Update app.py

app.py

@@ -22,9 +22,9 @@ def generate(
     top_k: int = 50,
     repetition_penalty: float = 1.2,
     ):
-    input_ids = tokenizer(prompt, return_tensors="pt").to("cuda")
+    input_ids = tokenizer(message, return_tensors="pt").to("cuda")
     outputs = model.generate(**input_ids,top_p=top_p,max_new_tokens=max_new_tokens,top_k=top_k,repetition_penalty=repetition_penalty,temperature=temperature)
-    return tokenizer.decode(outputs[0]);
+    return tokenizer.decode(outputs[0], skip_special_tokens=True);
     # return pipe(prompt)[0]['generated_text']
 
 gr.Interface(