wassname/CoT_rating
1
0
Fork 0
mirror of https://github.com/wassname/CoT_rating.git synced 2026-06-27 00:30:04 +08:00
Code Issues Packages Projects Releases Wiki Activity
Files
10444385aaf483d06fb483f615792b1f202dd0eb
CoT_rating/README.md
T
wassname 10444385aa readme
2025-08-20 14:21:57 +08:00

283 B
Raw Blame History

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

alt text

Reference in New Issue View Git Blame Copy Permalink