Update README.md
Browse files
README.md
CHANGED
@@ -34,7 +34,7 @@ How to use:
|
|
34 |
* use https://github.com/BlinkDL/ChatRWKV/blob/main/v2/benchmark_world.py to test it
|
35 |
* larger models are stronger even though not fully trained yet
|
36 |
|
37 |
-
The
|
38 |
* 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+)
|
39 |
* use Question/Answer or User/AI or Human/Bot for chat. **DO NOT USE Bob/Alice or Q/A**
|
40 |
* use **fp32** (will overflow in fp16 at this moment - fixable in future) or bf16 (slight degradation)
|
|
|
34 |
* use https://github.com/BlinkDL/ChatRWKV/blob/main/v2/benchmark_world.py to test it
|
35 |
* larger models are stronger even though not fully trained yet
|
36 |
|
37 |
+
The differences between World & Raven:
|
38 |
* 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+)
|
39 |
* use Question/Answer or User/AI or Human/Bot for chat. **DO NOT USE Bob/Alice or Q/A**
|
40 |
* use **fp32** (will overflow in fp16 at this moment - fixable in future) or bf16 (slight degradation)
|