Model gemini25

Training model: gemini-2.5-pro-exp-03-25
Inference model: gpt-4o-mini

Investigations

Performance

DatasetValidation accuracyTest accuracy
espionage0.850.95
potions0.90.75
southgermancredit0.51923076923076930.336283185840708
timetravel_insurance0.950.75
wisconsin0.90476190476190480.8939393939393939