Update README.md

README.md

@@ -74,7 +74,7 @@ llm = Llama(
     model_path="/content/models/Atlas-72B-SVT-Q4_K_M.gguf",
     n_gpu_layers=-1,
     n_ctx=4096,
-    chat_format="chatml",  # adjust if Atlas uses a different format
+    chat_format="chatml", # Atlas uses the chatML format
 )
 
 conversation = [{"role": "system", "content": SYSTEM}]