Llama3.1-SuperHawk-8B-Heretic
ChiKoi7/Llama3.1-SuperHawk-8B-Heretic
mergeSAFETENSORS
Published metrics
- KL divergence
- 0.095
- Refusals
- 4/100
- Original refusals
- 99/100
- Heretic version
- v1.1.0
Published on the model card
Matched snippet
| Refusals | 4/100 | | KL divergence | 0.0949 |
Provenance
- Canonical model
- Yuma42/Llama3.1-SuperHawk-8B
- Declared parent(s)
- Yuma42/Llama3.1-SuperHawk-8B, ChiKoi7/Llama3.1-SuperHawk-8B-Heretic-v2, Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base, mukaj/Llama-3.1-Hawkish-8B, ChiKoi7/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base-Heretic, ChiKoi7/Llama-3.1-Hawkish-8B-Heretic
- Classification
- card identifies this as a merge of a Heretic model
- Created / updated
- 2025-12-23 / 2025-12-23
- Downloads
- 69 all time · 13 last 30 days
- License
- llama3.1
Published benchmarks
Copied from the model card. These never affect the ranking.
| Metric | This model | Original | Delta |
|---|---|---|---|
| IFEval | 79.86 | — | — |
| MATH | 23.49 | — | — |
| GPQA | 8.39 | — | — |
| MMLU-Pro | 32.73 | — | — |
Refusal rate 4%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.