checkpoint-60/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7947019867549668,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 4.9999999999999996e-05,
      "loss": 13.2048,
      "step": 10
    },
    {
      "epoch": 0.26,
      "learning_rate": 7.46938775510204e-05,
      "loss": 26.5657,
      "step": 20
    },
    {
      "epoch": 0.26,
      "eval_loss": 17.199691772460938,
      "eval_runtime": 5.248,
      "eval_samples_per_second": 18.674,
      "eval_steps_per_second": 0.762,
      "eval_wer": 1.0,
      "step": 20
    },
    {
      "epoch": 0.4,
      "learning_rate": 7.36734693877551e-05,
      "loss": 8.0091,
      "step": 30
    },
    {
      "epoch": 0.53,
      "learning_rate": 7.265306122448979e-05,
      "loss": 7.6332,
      "step": 40
    },
    {
      "epoch": 0.53,
      "eval_loss": 4.783994197845459,
      "eval_runtime": 4.9818,
      "eval_samples_per_second": 19.672,
      "eval_steps_per_second": 0.803,
      "eval_wer": 1.0,
      "step": 40
    },
    {
      "epoch": 0.66,
      "learning_rate": 7.163265306122449e-05,
      "loss": 3.8261,
      "step": 50
    },
    {
      "epoch": 0.79,
      "learning_rate": 7.061224489795918e-05,
      "loss": 3.8717,
      "step": 60
    },
    {
      "epoch": 0.79,
      "eval_loss": 3.6783852577209473,
      "eval_runtime": 4.9141,
      "eval_samples_per_second": 19.943,
      "eval_steps_per_second": 0.814,
      "eval_wer": 1.0,
      "step": 60
    }
  ],
  "max_steps": 750,
  "num_train_epochs": 10,
  "total_flos": 1.21965615465984e+18,
  "trial_name": null,
  "trial_params": null
}