metadata
dataset_info:
features:
- name: A
dtype: string
- name: B
dtype: string
- name: SeqA
dtype: string
- name: SeqB
dtype: string
splits:
- name: train
num_bytes: 1867563
num_examples: 3959
download_size: 661079
dataset_size: 1867563
configs:
- config_name: default
data_files:
- split: train
path: data/train-*