| question-answering-on-piqa | PowerLM-3B | #62 | Accuracy: 79.9 |
| multi-task-language-understanding-on-mmlu | PowerLM-3B | #65 | Average (%): 49.2 |
| question-answering-on-boolq | PowerLM-3B | #65 | Accuracy: 72.0 |
| common-sense-reasoning-on-winogrande | PowerLM-3B | #81 | Accuracy: 70.0 |
| sentence-completion-on-hellaswag | PowerLM-3B | #95 | Accuracy: 74.6 |
| code-generation-on-humaneval | PowerLM-3B | #127 | Pass@1: 26.8 |
| code-generation-on-mbpp | PowerLM-3B | #147 | Accuracy: 33.6 |
| math-word-problem-solving-on-math | PowerLM-3B (4-shot) | #168 | Accuracy: 15.2 |
| arithmetic-reasoning-on-gsm8k | PowerLM-3B (8-shot CoT) | #239 | Accuracy: 49.7 |