🔥The 9th Round of Easy Loan, Earn $40 Reward is in progress❗️ ⏰ Promotion Period: January 15th - Feburary 15th, 2025 👉 Register now and check more details at gate.io/campaigns/358

"escape the matrix" Led by @Rudd80856040, we made LLMs go brrrrrrr (but efficiently). - 2.7B LLM with no matrix-multiplies - v efficient GPU implementation for training available on github - v v efficient implementation on FPGA @ 13W for human-readable throughput

This adds the missing bits from the 1.58 bits paper. The first company to propose a low-power cheap device with an hardcoded sota 100b+ model in 1.58bits combined with enough matmul for LoRas on top will make an absolute killing.

Well this exploded a little.

They built a 13 Watt (!) implementation of a billion parameter network running on an FPGA using this scheme.

If this is accurate, then NVIDIA's grip on the tech industry has just vanished. Matrix matrix multiplication (MatMul) is notoriously computationally difficult, which is why it's offloaded to GPUs. If MatMul can be avoided, then it's not just leveling the playing field. It's

While there are some conceptual similarities between the just released absolutely amazing paper "Scalable MatMul-free Language Modeling" which eliminates the regular Matrix Multiplication and the earlier released paper (in Feb-2024) "The Era of 1-bit LLMs: All Large Language

💡#BaskinEngineering Prof. Jason Eshraghian (Jason Eshraghian) & team developed an energy efficient way to run AI language models by eliminating matrix multiplication - which has important implications for the future of green computing. Via Ars Technica bit.ly/3VJq0BA

Telluride Neuromorphic Cognition & Engineering workshop time.

🚨📷We are organizing the 1st NeuroAI workshop at #NeurIPS2024! 📷neuroai-workshop.github.io Submission deadline: 7th September 2024

📢We're delighted to organise NeuroAI workshop at this year's NeurIPS Conference 2024 (excuse the typo in the graphic). Accepting submissions (4 pages + supp): neuroai-workshop.github.io Highlight: In-person panel with Yoshua Bengio and Karl Friston Follow NeuroAI Workshop @ NeurIPS2024 for details

An accompanying News & Views by Jason Eshraghian and colleagues is also available! nature.com/articles/s4358… ➡️rdcu.be/dQ8YF

🚨We have extended the paper submission deadline for our NeuroAI workshop NeurIPS Conference to 23:59, Sept 9 (AoE)! For more details and travel awards on offer: neuroai-workshop.github.io Nomination form for reviewers: forms.gle/sEAeHx1svWBgbS… Follow NeuroAI Workshop @ NeurIPS2024 for updates.

🚨NeuroAI researchers! ⏰5 days left for #NeuroAI workshop submissions! 🌟Awards: Spotlight papers: 1 student author gets free workshop registration, Top paper: 1 lucky student flies for free!✈️ 🔗submit here: bit.ly/3Z7YroY #NeurIPS2024 #CallForPapers

#Neuromorphic #computing just got more accessible! Our work on a Neuromorphic Intermediate Representation (NIR) is out in Springer Nature Communications. We demonstrate interoperability with 11 platforms. And more to come! nature.com/articles/s4146… A thread 🧵 1/5

It has been an incredible experience organizing the NeuroAI Workshop @ NeurIPS2024 workshop. Join us on December 14th at #NeurIPS2024 to explore the amazing research from the authors of our accepted papers, fusing Neuroscience and AI! For more details: neuroai-workshop.github.io

Excited to share that our paper Evaluation and mitigation of cognitive biases in medical language models was published in npj Digital Medicine In this work, we ask the question: can LLMs be easily fooled with simple cognitive bias-inducing prompts during patient diagnosis? To

🌟 Thrilled to be at NeurIPS Conference in Vancouver (Dec 9-15)! Don’t miss our NeuroAI Workshop @ NeurIPS2024 workshop on Dec 14 with an incredible lineup of keynotes. Stop by, join the discussions, and let’s connect! 🙌 neuroai-workshop.github.io

Excited for NeurIPS Conference in Vancouver (Dec 9-15)! Join us at the NeuroAI Workshop @ NeurIPS2024 workshop on Dec 14 for fascinating keynotes exploring the intersection of neuroscience & AI. Let’s chat, share ideas, and connect—see you there! 🙌 neuroai-workshop.github.io

A paper authored by #BaskinEngineering undergrads Ruhai Lin and Ridger Zhu has been selected for the Best Paper Award at the 17th IEEE International Symposium on Embedded Multicore/Many-core Systems-on-Chip (IEEE MCSoC Symposium) — they are advised by prof Jason Eshraghian. Congrats!

🥁 And the top #BaskinEngineering research story of 2024 is...? Developing a high-performing large language model (like ChatGPT) that runs on the energy needed to power a lightbulb! 💡 Led by Assistant Professor Jason Eshraghian. Read more: bit.ly/45AoNRJ