metadata
dataset_info:
features:
- name: sequence
dtype: string
splits:
- name: train
num_bytes: 2746414479
num_examples: 660003
download_size: 1172964527
dataset_size: 2746414479
configs:
- config_name: default
data_files:
- split: train
path: data/train-*