Superposition Is Not Necessary: A Mechanistic Interpretability Analysis of Transformer Representations for Time Series Forecasting

Documenti analoghi

• The Geometric Inductive Bias of Grokking: Bypassing Phase Transitions via Architectural Topology
  di: Yıldırım, Alper
  Pubblicazione: (2026)
• Language as a Wave Phenomenon: Semantic Phase Locking and Interference in Neural Networks
  di: Yıldırım, Alper, et al.
  Pubblicazione: (2025)
• DRFormer: Multi-Scale Transformer Utilizing Diverse Receptive Fields for Long Time-Series Forecasting
  di: Ding, Ruixin, et al.
  Pubblicazione: (2024)
• Evaluating the effectiveness of predicting covariates in LSTM Networks for Time Series Forecasting
  di: Davies, Gareth
  Pubblicazione: (2024)
• Inter-Series Transformer: Attending to Products in Time Series Forecasting
  di: Cristian, Rares, et al.
  Pubblicazione: (2024)