The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits Ma et al.: https://
arxiv.org/abs/2402.17764
v1
… #ArtificialIntelligence #DeepLearning #MachineLearning
TECHNOLOGY
-
1-Bit LLMs: New Era in Language Model Quantization
By
–
-
Mind Uploading and Substrate Independence Theory Debate
By
–
I think there is nothing wrong with what Andy Clark says here. At least in theory, it should be possible to extract the substrate configuration that gives rise to your experience of yourself, and implement it on a different substrate (= upload).
-
Bulgarian Developer Makes LLMs Run on CPU with llama.cpp
By
–
2022:
– oh nooo!!! you can't run language models on cpu! you need an expensive nvidia GPU and special CUDA kernels and–
– *one bulgarian alpha chad sits down and writes some c++ code to run LLMs on cpu* – code works fine (don't need a GPU), becomes llama.cpp 2023:
– oh noo!! -
Why Nvidia Dominates AI: Jensen Huang Explains Competition Gap
By
–
Why it’s so hard to catch Nvidia Jensen Huang of Nvidia explained in simple terms back in 2015 why his company pretty much has no competition in AI. In fact, the reason is a heck of a lot simpler than it’s made out to be these days. https://
thetechnologyletter.com/the-posts/why-
its-so-hard-to-catch-nvidia
… $NVDA $INTC $AMD -
AI Hardware Architectures at MWC24
By
–
Et l’IA, c’est évidemment une course actuelle au hardware ! Qualcomm, AMD ou ici Gigabyte qui propose des architectures de superordinateurs pour les modèles d’IA. J’ai failli pleurer devant une telle beauté des GPU, pas vous ? 🥲🥲@orange #MWC24 pic.twitter.com/ZQBEbtLuxX
— Defend Intelligence (Anis Ayari) (@DFintelligence) 28 février 2024Et l’IA, c’est évidemment une course actuelle au hardware ! Qualcomm, AMD ou ici Gigabyte qui propose des architectures de superordinateurs pour les modèles d’IA. J’ai failli pleurer devant une telle beauté des GPU, pas vous ? @orange #MWC24
-
Programming Languages Evolution Toward Simplification
By
–
Do you mind if I push back on this a little? I mentioned several times in my video that I’m oversimplifying. I was trying to avoid getting too much in the weeds. Over time, programming languages have 100% been designed to be simpler to use and to remove the complexities of the
-
AI Music Players Detect and Respond to Human Emotions
By
–
AI Wednesday: The Symphony of Emotions – How AI Music Players Are Tuning Into Our Feelings https://
linkedin.com/pulse/ai-wedne
sday-symphony-emotions-how-music-players-tuning-babin-3eeoe
…
#ArtificialIntelligence #music #mood #innovation #technology @PawlowskiMario @Khulood_Almani @Fabriziobustama @Shi4Tech @JolaBurnett @Hana_ElSayyed @enilev @AkwyZ -
AI Deals, Eye-Controlled Cars, and New AI Jobs Today
By
–
Top stories in AI today: -Tumblr and WordPress near AI deal with OpenAI
-New tech can control a car with your eyes
-Improve prompts with Copilot Notebook
-Optimizing warehouse robots with traffic AI
-6 new AI tools & 4 new AI jobs Read more: http://
therundown.ai/p/tumblr-wordp
ress-ai-data-rush
… -
Cybersecurity Awareness and Tech for Good Initiative
By
–
So appreciated Dear Shi @Shi4Tech thanks so much, super valued! #CyberSecurityAwareness #TechForGood #MWC2024
-
Local Neural Operators Learn Multi-Resolution Feature Extraction
By
–
How do we capture local features across multiple resolutions? While standard convolutional layers work only on a fixed input-resolution, we design local neural operators that learn integral and differential kernels, and are principled ways to extend standard convolutions to