--- dataset_info: features: - name: dataset dtype: string - name: url dtype: string splits: - name: train num_bytes: 84727113350 num_examples: 968000015 download_size: 51299133597 dataset_size: 84727113350 configs: - config_name: default data_files: - split: train path: data/train-* ---