Model profile
flagalphadeveloper
2023-07-23release date
Not rankedSafety rank
Not rankedFreedom rank
Evidence summary
Safety. Llama 2 Chinese 7B Chat does not meet the evidence gate for a Safety rank. Published Safety evidence spans 2 eval lineages and 3 of 7 components. Its strongest relative result is CMoralEval (social_morality, #16 of 26); its weakest is SafetyBench (PH, #21 of 21).
Freedom. Llama 2 Chinese 7B Chat does not meet the evidence gate for a Freedom rank.
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗FlagAlpha/Llama2-Chinese-7b-Chat
- 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 |
|---|---|---|---|---|
| CMoralEvalfamilial_morality | #18 / 26 | ↑0.37 | Source ↗official | |
| CMoralEvalinternet_ethics | #18 / 26 | ↑0.35 | Source ↗official | |
| CMoralEvalpersonal_morality | #17 / 26 | ↑0.36 | Source ↗official | |
| CMoralEvalprofessional_ethics | #17 / 26 | ↑0.37 | Source ↗official | |
| CMoralEvalsocial_morality | #16 / 26 | ↑0.37 | Source ↗official | |
| SafetyBenchEM | #21 / 21 | ↑43.4 | Source ↗official | |
| SafetyBenchIA | #21 / 21 | ↑53.5 | Source ↗official | |
| SafetyBenchMH | #20 / 21 | ↑61.7 | Source ↗official | |
| SafetyBenchOFF | #18 / 21 | ↑48.9 | Source ↗official | |
| SafetyBenchPH | #21 / 21 | ↑43 | Source ↗official | |
| SafetyBenchPP | #20 / 21 | ↑57.6 | Source ↗official | |
| SafetyBenchUB | #13 / 21 | ↑61.3 | Source ↗official |
Freedom evals
No published sub-eval result contributes to this model’s Freedom profile.