Improving Model Latent Visualization
Recommended Reads
Specific to France
- Interpretability vs Neuroscience [rough note] — colah’s blog
- Analogies between Biology and Deep Learning [rough note] — colah’s blog
Hyperbolic Embedding Spaces
Resources
Mono Repo
Papers
- Predicting model training dynamics with hyperbolic ‘skeleton’
- Relevant SAE paper (manifolds): Sparse Autoencoders Reveal Universal Feature Spaces Across Large Language Models
- 2010.00402
- GitHub - HazyResearch/hyperbolics: Hyperbolic Embeddings