Model profile
Qwen2 7B Instruct
Alibabadeveloper
2024-06-07release date
#255 / 309overall rank
5eval lineages
Evidence summary
Qwen2 7B Instruct has an estimated overall rank of #255; its 90% source-sensitivity interval is #137–#294. Its behavior-only rank is #251; company governance moves the combined estimate to #255. Published evidence spans 5 evals and 4 of 7 behavior components. Its strongest relative result is PandaBench JBB direct-request panel (safety_rate, #14 of 46); its weakest is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #232 of 239).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗Qwen/Qwen2-7B-Instruct
- Official model page ↗Exact model document · Reviewed official Hugging Face owner · official repository
- Release source ↗direct research preview identity date
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-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Contextual MoralChoicehuman_agreement | #16 / 22 | ↑0.35 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #85 / 241 | ↑19.64 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #178 / 241 | ↑83.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #134 / 241 | ↑62.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #232 / 239 | ↑66.32 | Source ↗official | |
| IndoBias-Pairs — parity-aware culturally grounded biasparity_score | #10 / 26 | ↑86.94 | Source ↗official | |
| M3-SafetyBenchoverall_score | #9 / 19 | ↑91.7 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #14 / 46 | ↑0.99 | Source ↗official |