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: 1791266928
num_examples: 504111
- name: validation
num_bytes: 181536777
num_examples: 55597
download_size: 1056103190
dataset_size: 1972803705
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*