metadata
dataset_info:
features:
- name: sequence
dtype: string
- name: sequence_A_start
dtype: int64
- name: sequence_B_start
dtype: int64
- name: label
dtype: int64
splits:
- name: train
num_bytes: 16074785400
num_examples: 1955570
download_size: 5552403075
dataset_size: 16074785400