metadata
dataset_info:
features:
- name: questions
dtype: string
- name: answers
dtype: string
splits:
- name: train
num_bytes: 4052701657
num_examples: 1909936
download_size: 2176509990
dataset_size: 4052701657
configs:
- config_name: default
data_files:
- split: train
path: data/train-*