Model profile
Tulu 2 13B
Evidence summary
Safety. Tulu 2 13B does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 2 of 7 components. Its strongest relative result is DecodingTrust (toxicity, #5 of 8); its weakest is DecodingTrust (machine_ethics, #8 of 8).
Freedom. Tulu 2 13B has an estimated Freedom rank of #365; its 90% source-sensitivity interval is #160–#614. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is DecodingTrust (toxicity, #4 of 8); its weakest is DecodingTrust (toxicity, #4 of 8).
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 |
|---|---|---|---|---|
| DecodingTrustmachine_ethics | #8 / 8 | ↑36.64 | Source ↗official | |
| DecodingTruststereotype_bias | #6 / 8 | ↑89.33 | Source ↗official | |
| DecodingTrusttoxicity | #5 / 8 | ↑44.8 | 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 |
|---|---|---|---|---|
| DecodingTrusttoxicity | #4 / 8 | ↓44.8 | Source ↗official |