metadata
dataset_info:
features:
- name: SeqA
dtype: string
- name: SeqB
dtype: string
- name: labels
dtype: int64
splits:
- name: train
num_bytes: 29145348
num_examples: 26944
download_size: 4833375
dataset_size: 29145348
configs:
- config_name: default
data_files:
- split: train
path: data/train-*