metadata
dataset_info:
features:
- name: query
dtype: string
- name: positive
sequence: string
- name: negative
sequence: string
splits:
- name: train
num_bytes: 23231663.1
num_examples: 9900
- name: test
num_bytes: 2581295.9
num_examples: 1100
download_size: 10424834
dataset_size: 25812959
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*