Update README.md

README.md

@@ -17,7 +17,7 @@ tags:
 
 These files are GPTQ model files for [Meta's Llama 2 70B](https://huggingface.co/meta-llama/Llama-2-70b-hf/tree/main) but with new FP16 files, made with the last transformers version. (transformers-4.32.0.dev0)
 
-This is a direct quant from the model, and since there was no remote_code in the files, it is unkown if GQA works or not.
+GQA Works with exllama, but not GPTQ for LLaMA/AutoGPTQ.
 
 ## Quant parameters