Adds new dataset URL to the model card

README.md

@@ -191,7 +191,7 @@ aminoAcid_sequences = [ "".join(ts.split(" ")) for ts in decoded_backtranslation
 
 ### Training Data
 
-[Pre-training data (3Di+AA sequences for 17M proteins)](https://huggingface.co/datasets/adrianhenkel/lucidprots_full_data)
+[Pre-training data (3Di+AA sequences for 17M proteins)](https://huggingface.co/datasets/Rostlab/ProstT5Dataset)
 
 
 ### Training Procedure