metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: input_ids
sequence: int32
- name: length
dtype: int64
splits:
- name: train
num_bytes: 234885131268
num_examples: 100000000
download_size: 10689166809
dataset_size: 234885131268