최신 글
4,471개정렬
7:21ENTesting Orca Mini 13B, an Unofficial Model Inspired by OrcaMatthew Berman · Forward FutureMatthew Berman loads a quantized Orca Mini 13B and tries coding, reasoning, arithmetic, and summarization prompts. His 2023 informal test produces several failures and a few successes.
13:50ENMPT-30B: Running MosaicML’s Model Locally and Testing Its Chat VariantMatthew Berman · Forward FutureMatthew Berman reviews MosaicML’s June 2023 release, demonstrates a local setup, and puts the hosted chat model through an informal prompt rubric.
1:07:41ENRob Go on Raising Venture Funds and Building LP Relationships게스트 Rob Go · 진행자 Harry StebbingsThe NextView co-founder explains fund construction, anchor investors, fundraising follow-up and the trade-offs in seed investing.
39:37ENNobody Plans Dinner in Keywords진행자 Sarah Guo, Elad GilInstacart CEO Fidji Simo tells No Priors why generative AI finally lets grocery software answer the questions people actually ask, and why 80,000 existing stores still beat a dark warehouse.
8:17ENUsing OpenLLM to Serve Open Models Through an APIMatthew Berman · Forward FutureMatthew Berman demonstrates a local model server, its interactive API and a Python client in a 2023 walkthrough.
1:09:33ENGina Gotthilf on Duolingo Growth, Experiments and Public Relations게스트 Gina Gotthilf · 진행자 Harry StebbingsThe Latitud co-founder shares lessons on retention, testing, pitching journalists and turning coverage into distribution.
6:49ENReddit’s API Changes, AI Training Data and the Open InternetMatthew Berman · Forward FutureMatthew Berman argues that the 2023 Reddit API dispute exposed tensions between AI data demand, third-party developers and the people who create online content.
3:11:35ENSearch was always a hack: Marc Andreessen on AI, truth, and the next internet게스트 Marc Andreessen · 진행자 Lex FridmanLex Fridman sits down with the Mosaic co-creator and Andreessen Horowitz founder to ask whether Google Search, the browser, and open-source AI survive the decade.
1:18:18ENFifteen Thousand Brain Numbers Were Enough to Reconstruct What Someone Saw진행자 Nathan LabenzTanishq Mathew Abraham walks through MindEye, the fMRI-to-image system that maps a handful of voxel values onto CLIP and diffusion models.
4:36ENTesting WizardCoder 15B as a Local Coding AssistantMatthew Berman · Forward FutureMatthew Berman runs the 2023 WizardCoder model locally, testing small Python exercises and a longer Snake game.
9:46ENInstalling Text Generation WebUI on Windows: A 2023 Local-Model WalkthroughMatthew Berman · Forward FutureMatthew Berman installs an early version of Text Generation WebUI, resolves setup errors and explains model downloads and interface settings. The steps and packages are historical; consult the project’s current documentation for a modern installation.
1:16:44ENAlex Lebrun on AI Assistants for Doctors, Model Limits and Building Nabla게스트 Alex Lebrun · 진행자 Harry StebbingsThe Nabla co-founder explains his approach to healthcare AI, lessons from Facebook and his 2023 views on competition and regulation.
3:15:07ENWikipedia Worked Once Writing Stopped Feeling Like Grad School게스트 Jimmy Wales · 진행자 Lex FridmanJimmy Wales tells Lex Fridman how Nupedia's seven-stage review died, why Wikipedia still refuses ads and governments, and what large language models should actually do with an encyclopedia.
1:02:21ENNoah Weiss on Product-Led Growth and Scaling Slack게스트 Noah Weiss · 진행자 Harry StebbingsSlack’s product leader discusses product principles, enterprise growth, speed and the lessons from products that did not work.
59:28ENFrom $500,000 to $50,000: MosaicML Made Training Your Own Model a Priced Product진행자 Nathan LabenzJonathan Frankle and Abhi Venigalla explain who is training custom LLMs, why extraction beats chat, and how ALiBi stretches a 65k context window.
44:19ENAI Will Write Ten Times More Code. Someone Still Has to Keep It Running.진행자 Sarah Guo, Elad GilDatadog co-founder Olivier Pomel on why faster generation moves value from writing software to understanding, securing, and operating it.
10:23ENTesting Hosted Falcon 40B and Document Chat in h2oGPTMatthew Berman · Forward FutureMatthew Berman tries H2O.ai’s 2023 hosted Falcon 40B interface on coding, writing and reasoning prompts, then demonstrates document chat.
6:26ENInstalling and Trying GPT-Engineer: A 2023 Snake Game WalkthroughMatthew Berman · Forward FutureMatthew Berman demonstrates the early project-generation workflow, from a Conda environment and prompt to clarification questions and generated Python files.
23:20ENRishi Sunak on UK AI Policy, Talent and Education in 2023게스트 Rishi Sunak · 진행자 Harry StebbingsThe then prime minister discusses an AI task force, skills and visa initiatives, and the political priorities he wanted to pursue.
1:03:40ENWho Banks the AI Wave: Startups or the Giants Already in Place?진행자 Nathan LabenzStability AI’s Emad Mostaque and Slow Ventures’ Sam Lessin debate whether generative AI is a reshuffler for new companies or an extender for Meta, Google, and Hollywood.
2:19:00ENLess Impressed, More Involved: Matthew McConaughey on Messy Love and a Live Life게스트 Matthew McConaughey · 진행자 Lex FridmanThe Oscar-winning actor talks with Lex Fridman about his parents' violent romance, the death of his father, green lights, and why some wounds should be kicked off the curb.
17:32ENSam Altman’s 2023 World Tour: AI Risks, Regulation and AmbitionSam Altman, Ilya Sutskever, Wojciech Zaremba, Greg Brockman · OpenAIA historical compilation of OpenAI leaders’ views on autonomy, misinformation, biological risk, climate and jobs.
8:09ENText-to-Video in 2023: Runway, Colab and a Local Zeroscope SetupMatthew Berman · Forward FutureMatthew Berman compares a hosted Runway Gen-2 demo with a Zeroscope notebook and a Windows GPU installation, showing the short-clip limitations he encountered.
47:05ENLarry Summers on Inflation, US Debt and the Global Economy게스트 Larry Summers · 진행자 Harry StebbingsIn a 2023 conversation, the former Treasury secretary critiques the Federal Reserve, debates taxation and shares his political and economic forecasts.
더 보기4,471개 중 24개 표시









