metadata
dataset_info:
features:
- name: data
dtype: string
splits:
- name: train
num_bytes: 13500564
num_examples: 3831
- name: validation
num_bytes: 4073661
num_examples: 1114
- name: test
num_bytes: 3568904
num_examples: 990
download_size: 2887435
dataset_size: 21143129
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*