metadata dataset_info: features: - name: conversation list: - name: content dtype: string - name: role dtype: string splits: - name: train num_bytes: 40157805 num_examples: 21164 download_size: 20525485 dataset_size: 40157805 configs:
- config_name: default
data_files:
- split: train path: data/train-