The fact that most individual neurons are uninterpretable presents a serious roadblock to a mechanistic understanding of language models. We demonstrate a method for decomposing groups of neurons into interpretable features with the potential to move past that roadblock.
What if Transformers are literally doomed?
1) Interpretability on transformers might not be feasible: We don't have evidence yet that it's even imaginable to have interpretability tools for this architecture to the degree that would allow us to rule out that a powerful system is…
We just released "Large Language Models with Semantic Search”, built with @cohere, and taught by @JayAlammar and @SerranoAcademy. Search is a key part of many applications. Say, you need to retrieve documents or products in response to a user query; how can LLMs help? You’ll…
Still use ⛓️Chain-of-Thought (CoT) for all your prompting? May be underutilizing LLM capabilities🤠
Introducing 🌲Tree-of-Thought (ToT), a framework to unleash complex & general problem solving with LLMs, through a deliberate ‘System 2’ tree search.
arxiv.org/abs/2305.10601
@hardmaru Benchmarks are boring - there should be an oLLMmpics, with shiny and/or edible medals for imagination, truthfulness, kindness, ethics/bias, style, security, gullibility (lack thereof), inclusiveness (i18n/l18n, affordability for running & retuning), interoperability, bookwiseness
Machine learning models are not going to “take over” and eradicate humanity. But rather, just like all technologies from before, humans will simply adapt to them, and absorb the new technology into our everyday lives, making these models part of our collective human intelligence.
I'm speechless.
Not peer-reviewed yet but a submitted paper.
The 'presented images' were shown to a group of humans. The 'reconstructed images' were the result of an fMRI output to Stable Diffusion.
In other words, #stablediffusion literally read people's minds.
Source 👇
🪐 Introducing Galactica. A large language model for science.
Can summarize academic literature, solve math problems, generate Wiki articles, write scientific code, annotate molecules and proteins, and more.
Explore and get weights: galactica.org
WILD! Working to achieve my goals! Lot's of test effort today.
Stable Diffusion in VR + touchdesigner = realtime immersive latent space. This proof of concept is the FUTURE!
#aiart#vr#stablediffusion#touchdesigner#deforum
I cannot stress enough how good the new beta is!
photograph close up portrait 62-year-old tough decorated general, CLEAN SHAVEN, serious, stoic cinematic 4k epic detailed 4k epic detailed photograph shot on kodak detailed bokeh cinematic hbo dark moody --ar 17:22 --beta --upbeta
As promised, a🧵on language and thought. The relationship between language and thought has long been pondered and debated. It may be one of the deepest and most exciting questions in cognitive science. 1/n
🤔¿Cuánto falta para ver un DALL·E de vídeo?
¡PUES YA EXISTE! Se llama CogVideo y el modelo 9.4 mil millones de parámetros... ¡LO HAN LIBERADO!
INPUT: "Mujer vistiendo un vestido rojo llorando" 💃
Este es el resultado que genera el modelo 👇
675 Followers 850 FollowingInterdisciplinary Institute Director || Professor of Machine Learning @aberdeenuni || opinions are my own || AberdeenML - formerly at @unilincoln
11K Followers 57 FollowingACM Conference on Fairness, Accountability, and Transparency (ACM FAccT). June 23rd to June 26th, 2025, in Athens, Greece. #FAccT2025
736 Followers 64 FollowingSenior Principal Researcher; Human-AI eXperiences (HAX); Microsoft Research. AI agent wrangler. Building @pyautogen, #MagenticOne, #MagenticUI, and #markitdown
218K Followers 427 FollowingThe latest rumors and developments in the world of artificial intelligence. DM to include your AI project in the newsletter.
80K Followers 1K FollowingInterdisciplinary researcher focused on shaping AI towards long-term positive goals. ML & Ethics.
Similar content in the Skies (this bird has flown).
1.4M Followers 1K FollowingBuilding @EurekaLabsAI. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets.
81K Followers 6K FollowingUKRI ESRC funds world-leading research, data and training in the economic and social sciences. We are part of UK Research and Innovation (@UKRI_News).
675 Followers 850 FollowingInterdisciplinary Institute Director || Professor of Machine Learning @aberdeenuni || opinions are my own || AberdeenML - formerly at @unilincoln
165K Followers 0 FollowingInvented principles of meta-learning (1987), GANs (1990), Transformers (1991), very deep learning (1991), etc. Our AI is used many billions of times every day.
195K Followers 1K Following🤖 Divulgador de Inteligencia Artificial (DotCSV)
✉️ Contacto comercial: [email protected]
📚 Enseño sobre IA en Youtube, Tiktok e Instagram
6.2M Followers 2K FollowingBest-Selling Author | Clinical Psychologist | #1 Education Podcast | Co-founder of @petersonacademy | Order my new book "We Who Wrestle With God" here:
216K Followers 2 FollowingI save your favorite Tweets and Threads to your Notion Workspace!
Just follow @SaveToNotion & check the pinned tweet to start,
Developed by: @Abdulhade_Ahmad
8K Followers 1 FollowingFollow the training of "BLOOM 🌸", the @BigScienceW multilingual 176B parameter open-science open-access language model, a research tool for the AI community.
46K Followers 1K FollowingWriter https://t.co/TquuQXlLOJ. O'Reilly Author https://t.co/Fl3uPAZHLg. LLM Builder @Cohere. Visualizing AI one concept at a time.
1.3M Followers 693 FollowingHIJO DEL SOL Y ENEMIGO DEL VIENTO. CREADOR DE ZUMO DE FETOS Y KARMAGGÄN - https://t.co/LDELXYD4Sk Twitch https://t.co/9ixKrWekIy
717K Followers 288 FollowingTogether with the AI community, we are pushing the boundaries of what’s possible through open science to create a more connected world.
454K Followers 77 FollowingTensors and neural networks in Python with strong hardware acceleration. PyTorch is an open source project at the Linux Foundation. #PyTorchFoundation
No recent Favorites. New Favorites will appear here.