AWQ 量化详解 - Zhang #226
Replies: 1 comment
-
真不错 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
AWQ 量化详解 - Zhang
从事 LLM 推理部署、视觉算法开发、模型压缩部署以及算法SDK开发工作,终身学习践行者。LLM_Compression先回顾了 SmoothQuant 算法的是三个核心观点,然后开始解读 AWQ 算法的两个核心观点(创新点)::LLM 权重并非同等重要,只有 0.1%~1% 的小部分显著权重对模型输出精度影响较大,又因为幅度较大的输入特征通常更重要,因此需要基于激活分布来挑选权重的显著通道,以及如何基于激活感知缩放保护关键权重。
https://www.armcvai.cn/2024-11-01/llm-quant-awq.html
Beta Was this translation helpful? Give feedback.
All reactions