Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude's system prompt is over 24k tokens with tools

github.com

Post reply on HN
Print view View original

291–300 of 350 posts

Re: Claude's system prompt is over 24k tokens with tools

#291

Post by mystified5016 »

nonethewiser wrote: post #124
freehorse wrote: post #95

I was a bit skeptical, so I asked the model through the claude.ai interface "who is the president of the United States" and its answer style is almost identical to the prompt linked https://claude.ai/share/ea4aa490-e29e-45a1-b157-9acf56eb7f8a Meanwhile, I also asked the same to sonnet 3.7 through an API-based interface 5 times, and every time it hallucinated that Kamala Harris is the president (as it should not "know…

I wonder why it would hallucinate Kamala being the president. Part of it is obviously that she was one of the candidates in 2024. But beyond that, why? Effectively a sentiment analysis maybe? More positive content about her? I think most polls had Trump ahead so you would have thought he'd be the guess from that perspective.

Being vice president and a presidential candidate in succession, the string 'Harris' probably appears next to 'president' more and more recently than 'Trump'. Assuming you exclude 'former president', which seems plausible.

I'm not convinced it's because she's a Democrat and the model is making some contorted connection to Democratic presidents. It's more likely just statistics. Harris is referred to more in texts talking about who is president, and Trump appears more in texts talking about who was president.

Re: Claude's system prompt is over 24k tokens with tools

#292

Post by int_19h »

dontlikeyoueith wrote: post #284
Mahn wrote: post #230

Earlier quoted context omitted.

At this point you have to start entertaining the question of what is the difference between general intelligence and a "sufficiently complicated" next token prediction algorithm.

A sufficiently large lookup table in DB is mathematically indistinguishable from a sufficiently complicated next token prediction algorithm is mathematically indistinguishable from general intelligence. All that means is that treating something as a black box doesn't tell you anything about what's inside the box.

Why do we care, so long as the box can genuinely reason about things?

Re: Claude's system prompt is over 24k tokens with tools

#293

Post by int_19h »

EvanAnderson wrote: post #264
jcims wrote: post #258

Earlier quoted context omitted.

And we get to learn all of the same lessons we've learned about mixing code and data. Yay!

That's what I was thinking, too. It would do some good for the people implementing this stuff to read about in-band signaling and blue boxes, for example.

They are well aware of it, which is why there's a distinction between "system" and "user" messages, for example.

The problem is that, at the end of the day, it's still a single NN processing everything. You can train it to make this distinction, but by their very nature the outcome is still probabilistic.

This is similar to how you as a human cannot avoid being influenced (one way or another, however subtly) by any text that you encounter, simply by virtue of having read it.

Re: Claude's system prompt is over 24k tokens with tools

#294

Post by olddustytrail »

Wowfunhappy wrote: post #257
jsnider3 wrote: post #253

Earlier quoted context omitted.

There's enough disagreement among native English speakers that you can't really say any pronoun is the obviously correct one for an AI.

"What color is the car? It is red." "It" is unambiguously the correct pronoun to use for a car. I'd really challenge you to find a native English speaker who would think otherwise. I would argue a computer program is no different than a car.

People often refer to their car and other people's as "she" ("she's a beauty") so you're is obviously wrong.

Re: Claude's system prompt is over 24k tokens with tools

#295

Post by int_19h »

chpatrick wrote: post #193
Jaxkr wrote: post #43

Earlier quoted context omitted.

That’s what this is. It’s caching the state of the model after the tokens have been loaded. Reduces latency and cost dramatically. 5m TTL on the cache usually.

Isn't the state of the model exactly the previous generated text (ie. the prompt)?

When the prompt is processed, there is an internal key-value cache that gets updated with each token processed, and is ultimately used for inference of the new token. If you process the prompt first and then dump that internal cache, you can effectively resume prompt processing (and thus inference) from that point more or less for free.

https://medium.com/@plienhar/llm-inference-series-3-kv-cachi...

Re: Claude's system prompt is over 24k tokens with tools

#296

Post by int_19h »

otabdeveloper4 wrote: post #167
mhmmmmmm wrote: post #65

Earlier quoted context omitted.

They certainly do, and also offer the tooling to the public: https://docs.anthropic.com/en/docs/build-with-claude/prompt-... They also recommend to use it to iterate on your own prompts when using Claude Code for example

By "rigorous" I mean peeking under the curtain and actually quantifying the interactions between different system prompts and model weights. "Chain of thought" and "reasoning" is marketing bullshit.

How would you quantify it? The LM is still a black box, we don't know what most of those weights actually do.

Re: Claude's system prompt is over 24k tokens with tools

#297

Post by darkteflon »

ludwigschubert wrote: post #177
jychang wrote: post #143

Earlier quoted context omitted.

> Combined with VS Code’s ‘# %%’-demarcated notebook cells in .py files What do you mean by this?

It’s a lighter-weight “notebook syntax” than full blown json based Jupyter notebooks: https://code.visualstudio.com/docs/python/jupyter-support-py...

Yep, lets you use normal .py files instead of using the .ipynb extension. You get much nicer diffs in your git history, and much easier refactoring between the exploratory notebook stage and library/app code - particularly when combined with the other stuff I mentioned.

Re: Claude's system prompt is over 24k tokens with tools

#298

Post by int_19h »

mcintyre1994 wrote: post #113
beardedwizard wrote: post #64

Earlier quoted context omitted.

But this is the solution the most cutting edge llm research has yielded, how do you explain that? Are they just willfully ignorant at OpenAI and anthropic? If fine tuning is the answer why aren't the best doing it?

I'd guess the benefit is that it's quicker/easier to experiment with the prompt? Claude has prompt caching, I'm not sure how efficient that is but they offer a discount on requests that make use of it. So it might be that that's efficient enough that it's worth the tradeoff for them? Also I don't think much of this prompt is used in the API, and a bunch of it is enabling specific UI features like Artifacts. So if the…

Prompt caching is functionally identical to snapshotting the model after it processed the prompt. And you need the KV cache for inference in any case so it doesn't even cost extra memory to keep it around, if every single inference task is going to have the same prompt suffix.

Re: Claude's system prompt is over 24k tokens with tools

#300

Post by lanstin »

johnthewise wrote: post #233
dymk wrote: post #222

Earlier quoted context omitted.

Like the old saying goes, a sufficiently complex next token predictor is indistinguishable from your average software engineer

A perfect next token predictor is equivalent to god

Not really - even my kids knew enough to interrupt my stream of words with running away or flinging the food from the fork.
Post reply on HN