trainer_state.json

{
  "best_metric": 75.27472527472527,
  "best_model_checkpoint": "./checkpoint-20",
  "epoch": 1.05,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 8.666666666666667e-05,
      "loss": 2.4473,
      "step": 10
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.3674653768539429,
      "eval_runtime": 44.1027,
      "eval_samples_per_second": 1.451,
      "eval_steps_per_second": 0.045,
      "eval_wer": 95.42124542124543,
      "step": 10
    },
    {
      "epoch": 1.0,
      "learning_rate": 2e-05,
      "loss": 1.256,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.974454939365387,
      "eval_runtime": 19.7011,
      "eval_samples_per_second": 3.249,
      "eval_steps_per_second": 0.102,
      "eval_wer": 75.27472527472527,
      "step": 20
    },
    {
      "epoch": 1.05,
      "step": 21,
      "total_flos": 1.654389080064e+16,
      "train_loss": 0.041942973931630455,
      "train_runtime": 33.8193,
      "train_samples_per_second": 18.924,
      "train_steps_per_second": 0.591
    }
  ],
  "max_steps": 20,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 1.654389080064e+16,
  "trial_name": null,
  "trial_params": null
}