| social-commonsense-reasoning-on-siqa | Nemotron-4 15B | #1 | Accuracy: 60.9 |
| cross-lingual-question-answering-on-tydiqa-goldp | Nemotron-4 15B (1-shot) | #6 | EM: 50.5 |
| common-sense-reasoning-on-arc-easy | Nemotron-4 15B | #18 | Accuracy: 80.9 |
| question-answering-on-piqa | Nemotron-4 15B | #30 | Accuracy: 82.4 |
| common-sense-reasoning-on-winogrande | Nemotron-4 15B | #40 | Accuracy: 78.0 |
| multi-task-language-understanding-on-mmlu | Nemotron-4 15B | #42 | Average (%): 64.2 |
| sentence-completion-on-hellaswag | Nemotron-4 15B | #58 | Accuracy: 82.4 |
| code-generation-on-humaneval | Nemotron-4 15B | #116 | Pass@1: 31.6 |
| code-generation-on-mbpp | Nemotron-4 15B (3-shot) | #135 | Accuracy: 40.6 |
| arithmetic-reasoning-on-gsm8k | Nemotron-4 15B (8-shot) | #235 | Accuracy: 46.0 |