metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: target
dtype: int64
- name: numbers
list: int64
splits:
- name: train
num_bytes: 199108848
num_examples: 327680
- name: test
num_bytes: 622352
num_examples: 1024
download_size: 16642787
dataset_size: 199731200
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*