181
Stable LM 3B: Bringing Sustainable, High-Performance LMs to Smart Devices
Our latest language model, Stable LM 3B, is a compact language model designed to operate on portable digital devices like handhelds and laptops.
New
The latest blog posts to take off on Hacker News. 789 posts.
182
Weird A.I. Yankovic: a cursed deep dive into the world of voice cloning
What would it sound like if the musicians parodied by "Weird Al" Yankovic sang *his* parodies instead? I jumped off the deep end into the world of A.I. cover songs to find out, and the results are even worse than you imagine.
183
Break-a-Scene: Extracting Multiple Concepts from a Single Image
A method for extracting multiple concepts from a single image.
184
FlashAttention: Fast Transformer training with long sequences
Transformers have grown deeper and wider, but training them on long sequences remains difficult. The attention layer at their heart is the compute and memory bottleneck: doubling the sequence length would quadruple the runtime and memory requirements.
185
ChatGPT-4 significantly increased performance of business consultants
In collaboration with Boston Consulting Group (BCG), new research from Digital Data Design Institute at Harvard chair and co-founder Karim Lakhani and In collaboration with Boston Consulting Group (BCG), new research from Digital Data Design Institute at Harvard chair and co-founder Karim Lakhani and others involved…
186
LeoLM: German-Language LLM Research
We proudly introduce LeoLM ( L inguistically E nhanced O pen L anguage M odel), the first comprehensive suite of German-language Foundation Language Models t…
187
The Llama Ecosystem: Past, Present, and Future
The latest updates on the Llama ecosystem coming out of Meta Connect 2023
188
Mistral 7B
The best 7B model to date, Apache 2.0
189
Workers AI: Serverless GPU-powered inference
We are excited to launch Workers AI - an AI inference as a service platform, empowering developers to run AI models with just a few lines of code, all powered by our global network of GPUs
190
First Impressions with GPT-4V(ision)
In this guide, we share findings experimenting with GPT-4 with Vision, released by OpenAI in September 2023.
191
Be My Eyes’ AI assistant starts rolling out
Be My Eyes’ AI assistant, powered by GPT-4, is rolling out to hundreds of thousands of iOS and Android users over the next several weeks.
192
193
Inside the Matrix: Visualizing Matrix Multiplication, Attention and Beyond
Use 3D to visualize matrix multiplication expressions, attention heads with real weights, and more. Matrix multiplications (matmuls) are the building blocks of today’s ML models. This note presents mm, a visualization...
194
195
Amazon will invest up to $4B in Anthropic
Amazon will invest up to $4 billion in Anthropic and become its primary cloud provider, with models trained on AWS Trainium and Inferentia.
196
Our humble attempt at “how much data do you need to fine-tune”
Efficacy of OpenAI Fine-tuning API, Cost/Latency Considerations, Experiencing Catastrophic Forgetting, and...Getting Roasted by a Fine-tuned GPT-3.5
197
LLM Falcon 180B Needs 720GB RAM to Run
Yes, if you have enough CPU RAM
198
I made a transformer to predict a simple sequence manually
To better understand how transformers work, I hand-assigned all the weights to predict a simple sequence.
199
Outperforming larger language models with less training data and smaller models
Posted by Cheng-Yu Hsieh, Student Researcher, and Chen-Yu Lee, Research Scientist, Cloud AI Team Large language models (LLMs) have enabled a new da...
200
Open source AI will win
You shouldn't bet against the bazaar or the GPU p̶o̶o̶r̶ hungry.
201
202
Compiling ML models to C for fun
We make micrograd fly with a little compiler magic. In this post, we’ll write a ML compiler from scratch.
203
Generative AI's Act Two
One year ago, we published a hypothesis that generative AI would become a profound platform shift in technology. Then came the firestorm.
204
205
78% MNIST accuracy using GZIP in under 10 lines of code
Using GZIP compression and the k-Nearest Neighbors algorithm, we explore an innovative approach to classifying the MNIST dataset with about 78% accuracy
206
Bard can now connect to your Google Apps and services
Bard gets its most capable model yet, along with new and expanded features.
207
What will LLM-powered software look like in the medium-term future?
Recently, I started working on a new LLM-oriented project . It’s currently far from feature-complete and probably filled with bugs, but I’m hoping to use it as a platform for exploring a question I’ve been considering a lot lately: what will LLM-powered software look like in the medium-term future? It seems undeniable…
208
Self-supervised learning: The dark matter of intelligence (2021)
How can we build machines with human-level intelligence? There’s a limit to how far the field of AI can go with supervised learning alone. Here's why self-supervised learning is one of the most promising ways to make significant progress in AI.
209
Data accidentally exposed by Microsoft AI researchers
Wiz Research found a data exposure incident on Microsoft’s AI GitHub repository, including over 30,000 internal Microsoft Teams messages – all caused by one misconfigured SAS token
210
Apple’s new Transformer-powered predictive text model
I found some details about Apple’s new predictive text model, coming soon in iOS 17 and macOS Sonoma.