← all tags

#interpretability

2 posts tagged interpretability.

Thoughts

Sparse autoencoders just turned interpretability into feature engineering

We're not making AI more transparent, we're just building better debugging tools for black boxes.

interpretability sparse-autoencoders model-analysis feature-engineering

Tools & Experiments

Qwen-Scope

Sparse autoencoders that turn LLM black-box internals into interpretable features you can actually use.

experimental
interpretability sparse-autoencoders llm-internals qwen