metadata
dataset_info:
features:
- name: inputs
dtype: string
- name: targets
dtype: string
splits:
- name: train
num_bytes: 155420
num_examples: 805
download_size: 91507
dataset_size: 155420
configs:
- config_name: default
data_files:
- split: train
path: data/train-*