metadata
dataset_info:
features:
- name: question_id
dtype: string
- name: category
dtype: string
- name: turns
sequence: string
- name: ground_truth
dtype: string
- name: task
dtype: string
splits:
- name: test
num_bytes: 304048
num_examples: 150
download_size: 148322
dataset_size: 304048
configs:
- config_name: default
data_files:
- split: test
path: data/test-*