metadata
dataset_info:
features:
- name: sequence
dtype: string
splits:
- name: train
num_bytes: 11800733793
num_examples: 3099333
download_size: 4619026250
dataset_size: 11800733793
configs:
- config_name: default
data_files:
- split: train
path: data/train-*