metadata
dataset_info:
config_name: pair-class
features:
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 138755142
num_examples: 942069
- name: dev
num_bytes: 3034127
num_examples: 19657
- name: test
num_bytes: 3142127
num_examples: 19656
download_size: 72651651
dataset_size: 144931396
configs:
- config_name: pair-class
data_files:
- split: train
path: pair-class/train-*
- split: dev
path: pair-class/dev-*
- split: test
path: pair-class/test-*