Update app.py

app.py

@@ -6,14 +6,14 @@ from transformers import AutoTokenizer, AutoModelForCausalLM
 # === STEP 1: Authenticate with Hugging Face ===
 # Make sure you set your HF token as an environment variable or paste it here temporarily
 # For security, prefer environment variable (recommended)
-hf_token = "TLpIICgZJrDCTgVTsaaydFFWbWyGKiGAPa"
-login(token="hf_" + hf_token)
+#hf_token = "TLpIICgZJrDCTgVTsaaydFFWbWyGKiGAPa"
+#login(token="hf_" + hf_token)
 
 # === STEP 2: Load base and adapter models ===
 base_model = "meta-llama/Llama-2-7b-chat-hf"
 adapter_model = "zimble-llama2-finetunedhybride"
 
-tokenizer = AutoTokenizer.from_pretrained(adapter_model, use_auth_token=hf_token)
+tokenizer = AutoTokenizer.from_pretrained(adapter_model)
 
 # Enable memory-efficient loading if needed
 device = "cuda" if torch.cuda.is_available() else "cpu"