{"id":6202,"date":"2025-02-14T05:10:00","date_gmt":"2025-02-13T20:10:00","guid":{"rendered":"https:\/\/devneko.jp\/wordpress\/?p=6202"},"modified":"2025-02-14T05:10:00","modified_gmt":"2025-02-13T20:10:00","slug":"smollm2-when-smol-goes-big-data-centric-training-of-a-small-language-model","status":"publish","type":"post","link":"https:\/\/devneko.jp\/wordpress\/?p=6202","title":{"rendered":"SmolLM2: When Smol Goes Big &#8212; Data-Centric Training of a Small Language Model"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>SmolLM2: When Smol Goes Big &#8212; Data-Centric Training of a Small Language Model\u00a0<\/strong>[33.9]<br>SmolLM2\u306f\u3001\u6700\u5148\u7aef\u306e&#8221;\u5c0f&#8221; (170\u5104\u306e\u30d1\u30e9\u30e1\u30fc\u30bf) \u8a00\u8a9e\u30e2\u30c7\u30eb\u3067\u3042\u308b\u3002 \u6211\u3005\u306fSmolLM2\u30921\u5146\u306e\u30c8\u30fc\u30af\u30f3\u3067\u30aa\u30fc\u30d0\u30fc\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u3001Web\u30c6\u30ad\u30b9\u30c8\u3068\u7279\u6b8a\u306a\u7b97\u6570\u3001\u30b3\u30fc\u30c9\u3001\u547d\u4ee4\u8ffd\u5f93\u30c7\u30fc\u30bf\u3068\u3092\u6df7\u5408\u3059\u308b\u591a\u6bb5\u968e\u306e\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30d7\u30ed\u30bb\u30b9\u3092\u7528\u3044\u305f\u3002 \u6211\u3005\u306f\u3001SmolLM2\u304cQwen2.5-1.5B\u3084Llama3.2-1B\u306a\u3069\u3001\u6700\u8fd1\u306e\u5c0f\u3055\u306aLM\u3088\u308a\u3082\u512a\u308c\u3066\u3044\u308b\u3053\u3068\u3092\u793a\u3057\u305f\u3002<br><a href=\"http:\/\/arxiv.org\/abs\/2502.02737v1\">\u8ad6\u6587<\/a>\u00a0\u00a0<a href=\"https:\/\/fugumt.com\/fugumt\/paper_check\/2502.02737v1\">\u53c2\u8003\u8a33\uff08\u30e1\u30bf\u30c7\u30fc\u30bf\uff09<\/a>\u00a0 \u00a0(Tue, 04 Feb 2025 21:43:16 GMT)<\/li>\n\n\n\n<li>Huggingface\u306b\u3088\u308bSLM\u3001\u300cSmolLM2 advances the state-of-the-art for open small LMs through a combination of careful dataset curation and multistage training.\u300d\u3068\u306e\u3053\u3068\u3002\u300cSmolLM2 outperforms other recent small LMs including Qwen2.5-1.5B and Llama3.2-1B.\u300d\u3092\u4e3b\u5f35<\/li>\n\n\n\n<li>\u30ea\u30dd\u30b8\u30c8\u30ea\u306f<a href=\"https:\/\/huggingface.co\/collections\/HuggingFaceTB\/smollm2-6723884218bcda64b34d7db9\">SmolLM2 &#8211; a HuggingFaceTB Collection<\/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":[223,293,365,368],"class_list":["post-6202","post","type-post","status-publish","format-standard","hentry","category-arxiv","tag-llm","tag-oss","tag-slm","tag-small-model"],"_links":{"self":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6202","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=6202"}],"version-history":[{"count":0,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/6202\/revisions"}],"wp:attachment":[{"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=6202"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=6202"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devneko.jp\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=6202"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}