metadata
dataset_info:
features:
- name: inputs
dtype: string
- name: choices_list
list: string
- name: labels
dtype: int64
- name: metadata
dtype: string
splits:
- name: train
num_bytes: 14771791
num_examples: 14190
- name: validation
num_bytes: 2083302
num_examples: 2020
- name: test
num_bytes: 3857360
num_examples: 3610
download_size: 4378934
dataset_size: 20712453
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*