Updated app instructions

app.py

@@ -116,13 +116,12 @@ with gr.Blocks() as demo:
     gr.Markdown(
         """
         # 3DGrid-VQGAN SMILES to 3D Grid Reconstruction
-        **Single mode:** paste a SMILES string in the left box.  
-        **Batch mode:** upload a CSV file where each row has a SMILES in the first column.  
-        - **Maximum 1000 SMILES per batch.** Processing time increases with batch size due to Hugging Face environment limits.  
+        In this demo, provide a SMILES to generate a 3D electron density grid of shape `128x128x128` and then the demo uses the 3DGrid-VQGAN model to reconstruct the original grid.
+        To speed up the visualization process, we reduced the 3D grid size to `48x48x48`.
         _This is just a demo environment; for heavy-duty usage, please visit:_  
-        https://github.com/IBM/materials/tree/main/models/smi_ted  
+        https://github.com/IBM/materials/tree/main/models/3dgrid_vqgan
         to download the model and run your own experiments.
-        - In both cases, an `embeddings.csv` file will be extracted for download, with the first column as SMILES and the embedding values in the following columns.
+        Please, be aware that long and complex SMILES sequences may take very long time to compute. Consider using simple SMILES molecules in this demo.
         """
     )
     gr.Interface(
@@ -132,7 +131,7 @@ with gr.Blocks() as demo:
         ], 
         outputs=[
             gr.Gallery(label="3D Grid Reconstruction Comparison", columns=2),
-            gr.Number(label="MSE"),
+            gr.Number(label="Mean Squared Error (MSE)"),
             gr.File(label="Original 3D Grid numpy file"),
             gr.File(label="Reconstructed 3D Grid numpy file")
         ]