RL4LLM: 2. PPO Algorithm and Implementation Details
This blog post introduces RLHF-PPO algorithm with code implementation.
Blogs
Zotero
This blog note how to use Zotero with iCloud as the storage.
RL4LLM: 1. A Brief Talk on DPO
This blog post notes my understanding of the algorithm DPO (Direct Preference Optimization) and the math derivation behind it.