--- dataset_info: features: - name: input_ids sequence: int32 - name: attention_mask sequence: int8 splits: - name: train num_bytes: 55751845968.0 num_examples: 5440266 - name: test num_bytes: 6194659800.0 num_examples: 604475 download_size: 23930738929 dataset_size: 61946505768.0 configs: - config_name: default data_files: - split: train path: data/train-* - split: test path: data/test-* ---