metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: answer_aliases
list: string
- name: documents
list: string
- name: rationale
dtype: string
splits:
- name: train
num_bytes: 208522592
num_examples: 19938
- name: validation
num_bytes: 26454221
num_examples: 2417
download_size: 100267790
dataset_size: 234976813
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*