--- dataset_info: features: - name: instruction dtype: string - name: completion dtype: string - name: conversations list: - name: from dtype: string - name: value dtype: string splits: - name: train num_bytes: 97503866 num_examples: 15819 download_size: 51514906 dataset_size: 97503866 configs: - config_name: default data_files: - split: train path: data/train-* ---