{"id":191220,"date":"2024-06-14T10:23:29","date_gmt":"2024-06-14T15:23:29","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2024\/06\/new-transformer-architecture-could-enable-powerful-llms-without-gpus"},"modified":"2024-06-14T10:23:29","modified_gmt":"2024-06-14T15:23:29","slug":"new-transformer-architecture-could-enable-powerful-llms-without-gpus","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2024\/06\/new-transformer-architecture-could-enable-powerful-llms-without-gpus","title":{"rendered":"New Transformer architecture could enable powerful LLMs without GPUs"},"content":{"rendered":"<p><a class=\"aligncenter blog-photo\" href=\"https:\/\/lifeboat.com\/blog.images\/new-transformer-architecture-could-enable-powerful-llms-without-gpus2.jpg\"><\/a><\/p>\n<p>The researchers compared two variants of their MatMul-free LM against the advanced Transformer++ architecture, used in Llama-2, on multiple model sizes.<\/p>\n<p>Interestingly, their scaling projections show that the MatMul-free LM is more efficient in leveraging additional compute resources to improve performance in comparison to the Transformer++ architecture.<\/p>\n<p>The researchers also evaluated the quality of the models on several language tasks. The 2.7B MatMul-free LM outperformed its Transformer++ counterpart on two advanced benchmarks, ARC-Challenge and OpenbookQA, while maintaining comparable performance on the other tasks.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>The researchers compared two variants of their MatMul-free LM against the advanced Transformer++ architecture, used in Llama-2, on multiple model sizes. Interestingly, their scaling projections show that the MatMul-free LM is more efficient in leveraging additional compute resources to improve performance in comparison to the Transformer++ architecture. The researchers also evaluated the quality of the [\u2026]<\/p>\n","protected":false},"author":367,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1523],"tags":[],"class_list":["post-191220","post","type-post","status-publish","format-standard","hentry","category-computing"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/191220","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/users\/367"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=191220"}],"version-history":[{"count":0,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/191220\/revisions"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=191220"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=191220"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=191220"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}