Ba2han commited on
Commit
3ebed8b
1 Parent(s): 6464237

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -1,3 +1,5 @@
1
  Should be working properly with KoboldCPP and other stuff. I changed EOS token to <|eot_id|> and this prevents endless generation.
2
 
3
- I also converted original weights back to fp32 before quantization.
 
 
 
1
  Should be working properly with KoboldCPP and other stuff. I changed EOS token to <|eot_id|> and this prevents endless generation.
2
 
3
+ I also converted original weights back to fp32 before quantization.
4
+
5
+ **Edit: Highly recommending you to download more recent quants.** 17/05/2024