I certainly analogize my behaviors to LLMs. How I learn, how I think - I see it reflected in the LLMs I use every day.
LLMorphism: When humans come to see themselves as language models
31–40 of 60 posts
Re: LLMorphism: When humans come to see themselves as language models
#32Before electric computers, the human mind was a steam engine: https://www.ezrabrand.com/p/releasing-the-pressure-a-dive-in...
Or clockwork.
Regardless of the degree to which the human mind works like an LLM, my reductionist tendency has always imagined that the human mind will be found to be built from simple enough principles (but at scale, of course). In that regard, LLM as model for the human brain (or at least one aspect of it) is attractive to me. I admit it.
Re: LLMorphism: When humans come to see themselves as language models
#33Before electric computers, the human mind was a steam engine: https://www.ezrabrand.com/p/releasing-the-pressure-a-dive-in...
It's interesting that it's easier to construct the argument † that a mind like an LLM would have an easier time capturing mind as steam engine than a mind like a steam engine would have capturing mind as LLM. †: come up with each token after the other that induces a graspable interpretation of a sequence of tokens representing a potential judgement
There's something really interesting here with Goguen Institutions. Also sometimes an argument just "clicks" into place fully-formed, rather than being generated token-by-token? Is that "knowing like a steam engine?"
Re: LLMorphism: When humans come to see themselves as language models
#34I mimic how LLM responds when I talk to my boss lol. Appear useful and present verbose facts. Works pretty well so far.
Try "caveman" mode and report back.
Re: LLMorphism: When humans come to see themselves as language models
#35A more insidious related pathology- marital induced projected LLMorphism... where your wife constantly accuses you of having the personality of a large language model.
Are you relentlessly optimistic and bit sycophantic? Maybe that’s what she means.
Re: LLMorphism: When humans come to see themselves as language models
#36Before electric computers, the human mind was a steam engine: https://www.ezrabrand.com/p/releasing-the-pressure-a-dive-in...
Or clockwork. Regardless of the degree to which the human mind works like an LLM, my reductionist tendency has always imagined that the human mind will be found to be built from simple enough principles (but at scale, of course). In that regard, LLM as model for the human brain (or at least one aspect of it) is attractive to me. I admit it.
Before clockwork, human mind was a city (Plato, Marcus Aurelius, i.a.)
Re: LLMorphism: When humans come to see themselves as language models
#37I teach students to use their own imagination like generative AI. Prompting works. They just need a bit of practice.
That's actually a really interesting thing to think about, learning how to "prompt" oneself.
Prompting the imagination (for example, writing prompts) was a thing before generative AI even emerged in a meaningful sense. And it does work.
This is one of the reasons why I do literate programming using org-mode. It's easy to lose track of what I was thinking when I wrote something, and what the original structure and goal was as I continue to write it. Org-mode helps me keep my thoughts in order in English and interlineate code in with them, then mash a key to spit out compilable source. I don't use it for everything, but it sure comes in handy when I do use it.
Re: LLMorphism: When humans come to see themselves as language models
#38Don't be too hard on yourself. If you've never walked to the car wash, then you are probably not an LLM.
Here's the thing though, unlike the old brain=computer analogy, this one may actually have a little truth to it. Not that your whole brain is an LLM, or even that the language part of your brain is just an LLM, but the language part may indeed be functioning in a similar way to an LLM to extent that it:
- Uses a hierarchy (cortical patch-panel) of parallel processing steps
- Is prediction based
- Is largely (but not 100%) auto-regressive
- Isn't actually specialized for language
The same is going to be true for all of our cortical areas/functions. The cortex is pretty much the same everywhere (it's 6 layers of neurons with a specific layer-to-layer interconnect pattern), and is therefore going to work the same everywhere.
What your cortex has that an LLM doesn't, and therefore makes your language cortex much more capable than an LLM, is that it learns incrementally and continually, based on prediction failure. An LLM/Transformer also learns from prediction failure, but needs the LLMs whole "life history" (training set) to be present at the same time, presented over and over, and learns via a special training algorithm. Your cortex in contrast doesn't have any magical external trainer, so has to learn for itself, and might be considered as 1/2 inference network and 1/2 prediction feedback/learning network.
The other major difference between an LLM and your language cortex is that the LLM is 100% auto-regressive, while your language cortex also has external inputs that bias/control generation, so that you can talk about things you are experiencing and what is going on in your head, not just generate a self-predicting sequence of words.
Re: LLMorphism: When humans come to see themselves as language models
#39I think it's meaningless anyway. A calculator doesn't multiply numbers like a human does. The important part is to develop systems that can do many human tasks
Early LLMs typically tried to do multiplication "in their head" by recall. Now most LLMs do multiplication using a tool call to a programming language, akin to a person reaching for a calculator rather than relying on a learned table or working the problem out mentally. The high level comparison between what LLMs do and what humans do" for this example is fairly parallel.
Humans drink water, and so do the datacenters! We’re just like AI!!!
Re: LLMorphism: When humans come to see themselves as language models
#40Nothing new under the sun. When clocks and precision mechanics started in the 17th century, there was a tendency to view humans as "machines". Computers came, suddenly human brains are "computers". Now we're LLMs. If scientists make green jelly that emits thoughtful judgements, humans will be compared to green jelly.
None of these analogies are entirely wrong, they're just incomplete. Humans are similar to machines for example in that our bodies convert energy to do work through a series of pumps and pipes and sensors and actuators. Life is not animated by some magic force but instead operates under the same physical laws that machines use to function.
Everything is the same if you zoom out far enough.