trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9345063538611926,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.98,
      "learning_rate": 9.604743083003954e-06,
      "loss": 0.6182,
      "step": 500
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.9411023259162903,
      "eval_runtime": 332.4611,
      "eval_samples_per_second": 10.94,
      "eval_steps_per_second": 1.369,
      "eval_wer": 0.4075747711867652,
      "step": 500
    },
    {
      "epoch": 1.96,
      "learning_rate": 9.110671936758893e-06,
      "loss": 0.5421,
      "step": 1000
    },
    {
      "epoch": 1.96,
      "eval_loss": 0.9645132422447205,
      "eval_runtime": 331.2591,
      "eval_samples_per_second": 10.979,
      "eval_steps_per_second": 1.374,
      "eval_wer": 0.4025198177153142,
      "step": 1000
    },
    {
      "epoch": 2.93,
      "learning_rate": 8.616600790513835e-06,
      "loss": 0.4882,
      "step": 1500
    },
    {
      "epoch": 2.93,
      "eval_loss": 1.0171494483947754,
      "eval_runtime": 334.3973,
      "eval_samples_per_second": 10.876,
      "eval_steps_per_second": 1.361,
      "eval_wer": 0.4008922758779152,
      "step": 1500
    }
  ],
  "max_steps": 10220,
  "num_train_epochs": 20,
  "total_flos": 6.161659150250572e+19,
  "trial_name": null,
  "trial_params": null
}