Update app.py

app.py

@@ -74,7 +74,7 @@ def model_prediction(model, text, device):
 
 # --- Llama 3.2 3B Model Setup ---
 LLAMA_MAX_MAX_NEW_TOKENS = 512
-LLAMA_DEFAULT_MAX_NEW_TOKENS = 512
+LLAMA_DEFAULT_MAX_NEW_TOKENS = 250
 LLAMA_MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "1024"))
 llama_device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 llama_model_id = "meta-llama/Llama-3.2-1B-Instruct"
@@ -135,7 +135,7 @@ Given the following issue description:
 ---
 {issue_text}
 ---
-Providing short explanation why this issue might be classified as a **{quality_name}** issue.
+Explain why this issue might be classified as a **{quality_name}** issue. Provide a concise explanation, relating it back to the issue description. Keep the explanation short and concise. Do not repeat the prompt or include any preamble in your response - just provide the explanation directly.
 """
     try:
         explanation = llama_generate(prompt)
@@ -279,4 +279,4 @@ interface = gr.Interface(
     examples=example_texts,
     css=css  # Apply the CSS
 )
-interface.launch(share=True)
+interface.launch( )