trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.88,
  "eval_steps": 500,
  "global_step": 27,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.96,
      "eval_loss": 1.076568841934204,
      "eval_runtime": 6.7526,
      "eval_samples_per_second": 37.171,
      "eval_steps_per_second": 0.592,
      "step": 9
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 16.64949310344252,
      "learning_rate": 5e-06,
      "loss": 1.2054,
      "step": 10
    },
    {
      "epoch": 1.92,
      "eval_loss": 1.0334553718566895,
      "eval_runtime": 6.6943,
      "eval_samples_per_second": 37.495,
      "eval_steps_per_second": 0.598,
      "step": 18
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 1.345117072531619,
      "learning_rate": 5e-06,
      "loss": 1.0745,
      "step": 20
    },
    {
      "epoch": 2.88,
      "eval_loss": 1.0039987564086914,
      "eval_runtime": 5.2974,
      "eval_samples_per_second": 47.381,
      "eval_steps_per_second": 0.755,
      "step": 27
    },
    {
      "epoch": 2.88,
      "step": 27,
      "total_flos": 45016625971200.0,
      "train_loss": 1.088523511533384,
      "train_runtime": 1185.3669,
      "train_samples_per_second": 12.027,
      "train_steps_per_second": 0.023
    }
  ],
  "logging_steps": 10,
  "max_steps": 27,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 45016625971200.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}