metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: meta
dtype: string
- name: question_type
dtype: string
- name: fact
dtype: string
- name: rationale
dtype: string
splits:
- name: train
num_bytes: 486877
num_examples: 803
download_size: 147529
dataset_size: 486877
configs:
- config_name: default
data_files:
- split: train
path: data/train-*