trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 14.981818181818182,
  "global_step": 405,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.69,
      "eval_loss": 3.536450147628784,
      "eval_runtime": 68.3549,
      "eval_samples_per_second": 24.095,
      "eval_steps_per_second": 0.761,
      "eval_wer": 1.0,
      "step": 100
    },
    {
      "epoch": 7.4,
      "eval_loss": 2.934115171432495,
      "eval_runtime": 67.8143,
      "eval_samples_per_second": 24.287,
      "eval_steps_per_second": 0.767,
      "eval_wer": 0.9998978653865795,
      "step": 200
    },
    {
      "epoch": 11.11,
      "eval_loss": 0.6993811130523682,
      "eval_runtime": 68.1937,
      "eval_samples_per_second": 24.152,
      "eval_steps_per_second": 0.763,
      "eval_wer": 0.68409764069043,
      "step": 300
    },
    {
      "epoch": 14.8,
      "eval_loss": 0.5623299479484558,
      "eval_runtime": 67.3052,
      "eval_samples_per_second": 24.471,
      "eval_steps_per_second": 0.773,
      "eval_wer": 0.5792053927075886,
      "step": 400
    },
    {
      "epoch": 14.98,
      "step": 405,
      "total_flos": 6.92765272380492e+18,
      "train_loss": 3.53821102189429,
      "train_runtime": 13547.1548,
      "train_samples_per_second": 3.851,
      "train_steps_per_second": 0.03
    }
  ],
  "max_steps": 405,
  "num_train_epochs": 15,
  "total_flos": 6.92765272380492e+18,
  "trial_name": null,
  "trial_params": null
}