run-1/checkpoint-500/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5723270440251573,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.0064516129032258064,
      "eval_loss": NaN,
      "eval_runtime": 5.354,
      "eval_samples_per_second": 579.009,
      "eval_steps_per_second": 12.141,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 143768.90625,
      "learning_rate": 1.606918238993711e-05,
      "loss": 0.0,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2544,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 130072209152340.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.41600283894248313,
    "num_train_epochs": 8,
    "temperature": 20
  }
}