Update app.py

app.py

@@ -3,7 +3,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, AutoConfig
 modelname="gpt2"
 config = AutoConfig.from_pretrained(modelname)
 tokenizer = AutoTokenizer.from_pretrained(modelname)
-model = AutoModelForCausalLM.from_pretrained(modelname,config=config).to("cuda")
+model = AutoModelForCausalLM.from_pretrained(modelname,config=config)
 
 
 def botsay(user_input):
@@ -29,7 +29,7 @@ def botsay(user_input):
     if length >limit:
       gen_tokens="⚠️sorry length limit. please reload the browser."
       return gen_tokens
-    outs=model(input_ids=input_ids.to("cuda"))
+    outs=model(input_ids=input_ids)
     topk = torch.topk(outs.logits.squeeze()[-1,:],k=j+1).indices
     if new_token =="that":
       that_id = 326