HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization Paper • 2503.04598 • Published 7 days ago • 17 • 8
HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization Paper • 2503.04598 • Published 7 days ago • 17 • 8
HybridNorm: Towards Stable and Efficient Transformer Training via Hybrid Normalization Paper • 2503.04598 • Published 7 days ago • 17 • 8
Scale-Distribution Decoupling: Enabling Stable and Effective Training of Large Language Models Paper • 2502.15499 • Published 20 days ago • 13 • 2