{"id":3696,"date":"2024-05-11T11:37:59","date_gmt":"2024-05-11T03:37:59","guid":{"rendered":"https:\/\/www.aqwu.net\/wp\/?p=3696"},"modified":"2024-05-11T11:37:59","modified_gmt":"2024-05-11T03:37:59","slug":"slicegpt%ef%bc%9a%e9%80%9a%e8%bf%87%e5%88%a0%e9%99%a4%e8%a1%8c%e5%92%8c%e5%88%97%e6%9d%a5%e5%8e%8b%e7%bc%a9%e5%a4%a7%e5%9e%8b%e8%af%ad%e8%a8%80%e6%a8%a1%e5%9e%8b","status":"publish","type":"post","link":"https:\/\/www.aqwu.net\/wp\/?p=3696","title":{"rendered":"SliceGPT\uff1a\u901a\u8fc7\u5220\u9664\u884c\u548c\u5217\u6765\u538b\u7f29\u5927\u578b\u8bed\u8a00\u6a21\u578b"},"content":{"rendered":"\n<p>Microsoft \u5728\u8fd9\u7bc7\u8bba\u6587\u4e2d\u63d0\u51fa\u4e86\u975e\u5e38\u597d\u7684\u5efa\u8bae &#8211; \u201cSliceGPT\uff1a\u901a\u8fc7\u5220\u9664\u884c\u548c\u5217\u6765\u538b\u7f29\u5927\u578b\u8bed\u8a00\u6a21\u578b\u201d<\/p>\n\n\n\n<p>\u5bf9\u4e8eLLAMA2-70B\u3001OPT 66B\u548cPhi-2\u6a21\u578b\uff0cSliceGPT\u53ef\u4ee5\u53bb\u9664\u9ad8\u8fbe25%\u7684\u6a21\u578b\u53c2\u6570\uff08\u5305\u62ec\u5d4c\u5165\uff09\uff0c\u540c\u65f6\u5206\u522b\u4fdd\u6301\u5bc6\u96c6\u6a21\u578b99%\u300199%\u548c90%\u7684\u96f6\u6837\u672c\u4efb\u52a1\u6027\u80fd\u3002<\/p>\n\n\n\n<p>\u5728 24GB \u6d88\u8d39\u7ea7 GPU \u4e0a\uff0cLLAMA2-70B \u4e0a\u7684\u63a8\u7406\u603b\u8ba1\u7b97\u91cf\u51cf\u5c11\u5230\u5bc6\u96c6\u6a21\u578b\u7684 64%;\u5728 40GB A100 GPU \u4e0a\u5c06\u5176\u964d\u4f4e\u5230 66%\u3002 &#8212;&#8212;&#8212;<\/p>\n\n\n\n<p>\u7a00\u758f\u5316\u662f\u8d44\u6e90\u7ea6\u675f\u95ee\u9898\u7684\u89e3\u51b3\u65b9\u6848\uff0c\u6700\u8fd1\u7684\u7814\u7a76\u8868\u660e\uff0c\u7ecf\u8fc7\u8bad\u7ec3\u7684\u6a21\u578b\u53ef\u4ee5\u5728\u4e8b\u540e\u8fdb\u884c\u7a00\u758f\u5316\u3002<\/p>\n\n\n\n<p>\u73b0\u6709\u7684\u7a00\u758f\u5316\u6280\u672f\u9700\u8981\u989d\u5916\u7684\u6570\u636e\u7ed3\u6784\uff0c\u5e76\u4e14\u4f7f\u7528\u5f53\u524d\u786c\u4ef6\u63d0\u4f9b\u6709\u9650\u7684\u52a0\u901f\u3002<\/p>\n\n\n\n<p>SliceGPT Paper \u63d0\u51fa\u4e86\u4e00\u79cd\u65b0\u7684\u8bad\u7ec3\u540e\u7a00\u758f\u5316\u65b9\u6848\uff0c\u8be5\u65b9\u6848\u5c06\u6bcf\u4e2a\u6743\u91cd\u77e9\u9635\u66ff\u6362\u4e3a\u66f4\u5c0f\uff08\u5bc6\u96c6\uff09\u7684\u77e9\u9635\uff0c\u4ece\u800c\u964d\u4f4e\u4e86\u7f51\u7edc\u7684\u5d4c\u5165\u7ef4\u5ea6\u3002<\/p>\n\n\n\n<p>\u8bba\u6587\u5730\u5740\uff1a<a href=\"https:\/\/arxiv.org\/abs\/2401.15024\">https:\/\/arxiv.org\/abs\/2401.15024<\/a><\/p>\n\n\n\n<p>github:<a href=\"https:\/\/github.com\/microsoft\/TransformerCompression\">https:\/\/github.com\/microsoft\/TransformerCompression<\/a><\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Microsoft \u5728\u8fd9\u7bc7\u8bba\u6587\u4e2d\u63d0\u51fa\u4e86\u975e\u5e38\u597d\u7684\u5efa\u8bae &#8211; \u201cSliceGPT\uff1a\u901a\u8fc7\u5220\u9664\u884c\u548c\u5217\u6765\u538b\u7f29\u5927 [&hellip;]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"site-sidebar-layout":"default","site-content-layout":"","ast-site-content-layout":"default","site-content-style":"default","site-sidebar-style":"default","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"","ast-breadcrumbs-content":"","ast-featured-img":"","footer-sml-layout":"","theme-transparent-header-meta":"","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"var(--ast-global-color-4)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[445,443,442,58],"tags":[459,458],"class_list":["post-3696","post","type-post","status-publish","format-standard","hentry","category-ainews","category-llm","category-llms","category-windows","tag-llama2-70b","tag-slicegpt"],"views":943,"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/3696","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3696"}],"version-history":[{"count":1,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/3696\/revisions"}],"predecessor-version":[{"id":3697,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/3696\/revisions\/3697"}],"wp:attachment":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3696"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3696"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3696"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}