metadata
dataset_info:
features:
- name: subject
dtype: string
- name: problem
dtype: string
- name: level
dtype: int64
- name: answer
dtype: string
splits:
- name: train
num_bytes: 2239391
num_examples: 8523
download_size: 1172379
dataset_size: 2239391
configs:
- config_name: default
data_files:
- split: train
path: data/train-*