Writing
-
🔗 J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
-
🔗 The Strange Science of Interpretability: Recent Papers and a Reading List for the Philosophy of Interpretability
-
Shazeer Typing
-
🔗 SAEBench: A Comprehensive Benchmark for Sparse Autoencoders
-
🔗 Standard SAEs Might Be Incoherent: A Choosing Problem & A “Concise” Solution
-
🔗 MDL-SAEs: Interpretability as Compression
-
Mamba Explained
-
The Impact of Mixtral
-
Descriptive Matrix Operations with Einops
-
Dictionary Learning with Sparse AutoEncoders
-
An Analogy for Understanding Mixture of Expert Models
-
From Sparse To Soft Mixtures of Experts
-
DeepSpeed's Bag of Tricks for Speed & Scale