--- dataset_info: features: - name: Context dtype: string - name: Knowledge dtype: string - name: Response dtype: string splits: - name: train num_bytes: 2418454 num_examples: 2126 download_size: 109855 dataset_size: 2418454 configs: - config_name: default data_files: - split: train path: data/train-* ---