checkpoint-21/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 1.7725496292114258,
      "learning_rate": 9e-05,
      "loss": 1.5322,
      "step": 10
    },
    {
      "epoch": 0.9580838323353293,
      "grad_norm": 1.9632666110992432,
      "learning_rate": 0.00019,
      "loss": 1.3222,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.2385461330413818,
      "eval_runtime": 0.824,
      "eval_samples_per_second": 23.058,
      "eval_steps_per_second": 3.641,
      "step": 21
    }
  ],
  "logging_steps": 10,
  "max_steps": 63,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 3654932565590016.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}