metadata
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: gold_label
dtype: string
splits:
- name: train
num_bytes: 2420098
num_examples: 21090
- name: validation
num_bytes: 460660
num_examples: 4033
- name: test
num_bytes: 504213
num_examples: 4212
download_size: 94548
dataset_size: 3384971
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*