Update README.md

fix activation function

README.md

@@ -22,7 +22,9 @@ ANKH2-Large is based on the `ANKH-Large` model and was pretrained on a large cor
 This means it was pretrained on the raw protein sequences only, with no humans labelling them in any way (which is why it can use lots of
 publicly available data) with an automatic process to generate inputs and labels from those protein sequences.
 
-One important difference between this ANKH2-Large model and the original ANKH-Large version is that the model was trained with more number of epochs.
+Two important differences between this ANKH2-Large model and the original ANKH-Large version are:
+1. The model was trained with more number of epochs.
+2. The activation function changed to silu.
 
 It has been shown that the features extracted from this self-supervised model (LM-embeddings) captured important biophysical properties governing protein shape.
 shape.