metadata
dataset_info:
features:
- name: id
dtype: string
- name: premise
dtype: string
- name: hypothesis
dtype: string
- name: label
dtype: string
- name: index
dtype: int64
splits:
- name: train
num_bytes: 21281556.832567178
num_examples: 51069
download_size: 12132925
dataset_size: 21281556.832567178
configs:
- config_name: default
data_files:
- split: train
path: data/train-*