Re:The Emergent Symbolic Structure of Artificial Neural Netw...
by eli_gottlieb on Wednesday September 02, 2026 @10:11AM
(#49536539)
> So at least some of what LLMs do internally seems to involve a bit more structure than just linear combinations of some overcomplete dictionary.
Well that's good to hear. Would have been rather exasperating to find out we could have somehow gotten Fable or Sol out of sufficiently advanced sparse-coding ages ago.
That in itself is of course not super useful for tasks that we don't already know to do symbolically, but it's nice to see that they can manipulate the keys and values of their representation somewhat independently. So at least some of what LLMs do internally seems to involve a bit more structure than just linear combinations of some overcomplete dictionary.
The interesting question is whether this can be used to interpret the operation of a single layer by lifting it to operate on key-value stores instead, replacing matrix-vector products with reading from some keys and then writing to others. That could potentially move the balance of power more towards CPU inference instead of GPUs.