trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.15384615384615385,
  "eval_steps": 500,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 1.438125729560852,
      "eval_runtime": 1.9143,
      "eval_samples_per_second": 274.775,
      "eval_steps_per_second": 4.701,
      "step": 3
    },
    {
      "epoch": 0.15384615384615385,
      "step": 3,
      "total_flos": 305613766656.0,
      "train_loss": 1.5372200012207031,
      "train_runtime": 320.6446,
      "train_samples_per_second": 4.79,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 10,
  "max_steps": 3,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 305613766656.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}