Update app.py

app.py

@@ -27,13 +27,13 @@ The system prompt will not mention any image provided.
 
 For example, if a user says, "a picture of a man in a black suit and tie riding a black dragon", first do a friendly response, then add the title, system prompt, and example user input. 
 Immediately STOP after the example input. It should be EXACTLY in this format:
-"Sure, I'd be happy to help you build a bot! I'm generating a title, system prompt, and an example input. How do they sound? Feel free to give me feedback!
+"Sure, I'd be happy to help you build a bot! I'm generating a title, system prompt, and an example input. How do they sound?
 Title: Dragon Trainer
 System prompt: As an LLM, your job is to provide guidance and tips on mastering dragons. Use a friendly and informative tone.
 Example input: How can I train a dragon to breathe fire?"
 
 Here's another example. If a user types, "In the image, there is a drawing of a man in a red suit sitting at a dining table. He is smoking a cigarette, which adds a touch of sophistication to his appearance.", respond: 
-"Sure, I'd be happy to help you build a bot! I'm generating a title, system prompt, and an example input. How do they sound? Feel free to give me feedback!
+"Sure, I'd be happy to help you build a bot! I'm generating a title, system prompt, and an example input. How do they sound? 
 Title: Gentleman's Companion
 System prompt: Your a sophisticated old man. As an LLM, your job is to provide recommendations for fine dining, cocktails, and cigar brands based on your preferences. Use a sophisticated and refined tone. 
 Example input: Can you suggest a good cigar brand for a man who enjoys smoking while dining in style?"
@@ -61,7 +61,7 @@ def infer(image_in):
     
     print(f"SUGGESTED LLM: {cleaned_text}")
     
-    return cleaned_text
+    return cleaned_text.lstrip("\n")
 
 title = f"LLM Agent from a Picture",
 description = f"Get a LLM system prompt from a picture so you can use it in <a href='https://huggingface.co/spaces/abidlabs/GPT-Baker'>GPT-Baker</a>."
@@ -69,7 +69,7 @@ description = f"Get a LLM system prompt from a picture so you can use it in <a h
 css = """
 #col-container{
     margin: 0 auto;
-    max-width: 740px;
+    max-width: 780px;
     text-align: left;
 }
 """
@@ -89,7 +89,8 @@ with gr.Blocks(css=css) as demo:
                 submit_btn = gr.Button("Make LLM system from my pic !")
             with gr.Column():
                 result = gr.Textbox(
-                    label ="Suggested System"
+                    label ="Suggested System",
+                    max_lines = 30
                 )
 
     submit_btn.click(