metadata
dataset_info:
features:
- name: question
dtype: string
- name: sentence
dtype: string
- name: label
dtype:
class_label:
names:
'0': entailment
'1': not_entailment
- name: idx
dtype: int32
splits:
- name: train
num_bytes: 25612443
num_examples: 104743
- name: validation
num_bytes: 250467.5086948563
num_examples: 1000
- name: test
num_bytes: 1368304
num_examples: 5463
download_size: 18562140
dataset_size: 27231214.508694857
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*