Artificial

Mimi and Chee discuss the design choices behind large language models, arguing that LLMs do not need to generate human-like language and that their personalities are a training choice, not a necessity. They explore alternative uses for LLMs, such as generating code for tools like Patchwork, and critique the current trend of making models sycophantic and expensive.

it works with the Humans" mimi submit a https://feelingof.london/demo demo consciously making a little joke when they say it? Claude is not a person. Right? really don't like it, but i don't know why yet mimi mimi claude program does this automatically when you use it to commit code. and i think github when you use copilot? mimi — , ; , proper capitalization also make me wary now seem right at first blush, but have subtle mistakes the kind a person never makes need to write something chee mimi felienne https://www.youtube.com/watch?v=-Br66SUjsdQ&t=12129s hermans make mistakes mimi have to be assistants, they don't have to talk in prose Natalia https://uxdesign.cc/when-ai-passes-the-capitalist-turing-test-18baacbcf18f Talmina came in to talk about ai and conversation she mentioned that there's linguistics research that people adapt to each other's mental states when they are talking to each other mimi chee from people outside of https://claude.ai and vscode part of why the models are SO compute expensive and therefore money expensive. The output of pre-training is a base-model. fine-tune the model by feeding it loads of conversations in the style that we're aiming for sycophantic simpering and retrain some layers of the model towards the target output. instruct models rather than base models. { "user message": "Generate a tool that does X", "model response": " { @patchwork: more javascript json stuff goes here, blah blah blah you get the idea}", ... } And then it would, hopefully, be really good at generating the javascript for patchwork tools. mimi LLMs don't have to generate human-like language AT ALL. It is a design / training choice that LLMs generally 'have' personalities. We could train an LLM to generate Patchwork tools or JavaScript and they wouldn't have any capabilities to generate human-like language like most of them do now. Our input could also be entirely different / richer than a text description of the thing we want out. exactly what i want We've had a go at another https://www.inkandswitch.com/patchwork/notebook/chitter-chatter/ way , but there are so many more to try : great . do we wanna talk about the stuff re: a new punk? punk computing? like, about craft vs art that's veneer and music with guttural screams and mistakes left in and... etc?