metadata
dataset_info:
features:
- name: document
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 510794
num_examples: 4709
download_size: 103825
dataset_size: 510794
configs:
- config_name: default
data_files:
- split: train
path: data/train-*