checkpoint-18/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.230769230769231,
  "eval_steps": 500,
  "global_step": 18,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.19,
      "learning_rate": 0.000125,
      "loss": 1.3745,
      "step": 5
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001942857142857143,
      "loss": 1.1943,
      "step": 10
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.00018,
      "loss": 1.0322,
      "step": 15
    }
  ],
  "logging_steps": 5,
  "max_steps": 78,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2560766288855040.0,
  "trial_name": null,
  "trial_params": null
}