this post was submitted on 18 Jun 2023
273 points (100.0% liked)
Programmer Humor
32557 readers
334 users here now
Post funny things about programming here! (Or just rant about your favourite programming language.)
Rules:
- Posts must be relevant to programming, programmers, or computer science.
- No NSFW content.
- Jokes must be in good taste. No hate speech, bigotry, etc.
founded 5 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Yep yep, statistical analysis as to the frequency of tokens in the training text.
Brand new, never-before-seen Windows keys have a frequency of zero occurrences per billion words of training data.
That isn't actually what's important. It's the frequency of the token, which could be as simple as single characters. The frequency of those is certainly not zero.
LLMs absolutely can make up new words, word combinations, or sentences.
That's not to say chatgpt can actually give you good windows keys, but it isn't a fundamental limitation of LLMs.
Okay, I'll take your word for it.
I've never ever, in many hours of playing with ChatGPT as a toy, had it make up a word. Hallucinate wildly, yes, but not stogulate a word out of nothing.
I'd love to know more, though. How does it combine new words? Do you have any examples of words ChatGPT has made up? This is fascinating to me, as it means the model is much less chained to the training data than I thought.
It can create new words, I just verified this. First word it gave me: flumjangle. Google gives me 0 results. Maybe Google is missing something and it exists in some data out there, Idk.
I'm not sure what is so impressive about this though. Language models can string words together in unique ways, why would it be different for characters?
I'm just surprised to hear that google hasn't found out about my flumjangles yet.
A lot of compound words are actually multiple tokens so there's nothing stopping the LLM from generating the tokens in a new order thereby creating a new word.