metadata
dataset_info:
features:
- name: promptID
dtype: int32
- name: pairID
dtype: string
- name: premise
dtype: string
- name: premise_binary_parse
dtype: string
- name: premise_parse
dtype: string
- name: hypothesis
dtype: string
- name: hypothesis_binary_parse
dtype: string
- name: hypothesis_parse
dtype: string
- name: genre
dtype: string
- name: label
dtype:
class_label:
names:
'0': entailment
'1': neutral
'2': contradiction
- name: input
dtype: string
splits:
- name: train
num_bytes: 489542376
num_examples: 392702
- name: validation
num_bytes: 12015470
num_examples: 9815
- name: test
num_bytes: 12677404
num_examples: 9832
download_size: 272608638
dataset_size: 514235250
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*