trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.3333333333333333,
      "eval_loss": 1.5153735876083374,
      "eval_runtime": 6.5784,
      "eval_samples_per_second": 2.28,
      "eval_steps_per_second": 0.304,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3333333333333333,
      "eval_loss": 1.3791700601577759,
      "eval_runtime": 8.0079,
      "eval_samples_per_second": 1.873,
      "eval_steps_per_second": 0.25,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3333333333333333,
      "eval_loss": 1.3326424360275269,
      "eval_runtime": 8.569,
      "eval_samples_per_second": 1.751,
      "eval_steps_per_second": 0.233,
      "step": 6
    },
    {
      "epoch": 3.0,
      "step": 6,
      "total_flos": 11840316410880.0,
      "train_loss": 1.5828933715820312,
      "train_runtime": 108.0308,
      "train_samples_per_second": 0.417,
      "train_steps_per_second": 0.056
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3333333333333333,
      "eval_loss": 1.3326424360275269,
      "eval_runtime": 6.5661,
      "eval_samples_per_second": 2.284,
      "eval_steps_per_second": 0.305,
      "step": 6
    }
  ],
  "logging_steps": 500,
  "max_steps": 6,
  "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": 11840316410880.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}