I would guess that every statement made is kind of true. It is a clock, a computer and a LLM,…
I would even go as far as LLM is the closest to a functioning brain we can produce from a functional perspective. And even the artificial brains are to complex to understand in detail.
I reckon we can get a lot closer than an LLM in time. For one thing, the mind has particular understanding of interim steps whereas, as I understand it, the LLM has no real concept of meaning between the inputs and the output. Some of this interim is, I think, an important part of how we assess truthfulness of generated ideas before we put them into words.
I experimented with rules like : “Summarize everything of our discussion into one text you can use as memory below your answer.” And “summarize and remove unnecessary info from this text, if contradictions occur act curious to solve them”… simply to mimic a short term memory.
It kind of worked better for problem solving but it ate tokens like crazy and the answers took longer and longer. The current GPT4 models seem to do something similar in the background.
I think that’s still different from what I’m thinking of of interim steps, though.
…but as I think how to explain I realize I’m about to blather about things I don’t understand, or at least haven’t had time to think about! So I’d better leave it there!
![[Thread] Mental Math](https://mander.xyz/pictrs/image/4a605b7b-7272-471f-ac11-37366ac7d8bd.png){kind=link}
I would guess that every statement made is kind of true. It is a clock, a computer and a LLM,…
I would even go as far as LLM is the closest to a functioning brain we can produce from a functional perspective. And even the artificial brains are to complex to understand in detail.
I reckon we can get a lot closer than an LLM in time. For one thing, the mind has particular understanding of interim steps whereas, as I understand it, the LLM has no real concept of meaning between the inputs and the output. Some of this interim is, I think, an important part of how we assess truthfulness of generated ideas before we put them into words.
I experimented with rules like : “Summarize everything of our discussion into one text you can use as memory below your answer.” And “summarize and remove unnecessary info from this text, if contradictions occur act curious to solve them”… simply to mimic a short term memory.
It kind of worked better for problem solving but it ate tokens like crazy and the answers took longer and longer. The current GPT4 models seem to do something similar in the background.
I would really like to get into LLM and AI development but the math…woosh right over my head.
I think that’s still different from what I’m thinking of of interim steps, though.
…but as I think how to explain I realize I’m about to blather about things I don’t understand, or at least haven’t had time to think about! So I’d better leave it there!