{"id":5785,"date":"2024-11-25T06:42:00","date_gmt":"2024-11-24T21:42:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=5785"},"modified":"2024-11-25T06:42:00","modified_gmt":"2024-11-24T21:42:00","slug":"hymba-a-hybrid-head-architecture-for-small-language-models","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=5785","title":{"rendered":"Hymba: A Hybrid-head Architecture for Small Language Models\u00a0"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>Hymba: A Hybrid-head Architecture for Small Language Models\u00a0<\/strong>[65.9]<br>Hymba\u306f\u3001\u30cf\u30a4\u30d6\u30ea\u30c3\u30c9\u30d8\u30c3\u30c9\u4e26\u5217\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3092\u7279\u5fb4\u3068\u3059\u308b\u3001\u5c0f\u3055\u306a\u8a00\u8a9e\u30e2\u30c7\u30eb\u306e\u30d5\u30a1\u30df\u30ea\u30fc\u3067\u3042\u308b\u3002 \u91cd\u8981\u306a\u60c5\u5831\u3092\u4fdd\u6301\u3059\u308b\u30d7\u30ed\u30f3\u30d7\u30c8\u306b\u5148\u7acb\u3063\u3066,\u5b66\u7fd2\u53ef\u80fd\u306a\u30e1\u30bf\u30c8\u30fc\u30af\u30f3\u3092\u5c0e\u5165\u3059\u308b\u3002 \u3053\u306e\u30e2\u30c7\u30eb\u306f\u3001\u5c64\u9593\u9375\u5024\u5171\u6709\u3068\u90e8\u5206\u7684\u30b9\u30e9\u30a4\u30c7\u30a3\u30f3\u30b0\u30a6\u30a3\u30f3\u30c9\u30a6\u30a2\u30c6\u30f3\u30b7\u30e7\u30f3\u3092\u7d44\u307f\u8fbc\u3080\u3053\u3068\u306b\u3088\u308a\u3001\u3055\u3089\u306b\u6700\u9069\u5316\u3055\u308c\u308b\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2411.13676v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2411.13676v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Wed, 20 Nov 2024 19:51:25 GMT)<\/li>\n\n\n\n<li>Transformer\u306eAttention\u3068SSM\u3092\u7d44\u307f\u5408\u308f\u305b\u305f\u30e2\u30c7\u30eb\u306e\u63d0\u6848\u3001\u5c0f\u578b\u30e2\u30c7\u30eb\u3067\u306f\u3068\u3066\u3082\u9ad8\u3044\u6027\u80fd\u3068\u7701\u30e1\u30e2\u30ea\u9ad8\u901f\u52d5\u4f5c\u3092\u5b9f\u73fe\u3068\u306e\u3053\u3068\u3002<\/li>\n\n\n\n<li>NVIDIA\u306b\u3088\u308b\u767a\u8868\u3067\u30e2\u30c7\u30eb\u304c\u516c\u958b\u3055\u308c\u3066\u3044\u308b\u3002<a href=\"https:\/\/huggingface.co\/nvidia\/Hymba-1.5B-Base\">nvidia\/Hymba-1.5B-Base \u00b7 Hugging Face<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[235,376],"class_list":["post-5785","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-mamba","tag-ssm"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5785","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=5785"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/5785\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=5785"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=5785"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=5785"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}