metadata
dataset_info:
features:
- name: content
dtype: string
splits:
- name: train
num_bytes: 15360567643
num_examples: 3000000
download_size: 6000794123
dataset_size: 15360567643
configs:
- config_name: default
data_files:
- split: train
path: data/train-*