metadata
dataset_info:
features:
- name: text1
dtype: string
- name: text2
dtype: string
- name: score
dtype: float64
splits:
- name: train
num_bytes: 885974
num_examples: 4809
download_size: 247402
dataset_size: 885974
configs:
- config_name: default
data_files:
- split: train
path: data/train-*