Small Vectors, Big Effects: A Mechanistic Study of RL-Induced Reasoning via Steering Vectors
Paper • 2509.06608 • Published
Gleb Gerasimov
gudleifrr
AI & ML interests
NLP, interpretability
Recent Activity
authored a paper 10 days ago
Small Vectors, Big Effects: A Mechanistic Study of RL-Induced Reasoning
via Steering Vectors authored a paper 10 days ago
Train One Sparse Autoencoder Across Multiple Sparsity Budgets to
Preserve Interpretability and Accuracy authored a paper 10 days ago
Teach Old SAEs New Domain Tricks with BoostingOrganizations
None yet
authored 4
papers 10 days ago
Train One Sparse Autoencoder Across Multiple Sparsity Budgets to Preserve Interpretability and Accuracy
Paper • 2505.24473 • Published
Teach Old SAEs New Domain Tricks with Boosting
Paper • 2507.12990 • Published • 12
Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders
Paper • 2606.12138 • Published • 8
upvoted a paper 10 days ago
upvoted a paper 16 days ago
liked a
Space 4 months ago
upvoted a paper 5 months ago
updated a
model 6 months ago
published a
model 6 months ago
published a
dataset 6 months ago
updated a
dataset 7 months ago
published a
dataset 7 months ago
updated a
dataset 10 months ago
published a
dataset 10 months ago
updated a
model 10 months ago
published 2
models 10 months ago
updated a
model 10 months ago
published a
model 10 months ago