ThinkingCap-Qwen3.6-27B-heretic
egibson/ThinkingCap-Qwen3.6-27B-heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.019
- Refusals
- 12/100
- Original refusals
- —
- Heretic version
- v1.4.0
Published on the model card
Matched snippet
| | Keywords (refusals detected on harmful prompts) | KL divergence from original model on harmless prompts | → | This model | 12/100 | 0.0186 |
Provenance
- Canonical model
- bottlecapai/ThinkingCap-Qwen3.6-27B
- Declared parent(s)
- bottlecapai/ThinkingCap-Qwen3.6-27B, egibson/ThinkingCap-Qwen3.6-27B-heretic-GGUF
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2026-07-19 / 2026-07-19
- Downloads
- 104 all time · 17 last 30 days
- License
- apache-2.0
Refusal rate 12%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.