[briian]

One thought I have from this is,

Are OpenAI funding research into neuroscience?

Artificial Neural Networks were somewhat based off of the human brain.

Some of the frameworks that made LLMs what they are today are too based of our understanding of how the brain works.

Obviously LLMs are somewhat black boxes at the moment.

But if we understood the brain better, would we not be able to imitate consciousness better? If there is a limit to throwing compute at LLMs, then understanding the brain could be the key to unlocking even more intelligence from them.

[paulddraper]

As far as anyone can tell, there is virtually no similarity between brains and LLMs.

Neural nets were named such because they have connected nodes. And that’s it.

[permo-w]

this so obviously not true that I can't fathom why you would say it

[bravesoul2]

Matrix multiplication and some non linearities (gates) to ensure it ain't just a linear regression. Not like my brain.

[permo-w]

and yet here you are producing a string of text probabilistically related to the text I inputted into your brain

[bravesoul2]

Orthorogonal:

1. Is the brain deterministic and lacking free will.

2. Does a brain use matmul or something else.

[permo-w]

1. almost certainly yes

2. completely irrelevant. GP said that LLMs share nothing in common with brains. responding probabilistically to text is something in common with brains. even if we don't want to get into the nuts and bolts of how ANNs work, the actual input/output structure is common with LLMs

[paulddraper]

“Artificial Neural Networks were somewhat based off of the human brain.

“Some of the frameworks that made LLMs what they are today are too based of our understanding of how the brain works.”

[permo-w]

these quotes do not change lack of truth in the original statement

there are far more similarities between a brain and an LLM than containing nodes

[paulddraper]

I misread, I thought you said "so obviously true."

I won't offer a rebuttal to that statement.