--- dataset_info: features: - name: abstract dtype: string - name: authors dtype: string - name: title dtype: string - name: __index_level_0__ dtype: int64 splits: - name: train num_bytes: 48215145 num_examples: 50000 download_size: 28489554 dataset_size: 48215145 configs: - config_name: default data_files: - split: train path: data/train-* ---