Update README.md

README.md

@@ -11,12 +11,13 @@ inference: false
 
 *(Not to be confused with [Pygmalion 13B](https://huggingface.co/TehVenom/Pygmalion-13b-GGML).)*
 
-# Converted with ggerganov/ggml's gpt-neox conversion script, and tested with KoboldCpp.
-## *(I can't promise that this will work with other frontends, if at all; I haven't had the most success myself. Use at your own risk!)*
-
 This is converted and quantized from [Pygmalion 1.3B](https://huggingface.co/PygmalionAI/pygmalion-1.3b), based on [an earlier version](https://huggingface.co/EleutherAI/pythia-1.4b-deduped-v0) of Pythia 1.4B Deduped.
 
-# RAM USAGE (on KoboldCpp w/ OpenBLAS)
+Notes:
+- Converted with ggerganov/ggml's gpt-neox conversion script, and tested with KoboldCpp.
+- I can't promise that this will work with other frontends, if at all. I've had problems with the tokenizer. Could be related to the ggml implementation of GPT-NeoX [(source)](https://github.com/ggerganov/ggml/tree/master/examples/gpt-neox#notes).
+
+### RAM USAGE (on KoboldCpp w/ OpenBLAS)
 Model | Initial RAM
 :--:|:--:
 ggml-pygmalion-1.3b-q4_0.bin | 1.1 GiB