metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
- name: documents
list: string
splits:
- name: train
num_bytes: 103431750
num_examples: 77400
- name: validation
num_bytes: 11310394
num_examples: 9535
download_size: 15184767
dataset_size: 114742144
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*