It's Not Just X. It's Y
mail.cyberneticforests.com
It's Not Just X. It's Y
1–10 of 155 posts
Re: It's Not Just X. It's Y
#2> RLVR is weirder, and I suspect it's why we see "It's not X, it's Y" so often.
This feels like an easy enough hypothesis to verify, for anyone in the business of training LLMs - does the not-X-but-Y rate increase after RLVR?
Re: It's Not Just X. It's Y
#3You’re absolutely right. This is the smoking gun. This changes everything.
Re: It's Not Just X. It's Y
#4Another bunch of dead give aways in code bases with READMEs is the repetitive:
- "No X, No Y, No Z." pattern
- "Here is X - it makes Y"
The worst and most obvious one is the constant over use of emoji ticks and crosses.
Re: It's Not Just X. It's Y
#5I like that these AI idioms exist. They're like watermarks for text. It's worth the cost of humans avoiding them. Companies will eventually train their models to be undetectable, but society would be better if they didn't.
Re: It's Not Just X. It's Y
#6Another bunch of dead give aways in code bases with READMEs is the repetitive: - "No X, No Y, No Z." pattern - "Here is X - it makes Y" The worst and most obvious one is the constant over use of emoji ticks and crosses.
For calibration purposes, I offer you a pre-LLM README I wrote that includes an em-dash* followed by "No X, No Y, No Z": https://github.com/DavidBuchanan314/stelf-loader
*actually a hyphen but it's functioning as an em dash.
Re: It's Not Just X. It's Y
#7This is how early forms of "reasoning" in LLMs worked: just literally inserting words like "Wait...", "Hmm...", "Let me reconsider...", "But is it really..." into the token stream.
Re: It's Not Just X. It's Y
#8You’re absolutely right. This is the smoking gun. This changes everything.
[flagged]
Re: It's Not Just X. It's Y
#9> RLVR is weirder, and I suspect it's why we see "It's not X, it's Y" so often. This feels like an easy enough hypothesis to verify, for anyone in the business of training LLMs - does the not-X-but-Y rate increase after RLVR?
It’s unlikely this is true. LLMs are way more mad-libs / templates than we like to admit, that’s (ironically) not a judgement about their capability, it’s primarily just an observation. But it’s also what plain old SFT, which I believe is the primary culprit, ends up imparting.