@kepano I just tried it this morning on the 245-page Mythos pdf and it failed badly and the outputs ... - 惯性聚合

推荐订阅源

Google DeepMind News

CERT Recently Published Vulnerability Notes

The Last Watchdog

LINUX DO - 热门话题

cs.AI updates on arXiv.org

Application and Cybersecurity Blog

cs.CV updates on arXiv.org

KPMG report finds enterprise disconnect between AI and its ROI | CIO

Security Archives - TechRepublic

TaoSecurity Blog

Heimdal Security Blog

Security Affairs

Lohrmann on Cybersecurity

Hacker News - Newest: "LLM"

Simon Willison's Weblog

WordPress大学

Security Latest

Apple Machine Learning Research

Engineering at Meta

阮一峰的网络日志

罗磊的独立博客

Full Disclosure

Schneier on Security

Know Your Adversary

Privacy International News Feed

Google Online Security Blog

Stack Overflow Blog

大猫的无限游戏

博客园 - 三生石上(FineUI控件)

The Register - Security

CTFtime.org: upcoming CTF events

博客园_首页

Fortinet All Blogs

The Exploit Database - CXSecurity.com

Hacker News: Ask HN

Andrej Karpathy(@karpathy)

Agree, it's beautiful, top tier fablemaxxing! :) I think with every new model tier there is somethin... Congrats!! I was impressed to learn about some of the engineering wizardry (e.g. *very* low voltage ... @leogau @kuroke01 @gallabytes Company Slack (I know, sad answer). Do you have recs? I’m not aware of... @kuroke01 @gallabytes I’ve been on Twitter for almost 2 decades now so I can say with confidence tha... This is correct, I think a number of people on the tl didn’t read past the title and made inferences... The basic idea is easy and v0 is a hackathon project. The product here is a lot closer to *it actual... This is a new paradigm for interacting with Claude that is significantly more "inline" with all the ... @josepha_mayo <-- working on the brains that glide our von neumann probes around, make contact, e... In awe of SpaceX and its story - past, present and the future. You can think about it in 10+ differe... This is a super exciting release - Claude Fable 5 is the same underlying model as Mythos but with ad... @yash1_ @shreyansj iirc Geoff Hinton’s official title at Google at one point was “intern” :D @shreyansj It’s refreshing to see a company of this size successfully call bs on the whole thing to ... x.com @jino_rohit I was recording my nanochat video when I realized that “first boot up an 8XH100 from you... This works really well btw, at the end of your query ask your LLM to "structure your response as HTM... @si_pbc @sonyatweetybird @MikowaiA @YasminRazavi @tszzl @_milankovac_ VPT (https://t.co/CSxHcXY6Vh) ... This is the the quote I've been citing a lot recently. Fireside chat at Sequoia Ascent 2026 from a ~week ago. Some highlights: The first theme I tried to ... @VBarsoum Haha yeah I call it cozy coding :) Usage: “This valentines, cozy code with someone you lov... Yes it's the tractable form of brain upload. There's a ton of scifi on brain uploads that requires w... Someone recently suggested to me that the reason OpenClaw moment was so big is because it's the firs... Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think ... Surprised with how good the comments on github gists are. A lot more helpful, insightful, constructi... Farzapedia, personal wikipedia of Farza, good example following my Wiki LLM tweet. I really like th... Something I've been thinking about - I am bullish on people (empowered by AI) increasing the visibil... @NirDiamantAI Peter Steinberger told me that he wants PR to be "prompt request". His agents are perf... Wow, this tweet went very viral! I wanted share a possibly slightly improved version of the tweet i... Oh and in the natural extrapolation, you could imagine that every question to a frontier grade LLM s... LLM Knowledge Bases Something I'm finding very useful recently: using LLMs to build personal knowle... @gvanrossum LLM = CPU (data: tokens not bytes, dynamics: statistical and vague not deterministic and... New supply chain attack this time for npm axios, the most popular HTTP client library with 300M week... - Drafted a blog post - Used an LLM to meticulously improve the argument over 4 hours. - Wow, feelin... When I built menugen ~1 year ago, I observed that the hardest part by far was not the code itself, i... (I cycle through all LLMs over time and all of them seem to do this so it's not any particular imple... One common issue with personalization in all LLMs is how distracting memory seems to be for the mode... Some links 2,112 total packages depend on litellm (https://t.co/T8xWdrBp9g), 1,403 of them directly... Software horror: litellm PyPI supply chain attack. Simple `pip install litellm` was enough to exfi... I'm not very happy with the code quality and I think agents bloat abstractions, have poor code aesth... Thank you Sarah, my pleasure to come on the pod! And happy to do some more Q&A in the replies. Andy Weir showing some of the spreadsheets underlying the calculations in the book https://t.co/3Cyp... Had to go see Project Hail Mary right away (it's based on the book of Andy Weir, of also The Martian... Ugh X breaks time links, it’s at 26:17 (link to blast from the past) https://t.co/gAP4YPV5uR Thank you Jensen and NVIDIA! She’s a real beauty! I was told I’d be getting a secret gift, with a hi... @Yulun_Du @ilyasut SGD is a ResNet too (the blocks of it are fwd+bwd), the residual stream is the we... My autoresearch labs got wiped out in the oauth outage. Have to think through failovers. Intelligenc... Human orgs are not legible, the CEO can’t see/feel/zoom in on any activity in their company, with re... All of these patterns as an example are just matters of “org code”. The IDE helps you build, run, ma... Expectation: the age of the IDE is over Reality: we’re going to need a bigger IDE (imo). It just lo...

@kepano I just tried it this morning on the 245-page Mythos pdf and it failed badly and the outputs ...

Andrej Karpa · 2026-04-10 · via Andrej Karpathy(@karpathy)

此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。原文来自 — 版权归原作者所有。