metadata
dataset_info:
features:
- name: text
dtype: string
- name: words
sequence: string
splits:
- name: train
num_bytes: 3568567511
num_examples: 1987794
download_size: 1085625772
dataset_size: 3568567511
configs:
- config_name: default
data_files:
- split: train
path: data/train-*