metadata
language:
- ko
dataset_info:
features:
- name: query
dtype: string
- name: document
dtype: string
- name: hard_negative
dtype: string
splits:
- name: train
num_bytes: 628315763
num_examples: 744862
download_size: 270060556
dataset_size: 628315763
configs:
- config_name: default
data_files:
- split: train
path: data/train-*