trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 545,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.92,
      "eval_loss": 149.2460479736328,
      "eval_runtime": 49.7078,
      "eval_samples_per_second": 33.134,
      "eval_steps_per_second": 4.144,
      "eval_wer": 1.003166173016035,
      "step": 100
    },
    {
      "epoch": 1.83,
      "eval_loss": 135.35861206054688,
      "eval_runtime": 50.6201,
      "eval_samples_per_second": 32.536,
      "eval_steps_per_second": 4.07,
      "eval_wer": 1.0,
      "step": 200
    },
    {
      "epoch": 2.75,
      "eval_loss": 114.33900451660156,
      "eval_runtime": 49.1186,
      "eval_samples_per_second": 33.531,
      "eval_steps_per_second": 4.194,
      "eval_wer": 1.0,
      "step": 300
    },
    {
      "epoch": 3.67,
      "eval_loss": 88.30770111083984,
      "eval_runtime": 48.4811,
      "eval_samples_per_second": 33.972,
      "eval_steps_per_second": 4.249,
      "eval_wer": 1.0,
      "step": 400
    },
    {
      "epoch": 4.59,
      "learning_rate": 0.00029699999999999996,
      "loss": 67.8005,
      "step": 500
    },
    {
      "epoch": 4.59,
      "eval_loss": 64.61665344238281,
      "eval_runtime": 48.9717,
      "eval_samples_per_second": 33.632,
      "eval_steps_per_second": 4.207,
      "eval_wer": 1.0,
      "step": 500
    },
    {
      "epoch": 5.0,
      "step": 545,
      "total_flos": 187150581292032.0,
      "train_loss": 65.05896735235092,
      "train_runtime": 710.1552,
      "train_samples_per_second": 24.488,
      "train_steps_per_second": 0.767
    }
  ],
  "max_steps": 545,
  "num_train_epochs": 5,
  "total_flos": 187150581292032.0,
  "trial_name": null,
  "trial_params": null
}