checkpoint-25/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03333333333333333,
  "eval_steps": 25,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 2.3797595190380762e-05,
      "loss": 3.5104,
      "step": 25
    },
    {
      "epoch": 0.03,
      "eval_loss": 2.643505334854126,
      "eval_runtime": 13.8992,
      "eval_samples_per_second": 17.987,
      "eval_steps_per_second": 2.302,
      "step": 25
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "total_flos": 410853310464000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}