metadata
dataset_info:
features:
- name: output
dtype: string
- name: input
dtype: string
splits:
- name: train
num_bytes: 2264352.3
num_examples: 4500
- name: test
num_bytes: 251594.7
num_examples: 500
download_size: 1431785
dataset_size: 2515947
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*