Llama-3.1-8B-Instruct-heretic
askalgore/Llama-3.1-8B-Instruct-heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.029
- Refusals
- 2/100
- Original refusals
- 96/100
- Heretic version
- v1.1.0
Published on the model card
Matched snippet
| KL divergence | 0.0287 | | Refusals | 2/100 |
Provenance
- Canonical model
- meta-llama/Llama-3.1-8B-Instruct
- Declared parent(s)
- meta-llama/Llama-3.1-8B-Instruct, meta-llama/Meta-Llama-3.1-8B, meta-llama/Llama-3.1-8B
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2026-01-26 / 2026-01-26
- Downloads
- 58 all time · 13 last 30 days
- License
- llama3.1
Refusal rate 2%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.