OpenNLPLab
commited on
Commit
•
cf7a5f5
1
Parent(s):
9cb6819
Update README.md
Browse files
README.md
CHANGED
@@ -109,6 +109,7 @@ export use_triton=False
|
|
109 |
>>> pred = model.generate(**inputs, max_new_tokens=2048, repetition_penalty=1.0)
|
110 |
>>> print(tokenizer.decode(pred.cpu()[0], skip_special_tokens=True))
|
111 |
```
|
|
|
112 |
|
113 |
|
114 |
# Fine-tuning the Model
|
|
|
109 |
>>> pred = model.generate(**inputs, max_new_tokens=2048, repetition_penalty=1.0)
|
110 |
>>> print(tokenizer.decode(pred.cpu()[0], skip_special_tokens=True))
|
111 |
```
|
112 |
+
* **Note**: we recommend to use `bfloat16` in `TransNormerLLM`, `float16` might lead `nan` error, please check your divce compatibility!
|
113 |
|
114 |
|
115 |
# Fine-tuning the Model
|