Update app.py

app.py

@@ -13,7 +13,7 @@ tokenizer = AutoTokenizer.from_pretrained(base_model_id)
 
 model = AutoModelForCausalLM.from_pretrained(
     base_model_id,
-    load_in_8bit=True,
+    load_in_4bit=True,
     torch_dtype=torch.float16,
     device_map="auto"
 )
@@ -56,7 +56,7 @@ def explain_dockerfile(instruction, code):
         max_new_tokens=256, 
         do_sample=True, 
         temperature=0.7,
-        return_full_text=False # We want only the new tokens generated after the prompt
+        return_full_text=False 
     )