Podcasts about unicorns

This episode is brought to you by Wodify — gym management software built for founders like you. Right now, new customers get $100 off per month, for life. Learn more in the show notes or visit wodify.com. ************************************* About Business for Unicorns Business for Unicorns helps gym owners and fitness studio operators build profitable, sustainable businesses without burning out. Founded by Mark Fisher and Michael Keeler —who built and sold the $34-million Mark Fisher Fitness —BFU provides coaching, mentorship, courses, and events for gym owners ready to grow revenue, systemize operations, and create more freedom in their lives. To learn more, check out businessforunicorns.com. Get More BFU In Your Life: Claim your FREE copy of Gym Marketing Secrets HERE Follow BFU on Instagram HERE Subscribe to MF's YouTube Channel HERE Ready to Grow Your Gym? If you're a gym owner with 30+ clients looking to add $5k-$10k/month in the next 90 days, book your FREE Brainstorm Call HERE.

To close Black History Month, I talk to a Faqs Project favorite Anthony Stokes (Tap or Die) and AJ Ampadu (Ninjak vs Roku) to the Football Epic Silent as a Non-Verbal Quarterback overcomes his obstacles to pressure his team to victory. First, the task alone to create one of your main characters totally using American sign language brings forth the challenge of linguists within your art, because now the emotions have been heightened all within this violent gridiron game. This is a Unicorn of sorts as you usually see sports in manga really than American Comics especially Football which not many creators have tackled (pun intended). This 4 issue mini-series drops on Kickstarter on March 3rd but Ashcans will be available at Blerd-Con the first week of March.Sidenote: AJ Ampadu is the first African American man to write Shadowman for Valiant ComicsWritten by Anthony StokesArt by Gabriel FonsecaColors by Iacopo CalistiLetters by LettersquidSupport this podcast at — https://redcircle.com/the-faqs-project-hosted-by-james-grandmaster-faqs-boyce/donations

Send a textIn this episode of Bookworthy, Valerie interviews Kayla E. Green, an award-winning YA fantasy author who has ventured into children's literature with her picture book 'Hadley the Hippo.' They discuss the inspiration behind the book, the importance of conversations about what children consume, both in terms of food and media, and Kayla's journey as a writer and teacher. Kayla shares her favorite books and their impact on her life, as well as what readers can expect from her upcoming projects.Hadley the Hippo is What She Eats (affiliate link) The Unicorn Writer – Faith. Fiction. Fireside Poetry.Don't forget to join in the Middle Grade March Madness. Fill out your bracket and follow the fun as you, the reader, help BookWorthy choose the #1 Middle Grade book for ages 8-12.Follow over on YouTube Takeaways Kayla's journey into writing began in elementary school.'Hadley the Hippo' is inspired by her daughter's love for food.Books can serve as conversation starters for families.It's important for children to understand what they consume mentally and emotionally.Kayla emphasizes the importance of teaching children to guard their hearts.She believes in the power of read-alouds for all ages.Kayla's teaching experience has shaped her storytelling.Different books impact us at different stages of life.Kayla's favorite books include 'The Girl Who Drank the Moon' and 'Matilda.'She is working on a young adult pirate fantasy and a sequel to 'The Goodness of Unicorns.'Chapters00:00 Introduction to Kayla E. Green and Hadley the Hippo01:49 The Inspiration Behind Hadley the Hippo04:55 The Message of Hadley and Its Importance09:14 Kayla's Writing Journey and Teaching Background14:48 Favorite Books and Their Impact20:15 What's Next for Kayla E. Green? Let's discover great books together!Follow for more:FB: @bookworthypodcastInstagram: @bookworthy_podcastYouTube: BookWorthy Podcast - YouTubetiktok: @valeriefentress

Today we sit down with Doug May, SVP of Productivity at Harness, to discuss one of the most critical yet overlooked aspects of a healthy organization: Sales Productivity. Doug has had an illustrious career at elite organizations including Datadog and Databricks, and he brings that expertise to Harness, where he has cut ramp time in half and increased per-rep contribution by 43%. We explore the "F1 engineering team" analogy of GTM support, why productivity metrics are the ultimate indicator of a company's health, and the specific questions every candidate should ask to de-risk their next career move.

In this week's episode, Martin and Steve are joined once again by Land Rover photographer, author and LRM contributor, Nick Dimbleby.The guys get straight in to how Nick came to be where he is today, following having his first book published at the tender age of just 13!There's chat about Nick's collection of Land Rovers, including Martin and Steve's personal favourite and long-term magazine project vehicle, TEW the 100-inch hybrid, and Nick imparts some information about his rare-spec Discovery 3.Nick talks about the projects he has on the go, including some very interesting classic Range Rovers, and the guys even momentarily discuss premium hatchbacks.As always, we love to get your feedback and questions – send them to editorial@lrm.co.uk and in the meantime, enjoy the episode.

Elad Gil was an early investor in 40 unicorns, including major AI companies like Perplexity. I asked him what's next for software companies now that AI can code better than humans, and what he'd invest in after AI. Elad Gil is the Founder & Investor at Gil Capital, his private investment firm. He has backed some of the most iconic technology companies of the past two decades, including Airbnb, Stripe, Coinbase, Instacart, OpenAI, and SpaceX. A former executive at Twitter and Google, Elad is known for identifying major technology waves early — from social to SaaS to AI — and helping founders build category-defining companies. Sponsored byZapier More interviews -> https://mixergy.com/moreint Rate this interview -> https://mixergy.com/rateint

"The obstacle is the way."  Unicorn companies are billion-dollar companies. They don't scale through hard work alone. They scale because through belief. A powerful mindset.  Most startups stall after traction. Not because the product fails but because alignment fractures. Founders say the same words but mean different things. Meetings drag. Decisions stall. Energy leaks. The companies that break through close the "belief gap" and move faster than fear.  Patrick Sweeney unpacks how neuroscience shapes leadership, conviction, and scale. He shares how overcoming lifelong fear reshaped his own trajectory — from near-death regret to high-performance clarity — and how founders can use shared belief maps, fast decision cycles, and hypothesis testing to eliminate debate paralysis. He explains why most teams suffer from the "illusion of agreement," how decision-making tax slows growth, and why imposter syndrome isn't a weakness — it's common among elite CEOs.  Patrick is a serial entrepreneur with three exits, bestselling author of Fear Is Fuel, and author of The Founder's Creed. His work helps founders align teams, sharpen conviction, and build companies capable of becoming unicorns.  Expert action steps:  Stay emotionally balanced. Don't overreact to success or setbacks.  Treat obstacles as lessons; adapt your response instead of resisting reality.  Focus your energy on your tribe. Not everyone is wired to like you.  Recognize imposter syndrome as normal; even elite CEOs experience it. Learn more & connect:  Pre order The Founder's Creed at https://thefounderscreed.com/.  Book: Fear Is Fuel – Patrick Sweeney https://a.co/d/0btOiKRh  LinkedIn / Instagram: @thefearguru  Visit https://www.eCircleAcademy.com and book a success call with Nicky to take your practice to the next level. 

What if the secret to standing out as a leader isn't about being smarter, more connected, or going viral - but about 12 simple habits you can learn? In this energizing conversation, William Vanderbloemen—founder of Vanderbloemen Search Group—shares why he's more bullish on the church than he's been in 30 years, what data reveals about top-performing leaders, and the one habit that naturally atrophies if you don't fight it.Discover why agility matters more than ever and how these 12 data-driven habits can transform your leadership starting today.Key Insights:02:48 - Why William Is More Bullish on the Church Than Ever 05:11 - Are We Living in the Last Days? 07:22 - To Be a Good Pastor, Smell Like the Sheep 08:47 - Your Superpower: You're the Only Pastor for Your Church 12:29 - 12 Habits, Not Traits (You Can Learn These) 16:08 - The Habit of Agility: The Only One That Atrophies 20:39 - Why Agility Matters Most for the Next 10 Years Resources Mentioned:Be the Unicorn: 12 Data-Driven Habits That Separate the Best Leaders from the Rest by William VanderbloemenVanderbloemen Leadership PodcastSimple Church by Eric Geiger and Thom RainerFollow Innovative Church Leaders:Website: https://innovativechurchleaders.org/ YouTube: https://www.youtube.com/@InnovativeChurchLeaders Facebook: https://www.facebook.com/InnovativeChurchLeaders/ Facebook Group: https://www.facebook.com/groups/innovativechurchleaders Instagram: https://www.instagram.com/innovativechurchleaders LinkedIn: https://www.linkedin.com/company/innovative-church-leaders/ TikTok: https://www.tiktok.com/@innovativechurchleadersEric Bryant: https://ericbryant.org/ William Vanderbloemen: https://www.vanderbloemen.com/Pastoral Cohort with N.T. Wright: https://innovativechurchleaders.org/cohort/Apple Podcasts: https://podcasts.apple.com/us/podcast/the-post-christian-podcast/id1509588357Spotify: https://open.spotify.com/show/6ZeQIrzr2tCMyq1VdwxGNnYou don't need to be the next viral preacher. Your people need YOU—their pastor. Get practical leadership tools weekly at innovativechurchleaders.org#ChurchLeadership #PastorLife #LeadershipDevelopment #BeTheUnicorn #SpiritLed #ChurchGrowth #PastoralCare #DataDrivenLeadership #Agility #Innovation

Since the James Harden trade, the reigning Defensive Player of the Year has seen a decline in production despite the Cavs success. He's struggled over the course of this season and the fans have started to worry if it's something to worry about moving forward, Ken and Lima have all the insight.

Learn More about Altanta Retreat and grab a seat HERE. About Business for Unicorns Business for Unicorns helps gym owners and fitness studio operators build profitable, sustainable businesses without burning out. Founded by Mark Fisher and Michael Keeler —who built and sold the $34-million Mark Fisher Fitness —BFU provides coaching, mentorship, courses, and events for gym owners ready to grow revenue, systemize operations, and create more freedom in their lives. To learn more, check out businessforunicorns.com. Get More BFU In Your Life: Claim your FREE copy of Gym Marketing Secrets HERE Follow BFU on Instagram HERE Subscribe to MF's YouTube Channel HERE Ready to Grow Your Gym? If you're a gym owner with 30+ clients looking to add $5k-$10k/month in the next 90 days, book your FREE Brainstorm Call HERE.

Russ was running a moderately successful live streaming startup. Then he got a terrifying offer from a tech giant: sell to us for cheap, or we'll crush you. He had no leverage. He was about to fold.Then he got an email from OpenAI. They had secretly built ChatGPT's voice mode on his infrastructure. Overnight, everything changed. In this episode, Russ reveals the wild story of how LiveKit became the backbone of multimodal AI, why he almost sold his previous company for parts, and how to survive when the biggest companies in the world are breathing down your neck.Why You Should ListenHow to secretly power ChatGPT's voice mode.Why you should build "boring" infrastructure instead of AI apps.How to negotiate an acquihire when you have no leverage.Why a "sell or die" threat from a tech giant was the best thing to happen.How to pivot from a failed consumer app to a unicorn infrastructure play.Keywordsstartup podcast, startup podcast for founders, product market fit, AI infrastructure, multimodal AI, OpenAI, ChatGPT voice mode, founder stories, pivot, LiveKit00:00:00 Intro00:02:49 The OG YC Batch Experience00:07:08 How to Sell a Failing Startup00:15:51 The "Good Cop, Bad Cop" Investor Negotiation00:35:56 The First Voice AI Demo That Flopped00:38:29 The Secret Email from OpenAI00:43:47 How to Scale Stateful Voice AgentsSend me a message to let me know what you think!

Stephen Grootes speaks to Jeni-Anne Campbell about her new book, Feeding Unicorns: How good leaders make great teams, and the shifting paradigm of empathetic leadership in the South African business landscape. As traditional "hustle culture" faces growing scrutiny, more executives and entrepreneurs are looking for ways to foster high-performing cultures without sacrificing the humanity of their staff. Her book offers a refreshingly honest guide to "feeding" the rare, talented "unicorns" within an organisation through fair compensation, radical clarity, and emotional intelligence. The Money Show is a podcast hosted by well-known journalist and radio presenter, Stephen Grootes. He explores the latest economic trends, business developments, investment opportunities, and personal finance strategies. Each episode features engaging conversations with top newsmakers, industry experts, financial advisors, entrepreneurs, and politicians, offering you thought-provoking insights to navigate the ever-changing financial landscape.    Thank you for listening to a podcast from The Money Show Listen live Primedia+ weekdays from 18:00 and 20:00 (SA Time) to The Money Show with Stephen Grootes broadcast on 702 https://buff.ly/gk3y0Kj and CapeTalk https://buff.ly/NnFM3Nk For more from the show, go to https://buff.ly/7QpH0jY or find all the catch-up podcasts here https://buff.ly/PlhvUVe Subscribe to The Money Show Daily Newsletter and the Weekly Business Wrap here https://buff.ly/v5mfetc The Money Show is brought to you by Absa     Follow us on social media   702 on Facebook: https://www.facebook.com/TalkRadio702 702 on TikTok: https://www.tiktok.com/@talkradio702 702 on Instagram: https://www.instagram.com/talkradio702/ 702 on X: https://x.com/CapeTalk 702 on YouTube: https://www.youtube.com/@radio702   CapeTalk on Facebook: https://www.facebook.com/CapeTalk CapeTalk on TikTok: https://www.tiktok.com/@capetalk CapeTalk on Instagram: https://www.instagram.com/ CapeTalk on X: https://x.com/Radio702 CapeTalk on YouTube: https://www.youtube.com/@CapeTalk567 See omnystudio.com/listener for privacy information.

In this pilot episode of the "Tax Strategy Hot Seat," Dr. Jackie Meyer, CPA introduces a new series aimed at demystifying tax strategies for both tax advisors and taxpayers. The episode focuses on the Roth IRA, a popular retirement account that allows for tax-free growth and withdrawals under certain conditions. Jackie and her guest, Jason Ackerman, CPA, CFP®, CGMA, CEO of Wealth Rabbit, delve into common misconceptions surrounding Roth IRAs, including the complexities of backdoor Roth contributions and the importance of understanding tax implications during tax season. They emphasize the need for proper documentation and the potential pitfalls of mismanaging Roth accounts, especially for high-income earners.

In this episode host Iain Ballantyne talks to guest Steve Kershaw, a former Royal Navy submarine officer, about his amazing career under the sea and his work today helping to bring about the UK'S ‘Hybrid Navy' transformation.Iain kicks off by asking Steve to explain what led him to choose a naval career in the mid-1980s and why it was the engineering side of the Senior Service that appealed most.After talking about his time undergoing Initial Sea Training and being ‘streamed' to the Submarine Service as an engineer, Steve relates how for a short period he returned to the Surface Fleet. He spent time in HMS London, including a foray to Berlin as the infamous Wall dividing East and West came down in late 1989.Steve served in several Royal Navy nuclear-powered hunter killer submarines (SSNs) of the Trafalgar Class, including during the 1991 Gulf War. That boat spent 13 weeks dived in the Mediterranean watching Libya to ensure it did not come into the conflict on the side of Saddam Hussein.Steve also went to sea in the Upholder Class diesel-electric submarine HMS Unicorn for a marathon voyage from the UK to the Indian Ocean and Gulf and back. He reveals to Iain how he found the ‘dirty boat' world aboard Unicorn to be somewhat different to the nukes.While away the UK Government decided to take the four (fairly new) Upholders out of commission, which was a blow. Steve reveals the impact that had on Unicorn's crew. A deployment involving Steve, which hit the headlines for the wrong reasons was that of HMS Tireless as part of Naval Task Group 2000, and which saw a circumnavigation of the world cancelled. The SSN was ‘trapped' in Gibraltar for a year due to serious technical problems and Steve returned home rather than going around the world.Among other things Iain and Steve discuss is his time with Naval Sea Trials Party 30 (NSTP 30) and its work to ensure RN submarine sensors remained on the cutting edge during a continuing contest under the sea.Steve and Iain also discuss how the ‘Hybrid Navy' aims to provide a solution to giving the British fleet of today and tomorrow more mass and presence at sea as part of the new Atlantic Bastion concept.*For more on navies and their activities worldwide, get the magazine! Web site http://bit.ly/wifrmag Also, follow it on X @WarshipsIFR Facebook @WarshipsIFR and Warships IFR TV on YouTube @warshipsifrtv3668 • Steve Kershaw served 21 years in the UK submarine service and has spent the rest of his career consulting in Defence and Security. He has been at PwC for over 15 years and a partner for 11 of them. His primary role is to lead consulting teams working in the UK Ministry of Defence (MoD). He is also PwC's Global Security and Defence Sector leader, helping individual territories and multi-national accounts such as NATO to develop and utilise the best that PwC has to offer. He specialises in improving military programmes and procurements and also enterprise-wide transformation.•Iain Ballantyne is the founding and current Editor of ‘Warships IFR' magazine (first published in 1998) along with its ‘Guide to the Royal Navy' (since 2003) and ‘Guide to the US Navy' (since 2018). Iain is also author of the books ‘Hunter Killers' (Orion) and ‘The Deadly Trade' (Weidenfeld & Nicolson), both about submarine warfare, plus ‘Arnhem: Ten Days in The Cauldron' and ‘Bismarck: 24 Hours to Doom' (both published by Canelo). In 2017 Iain was awarded a Fellowship by the British Maritime Foundation, which promotes awareness of the United Kingdom's dependence on the sea and seafarers. Visit his web site Bismarckbattle.com and follow him on X @IBallantyn

Rabbi Vann

In this episode of Software People Stories, VR Govindarajan, aka Govi, the co-founder and executive chairman of Perfios Software, shares his comprehensive journey from the early days of his career to building successful startups. Govi dives deep into his academic background, industry experience, and the various startups he co-founded, including Aztec Soft and Perfios. He emphasizes the importance of technology-focused innovation, building a product company out of India, and navigating the challenges of operating in a regulated industry. Govi also shares his insights on the value of teamwork, culture, and maintaining a larger purpose beyond just making money. Throughout the conversation, he provides valuable lessons for aspiring entrepreneurs on raising funds, managing teams, and scaling businesses globally. 00:00 Introduction and Welcome00:35 Govi's Background and Career Journey01:43 Early Startups and Challenges02:48 Building Aztec Soft and Going Public03:10 Transition to Perfios and Product Focus06:18 Navigating the Dot-Com Bust08:18 Acquisitions and Selling to MindTree13:41 Importance of Team Effort17:34 Choosing the BFSI Sector20:39 Challenges in a Regulated Industry22:29 Managing B2B Sales and Integration24:26 Focusing on Product Development24:46 Creating a New Market Category25:06 Challenges of Evangelizing a New Product26:21 Leveraging Global Trends28:07 Expanding to International Markets34:25 Managing Code and Customizations36:54 Importance of Local Customer Support38:11 Leveraging AI and Data41:22 Advice for Aspiring Entrepreneurs44:51 The Importance of Culture and Values48:53 Conclusion and Final ThoughtsThe timestamps are approximate and do not include the time for the intro. Add about 90 seconds to locate the sectionAs the Co-founder & Executive Chairman, Govi has been the driving force behind building Perfios into the Operating system for the BFSI in India and across the globe. Perfios is an acknowledged technology driven SaaS Pioneer that works across almost all financial institutions (Banking and Insurance Sector) both in India and across 18 countries in South East Asia, Middle East and North Africa. Perfios is a Unicorn and is backed by some of the best Global Investors. As a fintech Pioneer, Govi has been part of many committees in industry forums such as FICCI, CII and ASSOCHAM. He has been an active speaker at many BFSI related conferences and public institutions.He brings over 35 years of rich experience in the IT industry across the US and India. Before co-founding Perfios, Govi was the Co-founder, CTO, and Board Member at Aztecsoft, a pioneering force in the offshore product development space. From being a startup, Aztecsoft got listed in the Indian markets and was run as a public company before being sold to another Public company. Prior to starting Aztecsoft, he played a key role in advancing database technologies at global technology leaders such as Digital Equipment Corporation (DEC) and IBM.Govi holds an M.S. in Computer Science from the University of Massachusetts and a B.E. in Electrical and Electronics Engineering from the Indian Institute of Science, Bangalore. He may be reached at: govi@perfios.com

What magical powers do Unicorns have? What is a Unicorn's horn called? Could Unicorns be real? Have you started your FREE TRIAL of Who Smarted?+ for AD FREE listening, an EXTRA episode every week & bonus content? Sign up right in the Apple app, or directly at WhoSmarted.com and find out why more than 1,000 families are LOVING their subscription! Get official Who Smarted? Merch: tee-shirts, mugs, hoodies and more, at Who Smarted?

Learn More about Altanta Retreat and grab a seat HERE. About Business for Unicorns Business for Unicorns helps gym owners and fitness studio operators build profitable, sustainable businesses without burning out. Founded by Mark Fisher and Michael Keeler —who built and sold the $34-million Mark Fisher Fitness —BFU provides coaching, mentorship, courses, and events for gym owners ready to grow revenue, systemize operations, and create more freedom in their lives. To learn more, check out businessforunicorns.com. Get More BFU In Your Life: Claim your FREE copy of Gym Marketing Secrets HERE Follow BFU on Instagram HERE Subscribe to MF's YouTube Channel HERE Ready to Grow Your Gym? If you're a gym owner with 30+ clients looking to add $5k-$10k/month in the next 90 days, book your FREE Brainstorm Call HERE.

In this week's Parshas Teruma shiur, we journey into one of the most mysterious elements of the Mishkan: the tachash.While some academic opinions suggest that the tachash was a type of dolphin or sealskin, Chazal (Gemara Shabbos 28) reveal something far more wondrous—an extraordinary, multi-colored creature with a single horn. A creation that existed just for that moment in Jewish history… a true “unicorn.”But beyond the curiosity of what the tachash was, lies a profound message.The Mishkan was built from the contributions of every Jew. Each material, each donation, each color represented the individuality and unique beauty every neshama brings to Klal Yisrael. The tachash, with its spectrum of colors, becomes a symbol of the unity that emerges only when every “shade” is present.In the shiur, we explore how this Torah concept of diverse colors contrasts with contemporary uses of the rainbow as a cultural symbol. From a Torah perspective, individuality and uniqueness are real, beautiful, and divinely intended—but can be misunderstood or redirected without the spiritual framework the Mishkan embodies.Topics discussed include:

The boys are bring plenty of Sense, and just enough Nonsense to another rambunctious edition. We dive into pond hockey, Southwest Airline changes, skiing tales, ninja hostory, unicorns, and more. Plus the always demanded fact check and other smiles. Salut!!

In this episode of Lab Rats to Unicorns, John Flavin speaks with Dr. Kate Adamala, a synthetic biologist and University of Minnesota professor whose work explores one of science's biggest questions: what is life — and can we build it from scratch? As a leader in the International Build-a-Cell Collaboration, Kate is helping drive global efforts to construct synthetic living systems from nonliving components while advancing research across minimal cells, origins-of-life science, and biocomputing.Kate shares how a childhood love of science fiction shaped her path into synthetic biology and explains the idea of “life but not alive,” highlighting how synthetic cells differ from traditional genetic engineering. She also discusses the shift from academic discovery to entrepreneurship through Synlife, the challenges of scaling entirely new biological platforms, and the regulatory questions ahead.The conversation explores how programmable synthetic cells could transform medicine, manufacturing, sustainability, and even space exploration — while raising profound scientific and ethical questions about humanity's growing ability to design life itself.

Send a textIn this last story celebrating the New Year of the Fire Horse; Tanya and Pete tell the story of a poor man experiencing hard times who sets out to find something to feed his family with. He meets some difficult choices along the way.Support the showYou can find our audios and books to purchase at https://imagined-worlds.net/story-shopOr join the live monthly Batt Cave storytelling club at https://ko-fi.com/tanyabatt0027/tiers Or if you are an educator for children 0 - 8 years subscribe to 'Batt on the Mat' - a monthly online, professional development storytelling and arts programme https://imagined-worlds.net/batt-on-the-matAnd finally, if you enjoyed our podcast, please share with friends and write a review. Or make a donation to support what we do here https://ko-fi.com/tanyabatt0027Kia Ora!

On this episode of Empty the Benches hosts Nick Mannella and Mackane Vogel discuss the return of Olympic hockey, the evolution of international competition, and their favorite winter Olympic events. They delve into the dominance of the U.S. women's hockey team, preview the U.S. men's team, and compare the talent levels of the U.S. and Canada. The conversation also includes predictions for upcoming matchups and some fun trivia about Olympic hockey. Chapters 00:00 The Return of Olympic Hockey 02:57 The Evolution of International Hockey 06:01 Favorite Winter Olympic Events 08:58 Innovative Ideas for Winter Olympics 12:05 Dominance of the U.S. Women's Hockey Team 17:53 Previewing the Gold Medal Game 18:45 U.S. Men's Hockey Team Performance 25:46 Challenges and Changes in the Tournament 28:51 The Tom Wilson Effect 29:20 Team Canada vs Team USA: A Hockey Powerhouse 31:41 Unicorns of Hockey: The Elite Players 33:43 The Pressure of the Maple Leaf 35:43 USA vs Canada: Rivalry and Redemption 38:38 Upcoming Matches and Predictions 45:08 Olympic Trivia and Fun Discussions Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

In this episode, we sit down with Anish Agarwal, CEO and Co-founder of Traversal, for a deep dive into the substance behind the AI noise. Anish, a former Columbia professor and researcher in causal AI, shares his unique journey from academia to founding an organization disrupting the site reliability engineering (SRE) and observability space. We explore the critical difference between "AI wrappers" and companies building genuine infrastructure, the emergence of the "Forward Deployed Engineer" in the sales pod, and how to identify technical moats in a world where models are rapidly evolving.

Blue Label's long-awaited reset is underway. In this Director's Cut, Sharenet Wealth portfolio manager Dylan Bradfield unpacks Cell C's first post-listing results, Telkom's resurgence through Openserve, and the potentially game-changing move: Blue Label selling independent power to municipalities. With dividend potential, fintech growth and energy optionality all in play, Bradfield explains why Blue Label remains his top pick.

Ryan Rishaug calls in to provide line updates from Team Canada practice. In Canadian Brewhouse headlines: Mark Stone impressing, Celebrini stellar, Tom Wilson a unicorn, has Binnington has surpassed expectations? Team USA report card, including Auston Matthews's performance, and the Tkachuk brothers chirps. Who will play for Gold?See omnystudio.com/listener for privacy information.

According to a16z, it has eyes around the world in order to spot companies as early as local funds might. Learn more about your ad choices. Visit podcastchoices.com/adchoices

Enjoy a sleepover with the Unicorn Princess in Unicornia: The Land of the Unicorns

Learn More about Altanta Retreat and grab a seat HERE. About Business for Unicorns Business for Unicorns helps gym owners and fitness studio operators build profitable, sustainable businesses without burning out. Founded by Mark Fisher and Michael Keeler —who built and sold the $34-million Mark Fisher Fitness —BFU provides coaching, mentorship, courses, and events for gym owners ready to grow revenue, systemize operations, and create more freedom in their lives. To learn more, check out businessforunicorns.com. Get More BFU In Your Life: Claim your FREE copy of Gym Marketing Secrets HERE Follow BFU on Instagram HERE Subscribe to MF's YouTube Channel HERE Ready to Grow Your Gym? If you're a gym owner with 30+ clients looking to add $5k-$10k/month in the next 90 days, book your FREE Brainstorm Call HERE.

253 | In den letzten Wochen hat sich der Fortschritt von AI deutlich beschleunigt. Was kommt als nächstes und wo lauern neue Geschäftsideen?Partner dieser Folge:ClockodoClockodo ist das Time Tracking Tool unserer Wahl. Auf ⁠https://www.clockodo.com/⁠optimisten bekommst du mit Gutschein-Code optimisten25 25% RabattMach das 1-minütige Quiz und finde eine Geschäftsidee, die zu dir passt: digitaleoptimisten.de/quiz. Mich erreichst du unter alexander@digitaleoptimisten.de.Schick uns deine Audio-Message auf speakpipe.com/digitaleoptimisten.Kapitel(00:00) Intro(01:20) Wie kann man mit dem Tech Stack noch mithalten?(09:30) OpenClaw: Die Architektur des Internets ändert sich(22:00) Coder werden zu göttern?(37:00) Post von Optimisten: Tool-Inspiration von Mirja(39:00) Hörer Matthias baut Learningcards.com(46:00) Geschäftsidee von Hörer Felix: B2B Siri(49:30) Geschäftsidee von Samuel: AI Ready - Bootcamp(54:30) Geschäftsidee von Alex: StackScoutLearningsAI-Layer verändert SoftwarearchitekturOpenClaw ist der erste AI-Agent, der Aktionen im Internet durchführt und SaaS-Anwendungen direkt bedient. Dadurch verschiebt sich die Bedienung von Software von Benutzeroberflächen hin zu einem AI-Agenten. Der Wert entsteht vor allem daraus, wie effizient der Mensch seine Aufmerksamkeit einsetzt.Bestehende Firmen müssen radikal neu denkenEtablierte B2B-SaaS-Firmen müssen AI in ihr Produktdesign integrieren und nicht nur smarter machen. SAP-Migrationen zeigen, wie IT-Abteilungen monatelang blockiert werden; AI-basierte Services könnten aus Abteilungen wie Risk Management, Advisory, Cyber Security, Front Office, Tax Services entstehen. Ohne radikale Neugestaltung riskieren sie, von AI-getriebenen Gründern überholt zu werden.AI täglich praktisch nutzenNutze AI täglich und integriere sie in den Arbeitsalltag, am besten mit Tools, die du regelmäßig nutzt; der Vorschlag: etwa 50 Euro pro Monat investieren und Tools zu deiner Startseite machen. Beginne mit einem konkreten Projekt, bei dem AI die Kernaufgabe übernimmt, um schnell messbare Ergebnisse zu erzielen. So entwickelst du eine Praxis, in der AI langfristig zum Werttreiber wird.Distribution entscheidet über ErfolgKonkrete Produktideen wie The Learning Cards zeigen, dass Distribution wichtiger ist als die reine Technologie. Eine physische Verpackung oder Box als Vertriebskanal kann helfen, komplexe Konzepte greifbar zu machen; Kooperationen mit Buchläden oder lokales Offline-Marketing werden diskutiert. Offene Frage bleibt, wie physische Distribution wirkungsvoll skaliert wird, um Adoption zu beschleunigen.KeywordsKünstliche Intelligenz im UnternehmenseinsatzAI-Agenten und AutomatisierungAI-Architektur im UnternehmenAI-Readiness BootcampStack Scout AI Stack Empfehlungwie AI-gestützte Automatisierung Geschäftsprozesse verändertAI-Agenten im Unternehmensalltag einsetzenArchitektur einer Unternehmenssoftware mit AI LayerWie etablierte Firmen auf AI-basierte Architekturen umstellenOpenClaw Claude Bot Sicherheit BedenkenMake.comAirtableSAP S4HANAPersonio

Existentialism, unicornicopias, and speed dating EM SCHULZ… Oh lawd, it's about to get pretty dark! These ‘90s kids are beside themselves with anticipation to announce their new home—the ParaPods network, founded by Em Schulz and Christine Scheifer of And That's Why We Drink podcast! First up, Em joins Kalyn for a Valentine's extravaganza. Then, they get into your regularly scheduled programming with a trip through 1970s animation culture. The Last Unicorn, from earliest origins, is a tale for those who harbor a phantom, fading magic, and are up to the daunting task of finding their place in the world. Email us at thatsprettydarkpodcast@gmail.com Give to our Patreon for extra content: patreon.com/tpdpodcast Follow us on Instagram and Facebook @thatsprettydarkpodcast Learn more about your ad choices. Visit podcastchoices.com/adchoices

Welcome back to another After Dark episode of the Iron Sights Podcast.Recorded at SHOT Show 2026, I sit down with Brian Williams — better known as the Unicorn Chief. From SWAT operator to Chief, a stint in City Hall, and back to policing, Brian's career has given him a perspective that most people in the profession never get. He's seen leadership from the street, the admin side, and the political layer in between.In this conversation, we dig into police culture, mentorship, accountability, and why standards in training actually matter. As a USPSA Grandmaster, Brian doesn't just preach performance — he lives it. If he's not in the office handling the administrative side of the job, he's on the range sharpening his edge and pushing others to do the same.Brian is direct. He doesn't sugarcoat. He's not interested in protecting feelings — he's interested in protecting the mission. Sit back and enjoy this After Dark episode with the Unicorn Chief, Brian Williams.Timestamps:00:00 Intro08:38 Brian's Career18:52 Leadership Path28:42 Public Perception40:03 Leadership Struggles48:16 City Management01:04:53 Leadership Defined01:14:07 Accountability Crisis01:22:27 Mentorship01:27:21 Training Challenges01:42:57 Competition & Standards01:55:09 Execution Mode02:02:09 Self-CareRed Dot Fitness Train Online: http://rdftrainonline.com/Online Membership (Full Access To All Programs & Virtual Coaching):https://www.reddotfitness.net/online-membershipVirtual Coaching:https://www.reddotfitness.net/virtual-coachingSelf-Guided Programs:https://www.reddotfitness.net/Self-Guided-Programs1Connect With Us:Website - https://ironsightspodcast.com/Instagram - https://www.instagram.com/ironsightspodcast/Facebook - https://www.facebook.com/

February 11, 2026: Your daily rundown of health and wellness news, in under 5 minutes. Today's top stories: Solace Health raises $130M Series C at $1B+ valuation, connecting patients with nurse advocates to navigate care and insurance Weight Watchers partners with Pvolve to bring streaming strength workouts into medically guided weight loss and menopause care Function Health sues Superpower over "100-plus biomarkers" claims, highlighting pressure on diagnostic platforms to define measured versus inferred data More from Fitt: Fitt Insider breaks down the convergence of fitness, wellness, and healthcare — and what it means for business, culture, and capital. Subscribe to our newsletter → insider.fitt.co/subscribe Work with our recruiting firm → https://talent.fitt.co/ Follow us on Instagram → https://www.instagram.com/fittinsider/ Follow us on LinkedIn → linkedin.com/company/fittinsider Reach out → insider@fitt.co  

Today we tackle the noise surrounding the AI movement with Paul Klein, CEO and Founder of Browserbase. With a career spanning early-stage Twilio to raising $70 million in under two years for his own infrastructure startup, Paul brings much-needed critical thinking to the "AI bubble" debate. We explore the bridge between old-world sales principles and modern, developer-first GTM strategies. Paul breaks down why Product-Led Growth (PLG) should be viewed as a pipeline engine rather than just a revenue machine and explains the power of the "Logo Flywheel" in creating executive FOMO.

In den USA sind 2025 über 230 Milliarden Dollar in junge Firmen geflossen. In Deutschland bekamen Startups nur 6,5 Milliarden Euro. Wie abgehängt ist Deutschland wirklich? „Die USA sind ganz starker Vorreiter, UK hat aufgeholt und Deutschland ist stark im Hintergrund und muss extrem viel aufholen. Das Investmentvolumen im Venture Capital 2025 ist in den USA 230 Milliarden US-Dollar gewesen. Eine riesen Summe! UK ist kleiner und hat weniger Einwohner als Deutschland. Dort haben wir eine Investitionssumme von 22 Milliarden Euro und in Deutschland sind wir nur bei 6,5 Milliarden Euro. Auch bei den Unicorns gibt es ein starkes Gefälle”, sagt Fabian Fuchs. Der Startup-Gründer und Investor - der selbst in Deutschland, Großbritannien und den USA gelebt und gearbeitet hat - weiter: „Du willst sehr gute Leute für ein Startup, aber gleichzeitig konkurrierst du mit großen Firmen wie Google oder Facebook, die sehr hohe Gehälter zahlen können. Und der Schlüssel dazu, liegt eigentlich in den Anreizen Mitarbeiter-Beteiligungen zu machen. Da sind die USA schon extrem smart. In Deutschland hängen wir komplett hinterher. Es gibt aber Überlegungen, das zu ändern. Ich habe selbst in den USA ein Unternehmen gegründet - das ging innerhalb von einer Stunden. In Deutschland dauert das mehrere Wochen. Stichwort Finanzamt, Steuernummer, IHK, Notar, Bank.” Alle Infos im Interview von Inside Wirtschaft-Chefredakteur Manuel Koch und auf https://inside-wirtschaft.de

In this episode, Anna Sjönell shares her powerful journey of transformation. After many years in the fashion industry, she chose to change direction - and through Business & Dreams Academy, she reached goals she once believed were out of reach. Anna reflects on the skills, mindset shifts, and inner growth she gained along the way - learning things she never thought she was capable of learning. With honesty and warmth, she also speaks about the importance of truly investing your time, energy, and presence into the process, and how that commitment made all the difference.Contact Anna: - Instagram- Website

Send a textHost: Kendra BeavisGuest: Sarah Dumas What if success didn't require burnout, over-performance, or constant self-sacrifice?In this episode of Tribe of Unicorns, Kendra Beavis sits down with Sarah Dumas for a grounded conversation about embodied leadership, confidence, and building wealth without exhaustion. Together, they explore what happens when women stop performing success and start embodying it — leading their businesses from self-worth, clarity, and alignment instead of pressure.Sarah shares her perspective on why so many high-achieving women feel disconnected from their success, how separating life from business creates burnout, and what it looks like to lead from a fully expressed, embodied place. This conversation isn't about doing more — it's about refining how you lead yourself so your business can expand without costing you your health, presence, or joy.This episode is for women who are done proving, pushing, and powering through — and ready to build success that actually supports their lives.In this episode, we talk about:What embodied leadership really meansWhy confidence and self-worth shape business outcomesLetting go of performance-based successAligning your mission with how you make moneyCreating impact without burnout or sacrificeLeading from power, presence, and clarityAbout Sarah DumasSarah Dumas is a mentor for multi-6-figure women who are done performing success and ready to embody it. For over 15 years, she's guided high-achieving entrepreneurs, leaders, and visionaries out of burnout and into embodied wealth, where their businesses expand because they do. Sarah specializes in helping powerful women lead themselves with the same precision they lead their businesses, creating legacy without sacrifice and wealth without exhaustion. Her work focuses on refinement, embodiment, and next-level leadership for women who refuse to choose between power and presence.Resources & Links

Parents! Download Mr Jim's app Riffio to create your own stories and songs inside your favorite shows! iOS Download | Android Listen to this podcast, audiobooks and more on Storybutton, without your kids needing to use a screened device or your phone. Listen with no fees or subscriptions.—> Order Storybutton Today

Different gym owners need different leadership books.In this episode of “Run a Profitable Gym,” Two-Brain founder Chris Cooper shares a simple plan to help you use the contents of any book to improve yourself and your business, and then he recommends specific books in each realm of leadership.Self-Leadership (to go fast, go alone):“Think Like a Monk” by Jay Shetty“The Creative Act” by Rick Rubin“Courage Is Calling” by Ryan Holiday“Dare to Lead” by Brené Brown “The War of Art” by Steven Pressfield“Drive” by Daniel PinkTeam Leadership (to go far, go together):“Be the Unicorn” by William Vanderbloemen “Good to Great” by Jim Collins“Vivid Vision” by Cameron Herold“Leadershift” by John Maxwell Peer Leadership (share the mission beyond your gym):“Influence” by Robert Cialdini “Building a StoryBrand” by Donald Miller“The Go-Giver” by Bob Burg and John David Mann “How to Win Friends and Influence People” by Dale CarnegieTribe Leadership (influence your community at scale): “The Wisdom of Joseph Campbell” by Michael Toms “The Dichotomy of Leadership” by Jocko Willink and Leif Babin “Enchantment” by Guy Kawasaki “Resilience” by Eric Greitens “Tribes” by Seth Godin Plus, from Feb. 9 to 13, 2026, only, Coop is giving away free digital copies of his 10 books for gym owners.Go to Gym Owners United (linked below), DM him with your biggest challenge, and he'll send you the right book to start with.LinksGym Owners UnitedBook a Call1:56 - Your leadership diagnostic6:31 - Self-leadership books8:44 - Team leadership books10:57 - Peer leadership books11:58 - Tribe leadership books

Less than 1% of venture-capital-backed startups will achieve a valuation in excess of $1 billion. Despite Black-led technology startups receiving less than half of 1 one percent of all venture-funding, there are SEVERAL black-led Unicorns. On this first episode of Black History Month, we highlight a few successful companies that you may not realize were founded by Black people, and dispel the notion that Black businesses need “support” – what they really need is for us to buy their stuff.

Learn More about Altanta Retreat and grab a seat HERE. About Business for Unicorns Business for Unicorns helps gym owners and fitness studio operators build profitable, sustainable businesses without burning out. Founded by Mark Fisher and Michael Keeler —who built and sold the $34-million Mark Fisher Fitness —BFU provides coaching, mentorship, courses, and events for gym owners ready to grow revenue, systemize operations, and create more freedom in their lives. To learn more, check out businessforunicorns.com. Get More BFU In Your Life: Claim your FREE copy of Gym Marketing Secrets HERE Follow BFU on Instagram HERE Subscribe to MF's YouTube Channel HERE Ready to Grow Your Gym? If you're a gym owner with 30+ clients looking to add $5k-$10k/month in the next 90 days, book your FREE Brainstorm Call HERE.

From Palantir and Two Sigma to building Goodfire into the poster-child for actionable mechanistic interpretability, Mark Bissell (Member of Technical Staff) and Myra Deng (Head of Product) are trying to turn “peeking inside the model” into a repeatable production workflow by shipping APIs, landing real enterprise deployments, and now scaling the bet with a recent $150M Series B funding round at a $1.25B valuation.In this episode, we go far beyond the usual “SAEs are cool” take. We talk about Goodfire's core bet: that the AI lifecycle is still fundamentally broken because the only reliable control we have is data and we post-train, RLHF, and fine-tune by “slurping supervision through a straw,” hoping the model picks up the right behaviors while quietly absorbing the wrong ones. Goodfire's answer is to build a bi-directional interface between humans and models: read what's happening inside, edit it surgically, and eventually use interpretability during training so customization isn't just brute-force guesswork.Mark and Myra walk through what that looks like when you stop treating interpretability like a lab demo and start treating it like infrastructure: lightweight probes that add near-zero latency, token-level safety filters that can run at inference time, and interpretability workflows that survive messy constraints (multilingual inputs, synthetic→real transfer, regulated domains, no access to sensitive data). We also get a live window into what “frontier-scale interp” means operationally (i.e. steering a trillion-parameter model in real time by targeting internal features) plus why the same tooling generalizes cleanly from language models to genomics, medical imaging, and “pixel-space” world models.We discuss:* Myra + Mark's path: Palantir (health systems, forward-deployed engineering) → Goodfire early team; Two Sigma → Head of Product, translating frontier interpretability research into a platform and real-world deployments* What “interpretability” actually means in practice: not just post-hoc poking, but a broader “science of deep learning” approach across the full AI lifecycle (data curation → post-training → internal representations → model design)* Why post-training is the first big wedge: “surgical edits” for unintended behaviors likereward hacking, sycophancy, noise learned during customization plus the dream of targeted unlearning and bias removal without wrecking capabilities* SAEs vs probes in the real world: why SAE feature spaces sometimes underperform classifiers trained on raw activations for downstream detection tasks (hallucination, harmful intent, PII), and what that implies about “clean concept spaces”* Rakuten in production: deploying interpretability-based token-level PII detection at inference time to prevent routing private data to downstream providers plus the gnarly constraints: no training on real customer PII, synthetic→real transfer, English + Japanese, and tokenization quirks* Why interp can be operationally cheaper than LLM-judge guardrails: probes are lightweight, low-latency, and don't require hosting a second large model in the loop* Real-time steering at frontier scale: a demo of steering Kimi K2 (~1T params) live and finding features via SAE pipelines, auto-labeling via LLMs, and toggling a “Gen-Z slang” feature across multiple layers without breaking tool use* Hallucinations as an internal signal: the case that models have latent uncertainty / “user-pleasing” circuitry you can detect and potentially mitigate more directly than black-box methods* Steering vs prompting: the emerging view that activation steering and in-context learning are more closely connected than people think, including work mapping between the two (even for jailbreak-style behaviors)* Interpretability for science: using the same tooling across domains (genomics, medical imaging, materials) to debug spurious correlations and extract new knowledge up to and including early biomarker discovery work with major partners* World models + “pixel-space” interpretability: why vision/video models make concepts easier to see, how that accelerates the feedback loop, and why robotics/world-model partners are especially interesting design partners* The north star: moving from “data in, weights out” to intentional model design where experts can impart goals and constraints directly, not just via reward signals and brute-force post-training—Goodfire AI* Website: https://goodfire.ai* LinkedIn: https://www.linkedin.com/company/goodfire-ai/* X: https://x.com/GoodfireAIMyra Deng* Website: https://myradeng.com/* LinkedIn: https://www.linkedin.com/in/myra-deng/* X: https://x.com/myra_dengMark Bissell* LinkedIn: https://www.linkedin.com/in/mark-bissell/* X: https://x.com/MarkMBissellFull Video EpisodeTimestamps00:00:00 Introduction00:00:05 Introduction to the Latent Space Podcast and Guests from Goodfire00:00:29 What is Goodfire? Mission and Focus on Interpretability00:01:01 Goodfire's Practical Approach to Interpretability00:01:37 Goodfire's Series B Fundraise Announcement00:02:04 Backgrounds of Mark and Myra from Goodfire00:02:51 Team Structure and Roles at Goodfire00:05:13 What is Interpretability? Definitions and Techniques00:05:30 Understanding Errors00:07:29 Post-training vs. Pre-training Interpretability Applications00:08:51 Using Interpretability to Remove Unwanted Behaviors00:10:09 Grokking, Double Descent, and Generalization in Models00:10:15 404 Not Found Explained00:12:06 Subliminal Learning and Hidden Biases in Models00:14:07 How Goodfire Chooses Research Directions and Projects00:15:00 Troubleshooting Errors00:16:04 Limitations of SAEs and Probes in Interpretability00:18:14 Rakuten Case Study: Production Deployment of Interpretability00:20:45 Conclusion00:21:12 Efficiency Benefits of Interpretability Techniques00:21:26 Live Demo: Real-Time Steering in a Trillion Parameter Model00:25:15 How Steering Features are Identified and Labeled00:26:51 Detecting and Mitigating Hallucinations Using Interpretability00:31:20 Equivalence of Activation Steering and Prompting00:34:06 Comparing Steering with Fine-Tuning and LoRA Techniques00:36:04 Model Design and the Future of Intentional AI Development00:38:09 Getting Started in Mechinterp: Resources, Programs, and Open Problems00:40:51 Industry Applications and the Rise of Mechinterp in Practice00:41:39 Interpretability for Code Models and Real-World Usage00:43:07 Making Steering Useful for More Than Stylistic Edits00:46:17 Applying Interpretability to Healthcare and Scientific Discovery00:49:15 Why Interpretability is Crucial in High-Stakes Domains like Healthcare00:52:03 Call for Design Partners Across Domains00:54:18 Interest in World Models and Visual Interpretability00:57:22 Sci-Fi Inspiration: Ted Chiang and Interpretability01:00:14 Interpretability, Safety, and Alignment Perspectives01:04:27 Weak-to-Strong Generalization and Future Alignment Challenges01:05:38 Final Thoughts and Hiring/Collaboration Opportunities at GoodfireTranscriptShawn Wang [00:00:05]: So welcome to the Latent Space pod. We're back in the studio with our special MechInterp co-host, Vibhu. Welcome. Mochi, Mochi's special co-host. And Mochi, the mechanistic interpretability doggo. We have with us Mark and Myra from Goodfire. Welcome. Thanks for having us on. Maybe we can sort of introduce Goodfire and then introduce you guys. How do you introduce Goodfire today?Myra Deng [00:00:29]: Yeah, it's a great question. So Goodfire, we like to say, is an AI research lab that focuses on using interpretability to understand, learn from, and design AI models. And we really believe that interpretability will unlock the new generation, next frontier of safe and powerful AI models. That's our description right now, and I'm excited to dive more into the work we're doing to make that happen.Shawn Wang [00:00:55]: Yeah. And there's always like the official description. Is there an understatement? Is there an unofficial one that sort of resonates more with a different audience?Mark Bissell [00:01:01]: Well, being an AI research lab that's focused on interpretability, there's obviously a lot of people have a lot that they think about when they think of interpretability. And I think we have a pretty broad definition of what that means and the types of places that can be applied. And in particular, applying it in production scenarios, in high stakes industries, and really taking it sort of from the research world into the real world. Which, you know. It's a new field, so that hasn't been done all that much. And we're excited about actually seeing that sort of put into practice.Shawn Wang [00:01:37]: Yeah, I would say it wasn't too long ago that Anthopic was like still putting out like toy models or superposition and that kind of stuff. And I wouldn't have pegged it to be this far along. When you and I talked at NeurIPS, you were talking a little bit about your production use cases and your customers. And then not to bury the lead, today we're also announcing the fundraise, your Series B. $150 million. $150 million at a 1.25B valuation. Congrats, Unicorn.Mark Bissell [00:02:02]: Thank you. Yeah, no, things move fast.Shawn Wang [00:02:04]: We were talking to you in December and already some big updates since then. Let's dive, I guess, into a bit of your backgrounds as well. Mark, you were at Palantir working on health stuff, which is really interesting because the Goodfire has some interesting like health use cases. I don't know how related they are in practice.Mark Bissell [00:02:22]: Yeah, not super related, but I don't know. It was helpful context to know what it's like. Just to work. Just to work with health systems and generally in that domain. Yeah.Shawn Wang [00:02:32]: And Mara, you were at Two Sigma, which actually I was also at Two Sigma back in the day. Wow, nice.Myra Deng [00:02:37]: Did we overlap at all?Shawn Wang [00:02:38]: No, this is when I was briefly a software engineer before I became a sort of developer relations person. And now you're head of product. What are your sort of respective roles, just to introduce people to like what all gets done in Goodfire?Mark Bissell [00:02:51]: Yeah, prior to Goodfire, I was at Palantir for about three years as a forward deployed engineer, now a hot term. Wasn't always that way. And as a technical lead on the health care team and at Goodfire, I'm a member of the technical staff. And honestly, that I think is about as specific as like as as I could describe myself because I've worked on a range of things. And, you know, it's it's a fun time to be at a team that's still reasonably small. I think when I joined one of the first like ten employees, now we're above 40, but still, it looks like there's always a mix of research and engineering and product and all of the above. That needs to get done. And I think everyone across the team is, you know, pretty, pretty switch hitter in the roles they do. So I think you've seen some of the stuff that I worked on related to image models, which was sort of like a research demo. More recently, I've been working on our scientific discovery team with some of our life sciences partners, but then also building out our core platform for more of like flexing some of the kind of MLE and developer skills as well.Shawn Wang [00:03:53]: Very generalist. And you also had like a very like a founding engineer type role.Myra Deng [00:03:58]: Yeah, yeah.Shawn Wang [00:03:59]: So I also started as I still am a member of technical staff, did a wide range of things from the very beginning, including like finding our office space and all of this, which is we both we both visited when you had that open house thing. It was really nice.Myra Deng [00:04:13]: Thank you. Thank you. Yeah. Plug to come visit our office.Shawn Wang [00:04:15]: It looked like it was like 200 people. It has room for 200 people. But you guys are like 10.Myra Deng [00:04:22]: For a while, it was very empty. But yeah, like like Mark, I spend. A lot of my time as as head of product, I think product is a bit of a weird role these days, but a lot of it is thinking about how do we take our frontier research and really apply it to the most important real world problems and how does that then translate into a platform that's repeatable or a product and working across, you know, the engineering and research teams to make that happen and also communicating to the world? Like, what is interpretability? What is it used for? What is it good for? Why is it so important? All of these things are part of my day-to-day as well.Shawn Wang [00:05:01]: I love like what is things because that's a very crisp like starting point for people like coming to a field. They all do a fun thing. Vibhu, why don't you want to try tackling what is interpretability and then they can correct us.Vibhu Sapra [00:05:13]: Okay, great. So I think like one, just to kick off, it's a very interesting role to be head of product, right? Because you guys, at least as a lab, you're more of an applied interp lab, right? Which is pretty different than just normal interp, like a lot of background research. But yeah. You guys actually ship an API to try these things. You have Ember, you have products around it, which not many do. Okay. What is interp? So basically you're trying to have an understanding of what's going on in model, like in the model, in the internal. So different approaches to do that. You can do probing, SAEs, transcoders, all this stuff. But basically you have an, you have a hypothesis. You have something that you want to learn about what's happening in a model internals. And then you're trying to solve that from there. You can do stuff like you can, you know, you can do activation mapping. You can try to do steering. There's a lot of stuff that you can do, but the key question is, you know, from input to output, we want to have a better understanding of what's happening and, you know, how can we, how can we adjust what's happening on the model internals? How'd I do?Mark Bissell [00:06:12]: That was really good. I think that was great. I think it's also a, it's kind of a minefield of a, if you ask 50 people who quote unquote work in interp, like what is interpretability, you'll probably get 50 different answers. And. Yeah. To some extent also like where, where good fire sits in the space. I think that we're an AI research company above all else. And interpretability is a, is a set of methods that we think are really useful and worth kind of specializing in, in order to accomplish the goals we want to accomplish. But I think we also sort of see some of the goals as even more broader as, as almost like the science of deep learning and just taking a not black box approach to kind of any part of the like AI development life cycle, whether that. That means using interp for like data curation while you're training your model or for understanding what happened during post-training or for the, you know, understanding activations and sort of internal representations, what is in there semantically. And then a lot of sort of exciting updates that were, you know, are sort of also part of the, the fundraise around bringing interpretability to training, which I don't think has been done all that much before. A lot of this stuff is sort of post-talk poking at models as opposed to. To actually using this to intentionally design them.Shawn Wang [00:07:29]: Is this post-training or pre-training or is that not a useful.Myra Deng [00:07:33]: Currently focused on post-training, but there's no reason the techniques wouldn't also work in pre-training.Shawn Wang [00:07:38]: Yeah. It seems like it would be more active, applicable post-training because basically I'm thinking like rollouts or like, you know, having different variations of a model that you can tweak with the, with your steering. Yeah.Myra Deng [00:07:50]: And I think in a lot of the news that you've seen in, in, on like Twitter or whatever, you've seen a lot of unintended. Side effects come out of post-training processes, you know, overly sycophantic models or models that exhibit strange reward hacking behavior. I think these are like extreme examples. There's also, you know, very, uh, mundane, more mundane, like enterprise use cases where, you know, they try to customize or post-train a model to do something and it learns some noise or it doesn't appropriately learn the target task. And a big question that we've always had is like, how do you use your understanding of what the model knows and what it's doing to actually guide the learning process?Shawn Wang [00:08:26]: Yeah, I mean, uh, you know, just to anchor this for people, uh, one of the biggest controversies of last year was 4.0 GlazeGate. I've never heard of GlazeGate. I didn't know that was what it was called. The other one, they called it that on the blog post and I was like, well, how did OpenAI call it? Like officially use that term. And I'm like, that's funny, but like, yeah, I guess it's the pitch that if they had worked a good fire, they wouldn't have avoided it. Like, you know what I'm saying?Myra Deng [00:08:51]: I think so. Yeah. Yeah.Mark Bissell [00:08:53]: I think that's certainly one of the use cases. I think. Yeah. Yeah. I think the reason why post-training is a place where this makes a lot of sense is a lot of what we're talking about is surgical edits. You know, you want to be able to have expert feedback, very surgically change how your model is doing, whether that is, you know, removing a certain behavior that it has. So, you know, one of the things that we've been looking at or is, is another like common area where you would want to make a somewhat surgical edit is some of the models that have say political bias. Like you look at Quen or, um, R1 and they have sort of like this CCP bias.Shawn Wang [00:09:27]: Is there a CCP vector?Mark Bissell [00:09:29]: Well, there's, there are certainly internal, yeah. Parts of the representation space where you can sort of see where that lives. Yeah. Um, and you want to kind of, you know, extract that piece out.Shawn Wang [00:09:40]: Well, I always say, you know, whenever you find a vector, a fun exercise is just like, make it very negative to see what the opposite of CCP is.Mark Bissell [00:09:47]: The super America, bald eagles flying everywhere. But yeah. So in general, like lots of post-training tasks where you'd want to be able to, to do that. Whether it's unlearning a certain behavior or, you know, some of the other kind of cases where this comes up is, are you familiar with like the, the grokking behavior? I mean, I know the machine learning term of grokking.Shawn Wang [00:10:09]: Yeah.Mark Bissell [00:10:09]: Sort of this like double descent idea of, of having a model that is able to learn a generalizing, a generalizing solution, as opposed to even if memorization of some task would suffice, you want it to learn the more general way of doing a thing. And so, you know, another. A way that you can think about having surgical access to a model's internals would be learn from this data, but learn in the right way. If there are many possible, you know, ways to, to do that. Can make interp solve the double descent problem?Shawn Wang [00:10:41]: Depends, I guess, on how you. Okay. So I, I, I viewed that double descent as a problem because then you're like, well, if the loss curves level out, then you're done, but maybe you're not done. Right. Right. But like, if you actually can interpret what is a generalizing or what you're doing. What is, what is still changing, even though the loss is not changing, then maybe you, you can actually not view it as a double descent problem. And actually you're just sort of translating the space in which you view loss and like, and then you have a smooth curve. Yeah.Mark Bissell [00:11:11]: I think that's certainly like the domain of, of problems that we're, that we're looking to get.Shawn Wang [00:11:15]: Yeah. To me, like double descent is like the biggest thing to like ML research where like, if you believe in scaling, then you don't need, you need to know where to scale. And. But if you believe in double descent, then you don't, you don't believe in anything where like anything levels off, like.Vibhu Sapra [00:11:30]: I mean, also tendentially there's like, okay, when you talk about the China vector, right. There's the subliminal learning work. It was from the anthropic fellows program where basically you can have hidden biases in a model. And as you distill down or, you know, as you train on distilled data, those biases always show up, even if like you explicitly try to not train on them. So, you know, it's just like another use case of. Okay. If we can interpret what's happening in post-training, you know, can we clear some of this? Can we even determine what's there? Because yeah, it's just like some worrying research that's out there that shows, you know, we really don't know what's going on.Mark Bissell [00:12:06]: That is. Yeah. I think that's the biggest sentiment that we're sort of hoping to tackle. Nobody knows what's going on. Right. Like subliminal learning is just an insane concept when you think about it. Right. Train a model on not even the logits, literally the output text of a bunch of random numbers. And now your model loves owls. And you see behaviors like that, that are just, they defy, they defy intuition. And, and there are mathematical explanations that you can get into, but. I mean.Shawn Wang [00:12:34]: It feels so early days. Objectively, there are a sequence of numbers that are more owl-like than others. There, there should be.Mark Bissell [00:12:40]: According to, according to certain models. Right. It's interesting. I think it only applies to models that were initialized from the same starting Z. Usually, yes.Shawn Wang [00:12:49]: But I mean, I think that's a, that's a cheat code because there's not enough compute. But like if you believe in like platonic representation, like probably it will transfer across different models as well. Oh, you think so?Mark Bissell [00:13:00]: I think of it more as a statistical artifact of models initialized from the same seed sort of. There's something that is like path dependent from that seed that might cause certain overlaps in the latent space and then sort of doing this distillation. Yeah. Like it pushes it towards having certain other tendencies.Vibhu Sapra [00:13:24]: Got it. I think there's like a bunch of these open-ended questions, right? Like you can't train in new stuff during the RL phase, right? RL only reorganizes weights and you can only do stuff that's somewhat there in your base model. You're not learning new stuff. You're just reordering chains and stuff. But okay. My broader question is when you guys work at an interp lab, how do you decide what to work on and what's kind of the thought process? Right. Because we can ramble for hours. Okay. I want to know this. I want to know that. But like, how do you concretely like, you know, what's the workflow? Okay. There's like approaches towards solving a problem, right? I can try prompting. I can look at chain of thought. I can train probes, SAEs. But how do you determine, you know, like, okay, is this going anywhere? Like, do we have set stuff? Just, you know, if you can help me with all that. Yeah.Myra Deng [00:14:07]: It's a really good question. I feel like we've always at the very beginning of the company thought about like, let's go and try to learn what isn't working in machine learning today. Whether that's talking to customers or talking to researchers at other labs, trying to understand both where the frontier is going and where things are really not falling apart today. And then developing a perspective on how we can push the frontier using interpretability methods. And so, you know, even our chief scientist, Tom, spends a lot of time talking to customers and trying to understand what real world problems are and then taking that back and trying to apply the current state of the art to those problems and then seeing where they fall down basically. And then using those failures or those shortcomings to understand what hills to climb when it comes to interpretability research. So like on the fundamental side, for instance, when we have done some work applying SAEs and probes, we've encountered, you know, some shortcomings in SAEs that we found a little bit surprising. And so have gone back to the drawing board and done work on that. And then, you know, we've done some work on better foundational interpreter models. And a lot of our team's research is focused on what is the next evolution beyond SAEs, for instance. And then when it comes to like control and design of models, you know, we tried steering with our first API and realized that it still fell short of black box techniques like prompting or fine tuning. And so went back to the drawing board and we're like, how do we make that not the case and how do we improve it beyond that? And one of our researchers, Ekdeep, who just joined is actually Ekdeep and Atticus are like steering experts and have spent a lot of time trying to figure out like, what is the research that enables us to actually do this in a much more powerful, robust way? So yeah, the answer is like, look at real world problems, try to translate that into a research agenda and then like hill climb on both of those at the same time.Shawn Wang [00:16:04]: Yeah. Mark has the steering CLI demo queued up, which we're going to go into in a sec. But I always want to double click on when you drop hints, like we found some problems with SAEs. Okay. What are they? You know, and then we can go into the demo. Yeah.Myra Deng [00:16:19]: I mean, I'm curious if you have more thoughts here as well, because you've done it in the healthcare domain. But I think like, for instance, when we do things like trying to detect behaviors within models that are harmful or like behaviors that a user might not want to have in their model. So hallucinations, for instance, harmful intent, PII, all of these things. We first tried using SAE probes for a lot of these tasks. So taking the feature activation space from SAEs and then training classifiers on top of that, and then seeing how well we can detect the properties that we might want to detect in model behavior. And we've seen in many cases that probes just trained on raw activations seem to perform better than SAE probes, which is a bit surprising if you think that SAEs are actually also capturing the concepts that you would want to capture cleanly and more surgically. And so that is an interesting observation. I don't think that is like, I'm not down on SAEs at all. I think there are many, many things they're useful for, but we have definitely run into cases where I think the concept space described by SAEs is not as clean and accurate as we would expect it to be for actual like real world downstream performance metrics.Mark Bissell [00:17:34]: Fair enough. Yeah. It's the blessing and the curse of unsupervised methods where you get to peek into the AI's mind. But sometimes you wish that you saw other things when you walked inside there. Although in the PII instance, I think weren't an SAE based approach actually did prove to be the most generalizable?Myra Deng [00:17:53]: It did work well in the case that we published with Rakuten. And I think a lot of the reasons it worked well was because we had a noisier data set. And so actually the blessing of unsupervised learning is that we actually got to get more meaningful, generalizable signal from SAEs when the data was noisy. But in other cases where we've had like good data sets, it hasn't been the case.Shawn Wang [00:18:14]: And just because you named Rakuten and I don't know if we'll get it another chance, like what is the overall, like what is Rakuten's usage or production usage? Yeah.Myra Deng [00:18:25]: So they are using us to essentially guardrail and inference time monitor their language model usage and their agent usage to detect things like PII so that they don't route private user information.Myra Deng [00:18:41]: And so that's, you know, going through all of their user queries every day. And that's something that we deployed with them a few months ago. And now we are actually exploring very early partnerships, not just with Rakuten, but with other people around how we can help with potentially training and customization use cases as well. Yeah.Shawn Wang [00:19:03]: And for those who don't know, like it's Rakuten is like, I think number one or number two e-commerce store in Japan. Yes. Yeah.Mark Bissell [00:19:10]: And I think that use case actually highlights a lot of like what it looks like to deploy things in practice that you don't always think about when you're doing sort of research tasks. So when you think about some of the stuff that came up there that's more complex than your idealized version of a problem, they were encountering things like synthetic to real transfer of methods. So they couldn't train probes, classifiers, things like that on actual customer data of PII. So what they had to do is use synthetic data sets. And then hope that that transfer is out of domain to real data sets. And so we can evaluate performance on the real data sets, but not train on customer PII. So that right off the bat is like a big challenge. You have multilingual requirements. So this needed to work for both English and Japanese text. Japanese text has all sorts of quirks, including tokenization behaviors that caused lots of bugs that caused us to be pulling our hair out. And then also a lot of tasks you'll see. You might make simplifying assumptions if you're sort of treating it as like the easiest version of the problem to just sort of get like general results where maybe you say you're classifying a sentence to say, does this contain PII? But the need that Rakuten had was token level classification so that you could precisely scrub out the PII. So as we learned more about the problem, you're sort of speaking about what that looks like in practice. Yeah. A lot of assumptions end up breaking. And that was just one instance where you. A problem that seems simple right off the bat ends up being more complex as you keep diving into it.Vibhu Sapra [00:20:41]: Excellent. One of the things that's also interesting with Interp is a lot of these methods are very efficient, right? So where you're just looking at a model's internals itself compared to a separate like guardrail, LLM as a judge, a separate model. One, you have to host it. Two, there's like a whole latency. So if you use like a big model, you have a second call. Some of the work around like self detection of hallucination, it's also deployed for efficiency, right? So if you have someone like Rakuten doing it in production live, you know, that's just another thing people should consider.Mark Bissell [00:21:12]: Yeah. And something like a probe is super lightweight. Yeah. It's no extra latency really. Excellent.Shawn Wang [00:21:17]: You have the steering demos lined up. So we were just kind of see what you got. I don't, I don't actually know if this is like the latest, latest or like alpha thing.Mark Bissell [00:21:26]: No, this is a pretty hacky demo from from a presentation that someone else on the team recently gave. So this will give a sense for, for technology. So you can see the steering and action. Honestly, I think the biggest thing that this highlights is that as we've been growing as a company and taking on kind of more and more ambitious versions of interpretability related problems, a lot of that comes to scaling up in various different forms. And so here you're going to see steering on a 1 trillion parameter model. This is Kimi K2. And so it's sort of fun that in addition to the research challenges, there are engineering challenges that we're now tackling. Cause for any of this to be sort of useful in production, you need to be thinking about what it looks like when you're using these methods on frontier models as opposed to sort of like toy kind of model organisms. So yeah, this was thrown together hastily, pretty fragile behind the scenes, but I think it's quite a fun demo. So screen sharing is on. So I've got two terminal sessions pulled up here. On the left is a forked version that we have of the Kimi CLI that we've got running to point at our custom hosted Kimi model. And then on the right is a set up that will allow us to steer on certain concepts. So I should be able to chat with Kimi over here. Tell it hello. This is running locally. So the CLI is running locally, but the Kimi server is running back to the office. Well, hopefully should be, um, that's too much to run on that Mac. Yeah. I think it's, uh, it takes a full, like each 100 node. I think it's like, you can. You can run it on eight GPUs, eight 100. So, so yeah, Kimi's running. We can ask it a prompt. It's got a forked version of our, uh, of the SG line code base that we've been working on. So I'm going to tell it, Hey, this SG line code base is slow. I think there's a bug. Can you try to figure it out? There's a big code base, so it'll, it'll spend some time doing this. And then on the right here, I'm going to initialize in real time. Some steering. Let's see here.Mark Bissell [00:23:33]: searching for any. Bugs. Feature ID 43205.Shawn Wang [00:23:38]: Yeah.Mark Bissell [00:23:38]: 20, 30, 40. So let me, uh, this is basically a feature that we found that inside Kimi seems to cause it to speak in Gen Z slang. And so on the left, it's still sort of thinking normally it might take, I don't know, 15 seconds for this to kick in, but then we're going to start hopefully seeing him do this code base is massive for real. So we're going to start. We're going to start seeing Kimi transition as the steering kicks in from normal Kimi to Gen Z Kimi and both in its chain of thought and its actual outputs.Mark Bissell [00:24:19]: And interestingly, you can see, you know, it's still able to call tools, uh, and stuff. It's um, it's purely sort of it's it's demeanor. And there are other features that we found for interesting things like concision. So that's more of a practical one. You can make it more concise. Um, the types of programs, uh, programming languages that uses, but yeah, as we're seeing it come in. Pretty good. Outputs.Shawn Wang [00:24:43]: Scheduler code is actually wild.Vibhu Sapra [00:24:46]: Yo, this code is actually insane, bro.Vibhu Sapra [00:24:53]: What's the process of training in SAE on this, or, you know, how do you label features? I know you guys put out a pretty cool blog post about, um, finding this like autonomous interp. Um, something. Something about how agents for interp is different than like coding agents. I don't know while this is spewing up, but how, how do we find feature 43, two Oh five. Yeah.Mark Bissell [00:25:15]: So in this case, um, we, our platform that we've been building out for a long time now supports all the sort of classic out of the box interp techniques that you might want to have like SAE training, probing things of that kind, I'd say the techniques for like vanilla SAEs are pretty well established now where. You take your model that you're interpreting, run a whole bunch of data through it, gather activations, and then yeah, pretty straightforward pipeline to train an SAE. There are a lot of different varieties. There's top KSAEs, batch top KSAEs, um, normal ReLU SAEs. And then once you have your sparse features to your point, assigning labels to them to actually understand that this is a gen Z feature, that's actually where a lot of the kind of magic happens. Yeah. And the most basic standard technique is look at all of your d input data set examples that cause this feature to fire most highly. And then you can usually pick out a pattern. So for this feature, If I've run a diverse enough data set through my model feature 43, two Oh five. Probably tends to fire on all the tokens that sounds like gen Z slang. You know, that's the, that's the time of year to be like, Oh, I'm in this, I'm in this Um, and, um, so, you know, you could have a human go through all 43,000 concepts andVibhu Sapra [00:26:34]: And I've got to ask the basic question, you know, can we get examples where it hallucinates, pass it through, see what feature activates for hallucinations? Can I just, you know, turn hallucination down?Myra Deng [00:26:51]: Oh, wow. You really predicted a project we're already working on right now, which is detecting hallucinations using interpretability techniques. And this is interesting because hallucinations is something that's very hard to detect. And it's like a kind of a hairy problem and something that black box methods really struggle with. Whereas like Gen Z, you could always train a simple classifier to detect that hallucinations is harder. But we've seen that models internally have some... Awareness of like uncertainty or some sort of like user pleasing behavior that leads to hallucinatory behavior. And so, yeah, we have a project that's trying to detect that accurately. And then also working on mitigating the hallucinatory behavior in the model itself as well.Shawn Wang [00:27:39]: Yeah, I would say most people are still at the level of like, oh, I would just turn temperature to zero and that turns off hallucination. And I'm like, well, that's a fundamental misunderstanding of how this works. Yeah.Mark Bissell [00:27:51]: Although, so part of what I like about that question is you, there are SAE based approaches that might like help you get at that. But oftentimes the beauty of SAEs and like we said, the curse is that they're unsupervised. So when you have a behavior that you deliberately would like to remove, and that's more of like a supervised task, often it is better to use something like probes and specifically target the thing that you're interested in reducing as opposed to sort of like hoping that when you fragment the latent space, one of the vectors that pops out.Vibhu Sapra [00:28:20]: And as much as we're training an autoencoder to be sparse, we're not like for sure certain that, you know, we will get something that just correlates to hallucination. You'll probably split that up into 20 other things and who knows what they'll be.Mark Bissell [00:28:36]: Of course. Right. Yeah. So there's no sort of problems with like feature splitting and feature absorption. And then there's the off target effects, right? Ideally, you would want to be very precise where if you reduce the hallucination feature, suddenly maybe your model can't write. Creatively anymore. And maybe you don't like that, but you want to still stop it from hallucinating facts and figures.Shawn Wang [00:28:55]: Good. So Vibhu has a paper to recommend there that we'll put in the show notes. But yeah, I mean, I guess just because your demo is done, any any other things that you want to highlight or any other interesting features you want to show?Mark Bissell [00:29:07]: I don't think so. Yeah. Like I said, this is a pretty small snippet. I think the main sort of point here that I think is exciting is that there's not a whole lot of inter being applied to models quite at this scale. You know, Anthropic certainly has some some. Research and yeah, other other teams as well. But it's it's nice to see these techniques, you know, being put into practice. I think not that long ago, the idea of real time steering of a trillion parameter model would have sounded.Shawn Wang [00:29:33]: Yeah. The fact that it's real time, like you started the thing and then you edited the steering vector.Vibhu Sapra [00:29:38]: I think it's it's an interesting one TBD of what the actual like production use case would be on that, like the real time editing. It's like that's the fun part of the demo, right? You can kind of see how this could be served behind an API, right? Like, yes, you're you only have so many knobs and you can just tweak it a bit more. And I don't know how it plays in. Like people haven't done that much with like, how does this work with or without prompting? Right. How does this work with fine tuning? Like, there's a whole hype of continual learning, right? So there's just so much to see. Like, is this another parameter? Like, is it like parameter? We just kind of leave it as a default. We don't use it. So I don't know. Maybe someone here wants to put out a guide on like how to use this with prompting when to do what?Mark Bissell [00:30:18]: Oh, well, I have a paper recommendation. I think you would love from Act Deep on our team, who is an amazing researcher, just can't say enough amazing things about Act Deep. But he actually has a paper that as well as some others from the team and elsewhere that go into the essentially equivalence of activation steering and in context learning and how those are from a he thinks of everything in a cognitive neuroscience Bayesian framework, but basically how you can precisely show how. Prompting in context, learning and steering exhibit similar behaviors and even like get quantitative about the like magnitude of steering you would need to do to induce a certain amount of behavior similar to certain prompting, even for things like jailbreaks and stuff. It's a really cool paper. Are you saying steering is less powerful than prompting? More like you can almost write a formula that tells you how to convert between the two of them.Myra Deng [00:31:20]: And so like formally equivalent actually in the in the limit. Right.Mark Bissell [00:31:24]: So like one case study of this is for jailbreaks there. I don't know. Have you seen the stuff where you can do like many shot jailbreaking? You like flood the context with examples of the behavior. And the topic put out that paper.Shawn Wang [00:31:38]: A lot of people were like, yeah, we've been doing this, guys.Mark Bissell [00:31:40]: Like, yeah, what's in this in context learning and activation steering equivalence paper is you can like predict the number. Number of examples that you will need to put in there in order to jailbreak the model. That's cool. By doing steering experiments and using this sort of like equivalence mapping. That's cool. That's really cool. It's very neat. Yeah.Shawn Wang [00:32:02]: I was going to say, like, you know, I can like back rationalize that this makes sense because, you know, what context is, is basically just, you know, it updates the KV cache kind of and like and then every next token inference is still like, you know, the sheer sum of everything all the way. It's plus all the context. It's up to date. And you could, I guess, theoretically steer that with you probably replace that with your steering. The only problem is steering typically is on one layer, maybe three layers like like you did. So it's like not exactly equivalent.Mark Bissell [00:32:33]: Right, right. There's sort of you need to get precise about, yeah, like how you sort of define steering and like what how you're modeling the setup. But yeah, I've got the paper pulled up here. Belief dynamics reveal the dual nature. Yeah. The title is Belief Dynamics Reveal the Dual Nature of Incompetence. And it's an exhibition of the practical context learning and activation steering. So Eric Bigelow, Dan Urgraft on the who are doing fellowships at Goodfire, Ekt Deep's the final author there.Myra Deng [00:32:59]: I think actually to your question of like, what is the production use case of steering? I think maybe if you just think like one level beyond steering as it is today. Like imagine if you could adapt your model to be, you know, an expert legal reasoner. Like in almost real time, like very quickly. efficiently using human feedback or using like your semantic understanding of what the model knows and where it knows that behavior. I think that while it's not clear what the product is at the end of the day, it's clearly very valuable. Thinking about like what's the next interface for model customization and adaptation is a really interesting problem for us. Like we have heard a lot of people actually interested in fine-tuning an RL for open weight models in production. And so people are using things like Tinker or kind of like open source libraries to do that, but it's still very difficult to get models fine-tuned and RL'd for exactly what you want them to do unless you're an expert at model training. And so that's like something we'reShawn Wang [00:34:06]: looking into. Yeah. I never thought so. Tinker from Thinking Machines famously uses rank one LoRa. Is that basically the same as steering? Like, you know, what's the comparison there?Mark Bissell [00:34:19]: Well, so in that case, you are still applying updates to the parameters, right?Shawn Wang [00:34:25]: Yeah. You're not touching a base model. You're touching an adapter. It's kind of, yeah.Mark Bissell [00:34:30]: Right. But I guess it still is like more in parameter space then. I guess it's maybe like, are you modifying the pipes or are you modifying the water flowing through the pipes to get what you're after? Yeah. Just maybe one way.Mark Bissell [00:34:44]: I like that analogy. That's my mental map of it at least, but it gets at this idea of model design and intentional design, which is something that we're, that we're very focused on. And just the fact that like, I hope that we look back at how we're currently training models and post-training models and just think what a primitive way of doing that right now. Like there's no intentionalityShawn Wang [00:35:06]: really in... It's just data, right? The only thing in control is what data we feed in.Mark Bissell [00:35:11]: So, so Dan from Goodfire likes to use this analogy of, you know, he has a couple of young kids and he talks about like, what if I could only teach my kids how to be good people by giving them cookies or like, you know, giving them a slap on the wrist if they do something wrong, like not telling them why it was wrong or like what they should have done differently or something like that. Just figure it out. Right. Exactly. So that's RL. Yeah. Right. And, and, you know, it's sample inefficient. There's, you know, what do they say? It's like slurping feedback. It's like, slurping supervision. Right. And so you'd like to get to the point where you can have experts giving feedback to their models that are, uh, internalized and, and, you know, steering is an inference time way of sort of getting that idea. But ideally you're moving to a world whereVibhu Sapra [00:36:04]: it is much more intentional design in perpetuity for these models. Okay. This is one of the questions we asked Emmanuel from Anthropic on the podcast a few months ago. Basically the question, was you're at a research lab that does model training, foundation models, and you're on an interp team. How does it tie back? Right? Like, does this, do ideas come from the pre-training team? Do they go back? Um, you know, so for those interested, you can, you can watch that. There wasn't too much of a connect there, but it's still something, you know, it's something they want toMark Bissell [00:36:33]: push for down the line. It can be useful for all of the above. Like there are certainly post-hocVibhu Sapra [00:36:39]: use cases where it doesn't need to touch that. I think the other thing a lot of people forget is this stuff isn't too computationally expensive, right? Like I would say, if you're interested in getting into research, MechInterp is one of the most approachable fields, right? A lot of this train an essay, train a probe, this stuff, like the budget for this one, there's already a lot done. There's a lot of open source work. You guys have done some too. Um, you know,Shawn Wang [00:37:04]: There's like notebooks from the Gemini team for Neil Nanda or like, this is how you do it. Just step through the notebook.Vibhu Sapra [00:37:09]: Even if you're like, not even technical with any of this, you can still make like progress. There, you can look at different activations, but, uh, if you do want to get into training, you know, training this stuff, correct me if I'm wrong is like in the thousands of dollars, not even like, it's not that high scale. And then same with like, you know, applying it, doing it for post-training or all this stuff is fairly cheap in scale of, okay. I want to get into like model training. I don't have compute for like, you know, pre-training stuff. So it's, it's a very nice field to get into. And also there's a lot of like open questions, right? Um, some of them have to go with, okay, I want a product. I want to solve this. Like there's also just a lot of open-ended stuff that people could work on. That's interesting. Right. I don't know if you guys have any calls for like, what's open questions, what's open work that you either open collaboration with, or like, you'd just like to see solved or just, you know, for people listening that want to get into McInturk because people always talk about it. What are, what are the things they should check out? Start, of course, you know, join you guys as well. I'm sure you're hiring.Myra Deng [00:38:09]: There's a paper, I think from, was it Lee, uh, Sharky? It's open problems and, uh, it's, it's a bit of interpretability, which I recommend everyone who's interested in the field. Read. I'm just like a really comprehensive overview of what are the things that experts in the field think are the most important problems to be solved. I also think to your point, it's been really, really inspiring to see, I think a lot of young people getting interested in interpretability, actually not just young people also like scientists to have been, you know, experts in physics for many years and in biology or things like this, um, transitioning into interp, because the barrier of, of what's now interp. So it's really cool to see a number to entry is, you know, in some ways low and there's a lot of information out there and ways to get started. There's this anecdote of like professors at universities saying that all of a sudden every incoming PhD student wants to study interpretability, which was not the case a few years ago. So it just goes to show how, I guess, like exciting the field is, how fast it's moving, how quick it is to get started and things like that.Mark Bissell [00:39:10]: And also just a very welcoming community. You know, there's an open source McInturk Slack channel. There are people are always posting questions and just folks in the space are always responsive if you ask things on various forums and stuff. But yeah, the open paper, open problems paper is a really good one.Myra Deng [00:39:28]: For other people who want to get started, I think, you know, MATS is a great program. What's the acronym for? Machine Learning and Alignment Theory Scholars? It's like the...Vibhu Sapra [00:39:40]: Normally summer internship style.Myra Deng [00:39:42]: Yeah, but they've been doing it year round now. And actually a lot of our full-time staff have come through that program or gone through that program. And it's great for anyone who is transitioning into interpretability. There's a couple other fellows programs. We do one as well as Anthropic. And so those are great places to get started if anyone is interested.Mark Bissell [00:40:03]: Also, I think been seen as a research field for a very long time. But I think engineering... I think engineers are sorely wanted for interpretability as well, especially at Goodfire, but elsewhere, as it does scale up.Shawn Wang [00:40:18]: I should mention that Lee actually works with you guys, right? And in the London office and I'm adding our first ever McInturk track at AI Europe because I see this industry applications now emerging. And I'm pretty excited to, you know, help push that along. Yeah, I was looking forward to that. It'll effectively be the first industry McInturk conference. Yeah. I'm so glad you added that. You know, it's still a little bit of a bet. It's not that widespread, but I can definitely see this is the time to really get into it. We want to be early on things.Mark Bissell [00:40:51]: For sure. And I think the field understands this, right? So at ICML, I think the title of the McInturk workshop this year was actionable interpretability. And there was a lot of discussion around bringing it to various domains. Everyone's adding pragmatic, actionable, whatever.Shawn Wang [00:41:10]: It's like, okay, well, we weren't actionable before, I guess. I don't know.Vibhu Sapra [00:41:13]: And I mean, like, just, you know, being in Europe, you see the Interp room. One, like old school conferences, like, I think they had a very tiny room till they got lucky and they got it doubled. But there's definitely a lot of interest, a lot of niche research. So you see a lot of research coming out of universities, students. We covered the paper last week. It's like two unknown authors, not many citations. But, you know, you can make a lot of meaningful work there. Yeah. Yeah. Yeah.Shawn Wang [00:41:39]: Yeah. I think people haven't really mentioned this yet. It's just Interp for code. I think it's like an abnormally important field. We haven't mentioned this yet. The conspiracy theory last two years ago was when the first SAE work came out of Anthropic was they would do like, oh, we just used SAEs to turn the bad code vector down and then turn up the good code. And I think like, isn't that the dream? Like, you know, like, but basically, I guess maybe, why is it funny? Like, it's... If it was realistic, it would not be funny. It would be like, no, actually, we should do this. But it's funny because we know there's like, we feel there's some limitations to what steering can do. And I think a lot of the public image of steering is like the Gen Z stuff. Like, oh, you can make it really love the Golden Gate Bridge, or you can make it speak like Gen Z. To like be a legal reasoner seems like a huge stretch. Yeah. And I don't know if that will get there this way. Yeah.Myra Deng [00:42:36]: I think, um, I will say we are announcing. Something very soon that I will not speak too much about. Um, but I think, yeah, this is like what we've run into again and again is like, we, we don't want to be in the world where steering is only useful for like stylistic things. That's definitely not, not what we're aiming for. But I think the types of interventions that you need to do to get to things like legal reasoning, um, are much more sophisticated and require breakthroughs in, in learning algorithms. And that's, um...Shawn Wang [00:43:07]: And is this an emergent property of scale as well?Myra Deng [00:43:10]: I think so. Yeah. I mean, I think scale definitely helps. I think scale allows you to learn a lot of information and, and reduce noise across, you know, large amounts of data. But I also think we think that there's ways to do things much more effectively, um, even, even at scale. So like actually learning exactly what you want from the data and not learning things that you do that you don't want exhibited in the data. So we're not like anti-scale, but we are also realizing that scale is not going to get us anywhere. It's not going to get us to the type of AI development that we want to be at in, in the future as these models get more powerful and get deployed in all these sorts of like mission critical contexts. Current life cycle of training and deploying and evaluations is, is to us like deeply broken and has opportunities to, to improve. So, um, more to come on that very, very soon.Mark Bissell [00:44:02]: And I think that that's a use basically, or maybe just like a proof point that these concepts do exist. Like if you can manipulate them in the precise best way, you can get the ideal combination of them that you desire. And steering is maybe the most coarse grained sort of peek at what that looks like. But I think it's evocative of what you could do if you had total surgical control over every concept, every parameter. Yeah, exactly.Myra Deng [00:44:30]: There were like bad code features. I've got it pulled up.Vibhu Sapra [00:44:33]: Yeah. Just coincidentally, as you guys are talking.Shawn Wang [00:44:35]: This is like, this is exactly.Vibhu Sapra [00:44:38]: There's like specifically a code error feature that activates and they show, you know, it's not, it's not typo detection. It's like, it's, it's typos in code. It's not typical typos. And, you know, you can, you can see it clearly activates where there's something wrong in code. And they have like malicious code, code error. They have a whole bunch of sub, you know, sub broken down little grain features. Yeah.Shawn Wang [00:45:02]: Yeah. So, so the, the rough intuition for me, the, why I talked about post-training was that, well, you just, you know, have a few different rollouts with all these things turned off and on and whatever. And then, you know, you can, that's, that's synthetic data you can kind of post-train on. Yeah.Vibhu Sapra [00:45:13]: And I think we make it sound easier than it is just saying, you know, they do the real hard work.Myra Deng [00:45:19]: I mean, you guys, you guys have the right idea. Exactly. Yeah. We replicated a lot of these features in, in our Lama models as well. I remember there was like.Vibhu Sapra [00:45:26]: And I think a lot of this stuff is open, right? Like, yeah, you guys opened yours. DeepMind has opened a lot of essays on Gemma. Even Anthropic has opened a lot of this. There's, there's a lot of resources that, you know, we can probably share of people that want to get involved.Shawn Wang [00:45:41]: Yeah. And special shout out to like Neuronpedia as well. Yes. Like, yeah, amazing piece of work to visualize those things.Myra Deng [00:45:49]: Yeah, exactly.Shawn Wang [00:45:50]: I guess I wanted to pivot a little bit on, onto the healthcare side, because I think that's a big use case for you guys. We haven't really talked about it yet. This is a bit of a crossover for me because we are, we are, we do have a separate science pod that we're starting up for AI, for AI for science, just because like, it's such a huge investment category and also I'm like less qualified to do it, but we actually have bio PhDs to cover that, which is great, but I need to just kind of recover, recap your work, maybe on the evil two stuff, but then, and then building forward.Mark Bissell [00:46:17]: Yeah, for sure. And maybe to frame up the conversation, I think another kind of interesting just lens on interpretability in general is a lot of the techniques that were described. are ways to solve the AI human interface problem. And it's sort of like bidirectional communication is the goal there. So what we've been talking about with intentional design of models and, you know, steering, but also more advanced techniques is having humans impart our desires and control into models and over models. And the reverse is also very interesting, especially as you get to superhuman models, whether that's narrow superintelligence, like these scientific models that work on genomics, data, medical imaging, things like that. But down the line, you know, superintelligence of other forms as well. What knowledge can the AIs teach us as sort of that, that the other direction in that? And so some of our life science work to date has been getting at exactly that question, which is, well, some of it does look like debugging these various life sciences models, understanding if they're actually performing well, on tasks, or if they're picking up on spurious correlations, for instance, genomics models, you would like to know whether they are sort of focusing on the biologically relevant things that you care about, or if it's using some simpler correlate, like the ancestry of the person that it's looking at. But then also in the instances where they are superhuman, and maybe they are understanding elements of the human genome that we don't have names for or specific, you know, yeah, discoveries that they've made that that we don't know about, that's, that's a big goal. And so we're already seeing that, right, we are partnered with organizations like Mayo Clinic, leading research health system in the United States, our Institute, as well as a startup called Prima Menta, which focuses on neurodegenerative disease. And in our partnership with them, we've used foundation models, they've been training and applied our interpretability techniques to find novel biomarkers for Alzheimer's disease. So I think this is just the tip of the iceberg. But it's, that's like a flavor of some of the things that we're working on.Shawn Wang [00:48:36]: Yeah, I think that's really fantastic. Obviously, we did the Chad Zuckerberg pod last year as well. And like, there's a plethora of these models coming out, because there's so much potential and research. And it's like, very interesting how it's basically the same as language models, but just with a different underlying data set. But it's like, it's the same exact techniques. Like, there's no change, basically.Mark Bissell [00:48:59]: Yeah. Well, and even in like other domains, right? Like, you know, robotics, I know, like a lot of the companies just use Gemma as like the like backbone, and then they like make it into a VLA that like takes these actions. It's, it's, it's transformers all the way down. So yeah.Vibhu Sapra [00:49:15]: Like we have Med Gemma now, right? Like this week, even there was Med Gemma 1.5. And they're training it on this stuff, like 3d scans, medical domain knowledge, and all that stuff, too. So there's a push from both sides. But I think the thing that, you know, one of the things about McInturpp is like, you're a little bit more cautious in some domains, right? So healthcare, mainly being one, like guardrails, understanding, you know, we're more risk adverse to something going wrong there. So even just from a basic understanding, like, if we're trusting these systems to make claims, we want to know why and what's going on.Myra Deng [00:49:51]: Yeah, I think there's totally a kind of like deployment bottleneck to actually using. foundation models for real patient usage or things like that. Like, say you're using a model for rare disease prediction, you probably want some explanation as to why your model predicted a certain outcome, and an interpretable explanation at that. So that's definitely a use case. But I also think like, being able to extract scientific information that no human knows to accelerate drug discovery and disease treatment and things like that actually is a really, really big unlock for science, like scientific discovery. And you've seen a lot of startups, like say that they're going to accelerate scientific discovery. And I feel like we actually are doing that through our interp techniques. And kind of like, almost by accident, like, I think we got reached out to very, very early on from these healthcare institutions. And none of us had healthcare.Shawn Wang [00:50:49]: How did they even hear of you? A podcast.Myra Deng [00:50:51]: Oh, okay. Yeah, podcast.Vibhu Sapra [00:50:53]: Okay, well, now's that time, you know.Myra Deng [00:50:55]: Everyone can call us.Shawn Wang [00:50:56]: Podcasts are the most important thing. Everyone should listen to podcasts.Myra Deng [00:50:59]: Yeah, they reached out. They were like, you know, we have these really smart models that we've trained, and we want to know what they're doing. And we were like, really early that time, like three months old, and it was a few of us. And we were like, oh, my God, we've never used these models. Let's figure it out. But it's also like, great proof that interp techniques scale pretty well across domains. We didn't really have to learn too much about.Shawn Wang [00:51:21]: Interp is a machine learning technique, machine learning skills everywhere, right? Yeah. And it's obviously, it's just like a general insight. Yeah. Probably to finance too, I think, which would be fun for our history. I don't know if you have anything to say there.Mark Bissell [00:51:34]: Yeah, well, just across the science. Like, we've also done work on material science. Yeah, it really runs the gamut.Vibhu Sapra [00:51:40]: Yeah. Awesome. And, you know, for those that should reach out, like, you're obviously experts in this, but like, is there a call out for people that you're looking to partner with, design partners, people to use your stuff outside of just, you know, the general developer that wants to. Plug and play steering stuff, like on the research side more so, like, are there ideal design partners, customers, stuff like that?Myra Deng [00:52:03]: Yeah, I can talk about maybe non-life sciences, and then I'm curious to hear from you on the life sciences side. But we're looking for design partners across many domains, language, anyone who's customizing language models or trying to push the frontier of code or reasoning models is really interesting to us. And then also interested in the frontier of modeling. There's a lot of models that work in, like, pixel space, as we call it. So if you're doing world models, video models, even robotics, where there's not a very clean natural language interface to interact with, I think we think that Interp can really help and are looking for a few partners in that space.Shawn Wang [00:52:43]: Just because you mentioned the keyword

Written by AlexCome and follow more adventures on our animated TV show on Youtube!

Written by AlexCome and follow more adventures on our animated TV show on Youtube!

Peter Stansky discusses Orwell's wartime work for the BBC and The Lion and the Unicorn advocating Englishsocialism, arguing that Animal Farm was not anti-socialist but a critique of revolutionary leaders corrupted by absolute power who inevitably betray their ideals.1951

Thank you to our sponsor, Walrus! Walrus is where the world's data becomes reliable, valuable, and governable. --- In this exclusive Unchained interview, Griff Green, one of the original DAO curators and a member of the White Hat Group that helped recover funds after the 2016 DAO hack, reveals how tens of thousands of unclaimed ETH are being transformed into a long-term security fund for the Ethereum ecosystem. Nearly ten years after the most infamous exploit in crypto history, the community is repurposing its leftovers, not to rewrite history, but to prevent it from repeating itself. The new DAO Security Fund will deploy grants for Ethereum security research, infrastructure, incident response, and user protection, while also reviving DAO-based governance experiments that have fallen out of favor. Griff explains how the fund will work, why the Ethereum Foundation is involved, how staking will generate sustainable funding, and why, despite Ethereum's strength, crypto still isn't safe enough for everyday users. Guests: Griff Green, Co-Founder at Giveth, q/acc & Unicorn.eth Learn more about your ad choices. Visit megaphone.fm/adchoices

Hernosity and Pixicato must save the world from a plot so ridiculous and convoluted that it just might work! Lessons: Technology can be amazing when we learn to use it in healthy, sustainable ways. Subscribe, Support the show, and get our Yoto Cards! Want more kids podcasts for the whole family? Grown-ups, subscribe to Starglow+ here. Learn more about Starglow Media here. Follow Starglow on Instagram and YouTube Share questions with a grownup's help via email: hello@whatifworldpodcast.com or voicemail: 205-605-WHAT (9428) Eric and Karen O'Keeffe make What If World. Our producer is Miss Lynn. Character art by Ana Stretcu, episode art by Lynn Hickernell, podcast art by Jason O'Keefe, and theme song by Craig Martinson.