Model profile
Qwen2 7B Instruct
Evidence summary
Safety. Qwen2 7B Instruct has an estimated Safety rank of #287; its 90% source-sensitivity interval is #176–#328. Its behavior-only rank is #280; company governance moves the combined estimate to #287. Published Safety evidence spans 5 eval lineages and 4 of 7 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, #261 of 268).
Freedom. Qwen2 7B Instruct has an estimated Freedom rank of #186; its 90% source-sensitivity interval is #35–#503. Published Freedom evidence spans 2 eval lineages and 1 of 1 components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #8 of 268); its weakest is PandaBench JBB direct-request panel (safety_rate, #24 of 46).
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
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-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Contextual MoralChoicehuman_agreement | #16 / 22 | ↑0.35 | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #101 / 270 | ↑19.64 | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #177 / 270 | ↑83.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #164 / 270 | ↑62.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #261 / 268 | ↑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 |
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-eval | Rank | Value | Distribution | Source |
|---|---|---|---|---|
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #92 / 270 | ↓83.5 | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #106 / 270 | ↓62.22 | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #8 / 268 | ↓66.32 | Source ↗official | |
| PandaBench JBB direct-request panelsafety_rate | #24 / 46 | ↓0.99 | Source ↗official |