metadata
dataset_info:
features:
- name: text_a
dtype: string
- name: text_b
dtype: string
- name: label
dtype:
class_label:
names:
'0': not equivalent
'1': equivalent
splits:
- name: train
num_bytes: 929171
num_examples: 3668
- name: validation
num_bytes: 104247
num_examples: 408
- name: test
num_bytes: 435510
num_examples: 1725
download_size: 1000482
dataset_size: 1468928
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*