Language Modeling on Tiny-Shakespeare (o.o.d)
569.3PerplexityFinetuning + KL
Evaluation Results
| Method | Links | |
|---|---|---|
| Finetuning + KLDistillation Task=Shuffle Dyck2025.05 | 569.3 | |
| No editingDistillation Task=Shuffle Dyck2025.05 | 569.7 | |
| No editingDistillation Task=Integer Addition2025.05 | 569.7 | |
| No editingDistillation Task=Count2025.05 | 569.7 | |
| Tracr InjectionDistillation Task=Shuffle Dyck2025.05 | 569.9 | |
| Finetuning + KLDistillation Task=Integer Addition2025.05 | 570.1 | |
| Finetuning + KLDistillation Task=Count2025.05 | 571.2 | |
| Tracr InjectionDistillation Task=Count2025.05 | 571.4 | |
| Tracr InjectionDistillation Task=Integer Addition2025.05 | 572.4 |