AI engineering glossary · LLM training alignment
What is TRL (Hugging Face TRL)?
Hugging Face TRL (also called TRL): An open-source library, short for Transformer Reinforcement Learning, providing ready-made trainers for supervised fine-tuning, reward modelling, PPO, DPO and related alignment methods, built on top of the Hugging Face Transformers library.
Hugging Face TRL: an example
A practitioner runs an SFT trainer and then a DPO trainer from the same library on a chat dataset.
Related terms
More LLM training alignment terms
- Identity preference optimisation
- InstructGPT
- Instruction backtranslation
- Instruction tuning
- Kahneman-Tversky optimisation
- KL penalty
- Knowledge distillation
- LIMA
Where you learn it
Hugging Face TRL is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.