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: 957273323
num_examples: 17517
- name: validation
num_bytes: 56049345
num_examples: 973
- name: test
num_bytes: 51801048
num_examples: 973
download_size: 508173514
dataset_size: 1065123716
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*