metadata
dataset_info:
features:
- name: input
dtype: string
- name: choices
sequence: string
- name: expected_output
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 1303406
num_examples: 3009
download_size: 727895
dataset_size: 1303406
configs:
- config_name: default
data_files:
- split: train
path: data/train-*