code-davinci-002

Reported on 3 benchmarks across 3 tasks · 2 papers

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Methodology2 results

Transfer LearningonMGSM
Average (%)· 2022-10-20
35
best: 87 (PaLM 2 (few-shot, k=8, SC))
Scaling Instruction-Finetuned Language Models arXiv:2210.11416
Multi-Task LearningonMGSM
Average (%)· 2022-10-20
35
best: 87 (PaLM 2 (few-shot, k=8, SC))
Scaling Instruction-Finetuned Language Models arXiv:2210.11416

Miscellaneous1 result

General KnowledgeonTheoremQA
Accuracy· 2023-05-21
23.9
best: 52.4 (GPT-4 (PoT))
TheoremQA: A Theorem-driven Question Answering dataset arXiv:2305.12524