checkpoint-200/trainer_state.json

{
  "best_metric": 0.5542770299609205,
  "best_model_checkpoint": "./whisper-large-v3-quantized/checkpoint-200",
  "epoch": 0.12292562999385372,
  "eval_steps": 200,
  "global_step": 200,
  "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
    }
  ],
  "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": 7.326498816e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}