metadata
dataset_info:
features:
- name: train
dtype: string
splits:
- name: train
num_bytes: 2884975505
num_examples: 15650428
download_size: 1881975211
dataset_size: 2884975505
configs:
- config_name: default
data_files:
- split: train
path: data/train-*