Mathematical Reasoning on SimulEq out-of-domain (test)
77AccuracyMathCoder-L
Evaluation Results
| Method | Links | |
|---|---|---|
| MathCoder-LBase=Llama-2, Size=70B2023.10 | 77 | |
| MathCoder-CLBase=CodeLlama, Size=34B2023.10 | 65.8 | |
| MathCoder-LBase=Llama-2, Size=13B2023.10 | 62.3 | |
| MathCoder-CLBase=CodeLlama, Size=13B2023.10 | 60.7 | |
| MathCoder-CLBase=CodeLlama, Size=7B2023.10 | 49.6 | |
| MathCoder-LBase=Llama-2, Size=7B2023.10 | 47.5 | |
| WizardMathBase=Llama-2, Size=70B2023.10 | 37.9 | |
| WizardMathBase=Llama-2, Size=13B2023.10 | 14.9 | |
| GalacticaSize=30B2023.10 | 13.2 | |
| Llama-1 RFTBase=Llama-1, Size=34B2023.10 | 12.8 | |
| WizardMathBase=Llama-2, Size=7B2023.10 | 12.8 | |
| Llama-1 RFTBase=Llama-1, Size=7B2023.10 | 11 | |
| Llama-1 RFTBase=Llama-1, Size=13B2023.10 | 10.1 | |
| GalacticaSize=6.7B2023.10 | 4.2 |