Hacker Newsnew | past | comments | ask | show | jobs | submit | aklein's commentslogin

> Prefilling Kimi-K3's reasoning with the first 1% of tokens of Opus 4.8's reasoning moves its visible answer toward Opus's wording, even though the answer itself is never prefilled

is this supportive evidence for the distillation accusations in the news?


Prefilling any model with the first 1% of reasoning tokens from another model should always move the output towards the output of the other model directionally - that's just next token prediction doing its thing.


Same here! TI-85, and then HP48G series after that!


I have a CLAUDE.md symlinked to AGENTS.md


Longtime backblaze user. Time to vibecode myself a replacement.


I wrote a clone of this game for the HP-48 as a teen in the 90s. you can still find it if you google hard enough. good times.


I still have my GX from the late 90s which managed to outlast my Metakernel equipped, overly rubberized 49G. If I were to dig it out, the serial cable too.

And a printout of the sysrpl guide, it being quite thick of a print.


until you learn to trust the system and free mental capacity for more useful thinking. at some point compilers became better at assembly instructions than humans. seems inevitable this will happen here. caring about the details and knowing the details are two different things.


LLMs lie constantly. There should be no trust in that system. And no I don't think they will "get better".


How do they "lie constantly"? We are specifically talking about code here, not LLMs writing legal documents.


I've had the LLM "lie" to me about the code it wrote many times. But "lie" and "hallucinate" are incorrect anthropomorphisms commonly used to describe LLM output. The more appropriate term would be garbage.


Just a basic sanity check: did the LLM have the tools to check its output for lies, hallucinations and garbage? Could it compile, run tests, linters etc and still managed to produce something that doesn't work?


I've frankly given up on LLMs for most programming tasks. It takes just as much time (if not more) to coddle it to produce anything useful, with the addition of frustration, and I could have just written far better code myself in the time it takes to get the LLM to produce anything useful. I already have 40 years experience programming, so I don't really need a tin-can to do it for me. YMMV.


Compilers are deterministic tools. AI is not deterministic. It will tell you this if you ask it. AI then, is not a tool. It is an aide. It is not a tool like a compiler, IDE, editor, etc.


https://chatgpt.com/share/699346d3-fcc0-8008-8348-07a423a526...

interesting. if you probe it for its assumptions you get more clarity. I think this is much like those tricky “who is buried in grants tomb” phrasings that are not good faith interactions


This article highlights how experts disagree on the meaning of (non-human) intelligence, but it dismisses the core problem a bit too quickly imo -

“LLMs only predict what a human would say, rather than predicting the actual consequences of an action or engaging with the real world. This is the core deficiency: intelligence requires not just mimicking patterns, but acting, observing real outcomes, and adjusting behavior based on those outcomes — a cycle Sutton sees as central to reinforcement learning.” [1]

An LLM itself is a form of crystallized intelligence, but it does not learn and adapt without a human driver, and that to me is a key component of intelligent behavior.

[1] https://medium.com/@sulbha.jindal/richard-suttons-challenge-...


Humans are social-emotional beings who assign “irrational” value to things for social signaling and emotional (self-)gratification.


Not a lawyer, but my understanding is civilian casualties are not unlawful (according to international law) when the target is legitimate (on the theory that it is otherwise impossible to legally fight a war with an enemy that hides behind its citizens). To be clear this is not to say war crimes are not also happening.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: