Podcasts about KV

Bývalý šéf Motola promluvil. Miloslav Ludvík trvá na nevině a vzkazuje, že u soudu se ukáže. Policie podle něj bude překvapená. Kvůli čemu? V čem měla pochybit? Jak se Ludvík hájí? A proč ještě nebyl obžalován? Poví Artur Janoušek z investigativního týmu Radiožurnálu. Ptá se Matěj Skalický.

Bývalý šéf Motola promluvil. Miloslav Ludvík trvá na nevině a vzkazuje, že u soudu se ukáže. Policie podle něj bude překvapená. Kvůli čemu? V čem měla pochybit? Jak se Ludvík hájí? A proč ještě nebyl obžalován? Poví Artur Janoušek z investigativního týmu Radiožurnálu. Ptá se Matěj Skalický. Všechny díly podcastu Vinohradská 12 můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

This week on International Feature, we took a trip back to one of the most talked-about holiday rom-coms of all time: Love Actually. Often held up as a seasonal classic, the film has remained a staple of Christmas viewing for years — but for us, revisiting it raised a lot more questions than warm fuzzy feelings. We dug into why the movie's reputation doesn't match our experience, and how some of its most celebrated moments have aged… poorly.• Only really connecting with one or two of the relationship dynamics presented • Our genuine disbelief that this is considered a romantic comedy “classic” • The blatant fatphobia and the way several characters are mistreated for laughs • The uncomfortable and inappropriate dynamic between Keira Knightley and Andrew Lincoln's character • Too many characters, too little depth, and barely enough time spent with any storyline • The randomness of certain casting choices — and how Martin Freeman ends up in movies like this • Why ensemble rom-coms just don't work for us when the emotional payoff isn't thereA holiday staple for some… but for us, this was more frustrating than festiveLetterbox'd Synopsis: Eight very different couples deal with their love lives in various loosely interrelated tales all set during a frantic month before Christmas in London.

Kælvning er årets vigtigste øjeblik i kødkvægsproduktionen – og det starter længe før, der kommer to ben frem. Sammen med fodringskonsulent Melanie Poulsen og dyrlæge Heidi Bødker dykker vi ned i, hvordan du fodrer ammekøer rigtigt op mod kælvning, og hvorfor protein og mineraler kan være nøglen til en stærkere kalv. Vi taler tegnene du skal holde øje med, hvornår du bør gribe ind, og hvordan du gør det sikkert og hygiejnisk. Du får også konkrete råd til genoplivning, råmælk og “connection” mellem ko og kalv. Kødkvæg i fokus er et projekt, der er støttet af Kvægafgiftsfonden. I Kødkvæg i fokus sætter vi fokus på dansk kødkvæg – fra hverdagen i stalden til de store linjer i branchen. Serien tager fat på alt fra management, avl og kalvning til naturafgræsning, afsætning, økonomi og livet omkring kødkvæget. Med afsæt i praksis, erfaringer og faglig viden giver vi plads til både nørderi, holdninger og nuancer, så kødkvæget kan blive belyst fra flere vinkler. Kødkvæg i Fokus er støttet af Kvægafgiftsfonden.

Televize Nova řeší, jak naložit s víkendovou debatou Za pět minut dvanáct. Poté, co se neprosadila v neděli, přišel přesun na sobotu. Teď mění moderátora. „Koho jiného do prvního dílu než Andreje Babiše, “ říká nová tvář pořadu Bára Divišová.Původně nedělní debatu Nova spustila v říjnu 2023 s moderátorem Martinem Čermákem. Kvůli velké konkurenci Otázek Václava Moravce na ČT i Partie Terezie Tománkové na Primě, ale televize pořad přesunula na sobotu. Debatu navíc v pátek předtáčí, aby usnadnila zvaní atraktivních hostů.„Ten pořad měl možná velké ambice, ale naskočil do hodně konkurenčního prostředí. Když se ani nebudeme soustředit na sledovanost, tak jsme tam hrozně bojovali o hosty. Tři diskuze na neděli v jeden čas to si myslím, že už je moc. Tak jsme ustoupili na sobotu,“ vysvětluje Bára Divišová, která byla hostem podcastu Mediální cirkus.Ze Střepin do debatyDivišová, která na Nově pracuje už 18 let a v poslední době byla spojená především s pořadem Střepiny nebo Napřímo, se stala novou moderátorkou debaty Za pět minut dvanáct v únoru. Jako prvního hosta si do studia pozvala premiéra Andreje Babiše z hnutí ANO.„Nedávno jsme měli na rozhovoru pana prezidenta, takže jsem si říkala, koho jiného do prvního dílu než Andreje Babiše. Až potom mi došlo, co jsem si na sebe zase ušila,“ popisuje Divišová, která patří mezi pár novinářů, s nimiž Andrej Babiš mluví.„My se známe dlouho. Jednou, někdy v roce 2015, jsem měla s Andrejem Babišem živý vstup do poledních Televizních novin. A on si přinesl pokladničku a mluvil asi sedm minut. Kolegyně Emma Smetana mě za to tehdy strašně kritizovala. Po vstupu jsem Babišovi říkala: ‚Co to mělo být?‘ A on se začal hrozně smát. Ale od té doby si pamatoval, jak se jmenuji,“ vzpomíná Divišová v Mediálním cirkusu na vysílání, ve kterém Babiš vysvětloval elektronickou evidenci tržeb.Exkluzivní prostor dostala Divišová i krátce po loňských sněmovních volbách, když k ní Andrej Babiš přišel na rozhovor tehdy ještě na TN.cz a přinesl s sebou návrh programového prohlášení vlády ANO, SPD a Motoristů. Ten moderátorce ve vysílání i předal. „Kdybych to věděla, tak bych se na to lépe připravila. Kolegové si ze mě dělali legraci, že jsem ho normálně přitlačila, aby mi to dal. Že jsem mu řekla: ‚To mi dáte!‘ Já jsem to řekla, ale na konci věty byl otazník. Ale nečekala jsem to. Kdybych to čekala, tak ten rozhovor povedu trošku jinak. Ale Andrej Babiš takové nečekané věci rád dělá,“ říká novinářka.Babiše umí naštvat každýSama říká, že jí diváci občas vztahy s Babišem předhazují, třeba na sociálních sítích, což ji mrzí.„Diváci spekulují, proč to tak je, že tady nedostane otázky na tělo, že jsem k němu servilní, že má nějaké výhody. Ale na moji obranu - on takhle mluví i s některými dalšími novináři. Na plénu mluví třeba o Petru Vaškovi z České televize. Petr za ním může přijet pro vyjádření, odepíše mu na esemesku. Tak snad doufám, že když dám tyhle dva příklady vedle sebe, že mě lidé nebudou tolik obviňovat,“ říká Divišová a pokračuje:„Andreje Babiše umí naštvat každý. To není mým cílem. Chci z toho člověka dostat informace, hlavně ve chvíli, kdy těch informací má ze svého postu dost. My spolu nesouhlasíme, i se pohádáme, ale smějeme se u toho, je to taková výměna názorů. Na někoho je lepší takový ten sofistikovaný postup, že potom sám na sebe řekne všechno.“Jak se vyrábí zpravodajství pro Novu? Jak zábavné je dělat Střepiny? A jak se dívá na debatu o veřejnoprávních médiích? --Mediální cirkus. Podcast Marie Bastlové o dění na mediální scéně. Zajímá ji pohled do redakcí, za kulisy novinářské práce – s předními novináři i mediálními hráči.Sledujte na Seznam Zprávách, poslouchejte na Podcasty.cz a ve všech podcastových aplikacích.Archiv všech dílů najdete tady. Své postřehy, připomínky nebo tipy nám pište prostřednictvím sociálních sítí pod hashtagem #medialnicirkus nebo na e-mail: audio@sz.cz.

V dnešním Olympijském speciálu si probereme zákulisí olympiády, jak to tam funguje, včetně ikonického Nagana, rozebereme práci Roberta Záruby i předsedy ČOV Jiřího Kejvala.Kvůli časové vytíženosti hostů před Olympiádou, byla tato epizoda natáčena v prosinci. Proto tam nemusí některé informace znít aktuálně.

Kvůli nedostatku zlatých slitků se teď lidé začali orientovat hlavně na stříbro. Jen za poslední rok vzrostla cena zlata o 75 procent, u stříbra je to pak dokonce 140 procent.

CELÝ DÍL NAJDETE NA HEROHERO.CO/STUDION Nové dokumenty z kauzy Jeffreyho Epsteina ukázaly, jak rozsáhlá vlivová síť tohoto odsouzeného sexuálního predátora byla. S finančníkem se přátelili milardáři i politici, radila se s ním evropská šlechta, kromě toho získával utajované vládní informace. „Epstein vytvořil novodobý Olymp, svět, kde neplatila běžná pravidla. Překonal řadu konspirací,“ říkají ve Studiu N Jiří Sobota a Barbora Chaloupková z podcastu Amerika, bejby. Kvůli novým informacím ze složek Jeffreyho Epsteina se rozpadá britská vláda a zkompromitovaní muži se omlouvají či odstupují z vlivných pozic. „Sledujeme konec jedné generace elit,“ říká Barbora Chaloupková. Přesto to zatím nevypadá, že v Americe někdo další zamíří do vězení. Dočkají se oběti někdy spravedlnosti? A vystoupí Donald Trump ze stínu skandálu? „Podstatou kauzy je, že už v tuto chvíli naleptává důvěru celé společnosti v systém. Jestliže jsme se dříve smáli těm, kteří věřili v konspirace, jako je QAnon, dnes vidíme, že reálná situace může být ještě horší,“ říká Jiří Sobota. Na druhé straně ale materiály vycházejí na světlo a část elity, jež považovala samu sebe za nedotknutelnou a nad zákonem, teď čelí celospolečenskému odsouzení. Kauza Epstein je děsivá, může se ale stát i zárodkem budoucí očisty. K tomu ale vede v tuto chvíli ještě dlouhá cesta.

Strážci a odborníci z národního parku České Švýcarsko vyrážejí do terénu a monitorují netopýry v jeskyních. Kvůli neukázněným turistům opatřili jeskyně mřížemi.

Programledare: Fabian NorlundExperter: Anel Avdić & Marcelo Fernández Viva Fotboll görs i samarbete med ATG:Gå med i Viva Fotbolls Tillsammanslag på ATG, där vi varje helg skickar in en välkalibrerad Big 9-kupong där vi försöker fälla någon av dom stora favoriterna för att stå där med miljongarantin på ensam vinnare med 9 rätt. Här har ni laget: https://www.atg.se/tillsammans/inbjudan/XKZI-CGTW-319315/tDhBPMy5pbFG8uzq%3AaJrSG_tO82Uf1mO6Zm4Fpw%3A7b2V4nqE-g4m1k4fuwZJ3VAKVv-2dCMKgw?gameId=BIG9_2025-08-23_725344240_2060735806Du hittar alltid dom senaste tripplarna, andelarna, Big 9 och annat från oss på https://www.atg.se/tutto/18+ Regler & villkor gäller. Stödlinjen.seI samarbete med TV4 Play:Vinter-OS är runt hörnet och ni ser ju såklart vinters stora, stora sporthändelse på TV4 Play! Vi har en dunderdeal: 69kr för EN månad istället för 249kr/mån. Såklart kan ni se film, serier och en massa annan sport som Serie A och La Liga på abonnemanget. Signa upp här: https://www.tv4play.se/kampanj/viva eller så anger ni koden VIVAOS26 när ni köper paketet TV4 Play Sport.Kontakta redaktionen: linus@k26media.seVill ditt företag samarbeta med Viva fotboll? samarbete@tutto.seSociala Medier:Instagram - Viva_fotbollTwitter - VivafotbollTikTok - Vivafotboll#vivafotbollTIDSKODER:00:00 Intro05:19 Gårdagens matcher11:32 Roma22:02 Juventus - Lazio35:02 Swedes of the Week47:02 Kulusevskis skada51:02 Hugo Larsson52:22 TOTW1:01:12 Kvällens matcher Hosted on Acast. See acast.com/privacy for more information.

Kvůli změnám pořád není jasné, jak přesně bude nový sál vypadat a kdy se začne stavět, což vadí hlavně opozici. Stavební firma má na demolici 80 dní. Kdy zmizí zbytek budovy, radnice zatím neví.

V další epizodě série Anna v New Yorku sledujeme Annu během její první jízdy autobusem po městě. Tentokrát nejede pěšky ani metrem, ale rozhodne se vyzkoušet autobus, který ji doveze přímo do cíle. Přesto cítí lehkou nervozitu – veřejná doprava v nové zemi má svá vlastní pravidla a Anna si není jistá, jestli všechno zvládne správně.Uslyšíte, jak Anna hledá správnou zastávku, kontroluje mapu v telefonu, nastupuje do autobusu a snaží se zorientovat v neznámém prostředí. Sleduje ulice, počítá zastávky a připravuje se na vystoupení. Kvůli jedné drobné neznalosti ale svou zastávku mine a musí se zeptat řidiče, co udělala špatně. Díky klidnému vysvětlení pochopí, jak autobus funguje, a z malé chyby se stane cenná zkušenost.Epizoda ukazuje, že i obyčejná situace, jako je jízda autobusem, může být stresující, když neznáme místní zvyklosti. Zároveň ale připomíná, že lidé bývají ochotní pomoci a že každá chyba nás může posunout dál.V této epizodě potrénujete:– zjednodušený anglický poslech na úrovni A2–B1– porozumění příběhu pomocí otázek před i po poslechu– 10 praktických slovíček z oblasti dopravy a cestování– 7 užitečných frází pro situace ve veřejné dopravě– přirozené vnímání angličtiny v běžném městském životěTento formát vám pomůže učit se angličtinu přirozeně a bez tlaku. Sledujete příběh, rozumíte souvislostem a postupně získáváte jistotu v jazyce – přesně tak, jak se angličtina používá v každodenních situacích, třeba při jízdě autobusem ve velkém městě.____

Samé jedničky měla prý naposledy v první třídě. Nikdo jí teď neřekne jinak než Kvízová dáma. Úspěšná byla v AZ kvízu, v Riskuj, vyhrála Nejslabší máte padáka. Mým hostem ve studiu je Dagmar Jandová. Vzděláním i dlouholetou praxí meteoroložka.Všechny díly podcastu Zálety Aleny Zárybnické můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

La norma técnica “Uso de la Electricidad en Minas”, aprobada mediante la Resolución Ministerial N° 308-2001-EM/VME, establece los estándares mínimos de seguridad y requerimientos técnicos esenciales para el diseño, instalación, operación y mantenimiento de sistemas eléctricos en la actividad minera peruana. Su aplicación es complementaria al Código Nacional de Electricidad y abarca minas de superficie, canteras y minas subterráneas (exceptuando las de carbón).El propósito fundamental es la prevención de incendios y la protección de personas y propiedades frente a los riesgos eléctricos. La norma es de cumplimiento obligatorio para nuevas instalaciones, mientras que las existentes deben adecuarse según un plan de evaluación de riesgos.Supervisión y Personal: Solo personal autorizado puede operar equipos, y únicamente personal calificado en electricidad puede realizar reparaciones o cambios en las instalaciones.Bloqueo y Etiquetado (LOTO): Es obligatorio contar con sistemas de etiquetado y bloqueo para los medios de desconexión durante trabajos de mantenimiento. Las etiquetas deben ser de material no conductivo.Documentación Técnica: Toda mina debe mantener planos y diagramas eléctricos actualizados para la operación segura del sistema.Equipos de Emergencia: Las salas eléctricas deben contar con extintores aprobados para fuegos de origen eléctrico y, en casos de riesgo, sistemas de alarma contra incendios e iluminación de emergencia.Cables Portátiles y de Arrastre: Deben fabricarse bajo normas como NEMA WC-58. Para tensiones superiores a 750 V, se especifican tipos como SHD o SHD-GC. Los cables en piques y vías de escape subterráneas deben ser no propagadores de flama y de baja emisión de humos.Protección de Falla a Tierra: La norma exige limitar la tensión de falla a tierra a 100 V o menos mediante dispositivos de puesta a tierra del neutro. Además, se requiere el monitoreo constante del conductor de tierra, el cual debe desenergizar la fuente si se interrumpe la continuidad del circuito.Medios de Desconexión: En circuitos superiores a 300 V, se requieren dispositivos de apertura visible localizados cerca del punto de suministro.Protección Atmosférica: Se prescribe el uso de pararrayos, condensadores de protección y líneas de guarda, especialmente en alimentadores subterráneos conectados a líneas aéreas y estructuras expuestas como castillos de pique.Líneas Aéreas: Se establecen distancias mínimas de seguridad para el almacenamiento de materiales y el desplazamiento de vehículos debajo de líneas energizadas (por ejemplo, 10 metros para tensiones de 90 a 120 kV).Subestaciones Movibles: Deben tener estructuras aptas para terrenos rugosos, transformadores en cubiertas cerradas o cercados de al menos 2 metros de altura, y sistemas de puesta a tierra que eviten transferencias de potencial superiores a 100 V.Transformadores: Su ubicación depende del punto de inflamación del líquido aislante. Aquellos con punto menor a 300°C deben estar al menos a 15 m de estructuras combustibles o bocaminas. Se prohíbe el uso de transformadores y capacitores con PCB's.Winchas de Izaje: Deben contar con dispositivos de protección contra sobreenrollamiento, sobrevelocidad, baja tensión y sobrecarga, además de frenos automáticos de emergencia y un interruptor de parada manual al alcance del operador.Transporte Eléctrico (Trole): Las locomotoras deben incluir controles de "hombre muerto" y faros que iluminen al menos 30 metros. Los conductores de trole deben mantener una tensión mecánica adecuada y estar protegidos en galerías de transporte.Comunicación: Los sistemas de voz no deben exceder los 50 V, mientras que los sistemas de señales en piques pueden operar hasta 150 V y deben ser audibles y claros.La norma define el uso de cubiertas según el entorno:IP24: Para casetas de interruptores movibles.IP45: Para equipos instalados en vehículos móviles.IP68: Para cajas de empalme en piques con tensiones de hasta 750 V.

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

Chcete-li podpořit Studio Svobodného přístavu, můžete tak učinit v krypto i korunách! Pravidelná podpora a LN: https://opristavu.urza.cz/ BTC: bc1qwy8l3w0v826amd69h4awpt9hee6srxn4gk2cpg LTC: ltc1q2w2zezyj4anh3v428msf9kqvzelt76n62ys93h Číslo účtu: 2201359764/2010; variabilní symbol: 6 -------- V únorovém livestreamu Svobodného přístavu vystoupí Karolína K.; věnovat se budeme (nejen) veřejnoprávním médiím. Čím se liší od těch státních? Ta otázka může vést k demokratickým versus autokratickým režimům; a možná i k důvěře v instituce. A co stojí za rušením koncesionářských poplatků budoucí (v únoru už možná aktuální) vládou? Jaké to může mít dopady na naši společnost? Navíc se možná do termínu livestreamu objeví další závažná témata s novou vládou související. – Karolína Kváš (https://karolinakvas.cz/); tarotová rebelka; spisovatelka; cestovatelka; antropoložka; bloggerka; tvůrkyně; koučka – Urza (www.urza.cz); autor knihy Anarchokapitalismus; tvůrce Svobodného přístavu; spoluzakladatel a hlava Institutu Ludwiga von Misese; člen předsednictva Svobody učení; učitel ve svobodné škole Ježek bez klece

In einem Land wie Deutschland sollte das Thema dieser Woche eigentlich keine Rolle spielen. Das haben wir zumindest gedacht als wir begonnen haben und mit dem Thema zu befassen. Es ist schockierend zu sehen und zu lesen, dass in Deutschland bis zu eine Millionen Menschen leben die keinen adäquaten Krankenversicherungsschutz haben. Und hier geht es nicht nur um Asylbewerber oder Obdachlose. Der Großteil sind Menschen die Schulden bei ihrer Krankenversicherung haben. Alles was es brauch sind zwei Monate ohne Zahlung und schon ist man raus. Hier gibt es zwar Beratungsstellen aber viele von Ihnen finden den Weg nicht mehr zurück ins System. Zum Glück gibt es da noch die humanitären Sprechstunden. Wir haben das Glück diese Woche mit Dr. med. Stefanie Minkley sprechen zu dürfen, die in einer humanitären Sprechstunde im Gesundheitsamt Frankfurt arbeitet und ihre Erfahrungen mit uns teilt. Relevante Websites zum Thema Zugangsbarrieren/Krankenversicherungsschutz: Bundesweite Kontaktstellen und Grundsätzliches zum Thema: https://gesundheit-ein-menschenrecht.de/ Bundesverband Anonymer Behandlungsschein und Clearingstellen für Menschen ohne KV: https://anonymer-behandlungsschein.de/gemeinsame-politische-ziele-des-back/ Bundesarbeitsgemeinschaft Gesundheit&Illegalität: https://www.diakonie.de/informieren/infothek/2019/gesundheitsversorgung-fuer-menschen-ohne-papiere Bundesarbeitsgemeinschaft der Wohnungslosenhilfen, Thema Gesundheit: https://www.bagw.de/de/publikationen/pos-pap/pos-gesundheit Medibüros und MediNetze bundesweit/zivilgesellschaftliches Engagment: medibueros.org **Kapitelmarken** 00:00 Intro 01:00 Feedback 01:30 News I - Forschungsgelder zur Frauengesundheit 02:24 News II - Trauerreaktion nach Verlust von Tieren 03:26 News III - telefonische Krankschreibung vor dem Aus? 05:46 Main Part 20:07 Interview mit Dr. Stefanie Minkley 55:35 Take Home Message 57:55 Outro **Instagram:** AMS_Podcast **Email:** aufmessersschneidepodcast@gmail.com **Liken nicht vergessen! 5 Sterne bei Apple Podcast, Spotify oder der Podcastplattform eures Vertrauens helfen uns dabei unsere Reichweite zu erhöhen. Am 19.02.2026 geht es mit einer neuen, spannenden Folge weiter.** **Quellen** Mehr als 70.000 Menschen ohne Krankenversorgung https://www.tagesschau.de/inland/innenpolitik/krankenversicherung-statistik-102.html Menschen ohne Krankenversicherung: Ein oft übersehenes Problem www.aerzteblatt.de/archiv/menschen-ohne-krankenversicherung-ein-oft-uebersehenes-problem-0f03a705-055c-4855-9a36-d06040aef11f Hilfe für Menschen ohne Versicherung Ein Lückenfüller fürs System https://www.tagesschau.de/inland/gesellschaft/versorgung-ohne-krankenversicherung-100.html Medinetz Franfurt/Offenbach https://medinetz-frankfurt-offenbach.de Das Dilemma in der Versorgung von Menschen ohne Krankenversicherungsschutz https://www.laekh.de/fileadmin/user_upload/Heftarchiv/PDFs_ganze_Hefte/2025/HAEBL_12_2025.pdf Stefanie Minkley https://www.stefanieminkley.de Berufsordnung BÄK 2024 https://www.bundesaerztekammer.de/fileadmin/user_upload/BAEK/Themen/Recht/_Bek_BAEK_Musterberufsordnung-AE.pdf BMG gibt Startschuss für Forschungsförderung zu Frauengesundheit – Antragsphase gestartet https://www.bundesgesundheitsministerium.de/presse/pressemitteilungen/startschuss-forschungsfoerderung-frauengesundheit-pm-14-01-26.html Trauerreaktion nach Verlust eines Haustiers oft vergleichbar mit menschlichem Verlust https://www.aerzteblatt.de/news/trauerreaktion-nach-verlust-eines-haustiers-oft-vergleichbar-mit-menschlichem-verlust-6f96717b-146d-4feb-8298-49d76e188ae0 Telefonische Krankschreibung vor dem Aus? https://www.tagesschau.de/inland/gesellschaft/telefonische-krankschreibung-122.html

Professor Agnes Wold är tillbaka i Kvällspassets studio för att svara på lyssnarfrågor! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Ellen undrar om hon borde vaccinera sina barn mot vattkoppor, Tor har funderingar kring RS-vaccin och bältrosvaccin och Carina frågar om det är värt att dricka mycket honungsvatten när man är förkyld.

Vláda podle exministra kultury Martina Baxy nepracuje pro občany. Proto opozice vyvolala jednání o nedůvěře vládě. „Každý měsíc nedůvěra nebude, protože doufám, že Andrej Babiš odvolá Petra Macinku. Věřím tomu,“ řekl Baxa. Hostem Ptám se já byl bývalý ministr kultury, poslanec Martin Baxa (ODS). Poslanci pokračují v jednání o vyslovení nedůvěry vládě ANO, SPD a Motoristů druhým dnem. Opozice schůzi vyvolala kvůli kontroverzním krokům vládní koalice, zejména kvůli vyostření sporu mezi Motoristy a prezidentem Petrem Pavlem. Kvůli zprávám ministra zahraničí a lídra Motoristů Petra Macinky hlavě státu, které Pavel označil za pokus o vydírání, opoziční strany požadují také Macinkovo odvolání. Exministr kultury Martin Baxa věří, že je povinností opozice v takové situaci jednání o nedůvěře vyvolat. Zároveň doufá v Macinkův konec ve vládní funkci. „Pevně doufám, že Andrej Babiš odvolá Petra Macinku. Věřím tomu. Možná je to naivní, ale je naší povinností to říkat, protože místopředseda vlády nemá být člověk, který píše vyděračské esemesky prezidentovi.“Baxa v Ptám se já také komentoval poslední kroky dalšího zástupce Motoristů ve vládě, svého předchůdce v čele ministerstva kultury Oto Klempíře. Ten si po nedělní demonstraci spolku Milion chvilek na podporu prezidenta pozval na resort umělce, kteří na akci vystoupili, aby si to s ním přišli vyříkat na ministerstvo. Na ministrovo pozvání zástupci kulturní obce reagovali s výzvou, aby šéf resortu přišel na veřejnou debatu do divadla Palace v centru Prahy. Klempíř to odmítl s tím, že chtěl řešit budoucnost kultury, nikoliv politiku a pro umělce tedy stále platí pozvání na ministerstvo. Umělci ovšem trvají na setkání před veřejností.  „Do obdobné situace jsem se také dostal. Z částí umělecké obce jsem tímhle způsobem řešil téma statusu umělce. Dostal jsem možná obdobné pozvání jako pan ministr Klempíř  na takovou otevřenou debatu do nějakého kulturního prostoru. A já jsem to pozvání tehdy přijal,“prohlásl exministr. Kam míří česká kultura? Jak to dopadne s veřejnoprávními médii? A jaký smysl má hodiny trvající snaha o nedůvěru vládě s předem jasným výsledkem?--Podcast Ptám se já. Rozhovory s lidmi, kteří mají vliv, odpovědnost, informace.Sledujte na Seznam Zprávách, poslouchejte na Podcasty.cz a ve všech podcastových aplikacích.Archiv všech dílů najdete tady. Své postřehy, připomínky nebo tipy nám pište prostřednictvím sociálních sítí pod hashtagem #ptamseja nebo na e-mail: audio@sz.cz.

När stod du upp för din partner, en nära vän eller någon i din familj som älskar någon och behövde stöd? Hör lyssnarnas berättelser! Lyssna på alla avsnitt i Sveriges Radios app. Kvällens program handlar om modet att välja kärleken, även när omgivningen säger något annat. Vad drev dig, vilken reaktion fick du – och hur ser du på det idag? Vi öppnar Sveriges största samtalsrum där lyssnarnas berättelser är allt.Kanske gick du emot traditioner, stod upp inför släkt och vänner eller vågade säga ja fast allt kändes osäkert. Ditt samtal kan ge tröst, styrka och igenkänning för andra. Ring in, skriv till oss eller delta i samtalet på sociala medier. Ordet är ditt – ring och berätta. Om att stå upp för sin kärlek med Hanna SihlmanRing eller mejla oss, på karlavagnen@sverigesradio.se eller skriv till oss på Facebook och Instagram. Telefonslussen öppnar kl. 21.Programmet startar kl. 21:40.

Az előfizetők (de csak a Belső kör és Közösség csomagok tulajdonosai!) már szombat hajnalban hozzájutnak legfrissebb epizódunk teljes verziójához. A hétfőn publikált, ingyen meghallgatható verzió tíz perccel rövidebb. Itt írtunk arról, hogy tudod meghallgatni a teljes adást. Forradalmi követelés: drogmentes rendőrséget! Miért nem tudnak parkolni a kínaiak? Mit akar Lázár János a vécékefével? A választás titkos sztárjai a BlackRock támogatásával: Hiller haver, Jakab Péter, Humanisták, Vona Gábor. 00:53 A Blackrock szponzorálásával. A kínai néni és a tisztogatás. A szívószálpápa rendszáma.05:47 Breaking: Humanisták.07:21 Jakab Péter Borsod 01-ben. Life coach lennél inkább, vagy DK-s? Így múljon el minden náci! Nem maradt hely a Fidesztől jobbra.12:30 Hiller haver nem adja fel. A HVG MSZP-tesztje. Antiszemita plakátrongolás a XI. kerületben.17:09 Kínaiak, kecskék, birkák, hüvelyesek, kézjelek.22:28 Kvíz: TFR. A kínai egykepolitika vége.26:45 Ivan Krastev és a szláv népesedési háború. A kollektív parkolási képességek szerepe a geopolitikai játszmákban, különös tekintettel Tajvan lerohanására. Rommel és Guderian bezzeg tudtak párhuzamosan parkolni!34:52 Honosítások a 2030-as vébére.37:25 A legtávolabbi hallgató. Randevúk szociológusokkal és gyökerekkel.41:04 Visszatér a Heti hetes. Új idők új Bajor Imréje. Amikor Simicska szerint Orbán meg akarta venni az RTL-t, csak nem volt pénze.46:30 Vitézy Dávid és a KRESZ.50:01 Együgyű párt a drogmentes rendőrségért.See omnystudio.com/listener for privacy information.

Kvällspasset gästas av Sveriges Radios utrikeskorrespondenter, Simon Isaksson (USA), Andreas Liljeheden (Bryssel) och David Rasmusson (Norden) som svarar på lyssnarnas frågor om det spända världsläget. Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Neda undrar hur EU kan jobba mot styret i Iran och Anita undrar om man kan se en trend i bojkott mot amerikanska produkter i Skandinavien. Vi hör också Thomas som undrar vad gemene amerikan tycker om Trumps styre och Charlotte undrar vad Trumps nya fredsråd är för något?

Avant-dernière émission de la semaine spéciale Bilouki sur Planète Rap avec “18 Kara”, entouré de La Chine, KV, TVLM, Charly, Fuego et Fred Musa. Un rendez-vous clé pour accompagner la sortie de son nouveau projet !

What happens when the hobby you love starts feeling like a chore? We go straight at that question with Jim Bates, exploring how burnout creeps in, why favorite subjects can become fear targets, and what it takes to rediscover honest joy at the bench. Jim shares how a demanding year pushed modeling to the margins, why armor felt freer than aircraft, and the simple mindset shift that turned “perfect or quit” into “finish and learn.” Along the way, we unpack airbrush avoidance, photoetch dread, and the tiny victories that rebuild momentum—like stripping a botched primer, repainting, and choosing progress over paralysis.We also get practical. You'll hear how keeping short journal notes, and accepting weekend-only bench time can remove friction and make modeling sustainable again. We talk about the limits of step-by-step boilerplate articles, why video excels at teaching technique, and how personal writing can spark creativity in ways a camera can't. Jim's revived blog, A Scale Canadian, is his sandbox for that approach: short, thoughtful posts that value honesty over hype.There's fresh inspiration too. We walk through Model Mania at the Museum of Flight—a display-only, public-forward event with seminars, demos from Rick Lawler, and zero contest pressure—plus a quick tour of new kit announcements that caught our eye, from Airfix's Canberra and JU 52 to MiniArt's Opel Maultier. To close, we share bench updates: Shermans and Cromwells, a Hellcat edging toward weathering, a T-33 off the shelf of doom, and a KV-85 waiting on brass.If you've been stuck, second-guessing, or saving “the good kit” for a better version of yourself that never seems to arrive, this conversation is your nudge. Build for you. Finish something small. Protect your joy. Then tell us what you're tackling next. Subscribe, share with a friend who needs the push, and leave a quick review to help others find the show.Model Paint SolutionsYour source for Harder & Steenbeck Airbrushes and David Union Power ToolsSQUADRON Adding to the stash since 1968Model PodcastsPlease check out the other pods in the modelsphere!Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you.Give us your Feedback!Rate the Show!Support the Show!PatreonBuy Me a BeerPaypalBump Riffs Graciously Provided by Ed BarothAd Reads Generously Provided by Bob "The Voice of Bob" BairMike and Kentucky Dave thank each and everyone of you for participating on this journey with us.

Declan Rice a világ egyik legjobb középpályásává nőtte ki magát. A Real Madrid elleni szabadrúgásgóljai óta értékeli őt igazán a világ, de mi tette őt az Arsenal legfontosabb játékosává? Borbély Imivel és Brodarics Tomival megpróbáljuk megfejteni 30 percben. (0:00) Több kvíz (4:00) Hogyan illik az Arsenal játékába Rice? (7:30) Rice leigazolása (16:40) Ki hasonló Rice-hoz? (18:25) Rice gyenge pontja (20:30) Rice sokoldalúsága (24:00) Játékoskapcsolatok a középpályán (24:45) Clutch player (29:15) Kvíz megoldás

For our first International Feature of 2026, we dive into Anatomy of a Fall, the 2023 French courtroom drama, directed by Justine Triet. It became a major awards-season standout, winning the Palme d'Or at Cannes and sparking wide conversation for its unconventional approach to truth, perspective, and performance. Rather than leaning on flashy courtroom theatrics, the film builds tension through ambiguity, character study, and meticulous visual choices.In this episode, we break down what made Anatomy of a Fall resonate so strongly with audiences and critics alike, including what worked for us and why the film continues to linger long after the credits roll.The across-the-board performances, from the lead actors all the way down to the dogHow the film's open-ended storytelling challenges the audience to sit with uncertaintyDirect, intentional cinematography choices that quietly shape how we interpret eventsA courtroom drama focused on truth and perception, without over-the-top legal theatricsWhy restraint and realism make the emotional beats hit harderWhether you loved the ambiguity or found it frustrating, this was a film that gave us plenty to unpack.Letterbox'd Synopsis: A woman is suspected of her husband's murder, and their blind son faces a moral dilemma as the sole witness.

Welcome to episode 339 of The Cloud Pod, where the forecast is always cloudy! Justin and Matt are in the studio today to bring you all the latest in cloud and AI announcements, including more personnel shifts (and it doesn't seem like it was very friendly), a new way to get much needed copper, and Azure marketplace advertising 4,000 different models. What's the real story? Let's get into it and find out!  Titles we almost went with this week US-EAST-1: Still the Least Reliable Friend You Keep Inviting to Parties **OpenAI 0⃣ From Zero to Inference: BigQuery Makes Open Models a Two-SQL Problem AWS Goes Full Brandenburg Gate: Sovereign Cloud Opens for Business Seven Ate Nine: AWS Skips G7 and Goes Straight to G7e Instances From Crawling to Calling: Cloudflare Buys Human Native to Fix AI’s Data Problem Finally, an AI That Actually Listens to Your War Room Panic Tag, You’re Governed: AWS Automation Takes the Wheel Cloudflare Reaches for the Stars: Astro Framework Acquisition Lands Gemini Gets Personal: Google AI Finally Reads Your Email (With Permission) AWS Strikes Ore: Amazon Cuts Out the Middleman in Copper Supply Chain When Your Region Goes Down More Often Than Your Kubernetes Cluster ChatGPT Go: OpenAI’s New Middle Child Gets $8 Allowance Cloudflare’s Space-Age Acquisition: Astro Gets Jetsons-Level Upgrade Rosie the Robot Fired: Cloudflare Brings Astro Framework Into the Family It took 5 years, and now we have ads in our AI.  AI now with Ads EU says hands off my data General News  00:50 Heather's data is not unreliable  Maybe it's unreliable. I blame Matt for having screwed up his outtro (as he did today), in which case I no longer recognize his participation.  01:11 Astro is joining Cloudflare Cloudflare acquires The Astro Technology Company, bringing the popular open-source web framework in-house while maintaining its MIT license and multi-cloud deployment capabilities.  Major platforms like Webflow Cloud, Wix Vibe, and Stainless already use Astro on Cloudflare infrastructure to power customer websites. Astro 6 introduces a redesigned development server built on Vite Environments API that runs code locally using the same runtime as production deployment. When using the Cloudflare Vite plugin, developers can test against workerd runtime with access to Durable Objects, D1, KV, and other Cloudflare services during local development. The framework focuses on content-driven websites through its Islands Architecture, which renders most pages as static HTML while allowing

Most people assume AI “remembers everything” — every chat, every command, every conversation. But that's not how today's systems actually work. On this episode of Today in Tech, Keith Shaw talks with Manifest AI CEO Jacob Buckman about how AI memory really works under the hood, why chatbots feel so different from humans, and what has to change for true long-running digital agents to become reality. Jacob explains concepts like short-term vs. long-term AI memory, context windows, KV caches, and “scratchpad” summaries in plain language. He uses analogies from medicine and the movie Memento to show why current AI tools can ace a single conversation but struggle to stay on task over hours, days, or projects. They also dig into hallucinations, why simply “making models bigger” isn't enough, and how new architectures like power retention aim to give AI a more human-like ability to remember what actually matters over time. You'll learn: * Why AI remembers everything inside a chat window but almost nothing between sessions * How today's memory tricks (summaries, scratchpads, huge context windows) still fall short * How memory limits hold back reliable AI agents for coding, research, and creative work * Why better long-term memory could cut hallucinations and boost trust in business use cases * What “power retention” is — and how it could reshape the next generation of AI systems

Haka på när 2026 startar upp på riktigt med Apologia! • 21 januari 19.00 Webinar: Jesus & Josefus  Hör om det senaste i Jesus-forskningen, i detta webinar med Stefan Gustavsson. • 26 januari 19.00 Terminsstart Distanskurser Fyra kurser börjar nu i januari. Kvällsbibelskola Galaterbrevet Konstiga ställen i Bibeln Rusta dina barn Vänner eller fiender Ytterligare två kurser börjar i mars. Se allihopa i lärplattformen! • 29 januari 16.00 Ledarakademin - Apologetik 1  Kolla in för egen del, eller tipsa en kristen ledare i din närhet. • 14 mars Veritas för högstadie- och gymnasieungdomar Välkommen till Stockholm och Betlehemskyrkan. Inga frågor är off limits!

VOV1 - Trong bối cảnh khoa học công nghệ, chuyển đổi số là động lực then chốt để phát triển nhanh và bền vững, Tổng công ty Điện lực miền Bắc (EVNNPC) đã và đang tích cực đẩy mạnh ứng dụng khoa học công nghệ trong quản lý, vận hành lưới điện nhằm nâng cao độ tin cậy cung cấp điện.Miền Bắc hiện là khu vực có tốc độ tăng trưởng phụ tải điện cao, tập trung nhiều trung tâm công nghiệp, khu kinh tế và đô thị lớn. Đây cũng là địa bàn chịu áp lực lớn từ quá trình công nghiệp hóa nhanh, đô thị hóa mạnh và sự dịch chuyển của chuỗi cung ứng toàn cầu. Hiện Tổng công ty Điện lực miền Bắc (EVNNPC) đang quản lý, vận hành gần 94.131 km đường dây trung, cao áp và hơn 370 trạm biến áp 110 kV, trải dài trên địa bàn 17 tỉnh, thành phố miền Bắc.Trong bối cảnh nhu cầu điện của khu vực miền Bắc tăng cao, việc đẩy mạnh ứng dụng công nghệ số trong quản lý, giám sát, điều hành lưới điện được xem là giải pháp căn cơ để nâng cao độ tin cậy cung cấp điện, giảm thiểu sự cố và tối ưu hoá chi phí vận hành. Một trong những giải pháp được EVNNPC triển khai sớm và bài bản là ứng dụng thiết bị bay không người lái (UAV) trong kiểm tra lưới điện. Đến cuối năm 2025, EVNNPC đã đưa vào sử dụng hơn 400 UAV chuyên dụng, phục vụ công tác kiểm tra đường dây, trạm biến áp và khảo sát hiện trường. Điểm khác biệt nằm ở việc UAV không chỉ thực hiện chức năng chụp ảnh, ghi hình mà còn được tích hợp các thuật toán AI để tự động nhận diện khiếm khuyết. Nhờ đó, công tác kiểm tra trở nên nhanh hơn, chính xác hơn và an toàn hơn.Nhân viên Điện lực EVNNPC điều khiển thiết bị UAV ngoài hiện trường

Kvůli komiksu Zelený Raoul se s nimi kdysi soudil Jiří Paroubek, kvůli karikatuře na obálce to stejné dělal Tomio Okamura. Přesto si z politiků utahují dál. „Nejlépe prodávají časopis,“ říká šéfredaktor Reflexu Martin Bartkovský.„Tomio Okamura je samozřejmě Pitomio a my všichni to můžeme říkat. Jak konstatoval soud, musí snést vyšší míru kritiky. Navíc to, kdy jako první věc ve funkci předsedy Sněmovny podrží štafle, aby někdo jiný sundal ukrajinskou vlajku, která se mu nelíbí, je podle mě krystalickou ukázkou chování někoho, komu by se dalo říkat Pitomio,“ vysvětluje Martin Bartkovský, šéfredaktor časopisu Reflex, který byl hostem nejnovějšího dílu podcastu Mediální cirkus.Právě vyobrazení Tomia Okamury jako klauna s nápisem Pitomio je asi nejslavnější obálka Reflexu. A to i díky následnému soudnímu sporu.Vyšla 7. listopadu 2012 v době, kdy se Okamura stal senátorem a pracoval na znásobení známosti svého jména tím, že kandidoval v prvních přímých prezidentských volbách. Po vydání karikatury dal Okamura na Reflex žalobu. Tu ale po mnohaleté soudní tahanici předloni definitivně prohrál.Na obálce fungují Babiš, Zeman a OkamuraČasopis Reflex vychází v Česku už 35 let a na výrazných titulních stranách si zakládá.„Fungují čeští politici. Byla doba, kdy vládl Andrej Babiš s Milošem Zemanem a do toho tam jako třetí vzadu pobíhal Tomio Okamura nebo komunisté, o které se vláda tehdy opírala. Tam stačilo kohokoliv z téhle vlády dát na obálku a hned tam prodeje byly. Dneska už to takhle není,“ říká šéfredaktor Bartkovský. „Filip Turek a jeho volební blitzkrieg v eurovolbách zafungoval velmi dobře. Motoristé fungují. I Andrej Babiš. Funguje i prezident Petr Pavel a vždycky funguje Volodymyr Zelenskyj. Stejně tak Vladimir Putin nebo Donald Trump, ale to jsou jediné zahraniční persony,“ popisuje Bartkovský taktiku při výrobě titulních stran.Vůbec nejúspěšnější titulní strany Reflexu za rok 2025 se ale nakonec politiky vůbec netýkaly. „Byly to obálky in memoriam našich dlouholetých spolupracovníků. Tím jedním byl psycholog Cyril Höschl a tím druhým byl šéf karlovarského filmového festivalu pan Jiří Bartoška,“ dodává Martin Bartkovský.Ten se šéfredaktorem Reflexu stal v prosinci 2023, po půl roce nejistoty ve vedení časopisu. Před ním ho řídil Marek Stoniš a časopis se často dostával na hranu kritiky za texty a titulky zavánějící někdy až xenofobií.„Kvůli obálce s černým Hitlerem chtěla odejít polovina redakce. Občas už jsme byli prostě zbytečně zlí. Vtipné je být vtipní, satiričtí, jízliví, ale když jste vyloženě zlí, a dávali to vědět i čtenáři, tak to vtipné není,“ říká Bartkovský a naráží na titulní stranu s portrétem Adolfa Hitlera coby černocha s monstrózním afro účesem, která vyšla v roce 2020.Babiš jako Slabiš. Nevíme, z koho si utahovat dřívS novou vládou je podle Bartkovského stále těžší si z ministrů dělat legraci, protože v redakci neví, koho karikovat dříve.„Je to vlastně bezprecedentní situace, kdy byste mohli každý týden udělat na každého člena vlády jednu obálku, tedy kromě těch členů SPD, kteří nic neříkají. Ale ti ostatní jsou velmi plodní,“ žertuje novinář s tím, že na obálce tento čtvrtek by chtěl mít Andreje Babiše. „Bude na té obálce jako Slabiš, bude se opírat o Macinku s Turkem a Okamurou a bude se snažit udělat Česko lepší, nejlepší zemí na této planetě,“ směje se Bartkovský.I humor má ale v Reflexu hranice.„Nemáme hranice v tom, z kterého politika si udělat legraci, ale nechceme si dělat legraci ze všedních lidí, obyčejných Čechů, z někoho, kdo se nemůže bránit. My si vždycky děláme legraci z lidí, kteří mají nějakou moc nebo mají pocit, že drží nějakou moc a my je chceme trošku uzemňovat,“ říká šéf oblíbeného časopisu.Co čeká od vlády Andreje Babiše? Co kabinet změní a dotáhne za příští čtyři roky? A jak se vlastně dostal do čela Reflexu?--Mediální cirkus. Podcast Marie Bastlové o dění na mediální scéně. Zajímá ji pohled do redakcí, za kulisy novinářské práce – s předními novináři i mediálními hráči.Sledujte na Seznam Zprávách, poslouchejte na Podcasty.cz a ve všech podcastových aplikacích.Archiv všech dílů najdete tady. Své postřehy, připomínky nebo tipy nám pište prostřednictvím sociálních sítí pod hashtagem #medialnicirkus nebo na e-mail: audio@sz.cz.

A slow start, a full heart, and a clear plan. We kick off 2026 by resetting our modeling habits, sharpening the skills that matter most, and putting dates on the calendar to turn ideas into finished work. HeritageCon is pulling us forward, but it's the day-to-day that will make the difference: tighter bench time, better canopies, and bases that finish strong instead of phoning it in.One photo sent us down a rabbit hole—captured Soviet armor at Kummersdorf with mysterious inventory rectangles. We trace similar markings across other vehicles and share why the rectangle's color might be yellow, then ask armor specialists for hard provenance rather than AI guesses. That curiosity fuels the whole episode. The dojo keeps paying dividends, from canopy wax tips and stencil-cutter know-how to encouragement from modelers solving the same problems. We celebrate KitMask extending mojo30 for 30% off through HeritageCon and spotlight how small breaks in cost and friction can nudge more projects across the line.We lay out our goals for the year. Aircraft need spotless canopies—polished clear parts, confident masking, and frames that sit sharp and true. Speed is focus: fewer distractions, more finishes without losing joy. Armor projects get a base upgrade with cleaner edges, smarter terrain transitions, and groundwork that complements the model instead of competing with it. On the adjacent front, we commit to mastering a Cameo stencil cutter for crisp markings and layered paint effects, and we push to launch phase two of our website so the community can learn and share even more.On the bench, the Hellcat weathers the tiny-stencil storm, the Moosaroo rally build earns custom decals and a clever mixed-material interior, and the KV-85 stacks sub-assemblies toward primer. Our 2026 wish list is ambitious but grounded: MiniArt T-34/76 variants, a modern JSU-152, an early D3A1 Val, a 1/72 Privateer, and a 14-meter Daihatsu for Pacific dioramas. If you've got insight on Kummersdorf markings or a kit rumor we should track, jump in. Subscribe, share the show with a modeling friend, and leave a quick review—then tell us your top skill goal for 2026.SQUADRON Adding to the stash since 1968Model Paint SolutionsYour source for Harder & Steenbeck Airbrushes and David Union Power ToolsModel PodcastsPlease check out the other pods in the modelsphere!Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you.Give us your Feedback!Rate the Show!Support the Show!PatreonBuy Me a BeerPaypalBump Riffs Graciously Provided by Ed BarothAd Reads Generously Provided by Bob "The Voice of Bob" BairMike and Kentucky Dave thank each and everyone of you for participating on this journey with us.

VOV1 - Tổng kết các phong trào thi đua yêu nước của Ngành điện, Ngành Công Thương giai đoạn 2021-2025, Công trình đường dây 500kV mạch 3 Quảng Trạch - Phố Nối luôn được nhắc đến với những cảm xúc hết sức đặc biệt. “Thi đua là yêu nước” - hơn 75 năm kể từ ngày Chủ tịch Hồ Chí Minh ra “Lời kêu gọi thi đua ái quốc” (11/6/1948) đến nay, phong trào thi đua yêu nước phát triển ngày càng mạnh mẽ, sâu, rộng trong các tổ chức, từng cá nhân, mọi tầng lớp nhân dân, với nhiều nội dung phong phú, việc làm thiết thực, góp phần đưa cả nước vượt qua khó khăn, thách thức, phát triển toàn diện mọi mặt đời sống, kinh tế - xã hội đất nước.Công trình đường dây 500 kV mạch 3 Quảng Trạch - Phố Nối

Send us a textWe are headed to Shot Show in 2026.  We will be publishing episodes every day going over what we saw and some of the coolest things at the show.  Make sure you like, subscribe and share so you get all the latest info from the show.Intro/Outro Music: Quest by KV    / kvmusicprod  License: Creative Commons — Attribution 3.0 Unported  — CC BY 3.0 Free Download / Stream: https://audiolibrary.com.co/kv/questMusic promoted by Audio Library:    • DAILY No Copyright For You – Quest by KV

Welcome to your weekly UAS News Update. We have three stories for you this week; leaked pricing for the upcoming DJI Avata 360, the world record for the fastest drone has been shattered, and public safety is starting the year off with a ton of drones for good stories! Let's get to it.First up thanks to a leaked pricing table from a Chinese retail store, we have what appears to be the final pricing for the DJI Avata 360. And yes, the Avata 360 is already FCC approved. Now for the prices. In China, the base drone is listed at ¥2,988, which is about $426 USD. The Standard Combo with the Motion Controller 3 is about $569, and the Fly More Combo comes in at around $811. That puts the estimated US price for the base drone around $489, and the Fly More Combo will likely land right at that classic DJI price point of $999.This drone is rumored to brings true spherical 360 capture to an FPV platform, which is a huge deal. There are also rumors it could be under 250 grams. It seems to be a direct challenger to the Insta360 Antigravity A1, and DJI is betting that immersive 360 FPV is compelling enough for people to swallow that price tag for this new tech. Next up, for all you speed demons and FPV builders out there, the record for the world's fastest drone has been absolutely demolished. Luke Maximo Bell and his team have reclaimed the Guinness World Record with their Peregreen V4 drone, clocking an official top speed of 408.60 miles per hour, or 657.59 kilometers per hour. They took the record back from Benjamin Biggs, who had set it at 389 mph.What's really impressive here is the engineering. They meticulously tested three different motors—the AOS Supernova 3220, the AMX 2826, and the T-Motor 3120. They ended up choosing the T-Motor 3120 not because it had the most thrust, but because it was the most reliable and ran cooler. That shows it's not just about peak power, but about surviving the run! The frame itself was 3D printed, merging a hard PETG material with a softer TPU on the nose cone. To get that extra speed, they also bumped the motor KV up from 800 to 900. I want to pause for just a minute to discuss an upcoming webinar we are hosting. This webinar is all about how to land clients in 7 days, and it's on Tuesday, January 13th. If you're struggling to get your first client, this is perfect for you. Be sure to preregister if you want to attend. Check out the link in the comments, and we'll see you there! Last up this week, we have a bunch of drones for good stories, out of several places across the country:- A hiker was rescued using a drone in Chillicothe, Ohio using a drone, likely using a DJI Matrice series.- A hiker in Oregon was rescued, likely using an M30T or Matrice 4T.- A man with dementia in Campbellsville KY was located using a Matrice 30T. - A Skydio X10 was used to capture a man in Wichita after an armed robbery.k- A DJI Matrice 400 was instrumental in a rescue in Michigan, after a snowmobile broke through lake ice, sending the two riders into the water. - And a Matrice 4T in Fishers, Indiana located a firearm after it was dumped by a suspect during a chase.These stories are proof that drones have become like any other tool for Public safety departments, and that they do save lives. Alright, that's it for this week, Join us in the premium community for Post flight, our uncensored show where we share our opinions, which aren't always suitable for YouTube! See you on Monday for the live! https://dronexl.co/2026/01/04/luke-bells-peregreen-v4-new-fastest-drone/https://dronexl.co/2026/01/02/dji-avata-360-price-china-us/https://dronexl.co/2025/12/31/police-drone-missing-hiker-ohio-search/https://dronexl.co/2026/01/05/dji-drone-ice-rescue-saginaw-bay/https://dronexl.co/2026/01/04/wichita-police-drone-robbery/https://dronexl.co/2026/01/04/dji-matrice-drone-campbellsville-missing/https://dronexl.co/2026/01/03/drone-rescue-lost-hiker-oregon/

Märkliga ljud, tal eller felsägningar vi pratar om när man inte kunde tro sina öron! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Vi hör bland andra Kerstin som fick en mal i örat när hon var ute och plockade blåbär, Annika som vann en häst när hon var 12 år och så försöker vi reda ut vilket djur det var som väckte Rikard den där sommardagen egentligen. Var det lodjur som slogs eller kanske rådjur i brunst?I extramaterialet pratar vi stimpengar, Malcolm In The Middle och om det inte ändå är så att vi borde annordna Kvällspassets grisfest!

Och så tittar vi närmare på mediestödet och frågar oss hur mycket lokaljournalistik vi får för pengarna. Lyssna på alla avsnitt i Sveriges Radios app. Olika könsbeslut i rapporteringen om misstänkt styckmördareI slutet av 2025 kom nyheten om det bestialiska styckmordet på en ung kvinna i Rönninge. Ganska snart grips en misstänkt gärningsperson och i vanlig ordning gör olika medier olika bedömningar om namn, bild och detaljer. Kvällstidningarna namn- och bildpublicerar, Dagens Nyheter och public service gör det inte. Så långt, förväntat. Men en sak sticker ut, vilket pronomen som medier använder för den misstänkta. Där vissa skriver han undviker andra helt att nämna kön. Och en tidning väljer att istället skriva hon. Freddi Ramel intervjuar med Andreas Gustavsson, chefredaktör på ETC, Karin Schmidt, redaktionschef Aftonbladet, Hannes Lundberg Andersson, breakingchef Expressen, Karin Ekman, ansvarig utgivare på SVT och Signe Krantz, opinionsjournalist och förbundsordförande Transammans.Skral lokalbevakning trots miljoner i mediestödDet har gått två år sen det lappade och lagade presstödet gjordes om i grunden. Utöver det övergångsstöd som ska hjälpa de tidningar som förlorar på det nya systemet, så består mediestödet idag av ett allmänt redaktionsstöd, distributionsstöd och ett utökat redaktionsstöd. Det sistnämnda syftar till att stärka bevakningen av underbevakade områden och går ut på att lokaltidningar kan få upp till 600.000 kronor per underbevakad kommun. Men frågan är hur mycket journalistik medborgarna får för pengarna. Eller, det är åtminstone en fråga man ställt sig i den norrländska inlandskommunen Arjeplog.Joanna Korbutiak pratar med Marianne Hofman från ArjeplogNytt, Mari Berglund, chefredaktör för Piteå-Tidningen, Fredrik Westerlund, utbildnings- och kulturchef i Arjeplogs kommun och Kajsa Rohdin, avdelningschef vid Mediemyndigheten. Erik Peterson ringer även upp Göteborgs-Postens chefredaktör Christofer Ahlqvist.

Ibland kan man inte tro sina ögon vi hör några av lyssnarnas exempel! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Inger minns tillbaka till Rhodos och den stora grisfesten där hon mötte fyra sällskap från Fryksdalen, Amandas nyköpta kondompaket var helt tomt och Anders tåg blev inställt fyra gånger!Vi ringer också upp Håvard Lie, sportsligt ansvarig för backhoppning på svenska skidförbundet, för att prata om backhoppningsdräkter. I extramaterialet kan vi inte sluta tänka på grisfester, är det kanske dags för en ny Kvällspasset-tradition?

⚽Ki a hibás, ki a felelős? Amorim és a vezetőség felé is sok a kérdés, annyi bizonyos, hogy a kapcsolat visszafordíthatatlanul megromlott. (0:00) Kvíz (4:35) Hogyan építsünk fel egy klubot? (7:30) Üvöltéslánc (10:30) A Man United kijelölt útja (12:45) Menedzser vagy vezetőedző? (16:15) Hiányzott a rugalmasság Amorimból (21:25) Mindennel próbálkozott már a United (26:00) Miben javult a United Amorimmal? (28:30) Kvíz megoldás (30:30) Kinek jó hír Amorim távozása? (33:45) Jó igazolásokat kapott Amorim? (37:30) Párhuzam a Chelsea-vel (39:35) Maresca és Amorim jövője (42:30) A United jelöltjei

Kvėpavimas – tai vienas natūraliausių dalykų mūsų gyvenime, bet vis dažniau girdime, kad kvėpuoti galima ir „teisingai“, ir „neteisingai“. Ar iš tiesų mes visi mokame kvėpuoti? Kuo sąmoningas kvėpavimas skiriasi nuo įprasto, automatinio? Ir ar kvėpavimo įpročiai gali turėti įtakos mūsų savijautai, energijai, streso lygiui ar net atsparumui ligoms? Apie tai LRT RADIJO laidoje „10–12“ kalba kineziterapeutė, kvėpavimo trenerė Greta Puzonė.

Sausį, jau penktus metus iš eilės, vyksta iniciatyva, kviečianti išbandyti augalinę mitybą. Šiemet augalinio maisto naudomis besidominčią bendruomenę, kurią sudaro daugiau nei 35 tūkstančių entuziastų, organizatoriai kviečia sudalyvauti net keturiose nemokamose edukacinėse programose.Kvėpavimas – vienas natūraliausių dalykų mūsų gyvenime, bet vis dažniau girdime, kad kvėpuoti galima ir „taisyklingai“, ir „netaisyklingai“. Ar iš tiesų mes visi mokame kvėpuoti? Kuo sąmoningas kvėpavimas skiriasi nuo įprasto, automatinio? Ir ar kvėpavimo įpročiai gali turėti įtakos mūsų savijautai, energijai, streso lygiui ar net atsparumui ligoms?Klausantis muzikos, kartais gali nutikti taip, kad girdima daina primena kitą kūrinį, tarsi daina ar atskiri jos fragmentai kelia panašias asociacijas su kita daina. Kada kūriniai yra tik panašūs, o kada tai gali būti traktuojama kaip plagiatas? Tokių pasvarstymų ar net kaltinimų yra pasitaikę ir didžiojoje Eurovizijoje.Šiemet sueis 225-eri metai, kai gimė žemaičių vyskupas Motiejus Kazimieras Valančius. Primindamas tai, 2026-uosius Seimas paskelbė Valančiaus metais. Pasakojimas iš Varnių, kur Valančiui teko ir gyventi, ir dirbti kunigų seminarijoje. Ten likę nemažai šios ryškios asmenybės pėdsakų.Ved. Ignas Andriukevičius

VIVALDI: Concierto para violín, cuerda y continuo en Mi bemol Mayor, Op. 8 nº 5 RV 253 “La tempesta di mare” (8.41). A. Harnoncourt (vl.), Concentus Musicus Wien. Dir.: N. Harnoncourt. MOZART: Sinfonía nº 25 en Sol menor, KV 183 (26.21). Concentus Musicus Wien. Dir.: N. Harnonocourt. BACH: Allemande (Suite para violoncello solo nº 1 en Sol Mayor, BWV 1007) (5.11). N. Harnoncourt.Escuchar audio

Paměť národa na konci roku zaujala na sociálních sítích třemi výjimečnými příběhy: Květy Bartoňové, Miroslava Hampla a Františka Suchého. Tzv. vánoční kampaň Paměti národa měla nebývalý úspěch, dosah statisíců tisíců lidí. Co mají tyto příběhy společné?

Paměť národa na konci roku zaujala na sociálních sítích třemi výjimečnými příběhy: Květy Bartoňové, Miroslava Hampla a Františka Suchého. Tzv. vánoční kampaň Paměti národa měla nebývalý úspěch, dosah statisíců tisíců lidí. Co mají tyto příběhy společné?Všechny díly podcastu Příběhy 20. století můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Bedétlen ünnepi különkiadás, számozatlan, fríííííí: 00:30 A „Nőtlen tiszti” spontán formátum és az olvasónő. Hugo Johnson. Ál-marokkói AI-psychrock. 03:30 Kvíz 1, és a Zambia elleni diadal. Vitár Róbert emlékezete. 05:20: El Kaabi elkábította a rabati stadion közönségét. 10:00 Horgoló reality a Channel 4-n. 11:00 A Lumumba-imitátor. Kutyaherényi marokkói szappan. 14.00 Hogy énekelte föl a csordavokált három szólamban Winkler a második Moby Dick-lemezen? 15:00 Michael Monroe, Hanoi Rocks. Az igazi neve: Matti Antero Kristian Fagerholm, 18:00 Nigériából kéne hozni egy Orbánt. 20:00 Orbán országában már hiánycikk a Kinder Joy. A lakosság harácsol. Stranger Things. 23:00 Patria. (Másfél évvel ezelőtt már volt róla szó.) Nem hat rész valójában, hanem nyolc. Bede Márton cikke. 27:15 Espelette-i paprika, Capsicum annnum Baszkföldön 28:30 A Winkler-féle chiliszósz titka. 31:00 Miről érkezett a legtöbb olvasói levél 2025-ben? 32:15: A WD-40 törvény. 35:15 Kvíz: király csehül. (Csak az ország neve nem hangzik el. Egyébként – szpojler!!! – : Kambodzsa. 37:45 Amikor Bruno Cuccinelli a Sex Actionben basszgitározott… 42:00 A kanadai olimpikon droglord luxusmotorgyűjteménye, 45:00 VV Aurélió, az orbánista, rendpárti exdrogdealer, aki aranyköpéseiről lett híres, aztán összetűzésbe került a törvénnyel, édeasapja szerepét is élvezi. 47:50 A történelmi faszhelikoterezés. 50:00 Belső sávban ragadt tötymörgők az Egyesült Királyságban. 52:45 Tóth-Hódi Pamela és a gépi falvakolás. 63:00 Kézben tartott mobil. 65:00 Idegösszeroppanás sok kurvaanyáddal. 70:00 Vízkiöntésre alkalmatlan edények. 72:00 Szomjasak a madarak. 82:00 Ki indul Újlipótban? 85:00 Új kamupárt és sminkfilc. 86:00 Winkler újraéleszti az SZDSZ-t: Szédületes Dudák Szegeden. Egy kulcstartónk már van. 90:00 Milyen motorja van Seres Máriának? Kitelepítés. 94:00 Torxkulcs a kormányban. See omnystudio.com/listener for privacy information.

Kvällspasset önskar Gott Nytt År med favoritsamtal från de tio år som gått! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.12-årige Oscar blir glad av motocross! Ville feldoserade klorhalten i badtunnan så att gästerna tappade allt hår och en liten varning till er som tänkt skåla i champagne ikväll: håll hårt i glaset så att det inte blir stulet av en fiskmås!

K Vánocům patří hudba, stejně jako k životu Aleny Terezie Vítek, nevidomé maminky samoživitelky. Právě díky ní dokáže rozdávat radost i těm, kteří ji potřebují nejvíc. „V rámci projektu Spolu s vámi se věnuji seniorům, za kterými jezdím a zpívám jim lidové písničky. Ale sama mám ráda klasickou hudbu, ke které mě přivedla na konzervatoři moje paní profesorka zpěvu,“ vzpomíná maminka malé Šarlotky v rozhovoru s moderátorkou Terezou Kostkovou.Všechny díly podcastu Blízká setkání můžete pohodlně poslouchat v mobilní aplikaci mujRozhlas pro Android a iOS nebo na webu mujRozhlas.cz.

Vem vill du skicka en julhälsning till? Vi passar på att sprida lite extra värme så här dan före dopparedan! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Caroline vill skicka en hälsning till sonen och alla andra lastbilschaufförer som rullar på vägarna i juletid och Elå ringde från Tyskland hälsar till bästisen i Sverige.God jul önskar vi på Kvällspasset!

Julstressen har ibland en tendens att leda till tankspriddhet, glömska och hjärnsläpp. Vi pratar om alla gånger när det gick lite för fort! Lyssna på alla avsnitt i Sveriges Radios app. Ett nyfiket och underhållande aktualitetsprogram med lyssnaren i fokus.Sara råkade ge bort en museiresa till sitt ex, Mårten glömde tomtemasken och drog på sig en nylonstrumpbyxa istället och Harry deltog i en fårvallningstävling som bokstavligen gick ”skit”!Vi hör också Hugo som förväxlade vigvatten med handsprit och i extramaterialet säger vi hejdå till vår underbara kollega Yasmine som gör sin sista dag på Kvällspasset.

Filmbarátok Podcast #316 (December 2025) 318 perc Beszélgetnek: Sorter, Blacksheep, Gergő, freddyD Téma: -Felvezető (00:00:00) -Borítókép (00:10:18) -Villámkérdés féleség (00:21:05) -Zootropolis 2 (00:29:55) -Christy (01:02:27) -Üvegtigris (01:32:05) -Karácsonyi rovat felvezetés (01:57:48) -Aquaman (01:59:11) -Amerika kapitány (02:12:40) -Hamu és hó (02:29:25) -A Jackass bemutatja: Rossz nagyapó (02:44:48) -Egy ropi naplója (02:57:35) -Halálos iramban: Ötödik sebesség (03:14:00) -Perfect blue (03:32:55) -Strange Darling (03:53:16) -The Clovehitch Killer (04:02:10) -9 - A szám hatalma (04:13:00) -A Simpson család – 6 rész (04:25:35) -Terrifier 2 (04:53:00) Csatolmányok: Kvízest Márkkal (dec. 19) https://www.facebook.com/events/1443365101127623