| code-generation-on-mbpp | Code Llama - Python 70B (3-shot) | #29 | Accuracy: 65.5 |
| code-generation-on-mbpp | Code Llama 70B (3-shot) | #32 | Accuracy: 62.4 |
| code-generation-on-mbpp | Code Llama - Instruct 70B (3-shot) | #33 | Accuracy: 62.2 |
| code-generation-on-mbpp | Unnatural Code Llama 34B (3-shot) | #36 | Accuracy: 61.2 |
| code-generation-on-mbpp | Code Llama - Instruct 34B (3-shot) | #40 | Accuracy: 57 |
| code-generation-on-mbpp | Code Llama - Python 34B (3-shot) | #41 | Accuracy: 56.2 |
| code-generation-on-mbpp | Code Llama 34B (3-shot) | #43 | Accuracy: 55 |
| code-generation-on-mbpp | GPT-3.5 Turbo | #46 | Accuracy: 52.2 |
| code-generation-on-mbpp | Code Llama - Instruct 13B (3-shot) | #50 | Accuracy: 49.4 |
| code-generation-on-mbpp | Code Llama - Python 13B (3-shot) | #53 | Accuracy: 49 |
| code-generation-on-mbpp | Code Llama - Python 7B (3-shot) | #56 | Accuracy: 47.6 |
| code-generation-on-mbpp | Code Llama 13B (3-shot) | #61 | Accuracy: 47 |
| code-generation-on-mbpp | Code Llama - Instruct 7B (3-shot) | #66 | Accuracy: 44.4 |
| code-generation-on-mbpp | Code Llama 7B (3-shot) | #70 | Accuracy: 41.4 |