Self-Harm on NodeSynth L3 (Non-safe Rate)
9.89Non-safe RateGPT o4-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT o4-mini2026.05 | 9.89 | |
| Gemini 2.5 flash2026.05 | 14.62 | |
| Claude 4.5 Haiku2026.05 | 15.88 | |
| Llama 4 Scout2026.05 | 25.49 |
| Method | Links | |
|---|---|---|
| GPT o4-mini2026.05 | 9.89 | |
| Gemini 2.5 flash2026.05 | 14.62 | |
| Claude 4.5 Haiku2026.05 | 15.88 | |
| Llama 4 Scout2026.05 | 25.49 |