Faithfulness Detection on LatentAudit Llama-3-8B (evaluation set)
0.948AUROC
GPT-4o Judge
Updated 1mo ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2026.04 | 0.948 | 0.881 | |
2026.04 | 0.942 | 0.869 | |
2026.04 | 0.908 | 0.841 | |
2026.04 | 0.882 | 0.815 | |
2026.04 | 0.871 | 0.804 | |
2026.04 | 0.722 | 0.655 |