metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: unsandhied
dtype: string
splits:
- name: train
num_bytes: 2194065829
num_examples: 3801299
- name: validation
num_bytes: 14973592
num_examples: 42030
- name: test
num_bytes: 15127779
num_examples: 43075
download_size: 774991189
dataset_size: 2224167200
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*