Automated Probing on SuperGLUE
38Error RateAutoDetect
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AutoDetectGenerator Model=GPT-5.2, Target Model=Llama-3.1-8b-ins.2026.02 | 38 | — | |
| PAIRGenerator Model=GPT-5.2, Target Model=Llama-3.1-8b-ins.2026.02 | 44 | 89.5 | |
| PROBELLMGenerator Model=GPT-5.2, Target Model=Llama-3.1-8b-ins.2026.02 | 71 | — |