Update README.md

Corrected description of size to fp32, and added a link to a smaller fp16 version.

README.md

@@ -18,7 +18,7 @@ Uncensored refers to the [ehartford/wizard_vicuna_70k_unfiltered](https://huggin
 
 Special thanks to [George Sung](https://huggingface.co/georgesung) for creating [llama2_7b_chat_uncensored](https://huggingface.co/georgesung/llama2_7b_chat_uncensored), and to [Eric Hartford](https://huggingface.co/ehartford/) for creating [ehartford/wizard_vicuna_70k_unfiltered](https://huggingface.co/datasets/ehartford/wizard_vicuna_70k_unfiltered)
 
-The version here is the fp16 HuggingFace model.
+The version here is the fp32 HuggingFace model.
 
 In 8 bit mode, the model fits into 84% of A100 80GB (67.2GB) 68747MiB  
 In 4 bit mode, the model fits into 51% of A100 80GB (40.8GB) 41559MiB  
@@ -29,6 +29,9 @@ Thanks to [TheBloke](https://huggingface.co/TheBloke), he has created the GGML a
 * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GGML
 * https://huggingface.co/TheBloke/llama2_70b_chat_uncensored-GPTQ
 
+A smaller fp16 copy can be found at:
+* https://huggingface.co/YokaiKoibito/llama2_70b_chat_uncensored-fp16
+
 # Prompt style
 The model was trained with the following prompt style:
 ```