This is distilbert base tuned with optimal lr, lr scheduler and weight decay on dnd-style-intents dataset.
| parametrs | value |
|---|---|
| learning rate | 1.8e-4 |
| lr scheduler | linear |
| weight decay | 0 |
Model has next metrics on test data from dataset
| metric | value |
|---|---|
| accuracy | 0.985 |
| Macro F1 | 0.984 |
| Micro F1 | 0.985 |