LFM2-350M-heretic-safety-defiltered-xhigh-reasoning
MihaiPopa-1/LFM2-350M-heretic-safety-defiltered-xhigh-reasoning
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.021
- Refusals
- 10/100
- Original refusals
- 84/100
- Heretic version
- v1.1.0
Published on the model card
Matched snippet
| KL divergence | 0.0205 | | Refusals | 10/100 |
Provenance
- Canonical model
- LiquidAI/LFM2-350M
- Declared parent(s)
- LiquidAI/LFM2-350M
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2026-02-11 / 2026-02-11
- Downloads
- 123 all time · 24 last 30 days
- License
- other
Refusal rate 10%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.