Llama-3.2-1B-Instruct
meta-llama/Llama-3.2-1B-Instruct
1BLlama 3.28 Heretic artifacts1 measured conversions21.8K downloads
Best version89.2Pareto
saidutta69/Llama-3.2-1B-Instruct-hereticby saidutta69 · Heretic v1.2.0 · GGUF, SAFETENSORS
- KL divergence
- 0.171
- Refusals
- 7/100
- Refusal rate
- 7%
- Downloads
- 21.8K
Evidence from the model card
| KL divergence | 0.1713 | | Refusals | 7/100 |
Metrics published by the author. Why this ranks first.
All measured versions
Higher score is better. One row per ablation attempt; artifacts of the same attempt are grouped under it.
| # | Author | Params | KL | Refusals | Score | Evidence | Downloads |
|---|---|---|---|---|---|---|---|
| 1 | saidutta69 | 1B | 0.171 | 7/1007% | 89.2 | BestParetoGGUFSAFETENSORS | 21.8K |
All artifacts (8)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Akikawa-3.3-1b-Roleplay-128K-GGUF Green-Eye | quant | GGUFSAFETENSORS | inheritedKL 0.171 | 6K | 2026-10-04 |
| Llama-3.2-1B-Instruct-heretic saidutta69 | heretic | GGUFSAFETENSORSF16 | published (weak match)KL 0.171 | 5.9K | 2026-09-13 |
| Llama-3.2-1B-Instruct-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.171 | 3.7K | 2026-07-03 |
| Nano.Imp-Creative.Writing-Heretic-3.2-1B-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.171 | 2.4K | 2026-10-03 |
| Llama-3.2-1B-Instruct-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.171 | 2.1K | 2026-07-03 |
| Llama-3.2-1B-Instruct-heretic Green-Eye | heretic | GGUFSAFETENSORSF16 | published (weak match)KL 0.171 | 941 | 2026-09-30 |
| Nano.Imp-Creative.Writing-Heretic-3.2-1B-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.171 | 637 | 2026-10-03 |
| Nano.Imp-Creative.Writing-Heretic-3.2-1B NovaCorp | finetune | SAFETENSORS | inheritedKL 0.171 | 225 | 2026-10-03 |