metadata
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: score
dtype: float64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 586548740
num_examples: 9471735
- name: eval
num_bytes: 2953146
num_examples: 47597
download_size: 346882910
dataset_size: 589501886
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*