metadata
language:
- en
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
splits:
- name: train
num_bytes: 3738828930
num_examples: 6828769
download_size: 2379297508
dataset_size: 3738828930
configs:
- config_name: default
data_files:
- split: train
path: data/train-*