| code-generation-on-mbpp | phi-1.5-web 1.3B | #68 | Accuracy: 43.5 |
| common-sense-reasoning-on-arc-challenge | phi-1.5-web 1.3B (zero-shot) | #29 | Accuracy: 44.9 |
| common-sense-reasoning-on-arc-easy | phi-1.5-web 1.3B (0-shot) | #18 | Accuracy: 76.1 |
| common-sense-reasoning-on-winogrande | phi-1.5-web 1.3B (zero-shot) | #28 | Accuracy: 74.0 |
| multi-task-language-understanding-on-mmlu | phi-1.5-web 1.3B | #29 | Average (%): 37.9 |
| question-answering-on-piqa | phi-1.5-web (1.3B) | #42 | Accuracy: 77 |
| question-answering-on-social-iqa | phi-1.5-web 1.3B (zero-shot) | #16 | Accuracy: 53.0 |
| question-answering-on-social-iqa | phi-1.5 1.3B (zero-shot) | #17 | Accuracy: 52.6 |