Update app.py

app.py

@@ -5,8 +5,9 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
 # Initialize model and tokenizer
 model = AutoModelForCausalLM.from_pretrained(
     "microsoft/Phi-3.5-mini-instruct", 
+    device_map="cpu",
+    torch_dtype=torch.float16,
     low_cpu_mem_usage=True,
-    torch_dtype="auto", 
     trust_remote_code=True,
 )
 tokenizer = AutoTokenizer.from_pretrained("microsoft/Phi-3.5-mini-instruct")
@@ -16,11 +17,12 @@ pipe = pipeline(
     "text-generation",
     model=model,
     tokenizer=tokenizer,
+    device="cpu"
 )
 
 # Generation arguments
 generation_args = {
-    "max_new_tokens": 1024,
+    "max_new_tokens": 500,
     "return_full_text": False,
     "temperature": 0.0,
     "do_sample": False,