--- dataset_info: features: - name: translation dtype: string splits: - name: train num_bytes: 49128890 num_examples: 285584 download_size: 27861710 dataset_size: 49128890 configs: - config_name: default data_files: - split: train path: data/train-* ---