BlinkDL
/

rwkv-4-world

Text Generation

Model card Files Files and versions Community

BlinkDL commited on Jun 10, 2023

Commit

a6b6a11

•

1 Parent(s): 1067d68

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -36,12 +36,12 @@ How to use:
 The difference between World & Raven:
 * set pipeline = PIPELINE(model, "rwkv_vocab_v20230424") instead of 20B_tokenizer.json (EXACTLY AS WRITTEN HERE. "rwkv_vocab_v20230424" is included in rwkv 0.7.4+)
-* use Question/Answer or User/AI or Human/Bot prompt for Q&A. **DO NOT USE Bob/Alice or Q/A**
 * use **fp32** (will overflow in fp16 at this moment - fixable in future) or bf16 (slight degradation)
 NOTE: the new greedy tokenizer (https://github.com/BlinkDL/ChatRWKV/blob/main/tokenizer/rwkv_tokenizer.py) will tokenize '\n\n' as one single token instead of ['\n','\n']
-prompt (replace \n\n in xxx to \n):
 ```
 Instruction: xxx
 Input: xxx

 The difference between World & Raven:
 * set pipeline = PIPELINE(model, "rwkv_vocab_v20230424") instead of 20B_tokenizer.json (EXACTLY AS WRITTEN HERE. "rwkv_vocab_v20230424" is included in rwkv 0.7.4+)
+* use Question/Answer or User/AI or Human/Bot for chat. **DO NOT USE Bob/Alice or Q/A**
 * use **fp32** (will overflow in fp16 at this moment - fixable in future) or bf16 (slight degradation)
 NOTE: the new greedy tokenizer (https://github.com/BlinkDL/ChatRWKV/blob/main/tokenizer/rwkv_tokenizer.py) will tokenize '\n\n' as one single token instead of ['\n','\n']
+QA prompt (replace \n\n in xxx to \n):
 ```
 Instruction: xxx
 Input: xxx