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: 80615
num_examples: 232
- name: eval
num_bytes: 10938
num_examples: 29
download_size: 148612
dataset_size: 91553
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*