FuseChat-Qwen-2.5-7B-Instruct-Heretic
ChiKoi7/FuseChat-Qwen-2.5-7B-Instruct-Heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.070
- Refusals
- 3/100
- Original refusals
- 93/100
- Heretic version
- v1.0.1
Published on the model card
Matched snippet
| Refusals | 3/100 | | KL divergence | 0.07 |
Provenance
- Canonical model
- FuseAI/FuseChat-Qwen-2.5-7B-Instruct
- Declared parent(s)
- FuseAI/FuseChat-Qwen-2.5-7B-Instruct, ChiKoi7/FuseChat-Qwen-2.5-7B-Instruct-Heretic-GGUF
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2025-12-10 / 2025-12-10
- Downloads
- 159 all time · 27 last 30 days
- License
- —
Refusal rate 3%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.