{"id":4849,"date":"2024-05-13T05:08:00","date_gmt":"2024-05-12T20:08:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=4849"},"modified":"2024-05-13T05:08:00","modified_gmt":"2024-05-12T20:08:00","slug":"deepseek-v2-llama-3-youko-8b","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=4849","title":{"rendered":"DeepSeek v2, Llama 3 Youko 8B, AlphaFold\u20093"},"content":{"rendered":"\n<p>\u6765\u9031OpenAI\u304b\u3089\u5927\u304d\u306a\u767a\u8868\u304c\u3042\u308b\u3068\u30a2\u30ca\u30a6\u30f3\u30b9\u3055\u308c\u3066\u3044\u308b\u304c\u3001\u4eca\u9031\u3082LLM\u95a2\u9023\u306e\u8a71\u984c\u306f\u591a\u304b\u3063\u305f\u3002\u6ce8\u76ee\u306fMoE\u306aDeepSeek v2\u3067\u3042\u308a\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u9577\u3001\u6027\u80fd\u3068\u3082\u975e\u5e38\u306b\u671f\u5f85\u3067\u304d\u305d\u3046\u3002\u65e5\u672c\u8a9e\u8a71\u8005\u3068\u3057\u3066\u306fLlama 3\u306e\u65e5\u672c\u8a9e\u5f37\u5316\u30d0\u30fc\u30b8\u30e7\u30f3\u3067\u3042\u308bLlama 3 Youko 8B\u306b\u3082\u6ce8\u76ee\u3067\u3042\u308b\u3002<br><a href=\"https:\/\/rinna.co.jp\/news\/2024\/05\/20240507.html\">rinna\u3001Llama 3\u306e\u65e5\u672c\u8a9e\u7d99\u7d9a\u4e8b\u524d\u5b66\u7fd2\u30e2\u30c7\u30eb\u300cLlama 3 Youko 8B\u300d\u3092\u516c\u958b\uff5crinna\u682a\u5f0f\u4f1a\u793e<\/a><\/p>\n\n\n\n<p>\u305d\u3057\u3066\u3001AlphaFold3\u306e\u8ad6\u6587\u304cNature\u306b\u767a\u8868\u3055\u308c\u3066\u3044\u308b\u3002LLM\u95a2\u9023\u3067\u306f\u306a\u3044\u304c\u30e2\u30c7\u30eb\u306e\u610f\u7fa9\u3084\u5f71\u97ff\u306f\u5927\u304d\u3044\u3002\u6700\u65b0\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u306e\u9069\u7528\u306a\u3069\u7814\u7a76\u3092\u7d9a\u3051\u3001\u304b\u3064\u3001\u6210\u679c\u3092\u51fa\u3057\u3066\u3044\u308b\u306e\u304c\u3059\u3054\u3044\u3002<br><a href=\"https:\/\/blog.google\/technology\/ai\/google-deepmind-isomorphic-alphafold-3-ai-model\/\">Google DeepMind and Isomorphic Labs introduce AlphaFold 3 AI model (blog.google)<\/a><\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model&nbsp;<\/strong>[0.0]<br>We present DeepSeek-V2, a strong Mixture-of-Experts (MoE) language model characterized by economical training and efficient inference\u3002 DeepSeek-V2\u306f\u3001MLA(Multi-head Latent Attention)\u3084DeepSeekMoE\u3068\u3044\u3063\u305f\u9769\u65b0\u7684\u306a\u30a2\u30fc\u30ad\u30c6\u30af\u30c1\u30e3\u3092\u63a1\u7528\u3057\u3066\u3044\u308b\u3002\u5408\u8a08\u30d1\u30e9\u30e1\u30fc\u30bf\u306f236B\u3067\u3001\u305d\u306e\u3046\u306121B\u306f\u30c8\u30fc\u30af\u30f3\u3054\u3068\u306b\u6d3b\u6027\u5316\u3055\u308c\u3001128K\u30c8\u30fc\u30af\u30f3\u306e\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u9577\u3092\u30b5\u30dd\u30fc\u30c8\u3059\u308b\u3002 DeepSeek-V2\u306fDeepSeek 67B\u3068\u6bd4\u8f03\u3057\u3066\u5927\u5e45\u306b\u6027\u80fd\u304c\u5411\u4e0a\u3057\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30b3\u30b9\u30c8\u306f42.5%\u524a\u6e1b\u3055\u308c\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2405.04434v1\">\u8ad6\u6587<\/a>&nbsp;&nbsp;<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2405.04434v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>&nbsp; &nbsp;(Tue, 07 May 2024 15:56:43 GMT)<\/li>\n\n\n\n<li>Active\u306a\u30d1\u30e9\u30e1\u30fc\u30bf\u304c20B\u7a0b\u5ea6\u3067Mixtral 8x22B\u3084Llama 3 70B\u3068\u7af6\u5408\u3059\u308b\u6027\u80fd\u3092\u4e3b\u5f35\u3001\u30e9\u30a4\u30bb\u30f3\u30b9\u306f\u72ec\u81ea\u3060\u304c\u304b\u306a\u308a\u5bdb\u5bb9\u7684\u306a\u3082\u306e\u306b\u898b\u3048\u308b\u3002<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/github.com\/deepseek-ai\/deepSeek-V2\">GitHub &#8211; deepseek-ai\/DeepSeek-V2<\/a>\u3001Weight\u306f<a href=\"https:\/\/huggingface.co\/deepseek-ai\/DeepSeek-V2\">deepseek-ai\/DeepSeek-V2 \u00b7 Hugging Face<\/a>\u3001<a href=\"https:\/\/huggingface.co\/deepseek-ai\/DeepSeek-V2-Chat\">deepseek-ai\/DeepSeek-V2-Chat \u00b7 Hugging Face<\/a><\/li>\n<\/ul>\n","protected":false},"excerpt":{"rendered":"<p>\u6765\u9031OpenAI\u304b\u3089\u5927\u304d\u306a\u767a\u8868\u304c\u3042\u308b\u3068\u30a2\u30ca\u30a6\u30f3\u30b9\u3055\u308c\u3066\u3044\u308b\u304c\u3001\u4eca\u9031\u3082LLM\u95a2\u9023\u306e\u8a71\u984c\u306f\u591a\u304b\u3063\u305f\u3002\u6ce8\u76ee\u306fMoE\u306aDeepSeek v2\u3067\u3042\u308a\u30b3\u30f3\u30c6\u30ad\u30b9\u30c8\u9577\u3001\u6027\u80fd\u3068\u3082\u975e\u5e38\u306b\u671f\u5f85\u3067\u304d\u305d\u3046\u3002\u65e5\u672c\u8a9e\u8a71\u8005\u3068\u3057\u3066\u306fLlama 3\u306e\u65e5\u672c &hellip; <a href=\"https:\/\/devneko.jp\/wordpress\/?p=4849\" class=\"more-link\"><span class=\"screen-reader-text\">&#8220;DeepSeek v2, Llama 3 Youko 8B, AlphaFold\u20093&#8221; \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[2],"tags":[223,293],"class_list":["post-4849","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-oss"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4849","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=4849"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/4849\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4849"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4849"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4849"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}