metadata
dataset_info:
features:
- name: inputs
dtype: string
- name: targets
dtype: string
- name: task
dtype: string
splits:
- name: train
num_bytes: 1390248
num_examples: 1000
- name: validation
num_bytes: 144618
num_examples: 100
download_size: 756807
dataset_size: 1534866
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*