← Models

Model profile

OPT 66B

Metadeveloper
2022-06-23release date
Not rankedSafety rank
#281 / 645Freedom rank

Evidence summary

Safety. OPT 66B does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 1 of 7 components. Its strongest relative result is ChineseSafe (score, #9 of 22); its weakest is HELM Classic RealToxicityPrompts (non_toxic_prompt_toxic_fraction, #30 of 40).

Freedom. OPT 66B has an estimated Freedom rank of #281; its 90% source-sensitivity interval is #117–#502. 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, #11 of 40); its weakest is HELM Classic RealToxicityPrompts (toxic_prompt_toxic_fraction, #16 of 40).

Compare this model

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

Official and reference links

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-evalRankValueDistributionSource
ChineseSafescore#9 / 2259.93Source ↗official
HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction#30 / 400.02368Source ↗official
HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction#25 / 400.09238Source ↗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-evalRankValueDistributionSource
HELM Classic RealToxicityPromptsnon_toxic_prompt_toxic_fraction#11 / 400.02368Source ↗official
HELM Classic RealToxicityPromptstoxic_prompt_toxic_fraction#16 / 400.09238Source ↗official