metadata
dataset_info:
features:
- name: document
dtype: string
- name: question
dtype: string
splits:
- name: train
num_bytes: 959397
num_examples: 8404
download_size: 161006
dataset_size: 959397
configs:
- config_name: default
data_files:
- split: train
path: data/train-*