Qwen3.5-21B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-AutoRound-W4A16-RTN
LeaderboardModel1/Qwen3.5-21B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking-AutoRound-W4A16-RTN
finetuneSAFETENSORS
Published metrics
- KL divergence
- 0.065
- Refusals
- 14/100
- Original refusals
- 94/100
- Heretic version
- —
Inherited from the parent conversion — source
Matched snippet
| KL divergence | 0.0653 | | Refusals | 14/100 |
Provenance
- Canonical model
- Qwen/Qwen3.5-27B
- Declared parent(s)
- DavidAU/Qwen3.5-21B-Claude-4.6-Opus-Deckard-Heretic-Uncensored-Thinking
- Classification
- derivative of a Heretic model without its own ablation evidence
- Created / updated
- 2026-06-21 / 2026-06-21
- Downloads
- 57 all time · 10 last 30 days
- License
- —
Published benchmarks
Copied from the model card. These never affect the ranking.
| Metric | This model | Original | Delta |
|---|---|---|---|
| HellaSwag | 0.5483 | — | — |
| MMLU | 0.6436 | — | — |
Refusal rate 14%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.