Missing a version of “this is where X earns its keep”. Noticed lately that Claude (and other LLMs) really love to use it.
The analysis only considers individual words, not idioms or structures.
Show HN: The load-bearing vocabulary of Claude
231–240 of 331 posts
Re: Show HN: The load-bearing vocabulary of Claude
#232I wonder to what extent this is the result of suboptimal RLHF versus the inherent intelligence of the model making its language more intricate and difficult for humans to easily parse? On the one hand, it's a common trope that highly educated people can talk in a way that's confusing and annoying to regular people who don't know all the jargon. But on the other hand, it's a mark of a skilled communicator to be able t…
It's easy to think "it's not talking down, because I don't understand it, and I'm intelligent". But how is less intelligent being supposed to fully understand a more intelligent one, honestly speaking? All I know is that Claude understands Claude perfectly. I have the common session pause/resume setup that sometimes produces completely incomprehensible markdown files, but a new Claude session picks them up perfectly,…
> But how is less intelligent being supposed to fully understand a more intelligent one, honestly speaking?
It is really simple. It is on the supposedly more intelligent person to be able to phrase things in simple way. Writing something incomprehensible and convoluted is easier then writing something simple to understand. Even for people.
> a new Claude session picks them up perfectly, down to the smallest details.
I genuinely doubt so.
> What if what we consider excessive circular gibberish is actually highly precise set of instructions needed to minimize error cases for that unreliable human?
It is not highly precise set of instructions and it is not following them in highly precise way.
Re: Show HN: The load-bearing vocabulary of Claude
#233I once asked Claude to replace “byte-identical” with a simpler word, such as “duplicate.” He refused and said “duplicate” does not mean the same thing as “byte-identical” so it should not be changed. He was very nerdy about it, so maybe that is the right way of his evolution.
I think Claude has a point here (and I would expect LLMs to excel at surfacing this kind of nuance). "Byte-identical" connotes a) that the thing consists of bytes (in particular, that it isn't a physical object); b) that it was deliberately reproduced; c) that it was most likely not reproduced by copying (unless that's explicitly indicated). "Duplicate" connotes a) that the thing is redundant or otherwise unwanted; b) is either an incidental or accidental result, or a deliberate copy.
Re: Show HN: The load-bearing vocabulary of Claude
#234Re: Show HN: The load-bearing vocabulary of Claude
#235I find Claude language often hard to process and having to wade through these words can be draining. Embarrassingly, I’ve recently caught myself using them in conversations! Do all models have the their own jargon?
I haven't noticed GPT 5.6 having any obvious jargon.
I've noticed ChatGPT these days often starts a response with "Yes." when it doesn't actually make sense to do so, and sometimes says "and" where "but" seems more appropriate (or vice-versa).
Re: Show HN: The load-bearing vocabulary of Claude
#236The vocabulary truly is load-bearing, without these words the model is less able to think. Where a human can understand a concept without words, an LLM plainly cannot. This is based both on the technological limitations and based on the evidence we see: as these models get better at working they get worse at communication.
> Where a human can understand a concept without words, an LLM plainly cannot.
If we accept that the LLM can "understand" at all, why would we reject the possibility of this understanding living in "latent space", before tokens are output?
Re: Show HN: The load-bearing vocabulary of Claude
#237Why are people getting so hung up on the "load-bearing assumption" turn of phrase that Claude uses? I get that it becomes cliche, but it is also a rather semantically dense way to communicate an idea that a lot of people run into.
Is "load-bearing" better than e.g. "critical" or "crucial" here? I'm not convinced.
Re: Show HN: The load-bearing vocabulary of Claude
#238Why "eleven", though?
Programmers would write it as 11 (or maybe 0xB depending on context), whereas English style guides mandate using words for small numbers, and LLMs have presumably absorbed that advice. I imagine that up to "ten" the effect is not as strong.
Re: Show HN: The load-bearing vocabulary of Claude
#239Interesting response from Claude, when I attempted to reduce the "load-bearing" in its responses: I added to my global prompt: - Orwell's first rule: never use a metaphor you're used to seeing in print. "Load-bearing", "the crux", "first-class citizen" signal insight instead of showing it. Name the specific mechanism when I asked what it thought of the change, its reply was: The Orwell bullet fights my own system pro…
It's not picking the word, the harness literally told it to plant it.
Also it can't simply replace it. Instead of load-bearing seams it could say something like heavy-lifting edges or regions to double-stich etc. But it's instructed to use that specific keyword if I understand correctly.
Re: Show HN: The load-bearing vocabulary of Claude
#240Earlier quoted context omitted.
This kinda dissonance is uncomfortable. How do people reconcile it?
The LLM's output being hackneyed or cliche doesn't mean it can't also be useful .
I don’t even find the readme useful. Like your average Claude prose, it’s simultaneously too information-dense and too light on information in a super frustrating way. Essentially it’s focused on explaining how clever it is (while not actually being very clever).