HMoE: Heterogeneous Mixture of Experts for Language Modeling Paper • 2408.10681 • Published Aug 20, 2024 • 8
Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent Paper • 2411.02265 • Published Nov 4, 2024 • 24
Scaling Laws for Floating Point Quantization Training Paper • 2501.02423 • Published 20 days ago • 25