Out-of-Distribution Detection on 20NG (ID) vs MNLI (OOD)
0.9989AUROCGemma2-2b LR baseline
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Gemma2-2b LR baselineBackbone=Gemma2-2b, Backend=Logistic Regression, Training Protocol=ID-only training2026.05 | 0.9989 | 0.9988 | 0.27 | |
| Gemma2-9b LR baselineBackbone=Gemma2-9b, Backend=Logistic Regression, Training Protocol=ID-only training2026.05 | 0.9982 | 0.997 | 0.3 | |
| Gemma2-9b + 16k SAE + MarkovBackbone=Gemma2-9b, SAE Size=16k, Backend=Markov, Training Protocol=ID-only training2026.05 | 0.9763 | 0.9768 | 16.28 | |
| Gemma2-2b + 16k SAE + MarkovBackbone=Gemma2-2b, SAE Size=16k, Backend=Markov, Training Protocol=ID-only training2026.05 | 0.9699 | 0.968 | 19.84 |