metadata
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: labels
dtype:
class_label:
names:
'0': entailment
'1': neutral
'2': contradiction
splits:
- name: train
num_bytes: 10222003
num_examples: 30000
- name: validation
num_bytes: 49754
num_examples: 144
- name: test
num_bytes: 48945
num_examples: 144
download_size: 6272043
dataset_size: 10320702