To 2:4 Sparsity and Beyond: Neuron-level Activation Function to Accelerate LLM Pre-Training

Documenti analoghi

• Accelerating Transformer Inference and Training with 2:4 Activation Sparsity
  di: Haziza, Daniel, et al.
  Pubblicazione: (2025)
• Composer: A Search Framework for Hybrid Neural Architecture Design
  di: Acun, Bilge, et al.
  Pubblicazione: (2025)
• Masked Matrix Multiplication for Emergent Sparsity
  di: Wheatman, Brian, et al.
  Pubblicazione: (2024)
• MAD Max Beyond Single-Node: Enabling Large Machine Learning Model Acceleration on Distributed Systems
  di: Hsia, Samuel, et al.
  Pubblicazione: (2023)
• Quagmires in SFT-RL Post-Training: When High SFT Scores Mislead and What to Use Instead
  di: Kang, Feiyang, et al.
  Pubblicazione: (2025)