metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 2000239020
num_examples: 61035
- name: test
num_bytes: 19990920
num_examples: 610
download_size: 751085781
dataset_size: 2020229940
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*