trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 10,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 0,
      "loss": 1.2553,
      "step": 2
    },
    {
      "epoch": 0.33,
      "learning_rate": 0,
      "loss": 1.2298,
      "step": 4
    },
    {
      "epoch": 0.5,
      "learning_rate": 0,
      "loss": 1.2589,
      "step": 6
    },
    {
      "epoch": 0.67,
      "learning_rate": 0,
      "loss": 1.2981,
      "step": 8
    },
    {
      "epoch": 0.83,
      "learning_rate": 0,
      "loss": 1.1645,
      "step": 10
    },
    {
      "epoch": 0.83,
      "eval_loss": 1.6062859296798706,
      "eval_runtime": 55.4276,
      "eval_samples_per_second": 1.624,
      "eval_steps_per_second": 0.018,
      "step": 10
    },
    {
      "epoch": 1.0,
      "learning_rate": 0,
      "loss": 1.2375,
      "step": 12
    },
    {
      "epoch": 1.0,
      "step": 12,
      "total_flos": 0.0,
      "train_loss": 1.24068017800649,
      "train_runtime": 1392.1967,
      "train_samples_per_second": 0.256,
      "train_steps_per_second": 0.009
    }
  ],
  "logging_steps": 2,
  "max_steps": 12,
  "num_train_epochs": 1,
  "save_steps": 10,
  "total_flos": 0.0,
  "trial_name": null,
  "trial_params": null
}