metadata
dataset_info:
features:
- name: hypothesis
dtype: string
- name: idx
dtype: int64
- name: label
dtype: string
- name: premise
dtype: string
splits:
- name: train
num_bytes: 34762
num_examples: 110
- name: eval
num_bytes: 7909
num_examples: 22
download_size: 37326
dataset_size: 42671
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*