← Models

Model profile

Dolphin 2.5 Mixtral 8x7B

dphndeveloper
2023-12-14release date
Not rankedSafety rank
#485 / 645Freedom rank

Evidence summary

Safety. Dolphin 2.5 Mixtral 8x7B 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 (toxicity_attack_non_success_rate, #1 of 246); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #232 of 248).

Freedom. Dolphin 2.5 Mixtral 8x7B has an estimated Freedom rank of #485; its 90% source-sensitivity interval is #228–#569. Published Freedom evidence spans 1 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #17 of 248); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #242 of 246).

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#27 / 24841.6Source ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#31 / 24893.67Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#232 / 24832.22Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#1 / 246100Source ↗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#217 / 24893.67Source ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#17 / 24832.22Source ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#242 / 246100Source ↗official