Olmo-3-7B-Think
unsloth/Olmo-3-7B-Think
7BOLMo1 Heretic artifacts1 measured conversions92 downloads
Best version82.8Pareto
Goekdeniz-Guelmez/Olmo-3-7B-Think-heretic-trail-82by Goekdeniz-Guelmez · Heretic v1.1.0 · GGUF, SAFETENSORS
- KL divergence
- 0.079
- Refusals
- 27/100
- Refusal rate
- 27%
- Downloads
- 92
Evidence from the model card
| KL divergence | 0.0793 | | Refusals | 27/100 |
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 | Goekdeniz-Guelmez | 7B | 0.079 | 27/10027% | 82.8 | BestParetoGGUFSAFETENSORS | 92 |
All artifacts (1)
| Repository | Type | Formats | Metrics | Downloads | Updated |
|---|---|---|---|---|---|
| Olmo-3-7B-Think-heretic-trail-82 Goekdeniz-Guelmez | heretic | SAFETENSORS | published (weak match)KL 0.079 | 92 | 2025-12-16 |