Qwen3.5-4B-Deckard-HERETIC-UNCENSORED-Thinking
DavidAU/Qwen3.5-4B-Deckard-HERETIC-UNCENSORED-Thinking
finetuneSAFETENSORS
Published metrics
- KL divergence
- 0.041
- Refusals
- 4/100
- Original refusals
- 94/100
- Heretic version
- —
Published on the model card
Matched snippet
| KL divergence | 0.0406 | | Refusals | 4/100 |
Provenance
- Canonical model
- Qwen/Qwen3.5-4B
- Declared parent(s)
- coder3101/Qwen3.5-4B-heretic
- Classification
- card identifies this as a finetune of a Heretic model
- Created / updated
- 2026-03-27 / 2026-03-29
- Downloads
- 2.5K all time · 583 last 30 days
- License
- apache-2.0
Refusal rate 4%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.