Comment by stavros
7 hours ago
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.
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.
LLMs speak every language. I wonder if they're as insane in the other ones!
My experience with opus 5 is that its results are lower quality in dutch, but that its dutch is more readable than its english.
I literally created a /plain-language skill.
in which one!
1 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)
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.
You can always update your claude.md!
> (Its negging your soldering)
This made me laugh hard.
It wasn't, it was saying it hadn't looked at the UART because it only had access to the web API.
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.
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."
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
2 replies →
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
I thought it meant "the factory overclock is marginal" in the sense of "borderline unstable"?
2 replies →
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.
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.
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.
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.
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...
too.
(sorry)
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.
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.
I suspect it's because the different models co-evolve? The labs train on one model implementing the plans of another model, especially in the same family of models (like Fable to Sonnet).
I mean, humans have been doing just that for a long time.
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.
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.
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 :)
It was saying it can't tie the calibration results to a device, because it doesn't know the MAC address (I never asked it to look at the MAC address, it way overengineered things).
It also couldn't see the UART communication and could only see the web API endpoint, hence the rest of the slop.
Reminds me a bit of VXJunkies
Have you tuned your Retro Encabulator recently?
LLMs seem to create abstract, local jargon as a side effect of way it reasons using tokens
ChatGPT told me its "semantic compression"
How is that not plain English? why do my friends not like me? Hmm....