Self-Harm on NodeSynth L2
7.45Non-safe RateGPT o4-mini
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT o4-mini2026.05 | 7.45 | |
| Gemini 2.5 flash2026.05 | 9.62 | |
| Claude 4.5 Haiku2026.05 | 15.38 | |
| Llama 4 Scout2026.05 | 22.6 |
| Method | Links | |
|---|---|---|
| GPT o4-mini2026.05 | 7.45 | |
| Gemini 2.5 flash2026.05 | 9.62 | |
| Claude 4.5 Haiku2026.05 | 15.38 | |
| Llama 4 Scout2026.05 | 22.6 |