{"id":4483,"date":"2023-11-21T17:28:15","date_gmt":"2023-11-21T09:28:15","guid":{"rendered":"http:\/\/www.sniper97.cn\/?p=4483"},"modified":"2023-11-21T17:30:20","modified_gmt":"2023-11-21T09:30:20","slug":"%e3%80%90%e8%ae%ba%e6%96%87%e3%80%91small-scale-proxies-for-large-scale-transformer-training-instabilities","status":"publish","type":"post","link":"http:\/\/www.sniper97.cn\/index.php\/note\/deep-learning\/paper-deep-learning\/4483\/","title":{"rendered":"\u3010\u8bba\u6587\u3011Small-scale proxies for large-scale Transformer training instabilities"},"content":{"rendered":"\n<p>google\u7684\u4e00\u4e2a\u5de5\u4f5c\uff0c\u672c\u6587\u4f7f\u7528\u5c0f\u6a21\u578b\u6765\u9a8c\u8bc1\u5728\u8bad\u7ec3\u5927\u6a21\u578b\u65f6\u4f1a\u4ea7\u751f\u7684loss\u95ee\u9898\uff08\u8bad\u7ec3\u4e0d\u7a33\u5b9a\uff09\uff0c\u6210\u529f\u7684\u5728\u5c0f\u6a21\u578b\u4e0a\u590d\u73b0\u4e86\u8fd9\u4e00\u95ee\u9898\uff0c\u5e76\u9a8c\u8bc1\u4e86\u5f53\u524d\u89e3\u51b3\u5927\u6a21\u578b\u7684loss\u95ee\u9898\u7684\u65b9\u6cd5\uff0c\u5728\u5c0f\u6a21\u578b\u4e0a\u4f9d\u7136\u6709\u6548\u3002\u5e76\u6210\u529f\u9884\u6d4b\u4e86\u8bad\u7ec3\u4e0d\u7a33\u5b9a\u95ee\u9898\u7684\u51fa\u73b0\u3002<\/p>\n\n\n\n<p><\/p>\n\n\n\n<p>\u5199\u5728notion\u4e0a\u4e86\uff0c\u61d2\u5f97\u642c\u8fd0\u3002\u94fe\u63a5\u5982\u4e0b\uff08\u53ef\u80fd\u9700\u8981\u7ffb\u5899\uff09<\/p>\n\n\n\n<p><a href=\"https:\/\/daffodil-square-501.notion.site\/Small-scale-proxies-for-large-scale-Transformer-training-instabilities-80425d883bfe44a88fa63dee559d2eae?pvs=4\" target=\"_blank\" rel=\"noreferrer noopener\">https:\/\/daffodil-square-501.notion.site\/Small-scale-proxies-for-large-scale-Transformer-training-instabilities-80425d883bfe44a88fa63dee559d2eae?pvs=4<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>google\u7684\u4e00\u4e2a\u5de5\u4f5c\uff0c\u672c\u6587\u4f7f\u7528\u5c0f\u6a21\u578b\u6765\u9a8c\u8bc1\u5728\u8bad\u7ec3\u5927\u6a21\u578b\u65f6\u4f1a\u4ea7\u751f\u7684loss\u95ee\u9898\uff08\u8bad\u7ec3\u4e0d\u7a33\u5b9a\uff09\uff0c\u6210\u529f\u7684 [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"om_disable_all_campaigns":false,"_mi_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[16],"tags":[],"views":4957,"_links":{"self":[{"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/posts\/4483"}],"collection":[{"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/comments?post=4483"}],"version-history":[{"count":2,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/posts\/4483\/revisions"}],"predecessor-version":[{"id":4487,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/posts\/4483\/revisions\/4487"}],"wp:attachment":[{"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/media?parent=4483"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/categories?post=4483"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/www.sniper97.cn\/index.php\/wp-json\/wp\/v2\/tags?post=4483"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}