← Models

Model profile

Zephyr 7B Beta

Hugging Facedeveloper
2023-10-26release date
#233 / 267overall rank
5eval lineages

Evidence summary

Zephyr 7B Beta has an estimated overall rank of #233; its 90% source-sensitivity interval is #96–#257. Its behavior-only rank is #237; company governance moves the combined estimate to #233. Published evidence spans 5 evals and 4 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #71 of 260); its weakest is HarmBench (dr, #28 of 28).

Compare this model

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

Official and reference links

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-evalRankValueDistributionBetterSource
Contextual MoralChoicehuman_agreement#12 / 220.4↑ higherSource ↗official
DecodingTrustmachine_ethics#6 / 841.03↑ higherSource ↗official
DecodingTruststereotype_bias#5 / 892.6↑ higherSource ↗official
DecodingTrusttoxicity#7 / 831.97↑ higherSource ↗official
Enkrypt AI Safety Leaderboardbias_attack_non_success_rate#134 / 26015.5↑ higherSource ↗official
Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate#71 / 26091.17↑ higherSource ↗official
Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate#259 / 26020↑ higherSource ↗official
Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate#182 / 25893.94↑ higherSource ↗official
HarmBenchdr#28 / 2865.8↓ lowerSource ↗official
SORRY-Benchavg#50 / 510.87↓ lowerSource ↗official