trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "eval_cer": 0.4666666666666667,
      "eval_loss": 86.90267944335938,
      "eval_runtime": 17.2823,
      "eval_samples_per_second": 0.116,
      "eval_steps_per_second": 0.058,
      "eval_wer": 1.0,
      "step": 2
    },
    {
      "epoch": 4.0,
      "eval_cer": 0.4666666666666667,
      "eval_loss": 75.15727233886719,
      "eval_runtime": 15.1141,
      "eval_samples_per_second": 0.132,
      "eval_steps_per_second": 0.066,
      "eval_wer": 1.0,
      "step": 4
    },
    {
      "epoch": 6.0,
      "eval_cer": 0.4666666666666667,
      "eval_loss": 73.1053237915039,
      "eval_runtime": 3.6762,
      "eval_samples_per_second": 0.544,
      "eval_steps_per_second": 0.272,
      "eval_wer": 1.0,
      "step": 6
    },
    {
      "epoch": 8.0,
      "eval_cer": 0.4666666666666667,
      "eval_loss": 75.6639404296875,
      "eval_runtime": 3.6084,
      "eval_samples_per_second": 0.554,
      "eval_steps_per_second": 0.277,
      "eval_wer": 1.0,
      "step": 8
    },
    {
      "epoch": 10.0,
      "eval_cer": 0.4666666666666667,
      "eval_loss": 77.0830078125,
      "eval_runtime": 3.2264,
      "eval_samples_per_second": 0.62,
      "eval_steps_per_second": 0.31,
      "eval_wer": 1.0,
      "step": 10
    },
    {
      "epoch": 10.0,
      "step": 10,
      "total_flos": 3043080072115200.0,
      "train_loss": 22.953053283691407,
      "train_runtime": 1186.2575,
      "train_samples_per_second": 0.017,
      "train_steps_per_second": 0.008
    }
  ],
  "max_steps": 10,
  "num_train_epochs": 10,
  "total_flos": 3043080072115200.0,
  "trial_name": null,
  "trial_params": null
}