New year, new YouTube videos! We are resuming our regular interpretability seminar posts, with a fantastic talk by Deepti Ghadiyaram on interpreting diffusion models.

Watch the video: youtu.be/4eqvABPX5rA

nnterp by @butanium.bsky.social is now part of the NDIF ecosystem! nnterp standardizes transformer naming conventions, includes built-in best practices for common interventions, and is perfectly compatible with original HF model implementations.

Learn more: ndif-team.github.io/nnterp/

Happy Holidays from NDIF! Our new NNsight version improves performance and enhances vLLM integration, including support for tensor parallelism.

Ever wished you could explore what's happening inside a 405B parameter model without writing any code? Workbench, our AI interpretability interface, is now live for public beta at workbench.ndif.us!

Great research opportunity coming up!

Apply here by October 1st: forms.gle/BVHX8PVxcVvT...

Do you wish you could run experiments on any model remotely from your laptop? In a future release, NDIF users will be able to dynamically deploy any model from HuggingFace on NDIF for remote experimentation. But before this, we need your help!

Our YouTube channel is live! Our first video features @davidbau.bsky.social‬ presenting the ROME project:
www.youtube.com/watch?v=eKd...

We’re excited to announce a new series of applied "mini paper" tutorials! The goal of this series is to help researchers get hands-on experience with findings, methods, and results from recent papers in interpretability using NNsight and NDIF.

We are prereleasing NNsight 0.5 today! We've added many highly-requested new features, including easier access to intermediate values within the forward pass, ability to run any code within NNsight’s tracing context (no more proxies!), and easier debugging.

At #ICLR2025 in Singapore? Come visit our poster tomorrow (Saturday, April 26th)! This work addresses a critical gap in ML research: the ability to study the internals of very large neural networks.

Please help amplify ARBOR, a fantastic new research opportunity! If you’d like to start contributing, NDIF is now hosting DeepSeek R1 8B and 70B, open for all researchers to experiment on via our API.

Sign up for API access here: login.ndif.us

New version alert! We've officially released NNsight 0.4, which you can now install with pip.

If you haven’t yet tried out NNsight 0.4, here’s what’s new:

Thanks for the mention, @thebulletin.org! Great article on the importance of AI interpretability, highlighting the need for frameworks that make AI more transparent and help democratize research capabilities: thebulletin.org/2025/01/why-...

🔔 We have new features coming to #NNsight soon! NNsight 0.4 will be released on January 31st. Stay tuned and join our Discord for the latest updates: lnkd.in/eJkRGvGB.

👀 For a sneak peek at the new features, check out our walkthrough: colab.research.google.com/drive/1TtYi5...

Calling all students looking for summer opportunities! Apply to our Summer Engineering Fellowship today: ndif.us/fellowship.h...

Spend the summer with us! We're still accepting applications for our Summer Engineering Fellowship. Seeking candidates to help with full-stack tool development, systems reliability, and outreach and education! Apply by Feb 7, 2025 at ndif.us/fellowship.h...

Large language models show fascinating changes in capability with scaling parameters, but scaling also vastly increases the resources required for experimentation on model internals. NDIF is currently hosting the largest open sourced model, Llama 405b, for YOU to run research on!

Do you have a great experiment that you want to run on Llama 405b but not enough GPUs?

🚨 #NDIF is opening up more spots in our 405b pilot program! Apply now for a chance to conduct your own groundbreaking experiments on the 405b model. Details: 🧵⬇️