metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 2256953760
num_examples: 1099880
download_size: 753570748
dataset_size: 2256953760
configs:
- config_name: default
data_files:
- split: train
path: data/train-*