metadata
dataset_info:
features:
- name: split
dtype: string
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: score
dtype: float64
splits:
- name: test
num_bytes: 65999
num_examples: 500
download_size: 42481
dataset_size: 65999
configs:
- config_name: default
data_files:
- split: test
path: data/test-*