metadata
dataset_info:
features:
- name: text
dtype: string
- name: token_length
dtype: int64
splits:
- name: train
num_bytes: 5162046028
num_examples: 930514
download_size: 3008010520
dataset_size: 5162046028
configs:
- config_name: default
data_files:
- split: train
path: data/train-*