metadata
dataset_info:
features:
- name: set
struct:
- name: neg
sequence: string
- name: pos
sequence: string
- name: query
dtype: string
splits:
- name: train
num_bytes: 24794
num_examples: 58
download_size: 12195
dataset_size: 24794
configs:
- config_name: default
data_files:
- split: train
path: data/train-*