Update app.py

app.py

@@ -1,6 +1,6 @@
 import streamlit as st
 # Load model directly
-from transformers import AutoModel
+from transformers import AutoModel, AutoModelForCausalLM
 from huggingface_hub import login
 import os
 
@@ -9,11 +9,14 @@ file = 'llama-2-7b.Q5_0.gguf'
 
 NO_GPU = 0
 GPU_LAYERS = 50
-model = AutoModel.from_pretrained("valencar/llamm", 
-                                  model_file=file, model_type="llama", gpu_layers=NO_GPU)
 
-access_token = os.getenv('HF_TOKEN2')
-login(token = access_token)
+llm = AutoModelForCausalLM.from_pretrained(file, model_type="llama", gpu_layers=NO_GPU)
+
+# model = AutoModelForCausalLM.from_pretrained("valencar/llamm", 
+#                                   model_file=file, model_type="llama", gpu_layers=NO_GPU)
+
+# access_token = os.getenv('HF_TOKEN2')
+# login(token = access_token)
 
 prompt = "AI is going to"