run-1/checkpoint-1006/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1006,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 5.137186621867392e-06,
      "loss": 0.8553,
      "step": 500
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.438381522043518e-06,
      "loss": 0.7017,
      "step": 1000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2012,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 529879044096000.0,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.835991721691264e-06,
    "per_device_train_batch_size": 16
  }
}