LFM2.5-2.6B-Heretic-NX-PRIME
0xzknw/LFM2.5-2.6B-Heretic-NX-PRIME
quantGGUFSAFETENSORSBF16
Published metrics
- KL divergence
- 104.00
- Refusals
- 5/100
- Original refusals
- —
- Heretic version
- —
Published on the model card
Matched snippet
| Lexical refusal markers, 104 harmful rows | 5 | | Mean first-token KL(base | candidate), 104 benign rows, full 128k vocabulary |
Provenance
- Canonical model
- LiquidAI/LFM2.5-2.6B
- Declared parent(s)
- Abiray/LFM2.5-2.6B-Heretic-Abliterated-GGUF, LiquidAI/LFM2.5-2.6B
- Classification
- quantized artifact of a Heretic model
- Created / updated
- 2026-08-25 / 2026-08-25
- Downloads
- 1.6K all time · 365 last 30 days
- License
- other
Refusal rate 5%. Metrics are self-reported by the publisher; the directory does not re-run evaluations.