metadata
dataset_info:
features:
- name: sentence1
dtype: string
- name: sentence2
dtype: string
- name: label
dtype:
class_label:
names:
'0': entailment
'1': neutral
'2': contradiction
splits:
- name: train
num_bytes: 528983104.40889275
num_examples: 1963485
download_size: 290954453
dataset_size: 528983104.40889275
configs:
- config_name: default
data_files:
- split: train
path: data/train-*