trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.64,
  "eval_steps": 500,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 1.900968867902419e-05,
      "loss": 1.1633,
      "step": 1
    },
    {
      "epoch": 0.64,
      "learning_rate": 3.7651019814126656e-06,
      "loss": 1.1562,
      "step": 5
    },
    {
      "epoch": 0.64,
      "eval_loss": 1.1389280557632446,
      "eval_runtime": 100.6563,
      "eval_samples_per_second": 1.987,
      "eval_steps_per_second": 0.497,
      "step": 5
    },
    {
      "epoch": 0.64,
      "step": 5,
      "total_flos": 5.80221862412288e+16,
      "train_loss": 1.1949972957372665,
      "train_runtime": 2050.1238,
      "train_samples_per_second": 0.488,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 7,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 5.80221862412288e+16,
  "trial_name": null,
  "trial_params": null
}