metadata
dataset_info:
features:
- name: input
dtype: string
- name: question
dtype: string
- name: utility_category
dtype: string
- name: answers
list: string
splits:
- name: train
num_bytes: 67435227
num_examples: 27860
download_size: 24857081
dataset_size: 67435227
configs:
- config_name: default
data_files:
- split: train
path: data/train-*