metadata
dataset_info:
features:
- name: question
dtype: string
- name: facts
sequence: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 74207066
num_examples: 167617
download_size: 31919180
dataset_size: 74207066
configs:
- config_name: default
data_files:
- split: train
path: data/train-*