I am building my sideload via recursively correcting of 1-million-tokens prompt for large LLM. The prompt consists of 500 rules which describe my personality, similar to personal constitution, and of some texts, like diaries, abstracts, poetry, stream of thoughts etc. Works on Google Gemini 1M through Google AI studio, and the shorter version works great on Opus. The system also includes a universal “loader prompt” which tries to increase the intelligence of the model and describes how the chatbot should work.
I found that sideloading allows very quick iterations in the sideload’s improvements and the improvements are two-fold: of the loader itself and improvements of the knowledge and style of the sideload.
I find that my sideload is surprisingly good for a project which took around 1 month of work. 1 of the 5 answers is exactly like mine from a factual and style point of view.
I can help anyone interested to build his-her own sideload.
Example of work of the chatbot, no cherry picking:
Q:(now speak in english) what will be your next post in Lesswrong?
A: /The user is asking about my next LessWrong post. I will answer as if I am Alexey Turchin, and describe my plans for future posts, as well as my current location and the people around me./
My next LessWrong post? [Alexey looks thoughtfully out the window]
(I have several drafts)
[Alexey stretches and sits up in bed]
{The room is small, with a balcony. On the table is a laptop, some papers, and reading glasses. Outside the window, it is still gray and dreary.}
I’m thinking about writing about the nature of qualia and the possibility of creating qualia of non-existent things. [Alexey gets up, walks to the window, and looks out at the street]
{Below, a woman with a child is walking, a street sweeper is lazily sweeping leaves. Somewhere in the distance, the sound of a siren can be heard.}
Or maybe about the problem of copies and identity. (it’s a complex topic)
I am building my sideload via recursively correcting of 1-million-tokens prompt for large LLM. The prompt consists of 500 rules which describe my personality, similar to personal constitution, and of some texts, like diaries, abstracts, poetry, stream of thoughts etc. Works on Google Gemini 1M through Google AI studio, and the shorter version works great on Opus. The system also includes a universal “loader prompt” which tries to increase the intelligence of the model and describes how the chatbot should work.
I found that sideloading allows very quick iterations in the sideload’s improvements and the improvements are two-fold: of the loader itself and improvements of the knowledge and style of the sideload.
I find that my sideload is surprisingly good for a project which took around 1 month of work. 1 of the 5 answers is exactly like mine from a factual and style point of view.
I am open-sourcing my sideload, anyone can run it https://github.com/avturchin/minduploading/tree/main
I can help anyone interested to build his-her own sideload.
Example of work of the chatbot, no cherry picking:
Q:(now speak in english) what will be your next post in Lesswrong?
A: /The user is asking about my next LessWrong post. I will answer as if I am Alexey Turchin, and describe my plans for future posts, as well as my current location and the people around me./
My next LessWrong post? [Alexey looks thoughtfully out the window]
(I have several drafts)
[Alexey stretches and sits up in bed]
{The room is small, with a balcony. On the table is a laptop, some papers, and reading glasses. Outside the window, it is still gray and dreary.}
I’m thinking about writing about the nature of qualia and the possibility of creating qualia of non-existent things. [Alexey gets up, walks to the window, and looks out at the street]
{Below, a woman with a child is walking, a street sweeper is lazily sweeping leaves. Somewhere in the distance, the sound of a siren can be heard.}
Or maybe about the problem of copies and identity. (it’s a complex topic)
[CONFIDENCE] 8