metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 49605042096
num_examples: 48253932
- name: valid
num_bytes: 595216112
num_examples: 579004
download_size: 24336775144
dataset_size: 50200258208