metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: unsandhied
dtype: string
splits:
- name: train
num_bytes: 20735060
num_examples: 54069
- name: validation
num_bytes: 357468
num_examples: 1917
- name: test
num_bytes: 361517
num_examples: 1919
download_size: 13343984
dataset_size: 21454045
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*