{"id":4318,"date":"2024-07-18T21:44:23","date_gmt":"2024-07-18T13:44:23","guid":{"rendered":"https:\/\/www.aqwu.net\/wp\/?p=4318"},"modified":"2024-07-18T21:44:23","modified_gmt":"2024-07-18T13:44:23","slug":"%e4%b8%8d%e5%90%ab-matmul-%e7%9a%84-lm","status":"publish","type":"post","link":"https:\/\/www.aqwu.net\/wp\/?p=4318","title":{"rendered":"\u4e0d\u542b MatMul \u7684 LM"},"content":{"rendered":"\n<h1 class=\"wp-block-heading\">\u4ecb\u7ecd<\/h1>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"445\" src=\"https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1024x445.png\" alt=\"\" class=\"wp-image-4319\" srcset=\"https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1024x445.png 1024w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-300x130.png 300w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-768x334.png 768w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1536x668.png 1536w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-2048x890.png 2048w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1320x574.png 1320w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-600x261.png 600w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p><a href=\"https:\/\/github.com\/ridgerchu\/matmulfreellm?tab=readme-ov-file#introduction\"><\/a><\/p>\n\n\n\n<p>MatMul-Free LM \u662f\u4e00\u79cd\u8bed\u8a00\u6a21\u578b\u67b6\u6784\uff0c\u65e0\u9700\u77e9\u9635\u4e58\u6cd5 \uff08MatMul\uff09 \u8fd0\u7b97\u3002\u6b64\u5b58\u50a8\u5e93\u63d0\u4f9b\u4e86\u4e0e \ud83e\udd17 Transformer \u5e93\u517c\u5bb9\u7684 MatMul-Free LM \u5b9e\u73b0\u3002<\/p>\n\n\n\n<h1 class=\"wp-block-heading\">\u7f29\u653e\u5b9a\u5f8b<\/h1>\n\n\n\n<p><a href=\"https:\/\/github.com\/ridgerchu\/matmulfreellm?tab=readme-ov-file#scaling-law\"><\/a><\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"512\" src=\"https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-1024x512.png\" alt=\"\" class=\"wp-image-4320\" srcset=\"https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-1024x512.png 1024w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-300x150.png 300w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-768x384.png 768w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-1536x768.png 1536w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-2048x1024.png 2048w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-1320x660.png 1320w, https:\/\/www.aqwu.net\/wp\/wp-content\/uploads\/2024\/07\/\u56fe\u7247-1-600x300.png 600w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/figure>\n\n\n\n<p>\u6211\u4eec\u8bc4\u4f30\u4e86\u7f29\u653e\u5b9a\u5f8b\u5982\u4f55\u62df\u5408 Transformer++ \u548c\u6211\u4eec\u7684\u6a21\u578b\u4e2d\u7684 370M\u30011.3B \u548c 2.7B \u53c2\u6570\u6a21\u578b\u3002\u4e3a\u4e86\u516c\u5e73\u5730\u6bd4\u8f83\uff0c\u6bcf\u4e2a\u64cd\u4f5c\u7684\u5904\u7406\u65b9\u5f0f\u90fd\u76f8\u540c\uff0c\u5c3d\u7ba1\u6211\u4eec\u7684\u6a21\u578b\u5728\u67d0\u4e9b\u5c42\u4e2d\u4f7f\u7528\u4e86\u66f4\u6709\u6548\u7684\u4e09\u5143\u6743\u91cd\u3002\u6709\u8da3\u7684\u662f\uff0c\u4e0e Transformer++ \u76f8\u6bd4\uff0c\u6211\u4eec\u6a21\u578b\u7684\u7f29\u653e\u9884\u6d4b\u8868\u73b0\u51fa\u66f4\u9661\u5ced\u7684\u4e0b\u964d\uff0c\u8fd9\u8868\u660e\u6211\u4eec\u7684\u67b6\u6784\u5728\u5229\u7528\u989d\u5916\u7684\u8ba1\u7b97\u6765\u63d0\u9ad8\u6027\u80fd\u65b9\u9762\u66f4\u6709\u6548\u3002<\/p>\n\n\n\n<p>github: <a href=\"https:\/\/github.com\/ridgerchu\/matmulfreellm\">https:\/\/github.com\/ridgerchu\/matmulfreellm<\/a><\/p>\n\n\n\n<p>\u8bba\u6587\uff1a<a href=\"https:\/\/arxiv.org\/abs\/2406.02528\">[2406.02528] Scalable MatMul-free Language Modeling (arxiv.org)<\/a><\/p>\n\n\n\n<p><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u4ecb\u7ecd MatMul-Free LM \u662f\u4e00\u79cd\u8bed\u8a00\u6a21\u578b\u67b6\u6784\uff0c\u65e0\u9700\u77e9\u9635\u4e58\u6cd5 \uff08MatMul\uff09 \u8fd0\u7b97\u3002\u6b64\u5b58\u50a8\u5e93\u63d0\u4f9b\u4e86\u4e0e [&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":[444,445,443,442],"tags":[404,521],"class_list":["post-4318","post","type-post","status-publish","format-standard","hentry","category-ai","category-ainews","category-llm","category-llms","tag-llm","tag-matmul"],"views":900,"jetpack_sharing_enabled":true,"jetpack_featured_media_url":"","_links":{"self":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/4318","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=4318"}],"version-history":[{"count":1,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/4318\/revisions"}],"predecessor-version":[{"id":4321,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=\/wp\/v2\/posts\/4318\/revisions\/4321"}],"wp:attachment":[{"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4318"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4318"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.aqwu.net\/wp\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4318"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}