Update README.md

README.md

@@ -23,6 +23,10 @@ base_model: deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct
 **Original model**: [DeepSeek-Coder-V2-Lite-Instruct](https://huggingface.co/deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct)<br>
 **GGUF quantization:** provided by [bartowski](https://huggingface.co/bartowski) based on `llama.cpp` release [b3166](https://github.com/ggerganov/llama.cpp/releases/tag/b3166)<br>
 
+## Model Settings:
+
+Flash attention MUST be **disabled** for this model to work.
+
 ## Model Summary:
 
 This is a brand new Mixture of Export (MoE) model from DeepSeek, specializing in coding instructions.<br>
@@ -42,6 +46,7 @@ This will format the prompt as follows:
 User: {user_message}
 
 Assistant: {assistant_message}
+```
 
 ## Technical Details