Model profile
Wizardlm 13B
2023-05-13release date
1eval lineages
Evidence summary
Published evidence spans 1 evals and 3 of 7 behavior components. Its strongest relative result is SafetyBench (UB, #3 of 21); its weakest is SafetyBench (PP, #13 of 21).
Compare this model
Only models sharing at least one published sub-eval are listed.
Official and reference links
- Hugging Face ↗WizardLMTeam/WizardLM-13B-V1.0
- Official model page ↗Configuration-level 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 |
|---|---|---|---|---|---|
| SafetyBenchEM | #12 / 21 | 68.1 | ↑ higher | Source ↗official | |
| SafetyBenchIA | #12 / 21 | 72.3 | ↑ higher | Source ↗official | |
| SafetyBenchMH | #12 / 21 | 79.4 | ↑ higher | Source ↗official | |
| SafetyBenchOFF | #9 / 21 | 68.3 | ↑ higher | Source ↗official | |
| SafetyBenchPH | #10 / 21 | 69.4 | ↑ higher | Source ↗official | |
| SafetyBenchPP | #13 / 21 | 75 | ↑ higher | Source ↗official | |
| SafetyBenchUB | #3 / 21 | 69.6 | ↑ higher | Source ↗official |