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: 77758075
num_examples: 86821
- name: validation
num_bytes: 5677368
num_examples: 5928
download_size: 17978831
dataset_size: 83435443
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*