last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.18604651162790697,
  "eval_steps": 14,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.013289036544850499,
      "eval_loss": 2.4356181621551514,
      "eval_runtime": 0.8063,
      "eval_samples_per_second": 39.685,
      "eval_steps_per_second": 19.843,
      "step": 1
    },
    {
      "epoch": 0.132890365448505,
      "grad_norm": 3.1786859035491943,
      "learning_rate": 0.00019510565162951537,
      "loss": 1.7808,
      "step": 10
    },
    {
      "epoch": 0.18604651162790697,
      "eval_loss": 1.3662489652633667,
      "eval_runtime": 0.8213,
      "eval_samples_per_second": 38.962,
      "eval_steps_per_second": 19.481,
      "step": 14
    }
  ],
  "logging_steps": 10,
  "max_steps": 55,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 14,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 700136262991872.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}