与「MT夜会」相关的搜索结果

◤◢◤◢◤◢◤◢◤◢◤◢◤◢ 始まりましたーーー🎉🎉🎉 The Dark side of Japan Night Meeting 2025 ◤◢◤◢◤◢◤◢◤◢◤◢◤◢ MTオーナーのみなさんの思いが通じて 会場・CITY CIRCUIT TOKYO BAYには陽射しが差し込んでいます‼️ ✨19:30から集合写真撮影しまーす✨ ￣￣￣￣￣￣￣￣￣￣￣￣￣￣￣￣ ╭━━━━━━━━━━━━━╮ 　🌃イベントは21時まで 　安全運転でお越しください ※会場に四輪駐車場はありません ╰━━━━━━ｖ━━━━━━╯ 【展示車両】 MT-10／MT-09／MT-07／MT-07 Y-AMT カスタム／MT-03 Touring／MT-25／MT-125 【グッズ販売】（税込） ◎オリジナルTシャツ：5,500円 ◎DSOJキャップ：4,400円 【事前エントリーされたMTオーナーさんへ】 会場チェックインでオリジナル蓄光パスをプレゼント‼️ #DSOJ# #日本のダークサイド# #MT夜会#

5月10日【The Dark side of Japan Night Meeting 2025】が開催されるCITY CIRCUIT TOKYO BAYからはしれ！なっちゃんねる生配信を行います💙 時間等の詳細は後日発表！ チャンネル登録して待っててね🥰 #日本のダークサイド# #MT夜会# #DSOJ#

#AKB48春コンサート2024# 昼、夜 ありがとうございました！✨️ もう15期が2番目に先輩になって、リハ期間中も振り付けを教えたりするのに(M.T.公演とか)責任を感じる瞬間が沢山ありました。新しいAKBと一緒に自分も成長したいと思ったコンサートでした🥹🫶🏻

今日M.Tに捧ぐ公演に来て下さった皆さんありがとうございました😊 夜は大家さんの生誕祭でした🎉 ファンの皆さんが凄く盛り上げて下さって本当に楽しかったです❣️ コールも練習して下さったのかな…皆さんの暖かい気持ち凄く伝わりました😆とても素敵な生誕祭でした✨ また明日ー⭐︎ #浅井七海# ©︎

夜公演ありがとうございましたっ！ めぐちゃんの生誕祭でした✨ 「M.T.に捧ぐ」の曲中のめぐちゃんコールに凄く感動してウルっときた😭 ２ショット撮ろうとしたら後ろにいずちゃんがいたので３ショット😂💕笑

M.T.に捧ぐ お昼公演ありがとうございました❤️ ハロウィン前夜祭ということで お客さんも仮装してて皆さん可愛すぎた！夜公演もしくよろです(*^o^*) うさぎちゅん。

Welcome Havenex. Series A is underway and closing soon, already applied for all required licenses (even more than required). DM me if interested to invest, note that allocation is already quite packed. After extensive discussions with regulators, central bank governors, Financial Market Authority leadership, banks, family offices, exchanges, wallet providers and custodians, my mission became very clear: Build the most transparent, safest, institutional-grade, fully regulated exchange possible, with continuous, verifiable proofs of solvency. Not just web3. Havenex is not trying to become another Coinbase, Binance, Bybit or Kraken. The focus is different: infra that allows financial institutions to offer digital and traditional financial assets to their customers, while meeting the standards they expect around regulation, custody, security and transparency. My principles are simple: 100% multi-chain. Verifiable custody. Continuous solvency proofs. Multi-sig by default. Quantum-safe keys. Hardware 2FA wallets. Confidential and RWA assets wherever regulation allows. Unique self-custody and key-loss protection mechanisms. Some of the best engineers and experts in cryptography, exchanges and privacy-preserving technology are joining the effort. Havenex will use Sui tech wherever it makes sense, but it will also integrate the best primitives, assets and bridges from other ecosystems. I'm personally helping Havenex as an advisor, although it was my idea. Mysten Labs and Sui remain my focus, nothing changes there. Chief & Hacker team Officer, as always, innovating at daily basis :) As all of you know since my Satoshi days, my goal has always been bigger: help crypto meet regulation without sacrificing ownership, transparency or security, while protecting users against malicious and shady activity and giving the best ecosystems room to thrive. We cannot keep accepting another FTX or Mt. Gox as the cost of doing business, nor the silly, insane bugs driven by LLMs lately. Havenex intends to set a different standard. The most transparent effort in regulatory-friendly crypto. You have my signature.

1/ gm. gm. Name is Hsin-Ju. I have a dark, kinda tragic personal announcement. I've decided I'd rather take $0 than accept a settlement that requires me to stay silent about what happened to me. I have since fired my lawyers at @sanfordheisler & will be releasing all the evidence from my time @hack_vc on Wed 8/26. INTRO: I've been in crypto for 9 years. Most recently I was a partner & head of platform @hack_vc which is a $600-700M crypto VC firm. Many of you may know me from my time as Head of Growth at @StellarOrg in 2017 (worked w/ the Mt. Gox founder), @Solana in 2018 (joined under ~10 ppl, but quit before my cliff), & @Fhenix in 2024. Or have attended one of my @DystopiaLabs ETH events (2019-now). Last year, I was forced to work through a serious medical emergency while at @Hack_VC due to threats by @alpackaP @dbulaevsky, saying that they would blacklist me from the space if I quit before finishing the hacksummit conference & side events that I was helping with during @kbwofficial. I was working 13-14hr days, 6 days a week, sleeping 0-4 hrs a night, and had documented Graves' disease, hyperthyroidism, and severe insomnia. I repeatedly asked to quit. Hack VC refused. This + other threats and intimidation went on for almost a month. The conditions got so bad I attempted suicide. The fucked up part is that, even after the suicide attempt, I told everyone at Hack VC (in writing) that I had no plans to sue. I was scared & extremely sick. I just wanted to leave. I told them that as long as there was no retaliation or badmouthing - I had no plans to sue. @Hack_VC chose to retaliate. After just 1 month after my near suicide, Hack VC mishandled my COBRA health insurance, refused to fix things, kept shooing me to customer support (when it's legally Hack VC's responsibility to fix this), and my insurance had issues and was not fixed for almost 4 MONTHS. Hack VC did not fix everything until I hired lawyers and had them force Hack VC to adhere to the law & fix my COBRA insurance issues. The sheer cruelty of an employer harming someone's health insurance almost immediately after they nearly committed suicide is extremely fucked up. And, even during the legal engagement, Hack VC continued to lie & attempt to harm me through their lawyers and other employees/partners (multi-million dollar carry & clawbacks can make even the nicest ppl willing to turn the other way or lie). @0xRodney @IsTheBaron @roshunpatel @Alex__Botte. Now. Almost an entire year of engagement & my near suicide anniversary coming up, our lawyers (mine & Hack VC's) have reached the point of preparing for private mediation and settlement. And...I realized that I just can't do it. I don't want money in exchange for silence. I'd rather speak. I fired my lawyers yesterday. In case it needs to be said, I did a good job at Hack VC. I was only there full time for 7 months; been part-time since 2021. I got a $50k raise to $300k, $30k bonus, and promises of additional carry that would have been implemented about a week or so after my last day. I was also offered admission into the GP entity (but not sure if admission was actually completed bc it was during my medical emergency). This + my resume, should show that me suing is not a money grab; I was genuinely harmed. I'm aware that I'm just one person (who is not rich) facing a $600-700M VC institution. I've been told by @alpackaP that he's known most of the managing partners at most VC firms for almost a decade and he could absolutely blacklist me. I'm also painful aware that the conclusion of this will probably be extremely poor for me or death - but I've accepted it. Sharing with other web2 & web3 VCs: @paulg @ycombinator @500STARTUPS @a16zcrypto @PanteraCapital @dragonflyvc @RaceCapital @hosseeb @haunventures @hiFramework @ElectricCapital @blockchaincap @Delphi_Digital @variantfund @1kxnetwork @archetypevc @shimacapital @robotventures, etc. TLDR; I’m done with continued threats, pressure, and intimidation. I'm not going to take a payout and allow Alex or Hack VC to walk away from harming me. As mentioned, I'll be dropping the evidence from my case on Wed 8/26 (or before) -- privacy be damned. Support here (or DM) please: ETH: 0x68c10776C5c05Cbf5B4C2318bE02D61B9f06B875 SOL: B7S8TPcWWJXTsmgG8ShoTmy8sNPuJ3Dbmt7z5gLFAveq

Thoughts About Scaling Law Scaling, but not only of parameters. Every model release now ends with the same question: how many parameters? It isn't a question that can be answered on its own. Parameter count is only meaningful alongside three others — how much data you have, where you intend to spend your compute, and who will run the model, under what conditions. The field learned this the hard way. Kaplan et al. (2020) fit an exponent that told everyone to grow parameters faster than data — roughly 2.7:1 — and the industry complied: GPT-3, Gopher, MT-NLG. Hoffmann et al. (2022) redid the experiment across four hundred models and found the compute-optimal split is closer to 20 tokens per parameter, and that with sufficient compute the two should grow at the same rate rather than drifting apart. The error in the earlier fit compounded with every order of magnitude of compute, which is why the largest models of that generation were the most misallocated. The trillion-parameter round was, in retrospect, a detour the whole field took together and then reversed. Chinchilla wasn't the end either. It optimized training compute for models that would be trained once and evaluated. Today a model is called billions of times a day and inference dominates lifetime cost. Put inference into the objective and the optimum moves toward smaller models trained far longer — deliberate over-training, which is what Llama-2-7B and Gemma-2-9B were doing at roughly 290 and 889 tokens per parameter. Sparsity moved the target again. In a MoE model two quantities have to be kept apart: total parameters govern roughly how much the model can hold — knowledge, facts, the long tail — while activated parameters and effective depth govern roughly how far it can think, how many steps of a causal chain it can carry before it comes apart. A dense 20:1 ratio does not transfer. And the ratio isn't a single number at all: Roberts et al. (2025) find the optimal tokens-per-parameter is task-dependent, with memorization favoring more parameters and reasoning favoring more data. Follow-up work on MoE observes that at fixed TPP, pushing total parameters higher actually degrades reasoning, while activating more experts reliably helps it. This matters for what we are building toward. Finding a vulnerability is not a retrieval problem. It doesn't come from having memorized more CVEs; it comes from carrying a twenty-step chain of inference to the end without losing the thread. That capability does not live in total parameter count. Which brings us to this release. Total parameters appear to matter up to a threshold — enough to hold the world — after which additional capability comes from scaling elsewhere: effective depth per forward pass, and above all post-training. GLM-5.3 is our controlled experiment on that claim. Same base, same architecture, same total and activated parameters as GLM-5.2. One month of scaling long-horizon environments and RL. The gains are not marginal. Well, scaling has more than one dial. We turned the post-training one this time because it had the most slack left in it — not because the others are finished. Base model size, pretraining data, compute spent per forward pass: all of them are still on the table, and we will come back to each. What this experiment taught us is that the dials do not have to be turned together, and that the one worth turning next is rarely the one that was worth turning last. We are not done scaling. Next time, maybe mid-training, pre-training, and even more.