metadata
dataset_info:
features:
- name: id
dtype: int64
- name: question
dtype: string
- name: answer
dtype: string
- name: documents
list: string
splits:
- name: train
num_bytes: 2685450280
num_examples: 1979809
- name: validation
num_bytes: 13588176
num_examples: 10017
- name: test
num_bytes: 13546599
num_examples: 9999
download_size: 1732277965
dataset_size: 2712585055
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*