Update app.py

app.py

@@ -10,10 +10,10 @@ if is_available():
     )
 else:
     options = {
-        "low_cpu_mem_usage": True,
-        "offload_state_dict": True,
-        "offload_folder": "offload",
-        "device_map": "auto",
+        #"low_cpu_mem_usage": True,
+        #"offload_state_dict": True,
+        #"offload_folder": "offload",
+        #"device_map": "auto",
     }
 
 tokenizer = LlamaTokenizer.from_pretrained("openlm-research/open_llama_3b_v2")
@@ -21,7 +21,7 @@ model = LlamaForCausalLM.from_pretrained(
     "openlm-research/open_llama_3b_v2",
     **options
 )
-model = PeftModel.from_pretrained(model, "robinhad/open_llama_3b_uk", offload_state_dict=True, offload_folder=".")
+model = PeftModel.from_pretrained(model, "robinhad/open_llama_3b_uk", )#offload_state_dict=True, offload_folder=".")
 
 
 def generate_prompt(instruction, input=None, output=""):