Alexey Gorbatovski

Myashka

·

AI & ML interests

NLP Alignment

Recent Activity

liked a Space 4 days ago

t-tech/t-search-blog

upvoted a paper about 1 month ago

Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders

upvoted a paper about 1 month ago

Next Embedding Prediction Makes World Models Stronger

View all activity

Organizations

None yet

liked a Space 4 days ago

T-Search: an open agentic retriever

Open agentic retriever for hard multi-step search

liked a Space 5 months ago

Chasing the Counting Manifold in Open LLMs

Counting manifolds in open LLMs from behavior to SAEs.