metadata
dataset_info:
features:
- name: context
dtype: string
- name: question
dtype: string
- name: negative
dtype: string
splits:
- name: train
num_bytes: 460052736
num_examples: 58727
- name: test
num_bytes: 116757655
num_examples: 12597
download_size: 92566388
dataset_size: 576810391
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*