Update README.md

README.md

@@ -48,7 +48,7 @@ The "thinking/reasoning" tech (for the model at this repo) is from the original
 [ https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Llama-8B ] 
 
 This is Version 2, with some "tuning" to make the model more uncensored, plus I put back in some layers from the original Grand Horror
-model, which returns the model to 16.5 B parameters too. I found this version works better with Deepseek's "thinking/reasoning" system.
+model, which returns the model to 16.5 B parameters too. I found this version works better with Deepseek's "thinking/reasoning" system and seems to line up better with prompts too.
 
 You can download/access V1 here:
 
@@ -487,4 +487,5 @@ So we sat there, in a bar full of ghosts, tryin' to forget the things that weren
 
 Because sometimes, when the only thing still alive inside you is just the bottle, and the only friend you got is the bottom of it...
 
-Well, then I guess that's just how it is.
+Well, then I guess that's just how it is.
+