F2 - Energy-Efficient AI-Computing Systems for Large-Language Models.pdf

编号:154982 PDF 360页 23.03MB 下载积分:VIP专享
下载报告请您先登录!

F2 - Energy-Efficient AI-Computing Systems for Large-Language Models.pdf

1、ISSCC 2024Forum 2Energy-Efficient AI-Computing Systems for Large-Language ModelsISSCC 2024 Forum 2Energy-Efficient AI-Computing Systems for Large-Language ModelsEric Karl,Intel Corporation,Hillsboro,Oregon,USAJun-Seok Park,Samsung Electronics,Gyeonggi-do,KoreaJae-sun Seo,Cornell Tech,New York,New Yo

2、rk,USAYongpan Liu,Tsinghua University,Beijing,ChinaVivek De,Intel Corporation,Hillsboro,Oregon,USAAlicia Klinefelter,NVIDIA,Durham,North Carolina,USAISSCC 2024-Forum 2:Introduction 2024 IEEE International Solid-State Circuits ConferenceDuring the Forum:8 Talks1 Breakfast,2 Coffee Breaks&1 Lunch Brea

3、k45 minute talk followed by 5 minutes Q&A for each presenterPlease silence your mobile devicesPlease complete your speaker evaluations.Help us improve ISSCC.After the Forum:Forum presentations will be available as downloadable PDFs and streaming videos(not downloadable)to the registered forum attend

4、ees only.The material will not appear in IEEE Xplore or the conference digest.Forum BasicsISSCC 2024-Forum 2:Introduction 2024 IEEE International Solid-State Circuits ConferenceIncredible demand and interest in the varied capabilities of LLMs&Generative AI applicationsChatGPT 100M users in just 2 mo

5、nthsInstagram 2.5 years to 100MTikTok 9 months to 100MOn the Rise of Generative AI Applicationshttps:/ Electrician,from the series PSEUDOMNESIA,2022.Credit:Boris Eldagsen/Co-created with DALLE2/Courtesy of Photo Edition BerlinISSCC 2024-Forum 2:Introduction 2024 IEEE International Solid-State Circui

6、ts ConferenceAgendaStartTitleSpeakerAffiliation8:00 AMBreakfast8:15 AMForum 2 IntroductionEric KarlIntel8:25 AMA Brief History of Large Language Models and a glimpse into the futureLarry HeckGeorgia Tech9:15 AMLLM Training and Inference on GPU&HPC SystemsMohammad ShoeybiNVIDIA10:05 AMBreak10:20 AMCl

友情提示

1、下载报告失败解决办法
2、PDF文件下载后,可能会被浏览器默认打开,此种情况可以点击浏览器菜单,保存网页到桌面,就可以正常下载了。
3、本站不支持迅雷下载,请使用电脑自带的IE浏览器,或者360浏览器、谷歌浏览器下载即可。
4、本站报告下载后的文档和图纸-无水印,预览文档经过压缩,下载后原文更清晰。

本文(F2 - Energy-Efficient AI-Computing Systems for Large-Language Models.pdf)为本站 (2200) 主动上传,三个皮匠报告文库仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对上载内容本身不做任何修改或编辑。 若此文所含内容侵犯了您的版权或隐私,请立即通知三个皮匠报告文库(点击联系客服),我们立即给予删除!

温馨提示:如果因为网速或其他原因下载失败请重新下载,重复下载不扣分。
客服
商务合作
小程序
服务号
折叠