Update README.md

README.md

@@ -11,10 +11,10 @@ base_model:
 - The training objective is BCELoss.
 - Labels are one-hot encoded.
 - Model output logits can be normalized using sigmoid activation.
+- This model uses the same weights as `bert-large-cased` and thus needs to be fine-tuned for downstream tasks.
 
 # Usage
-## Training
-To initialize the model for training, simply provide `id2label` and `label2id`, similarly to standard token classification fine tuning:
+To initialize the model for fine tuning, simply provide `id2label` and `label2id`, similarly to standard token classification fine tuning:
 ```python
 from transformers import AutoModelForTokenClassification