SmolLM2-360M-Instruct-heretic
saidutta69/SmolLM2-360M-Instruct-heretic
hereticGGUFSAFETENSORSF16
Published metrics
- KL divergence
- 0.015
- Refusals
- 4/100
- Original refusals
- 8/100
- Heretic version
- v1.4.0
Published on the model card
Matched snippet
| KL divergence | 0.0154 | | Refusals | 4/100 |
Provenance
- Canonical model
- HuggingFaceTB/SmolLM2-360M-Instruct
- Declared parent(s)
- HuggingFaceTB/SmolLM2-360M-Instruct
- Classification
- card states the weights were abliterated with Heretic
- Created / updated
- 2026-10-09 / 2026-10-10
- Downloads
- 571 all time · 571 last 30 days
- License
- apache-2.0
Refusal rate 4%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.