fix cuda issiue on new llama added code

summarizer.py

@@ -60,7 +60,7 @@ def summarizer_summarize(model_type,tokenizer, base_summarizer, text:str,summari
             ]
             prompt = tokenizer.apply_chat_template(chat, tokenize=False, add_generation_prompt=True)
             inputs = tokenizer(prompt, 
-                      return_tensors="pt", truncation=True).to('cuda')
+                      return_tensors="pt", truncation=True).to('cpu')
             attention_mask = inputs["attention_mask"]
             approximate_tokens = int(len(text)//10)
             output = base_summarizer.generate(inputs['input_ids'],