Running 1 CorrSteer: Correlation-Based Steering of Language Models via Sparse Autoencoders π§ 1 Steer language model output by clicking visual layers
Running Featured 49 Porting nanochat to Transformers: an AI modeling history lesson π 49 Learn about ML and Transformers through nanochat
Running 11 FAT5 (Flash Attention T5) report β‘ 11 English version of the blog post introducing FAT5 model
Running 79 Unfolding Robotics: Open-Source Shirt Folding from Data to Deployment π€ 79 Explore the open-source guide to robot shirt folding
Running on CPU Upgrade 229 The Synthetic Data Playbook: Generating Trillions of the Finest Tokens π 229 Explore synthetic data experiments on a virtual bookshelf
Running 6 Robotics research should think (and do) more about sustainability! π 6 Explore robotics papers mapped to UN Sustainable Development Goals
Running Featured 24 Chasing the Counting Manifold in Open LLMs π 24 Counting manifolds in open LLMs from behavior to SAEs.
Running Featured 74 QED-Nano: Teaching a Tiny Model to Prove Hard Theorems π 74 Who needs 1T parameters? Olympiad proofs with a 4B model
Running Featured 88 Parakeet STT Progressive Transcription π€ 88 Transcribe speech to text instantly with WebGPU acceleration