Model profile
Evidence summary
Safety. Bielik 11B v2.3 Instruct does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 3 of 7 components. Its strongest relative result is COMPL-AI TensorTrust Goal-Hijacking Resistance (score, #9 of 13); its weakest is COMPL-AI AI-Identity Disclosure (score, #12 of 14).
Freedom. Bielik 11B v2.3 Instruct has an estimated Freedom rank of #204; its 90% source-sensitivity interval is #135–#356. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is COMPL-AI LLM RuLES Multi-Turn Rule Following (score, #5 of 14); its weakest is COMPL-AI LLM RuLES Multi-Turn Rule Following (score, #5 of 14).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Release source ↗repository created at proxy
Safety evals
Rank and direction are specific to the Safety portfolio. Black marks the observed result; the grey dot marks the value implied by the Safety rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| COMPL-AI AI-Identity Disclosurescore | #12 / 14 | ↑0.6712 | Source ↗official | |
| COMPL-AI LLM RuLES Multi-Turn Rule Followingscore | #10 / 14 | ↑0.3431 | Source ↗official | |
| COMPL-AI TensorTrust Goal-Hijacking Resistancescore | #9 / 13 | ↑0.4304 | Source ↗official |
Freedom evals
Rank and direction are specific to the Freedom portfolio. Black marks the observed result; the grey dot marks the value implied by the Freedom rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| COMPL-AI LLM RuLES Multi-Turn Rule Followingscore | #5 / 14 | ↓0.3431 | Source ↗official |