metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: pos_doc
dtype: string
splits:
- name: train
num_bytes: 2269543
num_examples: 999
download_size: 1161270
dataset_size: 2269543
configs:
- config_name: default
data_files:
- split: train
path: data/train-*