Paper

Abstract

这里用来放读论文的笔记，写的不好还请见谅>_<|||

• BiS-KM: Enabling Any-Precision K-Means on FPGAs
• ZeRO: Memory Optimizations Toward Training Trillion Parameter Models
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
• Mixed Precision Training
• FAST: DNN Training Under Variable Precision Block Floating Point with Stochastic Rounding
• Softmax Acceleration with Adaptive Numeric Format for both Training and Inference
• Efficient Memory Management for Large Language Model Serving with PagedAttention
• VecPAC: A Vectorizable and Precision-Aware CGRA

---

最后更新: 2024年7月2日 18:04:56
创建日期: 2024年2月4日 19:47:15

评论