Zero-shot Reasoning on Reasoning Suite (ARC-e, ARC-c, HellaSwag, PIQA, Winogrande)
0.7559ARC-e Accuracy
LLaMA-2 (FP16)
Updated 4d ago
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
2025.12 | 0.7559 | 0.4317 | 0.5706 | 0.7791 | 0.6985 | 0.6472 | |
2025.12 | 0.7273 | 0.3976 | 0.5333 | 0.7617 | 0.6803 | 0.62 | |
2025.12 | 0.6368 | 0.3276 | 0.4955 | 0.7476 | 0.6567 | 0.5728 | |
2025.12 | 0.5846 | 0.3106 | 0.4521 | 0.7149 | 0.5919 | 0.5308 | |
2025.12 | 0.5656 | 0.2482 | 0.3819 | 0.7008 | 0.5367 | 0.4866 | |
2025.12 | 0.5593 | 0.2415 | 0.3843 | 0.698 | 0.5517 | 0.487 | |
2025.12 | 0.5556 | 0.2884 | 0.4294 | 0.7138 | 0.6243 | 0.5223 | |
2025.12 | 0.5332 | 0.227 | 0.3557 | 0.6681 | 0.5264 | 0.4621 |