metadata
dataset_info:
features:
- name: input
dtype: large_string
- name: output
dtype: large_string
splits:
- name: train
num_bytes: 946400
num_examples: 3053
download_size: 516886
dataset_size: 946400
configs:
- config_name: default
data_files:
- split: train
path: data/train-*