News for hackers. Stuff that matters.
Re:Weird GPT-4 behavior for the specific string “ davidjl” by dragonwriter on Thursday June 08, 2023 @12:52PM (#36244969)
> What are the technical hurdles to removing tokenization?

In a sense, none, character (with whatever encoding scheme you want) models have been done. OTOH, you get effectively a smaller context window (since tokens represent one or more characters), and since tokens are semantic unions you push more work into the main network to acheive similar quality.

> Thinking about human brains, it seems we operate on an ever-evolving set of symbols, at various levels of abstraction.

Analogies between human brains and LLMs are often more obscuring than illuminating, but I think if you want to go down that route here, that’s equivalent to doing continuous fine tuning on both the tokenizer (that takes sensory input and maps to semantically meaningful symbols) and the main model.

> I believe LLMs capture higher level symbols in their weights, but would it make sense to try to include a dynamic, self-updating set of symbols into the “what is being predicted” part of the training?

“Dynamic, self-updating” simply isn’t part of the way modern LLMs work, independent of whether you are looking at the tokenizer or main model. If someone could make it practical, sure, there are lots of things it would potentially improve.

Re:Weird GPT-4 behavior for the specific string “ davidjl” by dwaltrip on Thursday June 08, 2023 @01:56PM (#36245931)
That makes sense, thanks for the detailed reply.

I’m curious about the latest research related to continuously improving the tokenizer while training the main model (or something equivalent).