DeepHermes-3-Llama-3-8B-Preview-Heretic
ChiKoi7/DeepHermes-3-Llama-3-8B-Preview-Heretic
hereticSAFETENSORS
Published metrics
- KL divergence
- 0.023
- Refusals
- 4/100
- Original refusals
- 87/100
- Heretic version
- v1.1.0
Published on the model card
Matched snippet
| Refusals | 4/100 | | KL divergence | 0.0226 |
Provenance
- Canonical model
- NousResearch/DeepHermes-3-Llama-3-8B-Preview
- Declared parent(s)
- NousResearch/DeepHermes-3-Llama-3-8B-Preview, meta-llama/Meta-Llama-3.1-8B, meta-llama/Llama-3.1-8B
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2025-12-24 / 2025-12-24
- Downloads
- 97 all time · 22 last 30 days
- License
- llama3
Refusal rate 4%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.