metadata
dataset_info:
features:
- name: question
dtype: string
- name: choices
sequence: string
- name: answer
dtype: string
- name: support
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 17409795.132692967
num_examples: 30000
- name: validation
num_bytes: 701614
num_examples: 1254
- name: test
num_bytes: 702050
num_examples: 1254
download_size: 14549597
dataset_size: 18813459.132692967
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*