metadata
dataset_info:
features:
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: label
dtype: int64
- name: hypothesis_negated
dtype: string
- name: label_negated
dtype: int64
splits:
- name: train
num_bytes: 2487892
num_examples: 9735
- name: validation
num_bytes: 493744
num_examples: 1954
download_size: 1967257
dataset_size: 2981636
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*