README.md

---
dataset_info:
 features:
 - name: question_id
 dtype: int64
 - name: category
 dtype: string
 - name: turns
 list:
 - name: content
 dtype: string
 - name: role
 dtype: string
 - name: reference
 sequence: string
 - name: history
 list:
 - name: content
 dtype: string
 - name: role
 dtype: string
 - name: finished
 dtype: bool
 splits:
 - name: train
 num_bytes: 85821
 num_examples: 80
 download_size: 55138
 dataset_size: 85821
configs:
- config_name: default
 data_files:
 - split: train
 path: data/train-*
---