Top
Best
New

Posted by Labo333 21 hours ago

Show HN: The load-bearing vocabulary of Claude(louisabraham.github.io)
432 points | 200 commentspage 3
whywhywhywhy 9 hours ago|
I'm shocked "shape" isn't near the top
Labo333 1 hour ago|
[delayed]
datadrivenangel 10 hours ago||
so the real turning point for Claude is around April, which is Opus 4.6/4.7, which is right around the point where I personally started thinking Claude was getting different in a way which feels worse and is more awkward to use. It's better but has lost something that made it feel better.
incrudible 10 hours ago|
Is it actually better though? Can you even measure it? Unguided, Claude will just produce mediocrity which by its own existence becomes almost worthless, and the newer versions become increasingly difficult to guide while also being obnoxious to read. If I did not have to interact with it, it would not be a big deal, but then also I could not get value out of it.
datadrivenangel 8 hours ago||
It's a little smarter and more effective in some ways, especially those that can be benchmarked. I suspect that they've lost something that is hard to benchmark for.
confusedbucket 15 hours ago||
That confirms the recent spike of Claude calling everything I was recently working on a 'spike'. I still don't know what that term is supposed to represent (apparently).
Kwpolska 15 hours ago||
In some software development methodologies, "spike" is a task whose goal is figuring something out instead of delivering shippable code. https://agiledictionary.com/209/spike/
smj-edison 14 hours ago|||
TIL that agile has its own words for everything.
swader999 5 hours ago||
That was a thing in the early 2000's. You had to have a blog and invent words that describe regular things everyone did.
Espressosaurus 14 hours ago|||
Why can't they call it a prototype or experiment? Sheesh.
bloomca 13 hours ago||
Prototype is usually something working, while spike can be pure research (e.g. validate that APIs are feasible and enough, or that something satisfies requirements). Prototype is usually more polished/usable. Experiment in my mind is something which user-facing but not in stable yet.

So for me they are differentiated enough, but could be that I am just used to it.

Espressosaurus 12 hours ago||
An experiment doesn’t say anything about user facing or not. I do tons of them when reverse engineering poorly documented hardware and APIs to determine if the thing I’m trying to do is even supported.
xg15 12 hours ago||
A spike is an early prototype that you're supposed to throw away after having figured out the real design.

I think it was subtly dissing you.

confusedbucket 11 hours ago||
I think you are subtly dissing me :)

I was half-joking, of course I could've just asked Claude, but the linked site shows there has been actual recent spikes in the use of the word 'spike'. The term does match what I was recently doing, but hacking around legacy ERP software, blackboxes and other enterprise abominations isn't that out of the ordinary for me.

Art9681 6 hours ago||
Claude talks like me. I'm so screwed. I fully anticipate being physically present and verbally saying something and being accused of using AI to say it one day. Nevermind we're in the coffee shop and neither on of us has looked at a screen the entire time. The accusation is coming.
sergey_v 6 hours ago|
So you're saying you're safe for when the machines take over? :|
swader999 5 hours ago||
I don't understand why we tolerate this. We'd never hire someone that interviewed with this communication style and if we did we'd probably pip them fast.
internet101010 4 hours ago|
I already pipped Opus 5. That thing is a tool.
fny 15 hours ago||
While Claude's style is obnoxious, I'm more frustrated by its inscrutable explanations.

You need a PhD to understand its explanation of a code snippet.

Labo333 15 hours ago||
I'm not even sure a PhD helps. It just overuses jargon that has NO meaning. Sometimes, it actually hand waves too much as well while trying to dumb down stuff for you.

I am not sure whether it's a consequence of learning to reason from its traces or some RLHF that trips it into using weird terms to sound smarter to the humans who rate it.

fny 14 hours ago|||
PhD was a joke.

My intuition is that Claude is trained to communicate to itself while coding. You see this in how bizarrely granular it is when explanation prior work, you also see this in the comments it leaves behinds.

hedgehog 14 hours ago|||
It's me, it's the reams of sessions I share back with a five star rating that are just Claude Code talking to itself about debugging its own generated code in jargon that has slowly diverged from anything a human would understand.
zbentley 9 hours ago|||
Isn’t that basically what thinking mode is?
black_knight 14 hours ago||||
I have a PhD and can confirm. Oftentimes, the stuff which comes out of Claude is just impenetrable because it invents jargon on the fly, and uses verbs in the most atrocious ways.

"The fibred side folded its capstone into the existing name, so the kinds are asymmetric."

What on earth does it mean to fold a capstone into a name‽

jimmaswell 14 hours ago||
Is that an actual Claude output or hyperbole? It feels like I'm trying to parse an equation in a new math class which makes me want to take a stab at it regardless.

So there's a "fibred side".. the most likely candidate seems to be "fibred categories" which I hadn't heard of before, and it's talking about one side of some mapping between two sets such that if f is the primary function and f(x)=y then there exists an inverse function g(y)=x? Was it something that converted some data bidirectionally with a different algorithm on both sides?

The capstone of the inverse function would be the most important thing about it maybe?

My best guess is "In the process of working on the inverse function, the existing name (of the inverse function itself maybe?) was made to reflect the operation of the inverse function, so now the name does not follow the same naming convention as the name of the primary function (which does not contain its 'capstone')."

Its original wording is certainly dense and harder to follow for us, but it's fascinating how the model finds this the best fit for what it's trying to express IMO. Like it arrives at its own ways of overloading words/concepts, and things we would refer to in different ways in different contexts all get compressed to the same more-useful/complete idea.

Codex has never said anything nearly so alien as the Claude examples I've seen floating around, interestingly. I wonder if it just has a better training on choosing its words to present to the user or if it inherently arrived at a somewhat different mapping that favors 'plain language' more.

black_knight 13 hours ago||
This was actual Claude output on my screen at the moment I was reading this thread. No hyperbole!

As far as I can tell "the capstone" is what Claude usually calls my current goal if it thinks it is a satisfying result.

jimmaswell 12 hours ago||
How close was my guess?
black_knight 12 hours ago||
I am not really sure what Claude meant, but you are not too far off, from what I understand.

I have several similar folders with variants of a construction, but taking differently structured input. They are named “plain”, “fibred” and “indexed”. So the fibred variant is clear enough.

The Claude speak I struggle with is “the capstone” and what name it could be talking about. And what folding means here. I think it just means:

“I changed an important result of the construction in the fibred variant, but kept the name. so the fibred variant is now different from the others.”

swader999 5 hours ago|||
It's like it has a built in e bullshit generator.

https://www.bullshitgenerator.com/

ziml77 12 hours ago|||
Sometimes I can't even tell if what it's saying actually makes any sense to someone who understands all the terms its using, or if it's just throwing together words in a way that only make sense to its own model of language.
zbentley 9 hours ago||
> or if it’s just throwing together words in a way that only make sense to its own model of language.

alwayshasbeen.jpg

stereolambda 11 hours ago|||
It's interesting that there must be a decision behind that, even if it's just appealing to the RLHF judges for some reason. Maybe there's an intention that if you cannot decipher what the chatbot is saying to you, you will have to ask and burn even more tokens.

Naively I would often expect it would talk to me about various niche topics like to a layman, which does occur about some topics an actual normal person would ask.

throwaway219450 1 hour ago||
I used to assume it was the fault of average human annotators. That the people who are paid peanuts to rank chat outputs preferred the pretentious sounding ones. It wasn’t until quite a bit into the LLM boom that companies started to pay for domain experts. Overt watermarking is another possibility.

I’ve noticed that Sol is pretty good most of the time, but with long contexts it’ll start to devolve into Claudish.

josefresco 14 hours ago|||
They just addressed this with "Output styles"

https://code.claude.com/docs/en/output-styles

torarnv 8 hours ago|||
My experiments with output styles have not been sufficient to keep the model in check. It still spits out incomprehensible gibberish and load-bearing-isms.

Does anyone have an output style nailed down that actually works? If so, please share!

redak 13 hours ago|||
This solution:

1. Adds a small prompt to each turn with the agent[1]. 2. Is like a band-aid on a bullet wound, properly solving it would mean retraining the model and they probably are already working on it.

[1] https://x.com/_can1357/status/2090360068529111530

fouc 14 hours ago|||
don't forget LLMs are great at translating between languages, and within the same language. depending on the problem it works on, it will often reach for terminology that tend to be more common or familiar within that problem set. which appears inscrutable, but there's many different ways to skin a cat. just remind it to translate it back to the terminology and subject matter you're already an expert in.
dave1999x 15 hours ago||
Is it the obnoxious style that causes this?
condiment 14 hours ago||
I think it's the hierarchies of agents summarizing each others' summaries before presenting a final answer to the user. The principal agent has the full context from all its workers, but when it distills this down to a message to the user it summarizes it into a mess of confident jargon that pertains to a conversation the user wasn't a part of and never saw.
fwip 4 hours ago||
It happens with a single chat session as well, but not as often. I suspect it's the same thing going on in the hidden thinking tokens.
malshe 9 hours ago||
About a year and half ago I frequently used ChatGPT for speech to text conversion followed by summarizing the text because I ramble. Although the people receiving the text knew this, I absolutely hated AI's writing style. So I fine-tuned GPT 4o on about 500 short paragraphs and created a desktop app only for my use. It worked fine until about three months ago. It just can't handle the atrocious writing by latest GPT and Claude models. I tried to fine-tune newer models on HF but there is no way I can get rid of the cringy writing style. I even tried converting GPT 5.6 Sol's writing to GPT 4o and then using my app. Nothing works.
MaxwellM 15 hours ago||
Really spectacular analysis – thank you for sharing, fun to scroll and easy to understand.

Is it possible to expand this analysis beyond words to other Claude ticks? Contrastive framings, sentence length, caveating, for instance.

Labo333 15 hours ago|
Author here, thank you so much! I really tried to make it nice to use, beyond the (quite original) modelling.

A prototype I did tried to detect some grammatical constructions, eg "it's not ..., it's ...", but I am not sure how to systematize that.

Also just a disclaimer: I am NOT tracking Claude tics, I am merely finding that a particular cluster of vocabulary increases. Tracking Claude requires labelled data IMO. I tried using model release dates in a structural model to constraint the clusters but the result was not compelling, so I ended up simplifying the model a lot!

simlevesque 15 hours ago||
I wish there was a search bar for the terms, I wanna see for "gate".
Labo333 15 hours ago||
I thought about that, I might add it if I can find a nice design!
khatkhati 15 hours ago||
Chrome's `find` finds it for me ;)
Labo333 15 hours ago|||
I have been using it as well, but I think adding a search bar will heighten the experience. I'm trying out some designs right now :)
shrikant 15 hours ago|||
Yeah Ctrl/Cmd+F works just fine on Firefox as well.
swader999 5 hours ago|
I wonder if this jargon is an attempt or strategy to use less output tokens? It sure is annoying.
More comments...