metadata
dataset_info:
features:
- name: inputs
dtype: string
- name: targets
dtype: string
splits:
- name: train
num_bytes: 1589650
num_examples: 541
download_size: 825356
dataset_size: 1589650
configs:
- config_name: default
data_files:
- split: train
path: data/train-*