metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: id
dtype: string
- name: relevant_passage_ids
sequence: string
splits:
- name: train
num_bytes: 1685914.2898803588
num_examples: 3817
- name: test
num_bytes: 276246.540311174
num_examples: 664
download_size: 1339277
dataset_size: 1962160.8301915328
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*