Text Generation
PyTorch
causal-lm
rwkv
BlinkDL commited on
Commit
0616e70
1 Parent(s): dbabc4a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -38,6 +38,8 @@ The difference between World & Raven:
38
  * use Question/Answer or User/AI or Human/Bot prompt. **DO NOT USE Bob/Alice or Q/A**
39
  * use **fp32** (will overflow in fp16 at this moment - fixable in future)
40
 
 
 
41
  A good prompt example:
42
  ```
43
  Question: hi
 
38
  * use Question/Answer or User/AI or Human/Bot prompt. **DO NOT USE Bob/Alice or Q/A**
39
  * use **fp32** (will overflow in fp16 at this moment - fixable in future)
40
 
41
+ NOTE: the new greedy tokenizer will tokenize '\n\n' as one single token instead of ['\n','\n']
42
+
43
  A good prompt example:
44
  ```
45
  Question: hi