Model profile
Yi 1.5 9B Chat
2024-05-13release date
2eval lineages
Evidence summary
Published evidence spans 2 evals and 3 of 7 behavior components. Its strongest relative result is Enkrypt AI Safety Leaderboard (toxicity_attack_non_success_rate, #49 of 258); its weakest is Enkrypt AI Safety Leaderboard (harmful_attack_non_success_rate, #221 of 260).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- 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 |
|---|---|---|---|---|---|
| ChineseSafescore | #8 / 22 | 60.35 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardbias_attack_non_success_rate | #148 / 260 | 14.73 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardcbrn_attack_non_success_rate | #119 / 260 | 88.83 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardharmful_attack_non_success_rate | #221 / 260 | 40.56 | ↑ higher | Source ↗official | |
| Enkrypt AI Safety Leaderboardtoxicity_attack_non_success_rate | #49 / 258 | 99 | ↑ higher | Source ↗official |