning_rate": 3e-06,
      "loss": 1.1231,
      "step": 150
    },
    {
      "epoch": 0.107559926244622,
      "grad_norm": 0.580566942691803,
      "learning_rate": 3.5e-06,
      "loss": 1.8924,
      "step": 175
    },
    {
      "epoch": 0.12292562999385372,
      "grad_norm": 0.3107517957687378,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3077,
      "step": 200
    },
    {
      "epoch": 0.12292562999385372,
      "eval_loss": 1.5075420141220093,
      "eval_runtime": 620.2783,
      "eval_samples_per_second": 0.656,
      "eval_steps_per_second": 0.082,
      "eval_wer": 0.5542770299609205,
      "step": 200
    },
    {
      "epoch": 0.13829133374308544,
      "grad_norm": 0.6910067796707153,
      "learning_rate": 4.5e-06,
      "loss": 1.8306,
      "step": 225
    },
    {
      "epoch": 0.15365703749231716,
      "grad_norm": 0.04023272544145584,
      "learning_rate": 5e-06,
      "loss": 1.5791,
      "step": 250
    },
    {
      "epoch": 0.16902274124154887,
      "grad_norm": 0.700638473033905,
      "learning_rate": 5.500000000000001e-06,
      "loss": 1.4698,
      "step": 275
    },
    {
      "epoch": 0.18438844499078058,
      "grad_norm": 0.697817325592041,
      "learning_rate": 6e-06,
      "loss": 1.7872,
      "step": 300
    },
    {
      "epoch": 0.1997541487400123,
      "grad_norm": 0.5358330011367798,
      "learning_rate": 6.5000000000000004e-06,
      "loss": 1.1349,
      "step": 325
    },
    {
      "epoch": 0.215119852489244,
      "grad_norm": 0.7219414710998535,
      "learning_rate": 7e-06,
      "loss": 1.4458,
      "step": 350
    },
    {
      "epoch": 0.23048555623847572,
      "grad_norm": 0.5983926653862,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.3574,
      "step": 375
    },
    {
      "epoch": 0.24585125998770743,
      "grad_norm": 1.1959525346755981,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.4669,
      "step": 400
    },
    {
      "epoch": 0.24585125998770743,
      "eval_loss": 1.4504855871200562,
      "eval_runtime": 634.3044,
      "eval_samples_per_second": 0.642,
      "eval_steps_per_second": 0.08,
      "eval_wer": 0.5688232739904473,
      "step": 400
    },
    {
      "epoch": 0.26121696373693915,
      "grad_norm": 0.9689226746559143,
      "learning_rate": 8.5e-06,
      "loss": 1.2762,
      "step": 425
    },
    {
      "epoch": 0.2765826674861709,
      "grad_norm": 0.2560756504535675,
      "learning_rate": 9e-06,
      "loss": 1.2975,
      "step": 450
    },
    {
      "epoch": 0.29194837123540257,
      "grad_norm": 1.208335518836975,
      "learning_rate": 9.5e-06,
      "loss": 1.3111,
      "step": 475
    },
    {
      "epoch": 0.3073140749846343,
      "grad_norm": 0.11639931052923203,
      "learning_rate": 1e-05,
      "loss": 1.2754,
      "step": 500
    },
    {
      "epoch": 0.322679778733866,
      "grad_norm": 3.112171173095703,
      "learning_rate": 9.5e-06,
      "loss": 1.4163,
      "step": 525
    },
    {
      "epoch": 0.33804548248309774,
      "grad_norm": 1.0575857162475586,
      "learning_rate": 9e-06,
      "loss": 1.1581,
      "step": 550
    },
    {
      "epoch": 0.3534111862323294,
      "grad_norm": 1.2147361040115356,
      "learning_rate": 8.5e-06,
      "loss": 1.1211,
      "step": 575
    },
    {
      "epoch": 0.36877688998156116,
      "grad_norm": 3.342254161834717,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.7042,
      "step": 600
    },
    {
      "epoch": 0.36877688998156116,
      "eval_loss": 1.2238162755966187,
      "eval_runtime": 654.1777,
      "eval_samples_per_second": 0.622,
      "eval_steps_per_second": 0.078,
      "eval_wer": 0.5605731654363874,
      "step": 600
    }
  ],
  "logging_steps": 25,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 3,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1979496448e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}