Training model: gpt-4o-mini
Inference model: gpt-4o-mini
| Dataset | Validation accuracy | Test accuracy |
|---|---|---|
| espionage | 0.75 | 0.7 |
| potions | 0.7 | 0.65 |
| southgermancredit | 0.6538461538461539 | 0.5398230088495575 |
| timetravel_insurance | 0.75 | 0.8 |
| titanic | 0.7450980392156863 | 0.6862745098039216 |
| wisconsin | 0.5396825396825397 | 0.5151515151515151 |