metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: uid
dtype: int64
- name: support_documents
sequence: string
splits:
- name: train
num_bytes: 1607575704
num_examples: 142968
download_size: 648841769
dataset_size: 1607575704