--- dataset_info: features: - name: conversations list: - name: from dtype: string - name: value dtype: string splits: - name: train num_bytes: 160339689 num_examples: 3974 download_size: 24332349 dataset_size: 160339689 configs: - config_name: default data_files: - split: train path: data/train-* ---