checkpoint-262/trainer_state.json

{
  "best_metric": 0.1635778844356537,
  "best_model_checkpoint": "YELP_full/checkpoint-262",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 262,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9328,
      "eval_loss": 0.17968642711639404,
      "eval_runtime": 78.2206,
      "eval_samples_per_second": 319.609,
      "eval_steps_per_second": 1.675,
      "step": 131
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.93844,
      "eval_loss": 0.1635778844356537,
      "eval_runtime": 78.3918,
      "eval_samples_per_second": 318.911,
      "eval_steps_per_second": 1.671,
      "step": 262
    }
  ],
  "logging_steps": 500,
  "max_steps": 1310,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.61296750592e+16,
  "train_batch_size": 192,
  "trial_name": null,
  "trial_params": null
}