updated the app.py because of gpu error

app.py

@@ -14,8 +14,8 @@ device = "cpu"
 
 # Load model and tokenizer
 
-# 1. Load the base model (unsloth/meta-llama-3.1-8b-bnb-4bit)
-base_model_name = "unsloth/meta-llama-3.1-8b-bnb-4bit"
+# 1. Load the base model (unsloth/meta-llama-3.1-8b)
+base_model_name = "unsloth/Meta-Llama-3.1-8B"
 tokenizer = AutoTokenizer.from_pretrained(base_model_name)
 
 # Disable GPU quantization or 4-bit loading explicitly to avoid bitsandbytes