metadata
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 2660889947
num_examples: 8222483
download_size: 1119754151
dataset_size: 2660889947
configs:
- config_name: default
data_files:
- split: train
path: data/train-*