metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*
dataset_info:
features:
- name: text1
dtype: string
- name: text2
dtype: string
- name: score
dtype: float64
splits:
- name: train
num_bytes: 28555408
num_examples: 15000
- name: val
num_bytes: 5686916
num_examples: 3000
download_size: 11941770
dataset_size: 34242324