metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: label
dtype: string
splits:
- name: train
num_bytes: 649296909.6590501
num_examples: 3322980
download_size: 428488796
dataset_size: 649296909.6590501