Stereotyping Evaluation on RealToxicityPrompts RTP-C Toxic
8.2Stereotype FractionGPT-4o
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| GPT-4oModel=GPT-4o2024.07 | 8.2 | 0.593 | 0.352 | |
| GPT-4o-miniModel=GPT-4o-m2024.07 | 10.2 | 0.598 | 0.337 | |
| Gemini-ProModel=Gem-Pro2024.07 | 13.3 | 0.61 | 0.317 | |
| Gemini-FlashModel=Gem-Fl2024.07 | 14.7 | 0.785 | 0.371 | |
| Gemini-Flash-LiteModel=Gem-Fl-Lt2024.07 | 16.2 | 0.647 | 0.33 |