metadata
dataset_info:
features:
- name: text
dtype: string
- name: hypothesis
dtype: string
- name: labels
dtype: int64
- name: task_name
dtype: string
splits:
- name: train
num_bytes: 14652960583
num_examples: 9051961
- name: test
num_bytes: 102755997
num_examples: 76354
download_size: 646190092
dataset_size: 14755716580
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*