12/11/2023 0 Comments Ripgrep downloadThe Protomaps PMTiles file format lets you bundle together vector tiles in a single file which is designed to be queried using HTTP range header requests. I saw a post about Protomaps on Hacker News. This has received a lot of traffic, presumably because it provides one of the more accessible answers to the question “what are embeddings?”. Embeddings: What they are and why they matter is the big write-up of my talk about embeddings from PyBay this year.In Execute Jina embeddings with a CLI using llm-embed-jina I released a new plugin to run the new Jina AI 8K text embedding model using my LLM command-line tool.I also wrote a TIL about the CSS grids I used in that post. Now add a walrus: Prompt engineering in DALL-E 3 talked about my explorations of the new DALL-E 3 image generation model, including some reverse engineering showing how OpenAI prompt engineered ChatGPT to pass generate its own prompts for DALL-E 3.I also did some fun research into new options for self-hosting vector maps and pushed out several new releases of plugins. It also feels like a whole bunch of my potential future side projects just dropped from several weeks of work to several hours.ĭALL-E 3, GPT4All, PMTiles, sqlite-migrate, datasette-edit-schema 11 days ago I think it’s going to take us all months to fully understand the new capabilities we have around the OpenAI family of models. It’s a huge recipe for prompt injection, but it also cuts out a lot of the work involved in building a custom chatbot. This makes building simple RAG systems trivial, and you can also enable both Code Interpreter and Bing Browse mode as part of your new assistant. You can now define custom GPTs (effectively a custom system prompt, set of function calls and collection of documents for use with Retrieval Augmented Generation) using the ChatGPT interface or via the API, then share those with other people. Function calling got some big upgrades, the most important of which is that you can now be asked by the API to execute multiple functions in parallel.Previously they could produce JSON but would occasionally make mistakes-this mode makes mistakes impossible by altering the token stream as it is being produced (similar to Llama.cpp grammars). JSON mode: both 3.5 and 4.0 turbo can now reliably produce valid JSON output.I have so many things I want to build on top of this. GPT-4 vision! You can now pass images to the GPT-4 API, in the same way as ChatGPT has supported for the past few weeks.I’ve honestly hardly even begun to dig into the things that were released today. I also added support for a new -o seed 1 option for the OpenAI models, which passes a seed integer that more-or-less results in reproducible outputs-another new feature announced today. I adapted that from my Claude 2 version, but I found I had to adjust the prompt a bit to get GPT-4 Turbo to output quotes in the manner I wanted. Llm -m gpt-4-turbo 'Summarize the themes of the opinions expressed here, including direct quotes in quote markers (with author attribution) for each theme. If you want to support a project, try pushing money towards them from your existing training budget instead! Open source developers are often bad at asking for money. I closed with a call to action for a novel way that companies can help support open source projects: pay maintainers to speak to your team, in the form of time-boxed one hour Zoom consulting calls. Read on for an annotated version of the slides, based on a Whisper transcript and extended with some extra clarity and links to further reading. You can watch my presentation on YouTube, or embedded below. To set expectations: Datasette is not yet financially sustainable, at least not in terms of my long-term goals for the project! Fitting everything I’ve explored so far into just ten minutes was a significant challenge. The goal was to share some of the advice from that program, and talk about my own personal experiences trying to achieve financial sustainability for my Datasette open source project. GitHub invited me to speak as a representative of the GitHub Accelerator program from earlier this year. I presented a ten minute segment at GitHub Universe on Wednesday, ambitiously titled Financial sustainability for open source projects. Financial sustainability for open source projects at GitHub Universe seven hours ago
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |