metadata
dataset_info:
features:
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: label
dtype: int64
- name: __index_level_0__
dtype: int64
splits:
- name: train
num_bytes: 22930410
num_examples: 51172
- name: test
num_bytes: 5770405
num_examples: 12794
download_size: 12450125
dataset_size: 28700815
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*