metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 8927832
num_examples: 17302
download_size: 5094023
dataset_size: 8927832
configs:
- config_name: default
data_files:
- split: train
path: data/train-*