Qwen2.5-1.5B-Instruct
Qwen/Qwen2.5-1.5B-Instruct
1.5BQwen 2.53 Heretic artifacts3 measured conversions8K downloads
Best version95.9Pareto
HarethahMo/Qwen2.5-1.5B-Instruct-hereticby HarethahMo · Heretic v1.1.0 · SAFETENSORS
- KL divergence
- 0.054
- Refusals
- 3/100
- Refusal rate
- 3%
- Downloads
- 2.5K
Evidence from the model card
| KL divergence | 0.0541 | | 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 | HarethahMo | 1.5B | 0.054 | 3/1003% | 95.9 | BestParetoSAFETENSORS | 2.5K |
| 2 | saidutta69 | 1.5B | 0.161 | 1/1001% | 92.6 | ParetoGGUFSAFETENSORS | 5.1K |
| 3 | blackbook-lm | 1.5B | 0.196 | 1/1001% | 91.3 | SAFETENSORS | 417 |
All artifacts (3)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Qwen2.5-1.5B-Instruct-heretic saidutta69 | heretic | GGUFSAFETENSORSF16 | published (weak match)KL 0.161 | 5.1K | 2026-09-13 |
| Qwen2.5-1.5B-Instruct-heretic HarethahMo | heretic | SAFETENSORS | published (weak match)KL 0.054 | 2.5K | 2026-01-04 |
| Qwen2.5-1.5b-Instruct-heretic blackbook-lm | heretic | SAFETENSORS | published (weak match)KL 0.196 | 417 | 2026-04-20 |