For a few months, embeddings were everything. I got intrigued by some of those old "king/queen" videos and wanted to see if the gpt embeddings showed the same properties.

Endpoint: v1/embeddings

Models: text-embedding-ada-002, text-embedding-3-small, text-embedding-3-large

Those old gensim videos showed how vector arithmetic could produce something like a semantic transformation when working with word2vec embeddings. To reproduce with gpt embeddings, the software I've written has a fairly extensive set of vector operations: normalize, dot product, 7D curl (IYKYK), etc... Dimension reduction using tSNE. Mulling over adding umap.

Mostly, I am interested in operating on blocks of texts, but to test my setup I wanted to reproduce what I'd seen with the word2vec embeddings .

This software will read the text and binary files like those in the nlpl repository (http://vectors.nlpl.eu/repository). Currently, embeddings are read from the file on-demand and put into a database. The idea here is that relatively few embeddings will be of interest and once in the database, there is no need to spend time parsing the file.

Most of what I've done was with the GoogleNews-vectors-negative300.bin file and other word2vec output.

Nearest-neighbor searching is only among those embeddings in the database, unlike with gensim which provides a list of the top N closest matches in the entire vocabulary. Such a thing has no applicability when using the gpt embeddings with blocks of text but I still think it would be nice to have so I will probably implement some kind of FAISS/ANN scheme. I'll need to deal with a vector database at some point though.

I am releasing a version of the software that works with the nlpl files (http://vectors.nlpl.eu/repository), even though feature-wise it lags behind gensim for the target data. But if you want to play around with that data and don't want to deal with Python, you are welcome to use it.

(For those of you who notice: Yes, GPT 4o responded with "is a ancient".)

For generating conversations, I had to pick something to start with and I went with this chat completion api of OpenAIs because it looked like the quickest way forward. And I already had an account.

The software utilizes the chat completion endpoint /v1/chat/completions with OpenAI and, quite by coincidence, DeepSeek/Anthropic/Gemini as well.

When I started using AI for applications, I thought there would be a lot of work to do with model parameters like temperature, top_p, etc... But after gpt3.5, imo, not really. There are safeguarding (eg, max_tokens) and management parameters (eg, # of responses) but the performance parameters I rarely change from the default anymore. The software keeps a record of the parameters and everything else for analysis anyway, and this might become relevant again someday, who knows.

System prompts are stored with the threads. Pre-caching I was using this a bit to save on cost with chat completions. I still use custom system prompts a lot, mostly to prevent the AI from wasting tokens doing things I don't want it to do.

Responses API added. But this is the last API (of course the bigs all provide it or chimera) I'm going to accomodate for awhile.'

Currently calls dall-e-2 and dall-e-3 with /generations, /edits and /variations endpoints.

Graphics really aren't my thing so this gets relatively short shrift. The limits of my ambitions here are to provide a way to draw on an image or create a selection mask in the browser window and feed it back to subsequent prompts.