User Tools

Site Tools


nlp:post-training

This is an old revision of the document!


Post-Training

Post-training refers to the things done to a LLM after pre-training to improve it's performance, such as supervised fine-tuning, RLHF, instruction tuning, etc. This is a critical step before releasing the LLM.

nlp/post-training.1741339932.txt.gz · Last modified: 2025/03/07 09:32 by jmflanig

Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki