trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.970760233918129,
  "eval_steps": 250,
  "global_step": 850,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.46,
      "learning_rate": 1.5e-05,
      "loss": 0.8043,
      "step": 250
    },
    {
      "epoch": 1.46,
      "eval_cer": 0.19486492447697548,
      "eval_loss": 0.5256313681602478,
      "eval_runtime": 226.5238,
      "eval_samples_per_second": 14.586,
      "eval_steps_per_second": 1.823,
      "eval_wer": 0.4023366474525127,
      "step": 250
    },
    {
      "epoch": 2.92,
      "learning_rate": 3e-05,
      "loss": 0.5435,
      "step": 500
    },
    {
      "epoch": 2.92,
      "eval_cer": 0.1960657773275714,
      "eval_loss": 0.4380650520324707,
      "eval_runtime": 227.8338,
      "eval_samples_per_second": 14.502,
      "eval_steps_per_second": 1.813,
      "eval_wer": 0.3965337150373322,
      "step": 500
    },
    {
      "epoch": 4.39,
      "learning_rate": 8.873239436619718e-06,
      "loss": 0.4827,
      "step": 750
    },
    {
      "epoch": 4.39,
      "eval_cer": 0.193239280141815,
      "eval_loss": 0.4322899878025055,
      "eval_runtime": 230.4509,
      "eval_samples_per_second": 14.337,
      "eval_steps_per_second": 1.792,
      "eval_wer": 0.398777515571202,
      "step": 750
    }
  ],
  "logging_steps": 250,
  "max_steps": 855,
  "num_train_epochs": 5,
  "save_steps": 850,
  "total_flos": 2.8872999396360806e+18,
  "trial_name": null,
  "trial_params": null
}