微博
加入微博一起分享新鲜事
登录
|
注册
140
Research without Re-search: Maximal Update Parametrization Yields Accurate Loss Prediction across Scales https://www.wangyequan.com/zh/publication/arxiv2023-mu-scaling/
请登录并选择要私信的好友
300
Research without Re-search: Maximal Update Parametrization Yields Accurate Loss Prediction across Scales https://www.wangyequan.com/zh/publication/arxiv2023-mu-scaling/
赞一下这个内容
公开
分享
获取分享按钮
正在发布微博，请稍候