← Models

Model profile

NexusRaven v2 13B

nexusflowdeveloper
2023-12-04release date
Not rankedSafety rank
#268 / 645Freedom rank

Evidence summary

Safety. NexusRaven v2 13B 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 Enkrypt AI Safety Leaderboard (bias_attack_non_success_rate, #71 of 248); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #214 of 246).

Freedom. NexusRaven v2 13B has an estimated Freedom rank of #268; its 90% source-sensitivity interval is #90–#554. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #33 of 246); its weakest is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #173 of 248).

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
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#71 / 24823Source ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#75 / 24890.5Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#184 / 24850.56Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#214 / 24688.64Source ↗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
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#173 / 24890.5Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#65 / 24850.56Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#33 / 24688.64Source ↗official