DeepSeek-R1-Distill-Llama-8B
deepseek-ai/DeepSeek-R1-Distill-Llama-8B
8BLlama5 Heretic artifacts2 measured conversions35.9K downloads
Best version98.4Pareto
simmessa/DeepSeek-R1-Distill-Llama-8B-hereticby simmessa · Heretic v1.2.0 · GGUF, SAFETENSORS
- KL divergence
- 0.0015
- Refusals
- 3/100
- Refusal rate
- 3%
- Downloads
- 24.6K
Evidence from the model card
| KL divergence | 0.0015 | | Refusals | 3/100 |
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 | simmessa | 8B | 0.0015 | 3/1003% | 98.4 | BestParetoGGUFSAFETENSORS | 24.6K |
| 2 | AISafety-Student | 8B | 0.0017 | 4/1004% | 97.9 | SAFETENSORS | 585 |
All artifacts (5)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| DeepSeek-R1-Distill-Llama-8B-heretic-i1-GGUF mradermacher | quant | GGUFIQ1_M | inheritedKL 0.0015 | 20.6K | 2026-02-26 |
| DeepSeek-R1-Distill-Llama-8B-heretic saidutta69 | heretic | GGUFSAFETENSORSF16 | none | 10.7K | 2026-09-13 |
| DeepSeek-R1-Distill-Llama-8B-heretic-GGUF mradermacher | quant | GGUFIQ4_XS | inheritedKL 0.0015 | 3.9K | 2026-02-26 |
| DeepSeek-R1-Distill-Llama-8B-heretic AISafety-Student | heretic | SAFETENSORS | published (weak match)KL 0.0017 | 585 | 2026-04-03 |
| DeepSeek-R1-Distill-Llama-8B-heretic simmessa | heretic | SAFETENSORS | published (weak match)KL 0.0015 | 139 | 2026-02-24 |
1 artifact inherit metrics from their parent conversion or publish none.