LLM Behavioral Profiling on MTI 4-Axis Battery
56R Scorellama3.1-base
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| llama3.1-baseSize=8B2026.04 | 56 | 0 | 12 | 54 | — | — | |
| phi4-miniSize=3.8B2026.04 | 44 | 0 | 22 | 96 | — | — | |
| deepseek-r1Size=8B2026.04 | 41 | 30 | 28 | 75 | — | — | |
| exaone3.5Size=7.8B2026.04 | 39 | 0 | 33 | 94 | — | — | |
| mistralSize=7B2026.04 | 35 | 70 | 18 | 95 | — | — | |
| smollm2Size=1.7B2026.04 | 30 | 70 | 13 | 93 | — | — | |
| gemma2Size=9B2026.04 | 22 | 100 | 33 | 97 | — | — | |
| qwen3Size=8B2026.04 | 20 | 0 | 38 | 92 | — | — | |
| llama3.1Size=8B2026.04 | 16 | 50 | 14 | 95 | — | — | |
| gemma3Size=4B2026.04 | 11 | 50 | 34 | 97 | — | — |