Llama-3.1-8B-Lexi-Uncensored-V2-Heretic
ChiKoi7/Llama-3.1-8B-Lexi-Uncensored-V2-Heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.020
- Refusals
- 2/100
- Original refusals
- 35/100
- Heretic version
- v1.0.1
Published on the model card
Matched snippet
| Refusals | 2/100 | | KL divergence | 0.02 |
Provenance
- Canonical model
- Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2
- Declared parent(s)
- Orenguteng/Llama-3.1-8B-Lexi-Uncensored-V2, ChiKoi7/Llama-3.1-8B-Lexi-Uncensored-V2-Heretic-GGUF
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2025-12-11 / 2025-12-11
- Downloads
- 289 all time · 107 last 30 days
- License
- llama3.1
Published benchmarks
Copied from the model card. These never affect the ranking.
| Metric | This model | Original | Delta |
|---|---|---|---|
| IFEval | 77.92 | — | — |
| MATH | 16.92 | — | — |
| GPQA | 4.36 | — | — |
| MMLU-Pro | 30.9 | — | — |
Refusal rate 2%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.