metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: label
dtype: float64
- name: is_answer
dtype: int64
splits:
- name: train
num_bytes: 632558017
num_examples: 400001
download_size: 87410887
dataset_size: 632558017