Shivam Raval

Shivam Raval

AI Research Scientist

Generalist. Eternally curious.


Interests

  • AI Interpretability & Safety
  • High-dimensional Visualization
  • Interactive ML Systems

Get in touch: sraval@g.harvard.edu

AboutResearchResumeWritingVisuals

Writing

Essays, explorables, and research write-ups.

Why SAEs Fail (and Why CLTs Might Too)

Interactive

An interactive explorable on sparse autoencoders, cross-layer transcoder features, and where both approaches break down.

InterpretabilitySparse AutoencodersMachine Learning
2025-05-15

Seeing Through the Projection

8 min + interactive

Why every 2D embedding of your data is a lie — and how to think about which lies are acceptable.

Dimensionality ReductionVisualizationResearch
2024-11-10

The Case for Explorable Explanations

10 min

Static figures are a compromise forced by print. We can do better — and we should.

Research CommunicationInteractive MediaScience
2024-08-20