metadata
dataset_info:
features:
- name: task_type
dtype: string
- name: problem
dtype: string
- name: solution
dtype: string
- name: problem_tokens
dtype: int64
- name: solution_tokens
dtype: int64
splits:
- name: train
num_bytes: 38997921.74082748
num_examples: 22000
- name: eval
num_bytes: 453793.9984387198
num_examples: 256
- name: test
num_bytes: 14290965.68520687
num_examples: 8062
download_size: 26238598
dataset_size: 53742681.42447307
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: eval
path: data/eval-*
- split: test
path: data/test-*