metadata
dataset_info:
features:
- name: text
struct:
- name: text
dtype: string
splits:
- name: train
num_bytes: 3117508
num_examples: 21679
download_size: 1960839
dataset_size: 3117508
configs:
- config_name: default
data_files:
- split: train
path: data/train-*