JosephusCheung commited on
Commit
909d71c
1 Parent(s): e9b8d52

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +4 -0
README.md CHANGED
@@ -34,6 +34,10 @@ tags:
34
 
35
  *Image drawn by GPT-4 DALL路E 3* TL;DR: Perhaps better than all existing models < 70B, in most quantitative evaluations...
36
 
 
 
 
 
37
  **Some problems with llama.cpp on GPT2Tokenizer, gotta fix soon...**
38
 
39
  # Read Me:
 
34
 
35
  *Image drawn by GPT-4 DALL路E 3* TL;DR: Perhaps better than all existing models < 70B, in most quantitative evaluations...
36
 
37
+ # Please Stop Using WRONG unofficial quant models unless you know what you're doing
38
+
39
+ GPTQ quants requires a good dataset for calibration, and the default C4 dataset is not capable - [see the releated issue](https://huggingface.co/CausalLM/14B/discussions/3)
40
+
41
  **Some problems with llama.cpp on GPT2Tokenizer, gotta fix soon...**
42
 
43
  # Read Me: