trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5,
  "global_step": 37,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 18.4118,
      "step": 10
    },
    {
      "epoch": 0.14,
      "eval_loss": 21.371702194213867,
      "eval_runtime": 184.1706,
      "eval_samples_per_second": 2.753,
      "eval_steps_per_second": 0.277,
      "eval_wer": 1.0001766472354707,
      "step": 10
    },
    {
      "epoch": 0.27,
      "learning_rate": 3.6e-06,
      "loss": 19.6613,
      "step": 20
    },
    {
      "epoch": 0.27,
      "eval_loss": 21.348163604736328,
      "eval_runtime": 182.855,
      "eval_samples_per_second": 2.773,
      "eval_steps_per_second": 0.279,
      "eval_wer": 1.0005299417064122,
      "step": 20
    },
    {
      "epoch": 0.41,
      "learning_rate": 5.600000000000001e-06,
      "loss": 18.9808,
      "step": 30
    },
    {
      "epoch": 0.41,
      "eval_loss": 21.291269302368164,
      "eval_runtime": 182.8526,
      "eval_samples_per_second": 2.773,
      "eval_steps_per_second": 0.279,
      "eval_wer": 1.0,
      "step": 30
    },
    {
      "epoch": 0.5,
      "step": 37,
      "total_flos": 1.448010543367872e+17,
      "train_loss": 19.20771748310811,
      "train_runtime": 1190.1942,
      "train_samples_per_second": 0.621,
      "train_steps_per_second": 0.031
    }
  ],
  "max_steps": 37,
  "num_train_epochs": 1,
  "total_flos": 1.448010543367872e+17,
  "trial_name": null,
  "trial_params": null
}