Hobbyist Attempts DIY RLHF on Llama2-13B to Make It Post by Itself

A netizen shared his DIY project of hand-performing RLHF on a Llama2-13B checkpoint to teach it to post on its own, using a classic pairwise preference labeling interface. He called the process "really hard" though the base model is his favorite.

2026-11-10 ~ 2026-11-10 · 2 related posts

• Manually RLHF-ing a Llama2-13B checkpoint to post is proving hilariously hard — cephaloform · 2026-11-10
• Hobbyist attempts DIY RLHF on a llama2-13b base model checkpoint: "this is so hard" — cephaloform · 2026-11-10