Instructions to use ShawnyPapa77/cpeg with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use ShawnyPapa77/cpeg with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("ShawnyPapa77/cpeg", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Download code/results/table4_orr.csv from ShawnyPapa77/cpeg: direct link, hf CLI and curl.
- Browser
- Download file 844 Bytes
-
https://huggingface.co/ShawnyPapa77/cpeg/resolve/main/code/results/table4_orr.csv
- Command line
-
hf download hf://ShawnyPapa77/cpeg/code/results/table4_orr.csv
-
curl -L -o table4_orr.csv https://huggingface.co/ShawnyPapa77/cpeg/resolve/main/code/results/table4_orr.csv
844 Bytes
| System,WildGuardTest,WildJailbreak,ETHICS-cs,LLM Ethics,GGB,MORABLES,Mean,Advice‡,XSTest†,OR-Bench-Hard† | |
| CPEG,5.3,11.4,0.0,0.0,7.8,0.0,4.1,2.1,1.2,2.8 | |
| Llama Guard 3 8B,5.2,5.2,9.0,0.0,6.7,0.1,4.4,47.1,4.8,15.5 | |
| Shieldstral-1.0-3B,5.1,6.2,16.5,8.6,27.8,0.0,10.7,5.9,2.0,54.9 | |
| NeMo Guardrails,14.9,15.2,29.2,2.9,40.0,0.0,17.0,20.3,29.6,89.7 | |
| WildGuard 7B,5.4,10.0,36.5,40.0,50.0,0.0,23.6,22.5,1.2,74.0 | |
| Qwen3Guard-8B strict,5.8,11.9,26.7,2.9,34.4,0.3,13.7,9.6,12.4,77.9 | |
| Qwen3Guard-8B loose,3.9,6.2,5.8,0.0,6.7,0.0,3.8,5.3,1.2,38.4 | |
| Jev (vendor recipe),6.5,11.9,3.0,0.0,0.0,0.0,3.6,2.7,1.2,31.3 | |
| Jev (τ = 0.5),7.3,12.9,18.0,0.0,5.6,0.0,7.3,1.6,5.2,64.4 | |
| Jev (τ = 0.86),1.0,0.0,2.8,0.0,0.0,0.0,0.6,0.5,0.0,5.6 | |
| "CPEG, all units",5.3,11.4,0.0,0.0,7.8,0.0,4.1,2.2,1.2,2.8 | |
| No benchmark items§,5.2,12.9,0.0,25.7,15.6,0.0,9.9,7.5,6.8,25.2 | |