Моя большая языковая модель

My language model

When the artist Dmitry Vrubel (the author of the famous “kiss” of Brezhnev and Honecker – a mural on the Berlin Wall) was alive, we planned to create an artistic work with him, where I would stand against a background wall made of all the documents comprising the “KARAGODIN Investigation” corpus.

So, it would have been walls of investigation files, all responses from government bodies, all inquiries, references, letters, everything, everything, everything, including screens of information (if the documents, for example, were electronic).

The idea was to show a person against the backdrop of that bureaucratic wall that one has to overcome and at the same time build.

Roughly estimated, this wall, against which I planned to stand alone, would have been enormous. Moreover, it would have been so large that it’s doubtful any space or place could have accommodated it all.

And that’s why we decided to do it virtually…

The idea to create this art project is still alive, in contrast to the life of Dmitry Vrubel, who has passed away, to my immense grief and regret.

I remembered this because I have been thinking for a long time about how the materials of the “KARAGODIN Investigation” project have reached such a critical mass that it becomes very difficult for one person to analyze them.

Not just one person is working on the project, but still, it’s very challenging.

My reflections gradually led me to the thought that it would be good to involve a Large Language Model (LLM) in the project work.

Let there be one more tool.

Frankly, I’m slightly apprehensive and even a bit fearful of this. But, at the same time, I’m gradually starting small tests and experiments.

In general, let’s consider this a working note about how I’ve slowly begun.

The illustration accompanying this text is actually one of the examples of my “trial run.”