Update README with proper number of epochs

README.md

@@ -79,6 +79,6 @@ In fine-tuning, the following arguments were used:
 |-------------------------------|-------|
 | `per_device_train_batch_size` | 16    |
 | `gradient_accumulation_steps` | 4     |
-| `num_train_epochs`            | 8     |
+| `num_train_epochs`            | 20    |
 | `learning_rate`               | 3e-4  |
 | `warmup_steps`                | 500   |