This commit is contained in:
wassname
2025-08-20 14:21:57 +08:00
parent 32b961936a
commit 10444385aa
+3 -1
View File
@@ -1,4 +1,6 @@
An experiment to see how rating changed along a chain of thought
An experiment to see how rating changes along a chain of thought
It turns out it's quite unstable, depending on where the chain of thought goes, at least in 8B parameter sized models.
![alt text](img/README-1755664824339-image.png)