Finding and using interpretable latents in a neutrino foundation model with sparse autoencoders
astro-ph.HE, cs.AI, cs.LG, hep-ex
Submitted: 2026-08-26
Updated: 2026-08-26
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Energy Reconstruction Methods in the IceCube Neutrino Telescope
- The IceCube Neutrino Observatory: Instrumentation and Online Systems
- Scaling Laws for Neural Language Models
- On the Opportunities and Risks of Foundation Models
- Pre-training strategy using real particle collision data for event classification in collider physics
- Finetuning Foundation Models for Joint Analysis Optimization
- Explainable AI for Jet Tagging: A Comparative Study of GNNExplainer, GNNShap, and GradCAM for Jet Tagging in the Lund Jet Plane
- Sparse Autoencoders Find Highly Interpretable Features in Language Models
- Scaling and evaluating sparse autoencoders
- Are Sparse Autoencoders Useful? A Case Study in Sparse Probing
- Neural Scaling Laws for Boosted Jet Tagging
- Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
- Axiomatic Attribution for Deep Networks
- The Physics Behind ML-based Quark-Gluon Taggers
- Open Problems in Mechanistic Interpretability
- Dissecting Jet-Tagger Through Mechanistic Interpretability
- Toy Models of Superposition
- Sparse Autoencoders Trained on the Same Data Learn Different Features
- Interpretable and Steerable Concept Bottleneck Sparse Autoencoders
- Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
Related papers
- Accreting Neutron Stars in Low-Mass X-ray Binaries
- Numerical Studies of Accretion Flows onto a Neutron Star Engulfed in a Massive Star
- Early panchromatic spectra reveal helium in the thermonuclear supernova 2024kce
- Multiwavelength Classification of Chandra Sources in Omega Centauri
- Impact of the equation of state on core collapse supernovae I: the low- T/W instability
- Secular Evolution of Quasiperiodic Eruptions: eRO-QPE2