Update app.py

app.py

@@ -21,9 +21,18 @@ def chatbot(message, history):
     prompt += f"User: {message}\nBot: "
     
     # Generate the response using the text-generation pipeline.
-    # We use max_new_tokens to limit the length of the response for efficiency.
-    # The return_full_text=False option ensures we only get the new text generated by the model.
-    response = pipe(prompt, max_new_tokens=50, return_full_text=False)
+    # We've increased max_new_tokens for longer responses and added other parameters
+    # to improve generation quality.
+    response = pipe(
+        prompt,
+        max_new_tokens=150,
+        return_full_text=False,
+        do_sample=True,
+        temperature=0.7,
+        top_k=50,
+        top_p=0.95,
+        repetition_penalty=1.2
+    )
     
     # Extract the generated text and clean up any extra whitespace or newlines.
     generated_text = response[0]['generated_text'].strip()