metadata
dataset_info:
features:
- name: query_id
dtype: int64
- name: query
dtype: string
- name: positive
dtype: string
- name: negative_1
dtype: string
- name: negative_2
dtype: string
- name: negative_3
dtype: string
- name: negative_4
dtype: string
splits:
- name: train
num_bytes: 1768793126
num_examples: 1003844
download_size: 1147992549
dataset_size: 1768793126
configs:
- config_name: default
data_files:
- split: train
path: data/train-*