metadata
dataset_info:
features:
- name: input
dtype: string
splits:
- name: train
num_bytes: 1569011376
num_examples: 2147248
download_size: 807147913
dataset_size: 1569011376
configs:
- config_name: default
data_files:
- split: train
path: data/train-*