metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: inputs
dtype: string
- name: targets
dtype: string
splits:
- name: train
num_bytes: 1201540790
num_examples: 1000000
- name: validation
num_bytes: 12049203
num_examples: 10000
download_size: 733055816
dataset_size: 1213589993