This might be happening because of the ‘elegant’ (incredibly hacky) way openai encodes multiple languages into their models. Instead of using all character sets, they use a modulo operator on each character, to make all Unicode characters represented by a small range of values. On the back end, it somehow detects which language is being spoken, and uses that character set for the response. Seeing as the last line seems to be the same mathematical expression as what you asked, my guess is that your equation just happened to perfectly match some sentence that would make sense in the weird language.
Do you have a source for that? Seems like an internal detail a corpo wouldn’t publish
Can’t find the exact source–I’m on mobile right now–but the code for the gpt-2 encoder uses a utf-8 to unicode look up table to shrink the vocab size. https://github.com/openai/gpt-2/blob/master/src/encoder.py
I suppose it’s conceivable that there’s a bug in converting between different representations of Unicode, but I’m not buying and of this “detected which language is being spoken” nonsense or the use of character sets. It would just use Unicode.
The modulo idea makes absolutely no sense, as LLMs use tokens, not characters, and there’s soooooo many tokens. It would make no sense to make those tokens ambiguous.
I completely agree that it’s a stupid way of doing things, but it is how openai reduced the vocab size of gpt-2 & gpt-3. As far as I know–I have only read the comments in the source code– the conversion is done as a preprocessing step. Here’s the code to gpt-2: https://github.com/openai/gpt-2/blob/master/src/encoder.py I did apparently make a mistake, as the vocab reduction is done through a lut instead of a simple mod.
Let me simplify it: proceeds to print the same expression
Typical AI behavior
Edit: and then it will gaslight you if you say the answer is the same.
Kind of looks like the writing system of Georgian language but I’m not sure
No, this is Glagolitic script, an alternative to Cyrillic. Mostly used in old Slavic scriptures, was later replaced by Cyrillic and Latin.
Most Slavs themselves don’t know how to read this
It’s a dead script that was not that common in the first place, in Kievan Rus’ it was even used as a form of encryption in XI—XVI centuries for how little spread it was. It is also very different from modern Cyrillic. So, saying “most Slavs don’t know how to read it” is a bit of an understatement. Noone knows how to read it, apart from some linguists and overzealous Witcher fans.
Which language uses these signs? It truly looks like some kind of alien language
Glagolitic script. Oldest known Slavic alphabet according to Wikipedia.
