Today's April Fools day joke from Microsoft Azure Devops is *hillarious*, way to go guys :) If you wouldn't mind finishing the joke now it'd be appreciated. @AzureSupport@AzureDevOps
I recreated an entire GPT architecture in a spreadsheet.
It is a nanoGPT designed by @karpathy with about 85000 parameters, small enough to be packed into a spreadsheet file.
It is great for learning about how transformer works as it shows all the data and parameters going through a transformer pipeline, and all the calculations are actually working inside those cells.
Here is a link to the project where you can download the file and play with it yourself. No coding needed, it is just a spreadsheet. 😉
Spreadsheet is all you need:
https://t.co/dreZlBDq4d
Also thanks to @BrendanBycroft with his LLM visualisation project which inspired this.
A ton of AI and Robotics developments this week.
Big announcements from Hugging Face, Kuaishou, Atom Limbs, Amazon, Apple WWDC, Microsoft, CMU Robotics, and Yale Faboratory.
Here's everything that happened and how to make sense out of it:
Love working in IT. Today I can't use https://t.co/xrzU5V7wWH to generate certificates for Azure VMs, because https://t.co/6o21L9fROK has changed underneath me. Arghh.
I hit a bug in the Attention formula that’s been overlooked for 8+ years. All Transformer models (GPT, LLaMA, etc) are affected.
Researchers isolated the bug last month – but they missed a simple solution…
Why LLM designers should stop using Softmax 👇
https://t.co/3TqSo13RKN
This paper is going viral.
Why?
Authors were able to find a way to enable the Recurrent Memory Transformer to retain information across up to 2 million tokens 🤯
In simpler words, it can process and remember vast amounts of data, significantly more than before.
Just so you have an idea, GPT-4 handles 32K tokens (~50 pages of documents), while the entire Harry Potter series is ~1.5M tokens.
Imagine the potential implications of this:
• AI could write entire novels, not just blog posts
• It could aid in complex scientific research by analyzing vast quantities of data
• Enhance customer service by retaining years of interaction history
• It could store and recall the entirety of your life experiences!!
I hacked into a @Bing CMS that allowed me to alter search results and take over millions of @Office365 accounts.
How did I do it? Well, it all started with a simple click in @Azure… 👀
This is the story of #BingBang 🧵⬇️
I'm excited to announce the release of GPT4All, a 7B param language model finetuned from a curated set of 400k GPT-Turbo-3.5 assistant-style generation.
We release💰800k data samples💰 for anyone to build upon and a model you can run on your laptop!
Real-time Sampling on M1 Mac
11 ways ChatGPT saves me hours of work every day, and why you'll never outcompete those who use AI effectively.
A list for those who write code:
1 of 16
Goaaaaaaaal (maybe?) ⚽️
Channel 4 suffer an outage on their website and app leaving football fans unable to watch the England vs Germany match.
https://t.co/k2SSUDTare #EnglandvsGermany#ENGGER#UefaNationsLeague@Channel4 @All4
My favorite bug among the vulnerabilities I presented today! 😆
The original intent was to compare the password. However, the developer copy-and-pasted the code but forgot to replace the variable name. That leads to the Authentication Bypass on IIS.