metadata
dataset_info:
features:
- name: id
dtype: int64
- name: text
dtype: string
- name: distractors
sequence: string
splits:
- name: train
num_bytes: 394268709
num_examples: 47464
- name: test
num_bytes: 108437350
num_examples: 12381
download_size: 111310172
dataset_size: 502706059
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*