metadata dataset_info:
features:
- name: question
dtype: string
- name: steps
sequence: string
- name: label
sequence: int64
splits:
- name: train
num_bytes: 391228918
num_examples: 443374
download_size: 179215148
dataset_size: 391228918
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
