Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders
Paper • 2606.12138 • Published • 8
Nikita Balagansky
elephantmipt
AI & ML interests
None yet
Recent Activity
authored a paper 7 days ago
Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders submitted a paper 7 days ago
Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse AutoencodersOrganizations
authored a
paper 7 days ago
upvoted a paper 7 days ago
submitted a
paper to Daily Papers 7 days ago
authored 4
papers 12 days ago
Small Vectors, Big Effects: A Mechanistic Study of RL-Induced Reasoning via Steering Vectors
Paper • 2509.06608 • Published
Steering LLM Reasoning Through Bias-Only Adaptation
Paper • 2505.18706 • Published
Train One Sparse Autoencoder Across Multiple Sparsity Budgets to Preserve Interpretability and Accuracy
Paper • 2505.24473 • Published
Interpreting and Steering a Text-to-Speech Language Model with Sparse Autoencoders
Paper • 2606.10029 • Published • 12
upvoted a paper 13 days ago
submitted a
paper to Daily Papers 13 days ago
authored a
paper 22 days ago
updated a
model 2 months ago
published a
model 2 months ago
updated a
model 2 months ago
published a
model 2 months ago
updated a
model 2 months ago
published a
model 2 months ago
updated a
model 2 months ago
published a
model 2 months ago
updated a
model 2 months ago
published a
model 2 months ago