metadata
dataset_info:
features:
- name: query
dtype: string
- name: positive
dtype: string
- name: negative
dtype: string
- name: label
dtype: float64
- name: task_name
dtype: string
splits:
- name: train
num_bytes: 1953335
num_examples: 13547
download_size: 882809
dataset_size: 1953335
configs:
- config_name: default
data_files:
- split: train
path: data/train-*