Hugging Face's logo

ValueFX9507
/
Tifa-DeepsexV2-7b-MGRPO-GGUF-Q4

Reinforcement Learning
Transformers
GGUF
Chinese
English
incremental-pretraining
sft
roleplay
cot
Inference Endpoints
conversational
Model card Files Community
22
ValueFX9507 commited on
Commit
e4846d3
·
verified ·
1 Parent(s): d032ed0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -0
README.md CHANGED
@@ -56,6 +56,9 @@ license: apache-2.0
56
 
57
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/650762d0eac45ee2e420a38b/xeCTmjGqtqxCMlC8lOAG4.png)
58
 
 
 
 
59
 
60
  ## 暂时没有遇到“啊哈”时刻
61
 
 
56
 
57
  ![image/png](https://cdn-uploads.huggingface.co/production/uploads/650762d0eac45ee2e420a38b/xeCTmjGqtqxCMlC8lOAG4.png)
58
 
59
+ ## 自我学习中的有趣时刻-强化学习中模型更加倾向于戏剧性输出-类似Carzy状态-算法需要优化
60
+
61
+ ![image/png](https://cdn-uploads.huggingface.co/production/uploads/650762d0eac45ee2e420a38b/pQDWfIttC7nGED7D-4_h-.png)
62
 
63
  ## 暂时没有遇到“啊哈”时刻
64