Fix problem with attention mask

app.py

@@ -32,13 +32,16 @@ streamer = TextIteratorStreamer(tokenizer, decode_kwargs=decode_kwargs)
 cplusplus = None
 def translate(python, progress=gr.Progress()):
     progress(0, desc="Starting")
-    inputs = tokenizer.apply_chat_template(
+    formatted_prompt = tokenizer.apply_chat_template(
                         messages_for(python),
+                        tokenize=False,
+                        add_generation_prompt=True,
                         return_tensors="pt").to(model.device)
+    inputs = tokenizer(formatted_prompt, return_tensors="pt", padding=True)
     attention_mask = inputs["attention_mask"]
 
     outputs = model.generate(
-        inputs,
+        inputs['input_ids'],
         attention_mask=attention_mask,
         max_new_tokens=1024,
         do_sample=False,