Hi, I'm Teïlo Millet 👋

I'm Currently Working On

Model Integration: Fine-tuning and merging diverse open-source models, spanning from role-playing applications to new language implementation.
- Merlin: Series of POC, aimed to create efficient, low-memory personal assistants. Merlin (5th best performing LLM in his category (3B) worldwide (as of dec. 23)
- Potato: Series, inspired by the adaptibility of Mr. Potato, focuses on the merging of model's task vectors to create versatile and capable AI systems.Potato
Dataset Development: Crafting specialized datasets for LLM training, tailored to specific tasks and goals.
Tool Creation: Building innovative tools to streamline the development and deployment of specialized datasets in AI.

I'm Currently Learning

LLM Architecture: Diving deep into the architecture of Large Language Models to understand and optimize their functionality.
Specialization Techniques: Researching methods to improve the specialization of LLMs, aiming for higher efficiency and accuracy.
Programming Languages: Expanding my coding arsenal with Rust and Mojo.

How to Reach Me

Twitter: teilomillet
Email: [email protected]
Website: teilo.xyz
Discord: teilom

Feel free to reach out for collaborations or just a chat about technology and AI. I'm always open to sharing ideas and learning from the community.

Teïlo M's Projects

ahdoc

Agent permettant de poser des questions et d'avoir des réponses sur des documents.

deep-rl-class

This repo contains the syllabus of the Hugging Face Deep Reinforcement Learning Course.

Lugia crée une infrastructure pour basculer entre modèles IA n'importe où. Offre une connexion sécurisée par code unique, sauvegarde les échanges, et utilise une base vectorielle pour optimiser les contenus.