Lewis Tunstall
Works on post-training at Hugging Face; first author of Zephyr and a listed author of TRL
- Machine learning engineer working on post-training, Hugging Face12
Lewis Tunstall works on post-training at Hugging Face, according to Tunstall's GitHub profile, and Hugging Face's LLM Course describes Tunstall as a machine learning engineer focused on developing open-source tools. Tunstall is first author of the Zephyr paper (2023), which aligned a smaller chat model with distilled direct preference optimization on AI-feedback data, and is listed first in the citation for the Alignment Handbook, a set of training recipes for aligning language models. Tunstall is also listed among the authors of TRL, Hugging Face's Apache-2.0 post-training library, and co-wrote the O'Reilly book Natural Language Processing with Transformers.12345
Work
TRL (Transformers Reinforcement Learning)Open-source code
Listed among the authors in TRL's citation file. TRL is an Apache-2.0 toolkit for post-training transformer models, with implementations of SFT, reward modeling, DPO, and GRPO.5
Zephyr (external site: arxiv.org)Published research
First author of "Zephyr: Direct Distillation of LM Alignment" (2023), which trained a smaller chat model with distilled supervised fine-tuning and distilled direct preference optimization on preference data ranked by a teacher model.3
The Alignment Handbook (external site: github.com)Open-source code
Listed first in the citation entry of the repository, which provides training recipes for continued pretraining and for aligning language models with human and AI preferences.4
Natural Language Processing with Transformers (O'Reilly)Published research
Co-author of the O'Reilly book, according to the author biographies on Hugging Face's LLM Course page.2
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.