metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 455501800
num_examples: 111098
- name: validation
num_bytes: 80228800
num_examples: 19568
download_size: 95652767
dataset_size: 535730600
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*