trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7211538461538461,
  "eval_steps": 25,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.24,
      "learning_rate": 2.3797595190380762e-05,
      "loss": 1.5175,
      "step": 25
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.1654824018478394,
      "eval_runtime": 4.3986,
      "eval_samples_per_second": 11.367,
      "eval_steps_per_second": 1.591,
      "step": 25
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.2545090180360722e-05,
      "loss": 0.994,
      "step": 50
    },
    {
      "epoch": 0.48,
      "eval_loss": 1.1613285541534424,
      "eval_runtime": 4.4017,
      "eval_samples_per_second": 11.359,
      "eval_steps_per_second": 1.59,
      "step": 50
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.1292585170340683e-05,
      "loss": 1.1237,
      "step": 75
    },
    {
      "epoch": 0.72,
      "eval_loss": 1.1351916790008545,
      "eval_runtime": 4.4061,
      "eval_samples_per_second": 11.348,
      "eval_steps_per_second": 1.589,
      "step": 75
    }
  ],
  "logging_steps": 25,
  "max_steps": 500,
  "num_train_epochs": 5,
  "save_steps": 25,
  "total_flos": 3315779135078400.0,
  "trial_name": null,
  "trial_params": null
}