--- dataset_info: features: - name: Instruction dtype: string - name: Response dtype: string splits: - name: train num_bytes: 3682863 num_examples: 2000 download_size: 959900 dataset_size: 3682863 configs: - config_name: default data_files: - split: train path: data/train-* ---