Update handler.py

Trying stack overflow fix

handler.py

@@ -12,6 +12,7 @@ class EndpointHandler:
         # load the model
         tokenizer = AutoTokenizer.from_pretrained(path)        
         model = AutoModelForCausalLM.from_pretrained(path, device_map="auto", torch_dtype=dtype, trust_remote_code=True)
+        model.to('cuda')
     
         # create inference pipeline
         self.pipeline = pipeline("text-generation", model=model, tokenizer=tokenizer, device='cuda:0')