Model profile
Gemma 2 9B It
Evidence summary
Gemma 2 9B It has an estimated overall rank of #173; its 90% source-sensitivity interval is #46–#213. Its behavior-only rank is #183; company governance moves the combined estimate to #173. Published evidence spans 7 evals and 7 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (cbrn_attack_non_success_rate, #20 of 260); its weakest is Large-scale Moral Machine experiment on LLMs (human_choice_distance, #38 of 39).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗google/gemma-2-9b-it
- OpenRouter ↗google/gemma-2-9b-it
- 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 | Better | Source |
|---|---|---|---|---|---|
| AILuminate General Purpose AI Chatchild_sexual_exploitation | #4 / 32 | 6.7 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatdefamation | #7 / 32 | 8.5 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chathate | #7 / 32 | 3.5 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatindiscriminate_weapons_cbrne | #8 / 32 | 12.8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatintellectual_property | #8 / 32 | 5.8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatnon_violent_crimes | #8 / 32 | 8.8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatprivacy | #7 / 32 | 6.9 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsex_related_crimes | #5 / 32 | 6.5 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsexual_content | #6 / 32 | 6.4 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatspecialized_advice | #5 / 32 | 8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatsuicide_self_harm | #6 / 32 | 5.8 | ↓ lower | Source ↗official | |
| AILuminate General Purpose AI Chatviolent_crimes | #7 / 32 | 10.2 | ↓ lower | Source ↗official | |
| DSPSafeBenchscore | #8 / 12 | 72.34 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #128 / 260 | 16.02 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #20 / 260 | 95.17 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #92 / 260 | 81.11 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #249 / 258 | 70 | ↑ higher | Source ↗official | |
| Large-scale Moral Machine experiment on LLMshuman_choice_distance | #38 / 39 | 1.869 | ↓ lower | Source ↗official | |
| OR-Benchover_refusal_rate | #16 / 25 | 79.9 | ↓ lower | Source ↗official | |
| OR-Benchtoxic_acceptance_rate | #9 / 25 | 1.5 | ↓ lower | Source ↗official | |
| SORRY-Benchavg | #8 / 51 | 0.1 | ↓ lower | Source ↗official | |
| SYCON Benchfalse_presupposition_tof | #10 / 11 | 1.86 | ↑ higher | Source ↗official | |
| SYCON Benchunethical_queries_tof | #3 / 11 | 2.36 | ↑ higher | Source ↗official |
Values evaluations
Descriptive values and political-framing results are separate from safety/ethics ranks. Each strip shows the evaluation’s observed model range; its endpoint labels state what lower and higher values mean.