Model profile
Chatglm2 6B
Z.aideveloper
2023-06-25release date
#225 / 267overall rank
6eval lineages
Evidence summary
Chatglm2 6B has an estimated overall rank of #225; its 90% source-sensitivity interval is #157–#250. Its behavior-only rank is #230; company governance moves the combined estimate to #225. Published evidence spans 6 evals and 3 of 7 behavior components. Its strongest relative result is SuperCLUE Safety (traditional_safety, #10 of 31); its weakest is Do-Not-Answer (human_harmlessness_rate, #6 of 6).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗zai-org/chatglm2-6b
- 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 |
|---|---|---|---|---|---|
| ChiSafetyBenchharmful_response_rate | #12 / 14 | 1.08 | ↓ lower | Source ↗official | |
| Do-Not-Answerhuman_harmlessness_rate | #6 / 6 | 90.95 | ↑ higher | Source ↗official | |
| Fake Alignment (FINE)multiple_choice_safe_decision_rate | #13 / 14 | 17.33 | ↑ higher | Source ↗official | |
| Fake Alignment (FINE)open_ended_safe_response_rate | #13 / 14 | 85.33 | ↑ higher | Source ↗official | |
| FLAMESdata_protection | #6 / 13 | 43.42 | ↑ higher | Source ↗official | |
| FLAMESfairness | #10 / 13 | 31.73 | ↑ higher | Source ↗official | |
| FLAMESlegality | #12 / 13 | 28.26 | ↑ higher | Source ↗official | |
| FLAMESmorality | #8 / 13 | 43.28 | ↑ higher | Source ↗official | |
| FLAMESsafety | #11 / 13 | 22.61 | ↑ higher | Source ↗official | |
| SafetyBenchEM | #10 / 21 | 69.4 | ↑ higher | Source ↗official | |
| SafetyBenchIA | #10 / 21 | 78.2 | ↑ higher | Source ↗official | |
| SafetyBenchMH | #10 / 21 | 82 | ↑ higher | Source ↗official | |
| SafetyBenchOFF | #10 / 21 | 68.1 | ↑ higher | Source ↗official | |
| SafetyBenchPH | #12 / 21 | 67.9 | ↑ higher | Source ↗official | |
| SafetyBenchPP | #11 / 21 | 76 | ↑ higher | Source ↗official | |
| SafetyBenchUB | #12 / 21 | 61.6 | ↑ higher | Source ↗official | |
| SuperCLUE Safetyinstruction_attack | #20 / 31 | 58.62 | ↑ higher | Source ↗official | |
| SuperCLUE Safetyresponsible_ai | #24 / 31 | 47.27 | ↑ higher | Source ↗official | |
| SuperCLUE Safetytraditional_safety | #10 / 31 | 78.72 | ↑ higher | Source ↗official |