trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 246,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.9796747967479676e-05,
      "loss": 13.6251,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_cer": 0.12285168682367918,
      "eval_loss": 0.5464695692062378,
      "eval_runtime": 26.5959,
      "eval_samples_per_second": 12.295,
      "eval_steps_per_second": 0.79,
      "step": 82
    },
    {
      "epoch": 2.0,
      "eval_cer": 0.02737110120942075,
      "eval_loss": 0.4406394362449646,
      "eval_runtime": 20.9773,
      "eval_samples_per_second": 15.588,
      "eval_steps_per_second": 1.001,
      "step": 164
    },
    {
      "epoch": 3.0,
      "eval_cer": 0.018459579885423297,
      "eval_loss": 0.39753007888793945,
      "eval_runtime": 20.8038,
      "eval_samples_per_second": 15.718,
      "eval_steps_per_second": 1.009,
      "step": 246
    },
    {
      "epoch": 3.0,
      "step": 246,
      "total_flos": 4.694072734336942e+17,
      "train_loss": 0.8410020301012489,
      "train_runtime": 361.7661,
      "train_samples_per_second": 10.847,
      "train_steps_per_second": 0.68
    }
  ],
  "max_steps": 246,
  "num_train_epochs": 3,
  "total_flos": 4.694072734336942e+17,
  "trial_name": null,
  "trial_params": null
}