dataset_info: | |
features: | |
- name: input_ids | |
sequence: int32 | |
splits: | |
- name: train | |
num_bytes: 168975915696.0 | |
num_examples: 20616876 | |
download_size: 71503236187 | |
dataset_size: 168975915696.0 | |
configs: | |
- config_name: default | |
data_files: | |
- split: train | |
path: data/train-* | |