Cites: The Linear Centroids Hypothesis: Features as Directions Learned by Local Experts

5 citation edges inside this catalog. A title match is weaker evidence than an identifier match. Zero entries does not mean no citations on the web.

• Searching for Activation Functions
  arxiv:1710.05941 · evidence: id
• DINOv2: Learning Robust Visual Features without Supervision
  arxiv:2304.07193 · evidence: title
• Sparse Autoencoders Find Highly Interpretable Features in Language Models
  arxiv:2309.08600 · evidence: title
• The Platonic Representation Hypothesis
  arxiv:2405.07987 · evidence: title
• The Llama 3 Herd of Models
  arxiv:2407.21783 · evidence: id