与「RRules」相关的搜索结果

Samahan natin ang #MNL48# and #RRules# this June 2 sa Tugatog Thursdays on Kumu! 🥳 May chikahan, performances, at ticket giveaways na nakahanda, so download Kumu na and follow ANT TV: Kitakits #MNLoves# and #RRulers#! 😘

A wrong summary is fixed in a second. A wrong order is money gone, and no explanation brings it back. So let the model research, screen and explain, and put fixed rules on the one step you cannot take back. Irreversibility changes the price of being wrong, and it should change where you draw the line, not just the wording of the prompt.

It took two weeks to end the career of the most successful Asian American in history. Jensen washed dishes at Denny‘s as a teenager, and by this year he was running the world’s most valuable company. This week a lot of Asian people on my timeline decided that story is over. That no matter how high you climb, a face like ours gets a different set of rules at the top. After Elizabeth Holmes, female founders spent years getting asked some version of “how do we know you’re not the next Holmes?“ One woman‘s fraud turned into a question every woman had to answer. I’m Chinese Canadian, I grew up in Vancouver, and now I live in the Bay Area. I already know what my version of that question is going to be and it‘s pathetic. Jensen was born in Taiwan and spent over thirty years building an American company. The second the word ”China“ landed next to his name, none of that counted. This country has done this before. In 1882 it banned Chinese immigrants. In 1942 it locked up 120,000 Japanese Americans, most of them citizens. In 1982 two Detroit autoworkers who blamed Japan for their jobs beat Vincent Chin to death with a baseball bat. He was Chinese. His killers never spent a day in prison. Yellow peril never checked which Asian you were, and that’s a big part of why “Asian American“ became one word. If they were going to lump us together anyway, we might as well stand together. Now look at Jensen‘s timeline. Oct 29, the spy story breaks. The next day Singapore’s government says it never said what the Wall Street Journal claimed. Nov 8, he admits an affair. Nov 12, he‘s gone. The spy part is still unproven. Two weeks start to finish, right before what could be the biggest IPO in AI. You can call that a coincidence. I don’t buy it. Meanwhile the powerful people who raped CHILDREN in the Epstein files are mostly still living their best life Jensen had an affair with an adult woman. That’s the only thing of what‘s actually been proven. So what exactly did Jensen do that was worse than them? Nothing. He was the best, he was Asian, and he was on the wrong side of the political fight. He did everything right. He built the company America needed most and showed up to every dinner in Washington. On Sunday night the president himself stood up for him and called him “an honorable guy.” It didn’t matter. By Thursday he was gone. If you can take down the president‘s favorite CEO in four days while the president is defending him, think about who that message was really for. They’ll use you for thirty years and still never count you as one of them. That’s what a political weapon is. You’re a tool while you’re useful. When the game gets serious, you’re the knife. And when it’s over, nobody comes back for the knife. Jensen was the best one they had. That's exactly why he made such a good "kitchenware".

I am actually disgusted by the timing of this drop. And this will get a bit more technical. Let’s do a real technical teardown of this "1,385-sigma Blackwell weight fingerprint" report that OpenAI and Anthropic are breathlessly promoting today, because banking on nobody reading past the headline is how DC narratives get manufactured. ​First, look at the actual CUDA maths: when PyTorch initialises model weights (torch.nn.init) with a fixed PRNG seed, curand/Philox parallelises random number generation across the GPU’s Streaming Multiprocessors (SMs). Because Hopper has 132 SMs and Blackwell has 148 or 152 SMs, the exact same seed lays out initial weights in a different spatial pattern depending on the GPU’s SM count. In huge MoE models, certain low-gradient null-spaces or unrouted slices preserve faint directional traces of those step-0 weights even after training. ​So yes, brute-forcing the seed and SM grid can show that a model’s step_0 weight initialisation script ran on a 148/152-SM Blackwell GPU. Now read what the summary explicitly admits in the fine print: 1. Initialisation is not full training: The report literally states that these signatures show how the weights were created, and "support (but do not prove) the inference that Blackwell also participated in model training". Initialising a model checkpoint takes seconds on a single rented dev node before loading the safetensors onto a standard Hopper cluster. The "2,000 to 3,000 Blackwell GPUs" and "$500 million" cluster numbers do not come from the weight fingerprint and Depue admits in his own thread that he simply "estimated" cluster sizes from release schedules and model details. ​2. Zero geolocation and zero proof of smuggling: The final sentence of the executive summary explicitly admits that the findings "do not establish the physical location of the chips, how they were acquired, or whether export control regulations were violated". Depue himself rented a cloud GB200 instance just to run his verification test. Commercial labs that are not on the Entity List renting bare-metal Blackwell nodes in offshore clouds (Singapore, Japan, or the Middle East) is completely legal under BIS rules and leaves the exact same SM fingerprint. Now ask yourself why a former OpenAI researcher dropped this on November 12, the exact day Jensen Huang was pushed out, with OpenAI and Anthropic immediately stepping in to "independently verify" it. US closed-source labs are watching their paid API moats get eaten alive by free Chinese open-weights models. Conflating a step-0 PRNG initialisation seed with a "$500M smuggled Blackwell cluster" hands DC hawks the exact pretext OpenAI and Anthropic want to regulate global cloud compute and kneecap open-source AI.

In 2014, Elon Musk stood in front of an audience at MIT and told the whole world exactly where AI was going. Everyone missed it. Because he used one word, and that's the only thing the internet remembered: "demon." The headlines called him scared of killer robots. Tech blogs called it hyperbole. People moved on. I went back and watched that moment. It's barely 2 minutes long. You'll never hear the word "AI" the same way again. Here's what I took away. 1. He called it before almost anyone. Someone in the audience asks about AI. His face goes serious. "If I were to guess like what our biggest existential threat is, it's probably that. So we need to be very careful with the artificial intelligence." This was 2014. No ChatGPT. Most people had never heard the word "chatbot." 2. He asked for rules 12 years ago. "I'm increasingly inclined to think that there should be some regulatory oversight, maybe at the national and international level, just to make sure that we don't do something very foolish." National AND international. In 2014. Governments are still arguing about it today. 3. The demon line wasn't a joke. "With artificial intelligence we are summoning the demon." And then the part almost nobody quotes: "In all those stories where there's the guy with the pentagram and the holy water, it's like, yeah, he's sure he can control the demon. Doesn't work out." Read that again. His warning wasn't about the demon. It was about the guy who thinks he can control it. 4. Now here's what got me. In 2014 he said: summoning the demon. In 2018 he told Joe Rogan: "Nobody listened." In 2023 he signed a letter asking AI labs to pause. In 2026 he told The Economist he can't see any way to stop it anymore. His new philosophy? "Let's enjoy the ride." And: "If you can't beat them, join them." Today he's building one of the most powerful AIs on Earth. 12 years later, the man who warned us about the guy with the pentagram is holding one. My takeaway: He wasn't wrong in 2014. He just realized nobody was going to stop it. So he decided to build it himself. Whether that's genius or the exact ending of his own story, I honestly don't know. So now I'm asking you: Was Musk right in 2014? Or did he become the guy in his own warning? Tell me below. I genuinely want to know where people land on this.

Your dot now has… ⚡️ Faster replies 💬 Less jargon 🫡 Better follow-through And, they can now follow your custom instructions. Tell your dot in the thread things like how to behave or which actions need your go-ahead. “always talk in lowercase.” “Show me a draft before sending any emails.” You can also set custom instructions and rules in Customize > Behavior on mobile, or through your dot’s menu on web and desktop.

Key takeaways on Nvidia CEO Jensen Huang’s resignation: 1. Near term, the bad news is priced in: For the stock market, this could be a one-time clearing event. Huang has now fully cut ties with Nvidia, so Congress summoning him and any follow-up investigations should have less impact on Nvidia’s stock from here. As for how Huang’s next moves might affect Nvidia, it’s too early to call before the investigation results are out. 2. Google’s TPU is likely the biggest beneficiary: With Nvidia facing succession uncertainty, Google’s TPU has a chance to win more trust from customers and the supply chain. Its key partners, MediaTek and Broadcom, could be the market’s near-term focus. 3. Succession impact shows up in 2028 at the earliest: CFO Colette Kress is now interim CEO. The formal CEO transition and the organizational changes that come with it will likely start affecting Nvidia’s product development and competitiveness in 2028 at the earliest. 4. Capacity pre-booking and spec upgrades are the key focus for supply chain checks: Huang’s product strategy pushed the supply chain hard in two ways: capacity pre-booking and spec upgrades. It’s critical to watch for any changes in capacity expansion and new tech development across the supply chain. These are among the core reasons Nvidia’s supply chain trades at high P/E multiples. 5. China demand and export controls: The share of Nvidia’s AI chip revenue that ultimately ends up with Chinese buyers may be higher than what the company has officially stated. That’s why some investors worry Nvidia’s revenue will drop sharply once the US government tightens export controls. But China’s end demand won’t disappear because of export controls. Even under tighter rules, as long as overseas resale channels exist, Chinese buyers may still pay the premium, even at several times the price, and get Nvidia chips/racks through indirect routes.

1confirmation has invested in every Polymarket round ever and we keep doubling down. We are a seed stage fund and investing late stage is not our mandate. Here’s why we broke our rules: - There’s a 100+ year history of people wanting skin in the game on things like politics and culture and the time to accelerate that is now in an AI world where the biggest problem young people have is knowing what’s true. Everyone is becoming a trader/investor, and a product that converts that energy into better information and more truth is truly unique. - Polymarket is much more than sports and global. In the U.S. right now sports is all the rage. A better sportsbook is fine but it’s not the most enduring aspect of a prediction market business. Better information and more truth on politics, geopolitics and culture is the why and most young people you talk to IRL know that Polymarket is in a class of its own on this front even though that discourse here can get lost. - The business is leveling up. I’ve seen a lot over the past 6 years and it’s far from perfect but the exec team in place now is exceptional, the US product is best in class and the revenue is making new highs. I have not loved the messaging about a token at times bc it’s always best to underpromise and overdeliver by quietly delivering value to users. But the idea that a token is dead is foolish. - Shayne has a will to win like none other and is the most direct and transparent person I know. He’s just himself and authenticity wins. Posting here for transparency. Talk soon.

这两天做了一个有意思的事情，把我的全部的 Skills 删除了将近 1/3，同时把所有项目的上下文也都优化了一下，感觉干净清爽了不少，我认为随着模型能力越来越厉害，我们有经常去更新裁剪我们的项目的一些给到模型的上下文信息，防止你的内容反而阻碍了模型的发挥。 于是我就把最近做的继续沉淀成一些大伙也用得上的提示词，可以直接给到你的 Agent 去帮你优化你的项目的内容要求和 Skills，已经安装 Waza 的小伙伴可以直接更新到最新然后说「帮我精简一下规则」就好，也可以 Copy 下面内容给你的Agent。 帮我精简这个项目给 Agent 的上下文和我装的 Skills，按下面的要求来，先整理出清单给我确认，确认前先不改任何文件。 1. 先盘点每轮对话都会加载的内容，包括根目录的 AGENTS.md / CLAUDE.md、全局规则、没写 paths: 的 .claude/rules 和每个 skill 的描述，看各占多少字，再看按需加载的部分，这里不少看着臃肿的内容可能不是写多了，而是放的位置不对。 2. 逐段给出处理方式，分成修、删、挪、改写、保留五种，挪的写清目的地，每条附一句原因，做成一张清单给我确认。 3. 逐句判断删掉后你的行为会不会变，把类似这种「先读再改」「跑测试」「注意边界」你本来就会做的描述删掉，拿不准的可以先保留找我确定。 4. 我们更加推崇是给 Agent 写目标，少些如何实现的路线，写清做完是什么样、怎么算完成、要什么证据，一个步骤删掉后你还能换条路做到同样结果就删掉，只有顺序本身是约束的才留下来，比如先验证再发布、先确认再删除这种需要保留。 5. 同一种要求说了三遍，换成一个你本来就懂的词语，比如「快、确定、开销小」可以写成 tight；「要简洁」这类口号换成一个更好理解的想想的词语比如「报告腔」，然后再配一组 NO / OK 例子。 6. 把对应内容放对位置，需要判断的进 skill，同样输入总是同样输出的改成脚本或 hook，项目命令和发布步骤进项目指南，跨项目的硬约束才放进全局规则，遇到「做 Y 之前先读 X」这类要求改成做 Y 时由 hook 把 X 注入进来更好。 7. skill 描述只写做什么、什么时候用、不用于什么，近义词移到 when_to_use，只有部分场景用到的内容移到 references；判断一个 skill 有没有人用，要看真实调用记录，不要只 grep 名字，要删的也先列给我看一下。 8. 这些内容不要删除： - 我的偏好、被否掉的方案和原因、机器环境。 - 不报错的坑，以及看着像出错其实正常的现象。 - 描述句里「从不写入」「不上传」这类半句约束。 - 删数据、需要花钱、要发布、提权、改写已推送历史这些不可逆操作的规则。 - 我们带日期的拍板决策记录。 9. 内容先移到对应正确的位置再来压缩，能整行搬到按需加载的文件里就整行搬，不要边搬边改写，不可逆的规则原样整段搬。 10. 把内容搬到能自己加载的地方可以参考这些技巧： - 提交、发布、回复、删除这类对话里随时会触发的规则留在根目录。 - 挪进 docs/ 的要在根目录留一句什么时候去读。 - 根目录不要建 CLAUDE.md，它会让 Claude Code 不再读 AGENTS.md。 - 项目里重复全局规则的句子，只有在我用的每个工具都读全局规则时才删。 11. 改完以后自己去查一遍：paths: 里的 glob 都要能匹配到真实文件，grep 一遍拆走的旧文件名，再列出所有消失的行，每一行都要对上清单里我确认过的删除。 12. 交给一个没参与精简的 Agent 查有没有删过度了，只给它消失的行和仓库，不给你的结论，让它搜关键名词而不是整句，意思在哪都找不到的原样恢复。 13. 验证移过去的内容真的会被加载：在新文件里埋一个暗号，开一个新会话，先读一个匹配的文件再问它暗号，然后读一个不匹配的文件做反向对照。 14. 最后全部做完以后，可以帮我沉淀一下这个思路，方便以后有模型升级的时候，我可以再来跑一遍。

As the Lee team, we are giving away MacBook Pro to 2 lucky fans! 🎁 No rules or conditions. Just like this post and comment "DONE" below. Winners will be announced tomorrow at 12:00 AM. Good luck! ☘️