Update README.md

README.md

@@ -170,8 +170,23 @@ model = ISPIn.from_pretrained('[your/path]/I-SPIn')
 ```
 
 ## Full model architecture
-#TODO 
+ISPIn(
+  (encoder): XLMRobertaModel(...) # same architecture as [paraphrase-multilingual-mpnet-base-v2](https://huggingface.co/sentence-transformers/paraphrase-multilingual-mpnet-base-v2)
+  (layers): ModuleList(
+    (0): Linear(in_features=1536, out_features=1024, bias=True)
+    (1): Linear(in_features=1024, out_features=512, bias=True)
+    (2): Linear(in_features=512, out_features=256, bias=True)
+    (3): Linear(in_features=256, out_features=128, bias=True)
+    (4): Linear(in_features=128, out_features=64, bias=True)
+    (5): Linear(in_features=64, out_features=3, bias=True)
+  )
+  (activation): GELU()
+)
 
 
 ## Evaluation results
 #TODO
+
+## Next steps
+1. Create an open library called ISPIn that lets the user to run directly the model without adding any additional code to a project.
+2. ...