arXiv 2026-08-26 EN Finding and using interpretable latents in a neutrino foundation model with sparse autoencoders Raphaël Bonnet-Guerrini · Johann Ioannou-Nikolaides · Inar Timiryasov · Vincenzo Piuri
