metadata
dataset_info:
features:
- name: _id
dtype: string
- name: base_operation
dtype: string
- name: target_operation
dtype: string
- name: a
dtype: int64
- name: b
dtype: int64
- name: prompt
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 2251042
num_examples: 20158
- name: validation
num_bytes: 2209038
num_examples: 20142
download_size: 2291376
dataset_size: 4460080
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*