metadata
dataset_info:
features:
- name: query
dtype: string
- name: pos
list: string
- name: neg
list: string
splits:
- name: train
num_bytes: 1072937590
num_examples: 25102
download_size: 1071787251
dataset_size: 1072937590
configs:
- config_name: default
data_files:
- split: train
path: data/train-*