The dataset includes: 60 small Python programs generated by OpenAI GPT-3.5 (10 functionally correct program variations for each of the selected 6 publicly available introductory programming assignments) 399 program comprehension questions that target the generated programs 798 answers to those questions by OpenAI GPT-3.5 and GPT-4 Researcher annotation on the correctness of each answer and developed codes for the error types In addition, scripts for processing and analyzing the data are included.
Date made available20 Jan 2024

Dataset Licences

  • CC-BY-4.0

Cite this