metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: unsandhied
dtype: string
splits:
- name: train
num_bytes: 5209305
num_examples: 20385
- name: validation
num_bytes: 663889
num_examples: 2962
- name: test
num_bytes: 637306
num_examples: 2938
download_size: 3726974
dataset_size: 6510500
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*