metadata
dataset_info:
features:
- name: task_id
dtype: string
- name: entry_point
dtype: string
- name: prompt
dtype: string
- name: suffix
dtype: string
- name: canonical_solution
dtype: string
- name: test
dtype: string
splits:
- name: train
num_bytes: 462594
num_examples: 300
download_size: 106995
dataset_size: 462594
configs:
- config_name: default
data_files:
- split: train
path: data/train-*