GLM 4.6 runs quite fast on an M3 Ultra with mlx-lm even at higher precision.
Pretty remarkable that it benchmarks competitive to the just-released Sonnet 4.5. Hope those benchmarks hold-up in day-to-day use.
Here's a run using 5.5 bpw quantized model, generating 5.3k tokens at…
GLM 4.6 runs quite fast on an M3 Ultra with mlx-lm even at higher precision.
Pretty remarkable that it benchmarks competitive to the just-released Sonnet 4.5. Hope those benchmarks hold-up in day-to-day use.
Here's a run using 5.5 bpw quantized model, generating 5.3k tokens at… https://t.co/ZMbdqSnWbv
GitHub site for the book "Mathematical Methods in Data Science (with Python)" (soon to be published):
mmids-textbook.github.io
You can pre-order a print copy of the book here (with Amazon price guarantee): amzn.to/3KrbKvu
.@rheimann's new book "Sutskever's List" is now #2 on @ManningBooks bestsellers list, in front of @fchollet's Deep Learning with Python and @rasbt's Build a Large Language Model (From Scratch).
An explanation of Sutskever's List: turingpost.com/p/ilya-sutskev…
Check out the book...⬇️
.@rheimann's new book "Sutskever's List" is now #2 on @ManningBooks bestsellers list, in front of @fchollet's Deep Learning with Python and @rasbt's Build a Large Language Model (From Scratch).
An explanation of Sutskever's List: turingpost.com/p/ilya-sutskev…
Check out the book...⬇️
Topological deep learning (TDL) merges topology's study of shape with deep learning. It analyzes the "shape" of data, like connectivity and holes, to uncover relationships that traditional methods miss. TDL applications include drug discovery, where it models molecular…
Our paper "Mathematical Foundations of Geometric Deep Learning" (co-authored with @mmbronstein ) has now been officially added to the Geometric Deep Learning Book website as the recommended preparation material! (special thanks to @PetarV_93)
geometricdeeplearning.com/book/
SSL pre-training in a vacuum is over.
search and continual-learning (codesign) remains.
sutton explains the bitter lesson again in detail, and seems like all the LLM-hackers were too keen to heckle a real turing award winner to get the nuance.
Researchers introduced the Energy-Based Transformer (EBT). EBTs score a candidate next token by “energy” and then iteratively lower that energy via gradient steps to verify and select the token.
In 44-million-parameter trials on RedPajama-Data-v2, EBT beat same-size vanilla…
The Databricks Virtual Learning Festival is back Oct 10–31!
Choose a self-paced pathway in Customer Academy and complete it during the festival to unlock exclusive rewards including:
-50% off any Databricks Certification
-20% off an annual Academy Labs subscription
Join us to…
A free book: A First Course on Data Structures in Python by Donald R. Sheehy
Provides building blocks you need for AI and machine learning:
- data structures
- algorithmic thinking
- complexity analysis
- recursion/dynamic programming
- search methods
donsheehy.github.io/datastructures…
Michael C. H. Choi, Youjia Wang, Geoffrey Wolfer. [math.PR]. Geometry and factorization of multivariate Markov chains with applications to MCMC acceleration. (Replacement). arxiv.org/abs/2404.12589
3K Followers 7K FollowingI write the bugs that future AIs will be paid to fix. AI Maximalist & Architect of Artisanal Technical Debt! Rust 🦀 supremacy!
658 Followers 1K FollowingDevReling at https://t.co/LasOGfmK57 | Data scientist by training. Podcaster, NLP and ML content generator, and Coffee connoisseur!
15K Followers 529 FollowingAsst. Prof. of CS at Stanford, Google DeepMind. Prev: Anthropic, Google Brain. Co-Creator of MoEs, AlphaChip, Test Time Scaling Laws.
3K Followers 7K FollowingI write the bugs that future AIs will be paid to fix. AI Maximalist & Architect of Artisanal Technical Debt! Rust 🦀 supremacy!
87K Followers 194 FollowingBuilding beautiful things like Mojo🔥 and MAX @Modular, lifting the world of production AI/ML software into a new phase of innovation. We’re hiring! 🚀🧠
105K Followers 318 FollowingDefending freedoms. Advancing equality. Ensuring justice for all. News from the U.S. Senate Judiciary Committee Democrats, led by Ranking Member @SenatorDurbin.
36K Followers 2K FollowingInformation Geometry, Information Theory, and Geometric Science of Information (GSI) for machine learning and AI, visual computing, HPC, pyBregMan lib @SonyCSL
3K Followers 220 FollowingxAI Head Legal Eagle: Lily is an adventurer, former rocket scientist, and now launcher of products at the innovative Elon Musk AI start-up, xAI.
23K Followers 858 FollowingMitochondrial Psychobiology. Bridging the science of energy and the human experience to create Healing Science. Upcoming book: ENERGY (2027).
5K Followers 376 Following@LanderAnalytics Chief Data Scientist
NY Open Stats Meetup @nyhackr & R Conference @rstatsnyc Organizer
@columbia_biz Adjunct Professor
R for Everyone Author
9K Followers 1K FollowingAssistant Professor at NUS. Scaling cooperation for an increasingly automated future. PhD @ MIT ProbComp / CoCoSci. Pronouns: 祂/伊
38K Followers 485 FollowingDigital Geometer, Assoc. Prof. of Computer Science & Robotics @CarnegieMellon @SCSatCMU and member of the @GeomCollective. There are four lights.
2K Followers 604 FollowingNowadays you must have a great combination of research skills and just-get-it-done attitude. #google #microsoft #ask #relx #iac #alumni (zur, lon,ams,waw,ita)
7K Followers 132 FollowingNYC-based trading and technology firm with offices around the globe. HRT is an equal opportunity employer; so whoever you are, we'd love to get to know you.
8K Followers 2K FollowingFounder/CEO @polar_sh – Monetization platform for developers and the next generation of software | Ex-Director of Product @shopify
9K Followers 0 FollowingJane Street is a global trading firm and liquidity provider with a unique focus on technology and collaborative problem solving.
4K Followers 752 FollowingAI researcher trying to make sense of all things cyberspace 🤖 Uni of Ox PhD (loading…) @oiioxford & @AISecurityInst. Prev @turinginst & @Cambridge_Uni.
3K Followers 440 FollowingIncoming Assistant Professor at Harvard and Kempner Institute. Postdoc at UC Berkeley. Former Ph.D. student at Cornell Tech. https://t.co/LyIdb5HmM9