hmm... putting back the cuda

app.py

@@ -32,8 +32,8 @@ As a derivate work of [Llama-2-7b-chat](https://huggingface.co/meta-llama/Llama-
 this demo is governed by the original [license](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/LICENSE.txt) and [acceptable use policy](https://huggingface.co/spaces/huggingface-projects/llama-2-13b-chat/blob/main/USE_POLICY.md).
 """
 
-if  torch.cuda.is_available():
-    DESCRIPTION += '\n<p>Running on GPU 🥶 Do not use GPU $$.</p>'
+if not torch.cuda.is_available():
+    DESCRIPTION += '\n<p>Running on GPU 🥶 Do not use CPU.</p>'
 
 
 def clear_and_save_textbox(message: str) -> tuple[str, str]:

model.py

@@ -9,7 +9,7 @@ from huggingface_hub import login
 # model_id = 'meta-llama/Llama-2-13b-chat-hf'
 model_id = 'meta-llama/Llama-2-7b-chat-hf'
 
-if not torch.cuda.is_available():
+if torch.cuda.is_available():
     tok = os.environ['HF_TOKEN']
     login(new_session=True,
           write_permission=False,
@@ -62,7 +62,7 @@ def run(message: str,
         top_p: float = 0.95,
         top_k: int = 50) -> Iterator[str]:
     prompt = get_prompt(message, chat_history, system_prompt)
-    inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False) #.to(torch.device) #.to('cuda')
+    inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda') #.to(torch.device) #
 
     streamer = TextIteratorStreamer(tokenizer,
                                     timeout=10.,