metadata
dataset_info:
features:
- name: problem
dtype: string
- name: is_correct
dtype: bool
- name: target_answer
dtype: string
- name: solution
dtype: string
- name: solution_steps
dtype: string
- name: attempts
dtype: string
- name: model_answer
dtype: string
splits:
- name: train
num_bytes: 190524394315
num_examples: 18469392
download_size: 20522956755
dataset_size: 190524394315
configs:
- config_name: default
data_files:
- split: train
path: data/train-*