{"id":34,"date":"2024-07-21T15:02:26","date_gmt":"2024-07-21T07:02:26","guid":{"rendered":"https:\/\/www.onetts.com\/ai\/?p=34"},"modified":"2024-07-21T16:04:30","modified_gmt":"2024-07-21T08:04:30","slug":"qwen1-5-1-8b","status":"publish","type":"post","link":"https:\/\/www.onetts.com\/ai\/models\/qwen1-5-1-8b\/","title":{"rendered":"Qwen1.5-1.8B"},"content":{"rendered":"<p>Qwen1.5-1.8B\u662f\u7531\u4e00\u7fa4\u6770\u51fa\u7684\u7814\u7a76\u4eba\u5458\u548c\u5de5\u7a0b\u5e08\u56e2\u961f\u5f00\u53d1\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u5b83\u662fQwen2\u7684beta\u7248\u672c\uff0c\u9884\u8bad\u7ec3\u5728\u5927\u91cf\u6570\u636e\u4e0a\uff0c\u65e8\u5728\u63d0\u4f9b\u66f4\u9ad8\u6548\u3001\u66f4\u667a\u80fd\u7684\u8bed\u8a00\u7406\u89e3\u4e0e\u751f\u6210\u80fd\u529b\u3002\u8be5\u6a21\u578b\u7684\u547d\u540d\u6765\u6e90\u4e8e\u5176\u53c2\u6570\u89c4\u6a21\uff0c1.8B\u8868\u793a\u6a21\u578b\u62e5\u670918\u4ebf\u4e2a\u53c2\u6570\u3002<\/p>\n<h2>\u6a21\u578b\u8bc4\u6d4b<\/h2>\n<h3>Transformer\u67b6\u6784<\/h3>\n<p>Qwen1.5-1.8B\u57fa\u4e8eTransformer\u67b6\u6784\uff0c\u8fd9\u662f\u4e00\u79cd\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u6700\u521d\u7531Vaswani\u7b49\u4eba\u57282017\u5e74\u63d0\u51fa\uff0c\u7528\u4e8e\u673a\u5668\u7ffb\u8bd1\u4efb\u52a1\u3002Transformer\u67b6\u6784\u4ee5\u5176\u81ea\u6ce8\u610f\u529b\u673a\u5236\uff08Self-Attention\uff09\u800c\u95fb\u540d\uff0c\u80fd\u591f\u6355\u6349\u8f93\u5165\u5e8f\u5217\u4e2d\u4e0d\u540c\u4f4d\u7f6e\u4e4b\u95f4\u7684\u4f9d\u8d56\u5173\u7cfb\uff0c\u65e0\u8bba\u8fd9\u4e9b\u4f4d\u7f6e\u4e4b\u95f4\u7684\u8ddd\u79bb\u6709\u591a\u8fdc\u3002<\/p>\n<h3>SwiGLU\u6fc0\u6d3b\u51fd\u6570<\/h3>\n<p>Qwen1.5-1.8B\u91c7\u7528\u4e86SwiGLU\u6fc0\u6d3b\u51fd\u6570\uff0c\u8fd9\u662f\u4e00\u79cd\u65b0\u9896\u7684\u975e\u5355\u8c03\u6fc0\u6d3b\u51fd\u6570\uff0c\u80fd\u591f\u5728\u4fdd\u6301\u6a21\u578b\u6027\u80fd\u7684\u540c\u65f6\uff0c\u51cf\u5c11\u6a21\u578b\u7684\u8ba1\u7b97\u590d\u6742\u5ea6\u3002<\/p>\n<h3>\u591a\u8bed\u8a00\u652f\u6301<\/h3>\n<p>Qwen1.5-1.8B\u7684\u4e00\u4e2a\u663e\u8457\u7279\u70b9\u662f\u5176\u5bf9\u591a\u8bed\u8a00\u7684\u652f\u6301\u3002\u8fd9\u5f97\u76ca\u4e8e\u5176\u6539\u8fdb\u7684\u5206\u8bcd\u5668\uff0c\u80fd\u591f\u9002\u5e94\u591a\u79cd\u81ea\u7136\u8bed\u8a00\u548c\u4ee3\u7801\uff0c\u4f7f\u5f97\u6a21\u578b\u80fd\u591f\u7406\u89e3\u548c\u751f\u6210\u4e0d\u540c\u8bed\u8a00\u7684\u6587\u672c\u3002<\/p>\n<h3>\u6a21\u578b\u5927\u5c0f\u4e0e\u6027\u80fd<\/h3>\n<p>Qwen1.5-1.8B\u63d0\u4f9b\u4e868\u79cd\u4e0d\u540c\u5927\u5c0f\u7684\u6a21\u578b\uff0c\u4ece0.5B\u523072B\u4e0d\u7b49\uff0c\u4ee5\u6ee1\u8db3\u4e0d\u540c\u5e94\u7528\u573a\u666f\u7684\u9700\u6c42\u3002\u6bcf\u79cd\u6a21\u578b\u90fd\u7ecf\u8fc7\u7cbe\u5fc3\u8bbe\u8ba1\uff0c\u4ee5\u5b9e\u73b0\u6700\u4f73\u7684\u6027\u80fd\u548c\u6548\u7387\u5e73\u8861\u3002<\/p>\n<h3>\u6027\u80fd\u63d0\u5347<\/h3>\n<h3>\u5bf9\u8bdd\u6a21\u578b\u7684\u4f18\u5316<\/h3>\n<p>Qwen1.5-1.8B\u5728\u5bf9\u8bdd\u6a21\u578b\u65b9\u9762\u8fdb\u884c\u4e86\u663e\u8457\u7684\u6027\u80fd\u63d0\u5347\u3002\u8fd9\u4f7f\u5f97\u6a21\u578b\u5728\u5904\u7406\u5bf9\u8bdd\u548c\u4ea4\u4e92\u5f0f\u4efb\u52a1\u65f6\u66f4\u52a0\u81ea\u7136\u548c\u6d41\u7545\u3002<\/p>\n<h3>\u4e0a\u4e0b\u6587\u957f\u5ea6\u7684\u652f\u6301<\/h3>\n<p>Qwen1.5-1.8B\u80fd\u591f\u7a33\u5b9a\u652f\u6301\u957f\u8fbe32K\u7684\u4e0a\u4e0b\u6587\u957f\u5ea6\uff0c\u8fd9\u4f7f\u5f97\u6a21\u578b\u5728\u5904\u7406\u957f\u6587\u672c\u65f6\u66f4\u52a0\u5f97\u5fc3\u5e94\u624b\u3002<\/p>\n<h3>\u5e94\u7528\u573a\u666f<\/h3>\n<h3>\u81ea\u52a8\u6587\u672c\u751f\u6210<\/h3>\n<p>Qwen1.5-1.8B\u53ef\u4ee5\u5e94\u7528\u4e8e\u81ea\u52a8\u6587\u672c\u751f\u6210\uff0c\u5305\u62ec\u65b0\u95fb\u64b0\u5199\u3001\u6545\u4e8b\u521b\u4f5c\u3001\u6280\u672f\u6587\u6863\u7f16\u5199\u7b49\u3002<\/p>\n<h3>\u804a\u5929\u673a\u5668\u4eba<\/h3>\n<p>\u5728\u804a\u5929\u673a\u5668\u4eba\u9886\u57df\uff0cQwen1.5-1.8B\u80fd\u591f\u63d0\u4f9b\u66f4\u52a0\u4eba\u6027\u5316\u548c\u81ea\u7136\u7684\u5bf9\u8bdd\u4f53\u9a8c\u3002<\/p>\n<h3>\u673a\u5668\u7ffb\u8bd1<\/h3>\n<p>Qwen1.5-1.8B\u7684\u591a\u8bed\u8a00\u652f\u6301\u4f7f\u5176\u6210\u4e3a\u673a\u5668\u7ffb\u8bd1\u4efb\u52a1\u7684\u7406\u60f3\u9009\u62e9\u3002<\/p>\n<h3>\u4ee3\u7801\u751f\u6210\u4e0e\u7406\u89e3<\/h3>\n<p>Qwen1.5-1.8B\u8fd8\u80fd\u591f\u7406\u89e3\u548c\u751f\u6210\u4ee3\u7801\uff0c\u4e3a\u7f16\u7a0b\u8f85\u52a9\u5de5\u5177\u548c\u6559\u80b2\u5e73\u53f0\u63d0\u4f9b\u652f\u6301\u3002<\/p>\n<h2>\u90e8\u7f72\u4f7f\u7528<\/h2>\n<p>\u867d\u7136Qwen1.5-1.8B\u662f\u4e00\u4e2a\u5f3a\u5927\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u4f46\u5f00\u53d1\u8005\u5efa\u8bae\u4e0d\u8981\u76f4\u63a5\u4f7f\u7528\u57fa\u7840\u8bed\u8a00\u6a21\u578b\u8fdb\u884c\u6587\u672c\u751f\u6210\u3002\u76f8\u53cd\uff0c\u5efa\u8bae\u5728\u8be5\u6a21\u578b\u4e0a\u5e94\u7528\u540e\u8bad\u7ec3\u6280\u672f\uff0c\u5982\u76d1\u7763\u5f0f\u5fae\u8c03\uff08SFT\uff09\u3001\u57fa\u4e8e\u4eba\u7c7b\u53cd\u9988\u7684\u5f3a\u5316\u5b66\u4e60\uff08RLHF\uff09\u6216\u7ee7\u7eed\u9884\u8bad\u7ec3\u7b49\uff0c\u4ee5\u8fdb\u4e00\u6b65\u63d0\u5347\u6a21\u578b\u7684\u6027\u80fd\u3002<\/p>\n<h3>\u5b89\u88c5\u4e0e\u914d\u7f6e<\/h3>\n<p>\u4e3a\u4e86\u4f7f\u7528Qwen1.5-1.8B\uff0c\u5efa\u8bae\u5b89\u88c5\u6700\u65b0\u7248\u672c\u7684Hugging Face transformers\u5e93\uff0c\u5373<code class=\"inlineCode___uUrJ9\">transformers&gt;=4.37.0<\/code>\u3002\u8fd9\u6837\u53ef\u4ee5\u786e\u4fdd\u517c\u5bb9\u6027\u5e76\u907f\u514d\u6f5c\u5728\u7684\u9519\u8bef\u3002<\/p>\n<h3>\u5f15\u7528\u4e0e\u8d21\u732e<\/h3>\n<p>\u5982\u679c\u4f60\u5728\u7814\u7a76\u6216\u5f00\u53d1\u4e2d\u4f7f\u7528\u4e86Qwen1.5-1.8B\uff0c\u5e76\u8ba4\u4e3a\u5b83\u5bf9\u4f60\u7684\u5de5\u4f5c\u6709\u6240\u5e2e\u52a9\uff0c\u8bf7\u5728\u76f8\u5173\u6587\u732e\u4e2d\u5f15\u7528Qwen\u6280\u672f\u62a5\u544a\u3002\u8fd9\u4e0d\u4ec5\u662f\u5bf9\u5f00\u53d1\u8005\u56e2\u961f\u7684\u8ba4\u53ef\uff0c\u4e5f\u662f\u5bf9\u5f00\u6e90\u793e\u533a\u7684\u8d21\u732e\u3002<\/p>\n<h3>\u672a\u6765\u5c55\u671b<\/h3>\n<p>\u968f\u7740\u4eba\u5de5\u667a\u80fd\u6280\u672f\u7684\u4e0d\u65ad\u8fdb\u6b65\uff0cQwen1.5-1.8B\u6709\u671b\u5728\u672a\u6765\u5b9e\u73b0\u66f4\u591a\u7684\u529f\u80fd\u548c\u4f18\u5316\u3002\u6a21\u578b\u7684\u6301\u7eed\u8fed\u4ee3\u548c\u793e\u533a\u7684\u79ef\u6781\u53c2\u4e0e\u5c06\u63a8\u52a8\u5176\u5728\u66f4\u5e7f\u6cdb\u7684\u5e94\u7528\u573a\u666f\u4e2d\u53d1\u6325\u4f5c\u7528\u3002<\/p>\n<h2 id=\"section3\">\u5e38\u89c1\u95ee\u9898<\/h2>\n<p><strong>Q: Qwen1.5-1.8B\u6a21\u578b\u652f\u6301\u54ea\u4e9b\u8bed\u8a00\uff1f<\/strong><br \/>\nA: Qwen1.5-1.8B\u652f\u6301\u591a\u8bed\u8a00\uff0c\u5305\u62ec\u4f46\u4e0d\u9650\u4e8e\u82f1\u8bed\u3001\u4e2d\u6587\u3001\u897f\u73ed\u7259\u8bed\u3001\u6cd5\u8bed\u7b49\u3002\u5176\u591a\u8bed\u8a00\u652f\u6301\u4f7f\u5f97\u5b83\u80fd\u591f\u7406\u89e3\u548c\u751f\u6210\u591a\u79cd\u8bed\u8a00\u7684\u6587\u672c\u3002<\/p>\n<p><strong>Q: \u5982\u4f55\u5904\u7406\u957f\u6587\u672c\u8f93\u5165\uff1f<\/strong><br \/>\nA: Qwen1.5-1.8B\u652f\u6301 32K \u7684\u4e0a\u4e0b\u6587\u957f\u5ea6\uff0c\u53ef\u4ee5\u5904\u7406\u8f83\u957f\u7684\u6587\u672c\u8f93\u5165\u3002\u7528\u6237\u53ef\u4ee5\u901a\u8fc7\u8c03\u6574\u8f93\u5165\u6587\u672c\u7684\u957f\u5ea6\u6765\u5145\u5206\u5229\u7528\u8fd9\u4e00\u7279\u6027\u3002<\/p>\n<p><strong>Q: \u662f\u5426\u9700\u8981\u7279\u6b8a\u7684\u786c\u4ef6\u914d\u7f6e\uff1f<\/strong><br \/>\nA: Qwen1.5-1.8B \u53ef\u4ee5\u5728\u666e\u901a\u786c\u4ef6\u4e0a\u8fd0\u884c\uff0c\u4f46\u4e3a\u4e86\u83b7\u5f97\u66f4\u597d\u7684\u6027\u80fd\uff0c\u5efa\u8bae\u4f7f\u7528\u5177\u6709\u8f83\u9ad8\u8ba1\u7b97\u80fd\u529b\u7684\u786c\u4ef6\uff0c\u5982 GPU\u3002<\/p>\n<p><strong>Q: \u5982\u4f55\u5728\u804a\u5929\u673a\u5668\u4eba\u4e2d\u4f7f\u7528 Qwen1.5-1.8B\uff1f<\/strong><br \/>\nA: \u53ef\u4ee5\u901a\u8fc7\u52a0\u8f7d Qwen1.5-1.8B\u7684\u804a\u5929\u6a21\u578b\uff0c\u5e76\u5c06\u5176\u96c6\u6210\u5230\u804a\u5929\u673a\u5668\u4eba\u7684\u5bf9\u8bdd\u7ba1\u7406\u7cfb\u7edf\u4e2d\u3002\u6a21\u578b\u53ef\u4ee5\u5904\u7406\u7528\u6237\u7684\u8f93\u5165\u5e76\u751f\u6210\u76f8\u5e94\u7684\u56de\u590d\u3002<\/p>\n<p><strong>Q: \u5982\u4f55\u8fdb\u884c\u6a21\u578b\u7684\u5fae\u8c03\uff1f<\/strong><br \/>\nA: \u7528\u6237\u53ef\u4ee5\u6839\u636e\u81ea\u5df1\u7684\u9700\u6c42\u5bf9 Qwen1.5-1.8B \u8fdb\u884c\u5fae\u8c03\uff0c\u4f8b\u5982\u901a\u8fc7\u76d1\u7763\u5b66\u4e60\u3001\u5f3a\u5316\u5b66\u4e60\u6216\u7ee7\u7eed\u9884\u8bad\u7ec3\u7b49\u65b9\u6cd5\u3002<\/p>\n<h2 id=\"section4\">\u76f8\u5173\u8d44\u6e90\u94fe\u63a5<\/h2>\n<ul>\n<li><strong>Hugging Face Transformers \u5e93<\/strong>\uff1a<a target=\"_blank\" rel=\"noopener noreferrer nofollow\" href=\"https:\/\/huggingface.co\/transformers\/\" target=\"_blank\" rel=\"noopener\">Hugging Face Transformers<\/a><\/li>\n<li><strong>Qwen \u6280\u672f\u62a5\u544a<\/strong>\uff1a<a target=\"_blank\" rel=\"noopener noreferrer nofollow\" href=\"https:\/\/arxiv.org\/abs\/2309.16609\" target=\"_blank\" rel=\"noopener\">Qwen Technical Report<\/a><\/li>\n<li><strong>GitHub \u4ed3\u5e93<\/strong>\uff1a<a target=\"_blank\" rel=\"noopener noreferrer nofollow\" href=\"https:\/\/github.com\/qwen-model\/qwen\" target=\"_blank\" rel=\"noopener\">Qwen GitHub<\/a><\/li>\n<\/ul>\n<h3>\u7ed3\u8bed<\/h3>\n<p>Qwen1.5-1.8B\u4f5c\u4e3a\u65b0\u4e00\u4ee3\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u4e0d\u4ec5\u5728\u6280\u672f\u4e0a\u5b9e\u73b0\u4e86\u7a81\u7834\uff0c\u66f4\u5728\u5e94\u7528\u5c42\u9762\u5c55\u73b0\u51fa\u5e7f\u9614\u7684\u524d\u666f\u3002\u968f\u7740\u5176\u4e0d\u65ad\u53d1\u5c55\u548c\u5b8c\u5584\uff0c\u6211\u4eec\u671f\u5f85\u5b83\u80fd\u591f\u5728\u81ea\u7136\u8bed\u8a00\u5904\u7406\u9886\u57df\u5e26\u6765\u66f4\u591a\u7684\u521b\u65b0\u548c\u4ef7\u503c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Qwen1.5-1.8B\u662f\u7531\u4e00\u7fa4\u6770\u51fa\u7684\u7814\u7a76\u4eba\u5458\u548c\u5de5\u7a0b\u5e08\u56e2\u961f\u5f00\u53d1\u7684\u8bed\u8a00\u6a21\u578b\uff0c\u5b83\u662fQwen2\u7684beta\u7248\u672c\uff0c\u9884\u8bad\u7ec3\u5728\u5927\u91cf\u6570\u636e\u4e0a\uff0c\u65e8\u5728\u63d0\u4f9b\u66f4\u9ad8\u6548\u3001\u66f4\u667a\u80fd\u7684\u8bed\u8a00\u7406\u89e3\u4e0e\u751f\u6210\u80fd\u529b\u3002\u8be5\u6a21\u578b\u7684\u547d\u540d\u6765\u6e90\u4e8e\u5176\u53c2\u6570\u89c4\u6a21\uff0c1.8B\u8868\u793a\u6a21\u578b\u62e5\u670918\u4ebf\u4e2a\u53c2\u6570\u3002 \u6a21\u578b\u8bc4\u6d4b Transformer\u67b6\u6784 Qwen1.5-1.8B\u57fa\u4e8eTransformer\u67b6\u6784\uff0c\u8fd9\u662f\u4e00\u79cd\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u6700\u521d\u7531Vaswani\u7b49\u4eba\u57282017\u5e74\u63d0\u51fa\uff0c\u7528\u4e8e\u673a\u5668\u7ffb\u8bd1\u4efb\u52a1\u3002Transformer\u67b6\u6784\u4ee5\u5176\u81ea\u6ce8\u610f\u529b\u673a\u5236\uff08Self-Attention\uff09\u800c\u95fb\u540d\uff0c\u80fd\u591f\u6355\u6349\u8f93\u5165\u5e8f\u5217\u4e2d\u4e0d\u540c\u4f4d\u7f6e\u4e4b<\/p>\n","protected":false},"author":1,"featured_media":35,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3],"tags":[],"collection":[42],"company":[6],"rank":[],"class_list":["post-34","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-llm","collection-qwen1-5","company-alibaba"],"_links":{"self":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/34","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/comments?post=34"}],"version-history":[{"count":3,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/34\/revisions"}],"predecessor-version":[{"id":38,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/34\/revisions\/38"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/media\/35"}],"wp:attachment":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/media?parent=34"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/categories?post=34"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/tags?post=34"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/collection?post=34"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/company?post=34"},{"taxonomy":"rank","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/rank?post=34"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}