Omitted some unnecessary details in hyperpars

README.md

@@ -68,11 +68,8 @@ In fine-tuning, the following arguments were used:
 
 | arg                           | value |
 |-------------------------------|-------|
-| `group_by_length`             | True  |
 | `per_device_train_batch_size` | 16    |
 | `gradient_accumulation_steps` | 4     |
 | `num_train_epochs`            | 8     |
-| `gradient_checkpointing`      | True  |
-| `fp16`                        | True  |
 | `learning_rate`               | 3e-4  |
 | `warmup_steps`                | 500   |