metadata
language:
- en
dataset_info:
features:
- name: context
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 557907208
num_examples: 1152000
download_size: 68525035
dataset_size: 557907208
configs:
- config_name: default
data_files:
- split: train
path: data/train-*