ValueFX9507
/

Tifa-Deepsex-14b-CoT

Reinforcement Learning

incremental-pretraining

Inference Endpoints

Not-For-All-Audiences

Model card Files Files and versions Community

ValueFX9507 commited on 14 days ago

Commit

50d38b2

·

verified ·

1 Parent(s): 6a7618b

Update README.md

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -93,8 +93,8 @@ prompt = """<system>进入Tifa角色引擎...</system>
 **参数推荐**：
 ```python
 generation_config = {
-    "temperature": 0.8,
-    "top_p": 0.8,
     "repetition_penalty": 1.17,
     "max_new_tokens": 1536,
     "do_sample": True

 **参数推荐**：
 ```python
 generation_config = {
+    "temperature": 0.4,
+    "top_p": 0.6,
     "repetition_penalty": 1.17,
     "max_new_tokens": 1536,
     "do_sample": True