Hi there!
I am a Software Engineer at Meta Superintelligence Labs, working on developing Large Language and Vision Models. I have more than twelve years of industry and research experience in machine learning and related areas. I occasionally blog about new things I learn and hope to write more in the near future.
Recent Work
I have contributed to the following releases / open-source models / projects:
- Multimodal reasoning in Meta AI
- Llama Guard Vision, Llama Guard 4 and Llama 3.2
- Core Contributor to Llama 3 research
- Meta AI assistant powered by Llama 3
- Llama Guard 2
- Learnings from a deep dive into recent LLM research
- Key takeaways from ICML 2023
Publications
- Diversity-driven Data Selection for Language Model Tuning through Sparse Autoencoder (link)
- The Llama 3 Herd of Models (link)
- An Intuitive Explanation of Convolutional Neural Networks (link, 330+ citations)
- A Quick Introduction to Neural Networks (link, 30+ citations)
- Llama Guard 2 (link, 70+ citations)
- Hand Gesture Recognition using Self Organizing Map (link)
- Machine Learning Tutorials (link, 17k+ stars on GitHub)
- xda: R package for exploratory data analysis (link, 100+ stars on GitHub)
More information about me
- Google Scholar profile
- GitHub profile
- ‘Competitions Expert’ on Kaggle
- Connect with me on LinkedIn
- Meta Homepage
Ujjwal Karn's blog 