metadata
dataset_info:
features:
- name: problem_id
dtype: int64
- name: problem_original
dtype: string
- name: problem_perturbed_simple
dtype: string
- name: problem_perturbed_hard
dtype: string
- name: answer_original
dtype: string
- name: answer_perturbed_simple
dtype: string
- name: answer_perturbed_hard
dtype: string
- name: level
dtype: string
- name: type
dtype: string
- name: original_split
dtype: string
splits:
- name: train
num_bytes: 121791
num_examples: 164
- name: test
num_bytes: 84196
num_examples: 115
download_size: 132739
dataset_size: 205987
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*