--- dataset_info: features: - name: act dtype: string - name: prompt dtype: string splits: - name: train num_bytes: 9816 num_examples: 170 download_size: 9196 dataset_size: 9816 configs: - config_name: default data_files: - split: train path: data/train-* ---