metadata
dataset_info:
features:
- name: question
dtype: string
- name: facts
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 1406908410
num_examples: 73346
download_size: 728478879
dataset_size: 1406908410
configs:
- config_name: default
data_files:
- split: train
path: data/train-*