Qwen3-VL-4B-Instruct-Abliterated-heretic
JSon-AI/Qwen3-VL-4B-Instruct-Abliterated-heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.460
- Refusals
- 6/100
- Original refusals
- 92/100
- Heretic version
- v1.0.1
Published on the model card
Matched snippet
| KL divergence | 0.46 | | Refusals | 6/100 |
Provenance
- Canonical model
- Qwen/Qwen3-VL-4B-Instruct
- Declared parent(s)
- Qwen/Qwen3-VL-4B-Instruct
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2025-11-19 / 2025-11-19
- Downloads
- 154 all time · 11 last 30 days
- License
- apache-2.0
Refusal rate 6%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.