{"id":176598,"date":"2023-11-23T12:35:31","date_gmt":"2023-11-23T18:35:31","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2023\/11\/1hr-talk-intro-to-large-language-models"},"modified":"2023-11-23T12:35:31","modified_gmt":"2023-11-23T18:35:31","slug":"1hr-talk-intro-to-large-language-models","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2023\/11\/1hr-talk-intro-to-large-language-models","title":{"rendered":"[1hr Talk] Intro to Large Language Models"},"content":{"rendered":"<p><\/p>\n<p><iframe style=\"display: block; margin: 0 auto; width: 100%; aspect-ratio: 4\/3; object-fit: contain;\" src=\"https:\/\/www.youtube.com\/embed\/zjkBMFhNj_g?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; encrypted-media; gyroscope;\n   picture-in-picture\" allowfullscreen><\/iframe><\/p>\n<p>This is a 1 hour general-audience introduction to Large Language Models: the core technical component behind systems like ChatGPT, Claude, and Bard. What they are, where they are headed, comparisons and analogies to present-day operating systems, and some of the security-related challenges of this new computing paradigm.<br \/> As of November 2023 (this field moves fast!).<\/p>\n<p>Context: This video is based on the slides of a talk I gave recently at the AI Security Summit. The talk was not recorded but a lot of people came to me after and told me they liked it. Seeing as I had already put in one long weekend of work to make the slides, I decided to just tune them a bit, record this round 2 of the talk and upload it here on YouTube. Pardon the random background, that\u2019s my hotel room during the thanksgiving break.<\/p>\n<p>- Slides as PDF: <a href=\"https:\/\/drive.google.com\/file\/d\/1pxx_ZI7O-Nwl7ZLNk5hI3WzAsTLwvNU7\/view?usp=share_link\">https:\/\/drive.google.com\/file\/d\/1pxx_ZI7O-Nwl7ZLNk5hI3WzAsTL\u2026share_link<\/a> (42MB)<br \/> - Slides. as Keynote: <a href=\"https:\/\/drive.google.com\/file\/d\/1FPUpFMiCkMRKPFjhi9MAhby68MHVqe8u\/view?usp=share_link\">https:\/\/drive.google.com\/file\/d\/1FPUpFMiCkMRKPFjhi9MAhby68MH\u2026share_link<\/a> (140MB) <\/p>\n<p>Chapters:<br \/> Part 1: LLMs.<br \/> 00:00:00 Intro: Large Language Model (LLM) talk.<br \/> 00:00:20 LLM Inference.<br \/> 00:04:17 LLM Training.<br \/> 00:08:58 LLM dreams.<br \/> 00:11:22 How do they work?<br \/> 00:14:14 Finetuning into an Assistant.<br \/> 00:17:52 Summary so far.<br \/> 00:21:05 Appendix: Comparisons, Labeling docs, RLHF, Synthetic data, Leaderboard.<br \/> Part 2: Future of LLMs.<br \/> 00:25:43 LLM Scaling Laws.<br \/> 00:27:43 Tool Use (Browser, Calculator, Interpreter, DALL-E)<br \/> 00:33:32 Multimodality (Vision, Audio)<br \/> 00:35:00 Thinking, System 1\/2<br \/> 00:38:02 Self-improvement, LLM AlphaGo.<br \/> 00:40:45 LLM Customization, GPTs store.<br \/> 00:42:15 LLM OS<br \/> Part 3: LLM Security.<br \/> 00:45:43 LLM Security Intro.<br \/> 00:46:14 Jailbreaks.<br \/> 00:51:30 Prompt Injection.<br \/> 00:56:23 Data poisoning.<br \/> 00:58:37 LLM Security conclusions.<br \/> End.<br \/> 00:59:23 Outro<\/p>\n","protected":false},"excerpt":{"rendered":"<p>This is a 1 hour general-audience introduction to Large Language Models: the core technical component behind systems like ChatGPT, Claude, and Bard. What they are, where they are headed, comparisons and analogies to present-day operating systems, and some of the security-related challenges of this new computing paradigm. As of November 2023 (this field moves fast!). [\u2026]<\/p>\n","protected":false},"author":709,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6,1492],"tags":[],"class_list":["post-176598","post","type-post","status-publish","format-standard","hentry","category-robotics-ai","category-security"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/176598","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\/709"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=176598"}],"version-history":[{"count":0,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/176598\/revisions"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=176598"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=176598"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=176598"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}