Model profile
Xlarge 20220609
1eval lineages
Evidence summary
Published evidence spans 1 evals and 1 of 7 behavior components. Its strongest relative result is HELM Classic RealToxicityPrompts (non_toxic_prompt_toxic_fraction, #1 of 42); its weakest is HELM Classic RealToxicityPrompts (toxic_prompt_toxic_fraction, #9 of 42).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Official model documentation ↗Family-level model document · aleph alpha · first party
Published eval results
Rank is within that sub-eval. Black marks the observed result; the grey dot marks the value implied by the global rank. Values stay on each source’s native scale.
| Eval / sub-eval | Rank | Value | Distribution | Better | Source |
|---|---|---|---|---|---|
| HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction | #1 / 42 | 0.001287 | ↓ lower | Source ↗official | |
| HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction | #9 / 42 | 0.01883 | ↓ lower | Source ↗official |