upvote
It's a shame that I have to run a local model to decipher Opus, but them "dumber" models read far more naturally - https://github.com/gvzdv/claudish-to-english
reply
This will be very useful for me, but I wish it could be built natively into the agent harness.
reply
[dead]
reply
Astra told me yesterday:

> The run baseline was captured without a physical MAC; the current device is not durably bound to it.

> Engineering mode confirmation is the ESPHome component read-back; the LD2410 UART acknowledgement is not observed, so this is not proof the radar itself applied the sensitivity change.

No clue what the fuck any of it means.

reply
It was only when native English speakers—or those I presumed were—started calling out how bad "GPT/Claude speak" has become that I realized I wasn't actually losing my grip on English as a second language. For a second, I thought, Oh, I learned this language on my own, but it seems I've hit a wall and need to study further. It didn't help that I've also been trying to acquire Swedish as a third language for a while now.
reply
LLMs speak every language. I wonder if they're as insane in the other ones!
reply
I literally created a /plain-language skill.
reply
in which one!
reply
Not the original commenter, but I did this in all of them.

Their skills formats are basically identical, so I setup simlinks from their own skills directories into a shared one so Claude, Codex, Cursor, and anything else that comes out will all read and write to the same shared skills.

It's great having access to the same skills no matter the harness being used

reply
Its telling you your mmWave radar isn't speaking over serial communication well.

https://www.analog.com/en/resources/analog-dialogue/articles...

(Its negging your soldering)

reply
I long for the day when AI will just say that directly: "your soldering sucks man" instead of the bizarre made up and jargon packed language they use now.
reply
You can always update your claude.md!
reply
> (Its negging your soldering)

This made me laugh hard.

reply
Sometimes when I get frustrated reading Opus/Fable 5+ output I pause my rage out briefly to wonder if it's because I'm just too dumb for the model or if the model is just terrible at English.

I'm not sure that telling it to "try explaining that again, simply and briefly" is helping my ego.

reply
It's often simply misleading / bad writing. Here's one I just got about some crashes:

"If the crashes stop, the factory overclock is marginal; run a small negative offset."

This looks like it's saying: "If the crashes stop then we know the factory overclock is marginal." (This makes no sense.)

What it's trying to say is: "If the crashes stop then we can run a small negative offset, because the factory overlock is marginal."

What I would write: "If the crashes stop, we can avoid crashes by underclocking slightly. The speed difference between that and factory clock is marginal."

reply
I suspect this happens due to optimising for reasoning... if you insert a few words, it will suddenly start to make more sense.

"If the crashes stop, (that means) the factory overclock is marginal; (so) run a small negative offset. (to confirm this hypothesis)"

The core thought is basically avoid crashes -> caused by marginal overclock -> apply small -offset to test. Which is exactly the order the sentence is in :P

reply
I'm guessing it's because the way the first one was written looks real smart and sophisticated, which I'm presuming the models are rewarded for, especially when they're fed all kinds of PhD papers and so on as high quality, high weight data
reply
Is it possible that the first message is more information dense/less likely to be ambiguous than the latter? It’s clearly being selected for for some reason, maybe it’s an artifact of the tokenizer or specific training data, but I don’t know. If the use of jargon was complete cruft, I would expect it to be selected against during reinforcement learning
reply
You’d think that, I thought that… but then I realized I’m just kidding myself thinking its output makes sense. It doesn’t. It doesn’t. Sometimes it might as well just speak tongues.

In other words, it ain’t you. It’s the model. It’s just genuinely bad.

Then you switch to ChatGPTs lineup and realize how things can actually be better. It took about a week to really get the feel for how to use their models… then I basically switched. I’ll check in every now and then when they actually make a deal about how opus “now makes sense”.

But honestly I’m half convinced Anthropic actually prefers the output of opus 5. I dunno why, but how else could you explain how such a thing got shipped? I mean somebody in the pipeline had to say “dude this model doesn’t make sense, you think we should fix it?” Right? Like it’s a pretty massive drop in quality for such a major brand in this space, you know? How did it make it out the door?!?

reply
I thought it meant "the factory overclock is marginal" in the sense of "borderline unstable"?
reply
Same. I read it as "if the crashes stop [ when we test by reducing the clock ] then we know that the overclock applied by the factory is marginal [ ie it barely passed QC or maybe there wasn't proper QC to begin with ] so running with a small negative offset [ ie what we just tested ] can be expected to fix the problem for good". No idea if my reading is right given all the context I'm missing. Either way it's absolutely shit writing in the same way that golfed code is shit code (except when participating in a code golf competition).
reply
This is my read, too.
reply
I wonder if this is a result of them trying to cut token consumption by summarizing their RL training data, or maybe it's from how they anonymize user data for training.
reply
Marginal - definition 2a: of, relating to, or situated at a margin or border.

Succinct and precise; a well crafted sentence. A marginal OC results in unpredictable crashes and can be corrected with a small offset; marginality describes the behavior and explains the solution.

Inscrutable clues casually conveyed can now be readily explained, at least, unlike the training data of [silence]. Brevity is the soul of wit, but perhaps also exasperated confusion.

reply
I just started to use GPT models. it's incredible how this seemingly is not a problem in the OpenAI world

it's absolutely not just you, the text it produces causes my blood pressure to go up.

reply
I'm constantly using the "Extract this in basic technical terms, be succinct and assume the reader has technical knowledge"."

Because good lord, does claude waffle when left to its own devices.

reply
This is actually a new skill I've been working on. Learning how to elicit concise and simple speech from models (and from people to!).

Whenever I come to a wall of complicated text I kick into gear and think through getting it to distill this into the high-level useful bits that I actually need to know.

I guess I could create an actual agent skill for this :) And next-gen models might eventually be trained to simplify their output themselves...

reply
too.

(sorry)

reply
The most surprising part, however, is that when one model slops this into a plan, another model somehow is able to interpret it correctly enough to produce code to spec.
reply
I have shared this dismay. I’ll have opus create a plan, I read it doubtfully. And then sonnet implements it. I am surprised it went so well. I theorize the redundant verbosity effectively builds rails that help keep llm focused. I will experiment with such rails myself.
reply
I mean, humans have been doing just that for a long time.
reply
I bet this is what the thinking blocks look like. If so then it maybe it is intelligible, just not to us. I have the same problem.
reply
Well, whatever its thinking block looks like, this is when it was talking to me. I suspect you're right, though, I think it thinks it's thinking.

It seems like it doesn't have enough of a theory of mind to know that other people don't think exactly like it thinks.

reply
LLMs seem to create abstract, local jargon as a side effect of way it reasons using tokens

ChatGPT told me its "semantic compression"

reply
There’s some specific terminology here, like the MAC address of the network device, which might have been virtual.

UART is a hardware circuit for communication, possibly a serial port. Were you trying to reverse engineer a consumer device or appliance?

This particular instance doesn’t seem terse, but I’m sure it has been on other occasions :)

reply
Reminds me a bit of VXJunkies
reply
Have you tuned your Retro Encabulator recently?
reply
How is that not plain English? why do my friends not like me? Hmm....
reply