metadata
dataset_info:
features:
- name: question
dtype: string
- name: actual_answer
dtype: string
- name: incorrect_answer
dtype: string
splits:
- name: train
num_bytes: 26646048
num_examples: 21325
download_size: 14500050
dataset_size: 26646048
configs:
- config_name: default
data_files:
- split: train
path: data/train-*