--- dataset_info: features: - name: input_ids sequence: int32 - name: token_type_ids sequence: int8 - name: attention_mask sequence: int8 - name: special_tokens_mask sequence: int8 splits: - name: train num_bytes: 176400.0 num_examples: 49 download_size: 34035 dataset_size: 176400.0 configs: - config_name: default data_files: - split: train path: data/train-* ---