March 2023
March 27, 2023
I think it’s likely that soon all computer users will have the ability to develop small software tools from scratch, and to describe modifications they’d like made to software they’re already using.
AI-enhanced development makes me more ambitious with my projects
The thing I’m most excited about in our weird new AI-enhanced reality is the way it allows me to be more ambitious with my projects.
[... 3,334 words]Every wave of technological innovation has been unleashed by something costly becoming cheap enough to waste. Software production has been too complex and expensive for too long, which has caused us to underproduce software for decades, resulting in immense, society-wide technical debt. This technical debt is about to contract in a dramatic, economy-wide fashion as the cost and complexity of software production collapses, releasing a wave of innovation.
Simple Push Demo (via) Safari 16.4 is out (upgrade to iOS 16.4 to get it) and the biggest feature for me is mobile support for Web Push notifications. This little demo tool was the first I found that successfully sent a notification to my phone: frustratingly you have to add it to your home page first in order to enable the feature. The site also provides a curl command for sending push notifications through the Apple push server once a token has been registered, which is the crucial step to figuring out how to build applications that can send out notifications to users who have registered to receive them.
LLaMA voice chat, with Whisper and Siri TTS. llama.cpp author Georgi Gerganov has stitched together the LLaMA language model, the Whisper voice to text model (with his whisper.cpp library) and the macOS “say” command to create an entirely offline AI agent that he can talk to with his voice and that can speak replies straight back to him.
March 28, 2023
By gaining mastery of language, A.I. is seizing the master key to civilization, from bank vaults to holy sepulchers.
What would it mean for humans to live in a world where a large percentage of stories, melodies, images, laws, policies and tools are shaped by nonhuman intelligence, which knows how to exploit with superhuman efficiency the weaknesses, biases and addictions of the human mind — while knowing how to form intimate relationships with human beings?
Announcing Open Flamingo (via) New from LAION: “OpenFlamingo is a framework that enables training and evaluation of large multimodal models (LMMs)”. Multimodal here means it can answer questions about images—their interactive demo includes tools for image captioning, animal recognition, counting objects and visual question answering. Theye’ve released the OpenFlamingo-9B model built on top of LLaMA 7B and CLIP ViT/L-14—the model checkpoint is a 5.24 GB download from Hugging Face, and is available under a non-commercial research license.
Cerebras-GPT: A Family of Open, Compute-efficient, Large Language Models (via) The latest example of an open source large language model you can run your own hardware. This one is particularly interesting because the entire thing is under the Apache 2 license. Cerebras are an AI hardware company offering a product with 850,000 cores—this release was trained on their hardware, presumably to demonstrate its capabilities. The model comes in seven sizes from 111 million to 13 billion parameters, and the smaller sizes can be tried directly on Hugging Face.
Quicker serverless Postgres connections. Neon provide “serverless PostgreSQL”—autoscaling, managed PostgreSQL optimized for use with serverless hosting environments. A neat capability they provide is the ability to connect to a PostgreSQL server via WebSockets, which means their database can be used from environments such as Cloudflare workers which don’t have the ability to use a standard TCP database connection. This article describes some clever tricks they used to make establishing new connections via WebSockets more efficient, using the least possible number of network round-trips.
I would say ChatGPT (mostly the new GPT-4 model), with a lot of hand-holding and cajoling from me, wrote 60-70% of the code (PHP, Javascript, CSS, SQL) for this AMA site. And we easily did it in a third of the time it would have taken me by myself, without having to look something up on Stack Overflow every four minutes or endlessly consulting CSS and PHP reference guides or tediously writing tests, etc. etc. etc. In fact, I never would have even embarked on building this little site-let had ChatGPT not existed...I would have done something much simpler and more manual instead. And it was a blast. I had so much fun and learned so much along the way.
March 29, 2023
gpt4all. Similar to Alpaca, here’s a project which takes the LLaMA base model and fine-tunes it on instruction examples generated by GPT-3—in this case, it’s 800,000 examples generated using the ChatGPT GPT 3.5 turbo model (Alpaca used 52,000 generated by regular GPT-3). This is currently the easiest way to get a LLaMA derived chatbot running on your own computer: the repo includes compiled binaries for running on M1/M2, Intel Mac, Windows and Linux and provides a link to download the 3.9GB 4-bit quantized model.
Making SQLite extensions npm install’able for Node.js, and on deno.land/x for Deno (via) Alex Garcia figured out how to get his “pip install X” trick for distributing compiled SQLite extensions to work for Node too! Now you can “npm install” 10 of his extensions, including sqlite-regex and sqlite-xsv and sqlite-http and sqlite-html and more, and attach them to a node-sqlite3 or better-sqlite3 connection. He’s bundled them for Deno too!
March 30, 2023
Schillace Laws of Semantic AI (via) Principles for prompt engineering against large language models, developed by Microsoft’s Sam Schillace.
March 31, 2023
Downloading and converting the original models (Cerebras-GPT) (via) Georgi Gerganov added support for the Apache 2 licensed Cerebras-GPT language model to his ggml C++ inference library, as used by llama.cpp.
How to use AI to do practical stuff: A new guide (via) Ethan Mollick’s guide to practical usage of large language model chatbot like ChatGPT 3.5 and 4, Bing, Claude and Bard is the best I’ve seen so far. He includes useful warnings about common traps and things that these models are both useful for and useless at.
You’ll often find prompt engineers come from a history, philosophy, or English language background, because it’s wordplay. You're trying to distill the essence or meaning of something into a limited number of words.