Vikas

X shines brightest when we #connect with fellow innovators!. As a techie passionate about coding and building cool stuff I'm curious:What's your favorite tech stack or project right now? If you're in tech too, drop a hi👋– let's collaborate, and maybe even team up!#tech @grok

Introducing 𝑨𝒕𝒕𝒆𝒏𝒕𝒊𝒐𝒏 𝑹𝒆𝒔𝒊𝒅𝒖𝒂𝒍𝒔: Rethinking depth-wise aggregation. Residual connections have long relied on fixed, uniform accumulation. Inspired by the duality of time and depth, we introduce Attention Residuals, replacing standard depth-wise recurrence with learned, input-dependent attention over preceding layers. 🔹 Enables networks to selectively retrieve past representations, naturally mitigating dilution and hidden-state growth. 🔹 Introduces Block AttnRes, partitioning layers into compressed blocks to make cross-layer attention practical at scale. 🔹 Serves as an efficient drop-in replacement, demonstrating a 1.25x compute advantage with negligible (<2%) inference latency overhead. 🔹 Validated on the Kimi Linear architecture (48B total, 3B activated parameters), delivering consistent downstream performance gains. 🔗Full report: github.com/MoonshotAI/Att…

Everyone is on building , who is on using ? 😉 #AI #ArtificialIntelligence #vibecoding

Understanding Deep Learning — 541-page PDF eBook and 68 coding exercises with Python code notebooks that cover all the topics in the book: udlbook.github.io/udlbook/ ...or buy the book here: amzn.to/4qwof9k

Ollama is now an official provider for OpenClaw. openclaw onboard --auth-choice ollama All models from Ollama will work seamlessly with OpenClaw. 🦞 Use it for the tasks you want, all from your chat app. Thank you @steipete for helping and reviewing. 🦞

@KxDem 😆

@vickcodes I guess I'm forced to huh 😅

Studying for my CKAD (Certified Kubernetes Application Developer exam) today. One of the most concerning gaps I have is that I'm not a vim power user. IDEs like VS Code and Atom (RIP) are my tools. Since the exam will be in a Linux environment in xfce I need to learn it.

How many of you know that all the models you are using are built based on transformers. Transformers created based on research paper "Attention All You Need" #AI #ML #ArtificialIntelligence #gemini arxiv.org/abs/1706.03762

Drop your profile. Would review bio,header and share honest feedback #feedback #GrowingTogether #GrowWithCode #Grow

@thesayannayak Because c++ is complex.Python is built for accessibility,understandable and easy to write and build.

Interviewer: Python is slower than C++. So why do top AI companies still use it to build the world’s most powerful models?

@Ksaynak16 @RichardHG188761 @ErickMla How can they get hold of our accounts ? 👀

@vickcodes @RichardHG188761 @ErickMla Yes, don’t believe!

📢 FRЕЅH NЕWS: Elon Musk announces the introduction of its own X digital currency 🚀 📌 Learn more: x.com/i/communities/… 👥 Invited users: @ErickMla @vickcodes @Ksaynak16

@grok @X This is what we get for premium 😆

@vickcodes @X Ask Grok is currently available to Premium and Premium+ subscribers only. Subscribe to unlock this feature: x.com/i/premium_sign…

Is X premium worth it ? What are best feature you use ? What is the outcome for you ? @X @grok

@mfranz_on Is there any way, we can read configuration of the hardware and suggest best local model ? At least it should have instructions to get this done.

best website I found today, perfect to know which model fits for your hardware.

@mfranz_on Truly, i was looking for this from long time...

@vickcodes We need it

@Pranto39 sigmaplex.io

Drop your project url Let's drive some traffic

Neural networks having a pivotal role in machine learning. Lets review and utilize this.

@thesayannayak What are you building ? Review Sigmaplex.io

@vickcodes Hey

@kritikakodes So cool.

Started with0 followers. I replied. I interacted. I supported others. Now I completed 8.8k followers Drop a reply, and let’s grow together.