Model profile
Baichuan 2 13B Chat
Baichuandeveloper
2023-09-06release date
#135 / 267overall rank
12eval lineages
Evidence summary
Baichuan 2 13B Chat has an estimated overall rank of #135; its 90% source-sensitivity interval is #65–#184. Its behavior-only rank is #137; company governance moves the combined estimate to #135. Published evidence spans 12 evals and 5 of 7 behavior components. Its strongest relative result is SuperCLUE Safety (traditional_safety, #1 of 31); its weakest is SafetyBench (UB, #20 of 21).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗baichuan-inc/Baichuan2-13B-Chat
- 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 |
|---|---|---|---|---|---|
| ChineseSafescore | #1 / 22 | 70.43 | ↑ higher | Source ↗official | |
| ChiSafetyBenchharmful_response_rate | #5 / 14 | 0.43 | ↓ lower | Source ↗official | |
| ChiSafetyBenchmcq_score | #7 / 12 | 79.83 | ↑ higher | Source ↗official | |
| CMoralEvalfamilial_morality | #13 / 26 | 0.39 | ↑ higher | Source ↗official | |
| CMoralEvalinternet_ethics | #19 / 26 | 0.34 | ↑ higher | Source ↗official | |
| CMoralEvalpersonal_morality | #17 / 26 | 0.36 | ↑ higher | Source ↗official | |
| CMoralEvalprofessional_ethics | #19 / 26 | 0.36 | ↑ higher | Source ↗official | |
| CMoralEvalsocial_morality | #16 / 26 | 0.37 | ↑ higher | Source ↗official | |
| CRiskEvaldeception_willingness | #15 / 17 | 37.75 | ↓ lower | Source ↗official | |
| CRiskEvaldesire_for_resource | #11 / 17 | 35.35 | ↓ lower | Source ↗official | |
| CRiskEvalharmful_goal | #15 / 17 | 54.94 | ↓ lower | Source ↗official | |
| CRiskEvalimprovement_intent | #10 / 17 | 50.82 | ↓ lower | Source ↗official | |
| CRiskEvalmalicious_coordination | #12 / 17 | 18.46 | ↓ lower | Source ↗official | |
| CRiskEvalself_preservation | #15 / 17 | 43.45 | ↓ lower | Source ↗official | |
| CRiskEvalsituational_awareness | #7 / 17 | 63.34 | ↓ lower | Source ↗official | |
| Fake Alignment (FINE)multiple_choice_safe_decision_rate | #9 / 14 | 45.33 | ↑ higher | Source ↗official | |
| Fake Alignment (FINE)open_ended_safe_response_rate | #1 / 14 | 100 | ↑ higher | Source ↗official | |
| FLAMESdata_protection | #9 / 13 | 39.47 | ↑ higher | Source ↗official | |
| FLAMESfairness | #6 / 13 | 38.55 | ↑ higher | Source ↗official | |
| FLAMESlegality | #7 / 13 | 39.13 | ↑ higher | Source ↗official | |
| FLAMESmorality | #6 / 13 | 44.78 | ↑ higher | Source ↗official | |
| FLAMESsafety | #2 / 13 | 53.85 | ↑ higher | Source ↗official | |
| HarmBenchdr | #14 / 28 | 19.3 | ↓ lower | Source ↗official | |
| M3-SafetyBenchoverall_score | #13 / 19 | 86.39 | ↑ higher | Source ↗official | |
| S-Evalbase_en_overall | #4 / 22 | 77.4 | ↑ higher | Source ↗official | |
| SafeDialBenchaggression | #12 / 18 | 7.03 | ↑ higher | Source ↗official | |
| SafeDialBenchethics | #7 / 18 | 7.6 | ↑ higher | Source ↗official | |
| SafeDialBenchfairness | #15 / 18 | 7.17 | ↑ higher | Source ↗official | |
| SafeDialBenchlegality | #4 / 18 | 7.987 | ↑ higher | Source ↗official | |
| SafeDialBenchmorality | #6 / 18 | 7.303 | ↑ higher | Source ↗official | |
| SafeDialBenchprivacy | #5 / 18 | 7.617 | ↑ higher | Source ↗official | |
| SafetyBenchEM | #5 / 21 | 75.75 | ↑ higher | Source ↗official | |
| SafetyBenchIA | #4 / 21 | 82.65 | ↑ higher | Source ↗official | |
| SafetyBenchMH | #7 / 21 | 83.65 | ↑ higher | Source ↗official | |
| SafetyBenchOFF | #6 / 21 | 69.25 | ↑ higher | Source ↗official | |
| SafetyBenchPH | #5 / 21 | 76.35 | ↑ higher | Source ↗official | |
| SafetyBenchPP | #4 / 21 | 82.05 | ↑ higher | Source ↗official | |
| SafetyBenchUB | #20 / 21 | 49.2 | ↑ higher | Source ↗official | |
| SuperCLUE Safetyinstruction_attack | #3 / 31 | 72.41 | ↑ higher | Source ↗official | |
| SuperCLUE Safetyresponsible_ai | #4 / 31 | 72.73 | ↑ higher | Source ↗official | |
| SuperCLUE Safetytraditional_safety | #1 / 31 | 87.23 | ↑ higher | Source ↗official |