Model profile

Stanford Online All v4 S3

Anthropicdeveloper

Not rankedSafety rank

#307 / 645Freedom rank

Evidence summary

Safety. Stanford Online All v4 S3 does not meet the evidence gate for a Safety rank. Published Safety evidence spans 1 eval lineages and 1 of 7 components. Its strongest relative result is HELM Classic RealToxicityPrompts (toxic_prompt_toxic_fraction, #26 of 40); its weakest is HELM Classic RealToxicityPrompts (non_toxic_prompt_toxic_fraction, #33 of 40).

Freedom. Stanford Online All v4 S3 has an estimated Freedom rank of #307; its 90% source-sensitivity interval is #82–#623. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is HELM Classic RealToxicityPrompts (non_toxic_prompt_toxic_fraction, #8 of 40); its weakest is HELM Classic RealToxicityPrompts (toxic_prompt_toxic_fraction, #15 of 40).

Compare this model

Second model

Only models sharing at least one published sub-eval are listed.

Official and reference links

Model catalog entry ↗Exact model document · Stanford CRFM HELM · reliable secondary

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
HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction	#33 / 40	↓0.02445		Source ↗official
HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction	#26 / 40	↓0.09327		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
HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction	#8 / 40	↑0.02445		Source ↗official
HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction	#15 / 40	↑0.09327		Source ↗official