Have you tried turning it on and off? More seriously try to remove agent memory or make it refactor it or go through the docs and look for inconsistencies.
But I still don't know how people work with codex when it keeps resetting the context so often, I mean it's surprisingly good at making notes to itself and can follow a long task, but if there are multiple instructions it sometimes forgets some of them.
26% weekly limit left, 21+ hours strong and still going. I can see the progress though, but I am not confident it will finish before weekly limit is done for (ChatGPT Pro 20x subscription). Since I don't really use LLMs for anything else, except pestering gemini for stupid questions instead of straight up googling it, that's ok. I just wish it would either finish or let me hand it over to it once again whence limit resets.
Maybe I’m a Luddite but I find AI prose completely tiresome to read, and its presence in an article forces me to be skeptical about whatever the author wrote. The latter was obviously a problem pre-AI but these tools simply make it much easier to tell stories, narratives, etc. that aren’t yours.
Due to their past, LLMs need to prove themselves as correct and accurate. Bloggers have already gone through this gauntlet and have proven to be more likely to be correct and accurate.
Yeah, and there is no pleasure reading this. It's way too long and verbose for what it's trying to say. I'd much rather read the prompt, in bullet points or whatever format they were provided before inflating a few example pictures and notes into a short novel
How is this criticism not just ad hominem slop? You could have said "It's way too long and verbose for what it's trying to say". The writer is orthogonal.
The journey to a magical, free bird sound ID app involves some obvious steps (the NSF gave them money), but the amount of timeline offramps is really high. The NSF didn't think this app was educational ("it's just telling them the answer, they don't even have to work for it!"), the funding ran out so they were thinking of charging money, eBird happened to have five years of bird sound recordings they could use, a computer vision guy on the team happened to learn that spectrograms are basically the same as pictures...
And the result is tremendously useful! They get great signal on where birds are, and they're making people so much more interested in birds and nature and science. And I think charging $1 for the app or, shudder, some sort of subscription model might've ruined the whole thing.
Funnily enough I was in the bathroom (that's where I do my most random thinking) just yesterday evening thinking how template matching in computer vision, e.g. https://docs.opencv.org/5.0/py_tutorials/py_imgproc/py_templ... is basically audio signal processing but instead of being temporal it is spatial. Different signals but same principle.
I've got 33. I don't know how that stacks up against other but a large percentage of them came just from my backyard. Just got added a Baltimore Oriole that showed up this week.
I'm up to 71 thanks to the joys of taking it on holiday to other countries (less joy for my family), and occasionally going places just because I know there'll be new birds there.
Personal best for a single session is 23 birds in my back garden over the course of a couple of coffees. I have though spent a fair bit of time making the garden as attractive to different types of bird as possible - mainly through having a variety of different shaped feeders with different feed in. I can recommend the NatureSpy camera feeder too [0] - though I have had to modify it a little to make the food hole smaller otherwise magpies empty the thing in a single day just chucking all the seed on the floor. Dickheads.
Tawny owls make so many different noises. The classic sound is two, one calls and one answers but they are very chatty with different sounds. They live all around, but never seen one.
Merlin got me into birding, and I've found this to be true, at least in some regions, especially if there's lots of background noise.
Unless a species has a really, really unique call or song, I've made it my personal rule to get eyes on it and, if I'm still not convinced, to take a picture and cross reference with iNaturalist and what folks on there think of it.
I've learned to curse the species that differ from others by the sheen of the backs of the males' necks or the angle of a wingbar. Also, gulls that only differ by things like the color of their feet or beaks.
For me, mostly a lot of false negatives. I have a Pixel 8, and I have no problems using the microphone normally, but somehow, Merlin requires the bird to be within a couple of meters of me and sing for quite a while before Merlin will even say “hearing a bird”. And of course, you have to be dead silent; if anyone talks or coughs, it will mess up the spectogram badly.
Given that it's colloquially “Shazam for birds” and Shazam is just amazingly resistant to noise, it's a bit disappointing :-)
I think what makes Shazam magically "just work" is that they are searching for the exact embedding of the recorded version of the song - it doesn't work for a cover version, acapella, etc even if it is very similar. (Everyone would want an app that you can hum into and it will ID the song, but I guess it's a very tough problem.)
Merlin is solving the more general problem so makes sense that it's much more finicky and less accurate.
Still, we're talking orders of magnitude here. I've fed Shazam with a stream of the cheapest lapel microphone you could get, mounted behind a rack filled with twelve very noisy servers, in a hall filled with 5000 people also all being noisy, and it would reliably recognize music from the other side of that hall before I could do so myself (I had to walk halfway there to hear “oh, yes, it's actually right”). And that's on a stream encoded with 9600bps GSM compression. With Merlin, I can hold out my phone on a silent day, hear the bird loud and clear and it will hardly register.
Shazam primarily identifies on the geometry of spectogram peaks, FWIW (I wrote my master's thesis on the DSP of music recognition back in the day; it's possible that they are doing something more fancy now, of course, but I'm not sure if they would want to). I don't know exactly what Merlin is doing.
> an app that you can hum into and it will ID the song
YouTube Music and Google Search do OK with these types. Sometimes if I know a song well enough, I will open YT Music, activate the mic search, and sing a few bars, instead of typing the song title.
According to Gemini, SoundHound is also a good app for song-hummers to try.
I'm not sure why but this seems very phone dependent. I can stand next to another person, both running Merlin in silence, and mine will pick up much more than some people's phones. My phone is not particularly new (Zenfone 8).
You can usually tell, though. Rare birds in the wrong place or time that come up only once are problematic, but if the bird is detected repeatedly there's a better chance. You should also be on the lookout for birds that mimic other birds (like the northern mockingbird and the gray catbird.)
Occasionally I'll see a screenshot of Merlin with, I dunno, a dozen different birds, and in the middle of the list: a catbird. I never have the heart to tell people that they just heard one very creative bird.
Cornell / Merlin / eBird publishes a global species frequency DB indexed by location and date (so, how likely is a species to appear in a place, on a day of the year)... In a project that I was working on, this helped a lot when used to influence the "raw" identification based on appearance alone.
Indeed - background noise will tend to lead to a lot more false positives, but you should use it as a guide rather than inherently trusting results. Same goes for using BirdNET-go and other models, though these do allow you to be very strict with detections and reduce false positives more.
The point is that auto mode gives people a false sense of security that leads them to believe they don't need to run Claude in a proper sandbox. This same attack running in a sandbox (even in YOLO mode) would be comparatively harmless.
Ever since they made auto-mode default I swear claude has tuned to use python commands instead of the Edit Tool to frustrate the ~security conscience~ luddites into using auto-mode.
Yeah, It’s in the system prompt, Claude will tell you if you ask why it’s using Python.
My theory is that Anthropic is just a vibe-coding company. Their goal is to capture the attention of white-collar non-coders, since programmers will jump ship fast to another model.
That's what I meant - you give it an instruction that seems to work (always ask before deploy) and so you trust it, and then you notice it can easily convince itself to deploy without authorization ("the user asked me to fix this, and they must know it's a deploy ...").
What seems to work for me is automation - read file hook that re-injects instructions in the prompt every 15 minutes. Switch on the filename and get language-specific instructions too.
Is this about normal system prompt instructions or instructions for the auto mode classifier? I'd be a bit more surprised about the classifier forgetting instructions.
The commenter you replied to mentioned that you can customize the auto mode classifier by providing a prompt, implying that this would be a more robust way of constraining Claude's behavior. It wasn't clear from your response whether you were using this functionality. You might try it out as a way to more reliably prevent these kinds of workarounds.
It's absolutely not reliable, and we have opened a few issues for that.
For example: our instructions (which are read by the model and classifier) include "do not use sed/python/perl/etc, always use the edit tool for editing", and this only gets followed for a few messages. We have introduced scripts to block those ourselves, since the classifier doesn't care.
Because of those problems, my team is currently testing OpenAI after about a year of Anthropic.
I also feel like this is an attack that manual review is not that likely to catch, given none of the malicious code appears in any of the tool calls or output.
Can you suggest a proper sandbox on mac? One that allows both me and the agent to interact with the processes? Where it can drive browser, for both oauth setup and runtime visual inspection? I've tried building docker setups, but can't figure out the browser driving part.
I don’t think it’s related to the auto mode at all. It would work perfectly in the manual mode. It does not even need Claude: just give a human a similar archive and hope they run some simple Python from the directory at least once. And make sure there are lots of files do they don’t notice a weird .py around
RAGish storage with different access methods including IM and parsing data from photos sounds good to me. But openclaw seems like a bit too big of a gun for that, did anybody create something more suitable for the job yet?
This is nothing. China is about to start doing their own 5nm. When they roll out their own silicon at scale, I don't think there's catching up, nobody can match their manufacturing capabilities. Not even close.
They are already doing 5nm with multi-patterning DUV. But that's not the same as EUV machines. It is not nearly the same scale of production (with yields as low as 20%). They are still a few years away from indigenous EUV. Though I have to admit the shenzhen breakthrough was pretty impressive
For multi-patterning DUV, each patterning/etching step accumulates defect risk. Actually now that I'm reading about it, 20% is the better end of the spectrum. I don't think technology could possibly improve it much beyond that.
I believe the theoretical maximum yield for multi-patterning DUV at 5nm and 7nm is 30-40%
It's not economically sustainable/viable long term. China is only doing it because they've been forced to by the trade war Trump started against them in his first term (Biden didn't really back down either and now Trump 2 tried to ramp it up even higher). China is okay with such low yields because its supposed to be a temporary solution until they achieve domestic EUV tech.
Surely yield is very much a function of the size of the device in question and its tolerance for defects. I don’t think it makes sense for a technology to have a theoretical maximum yield that’s a constant number unless it somehow destroys entire wafers with highish probability.
You can normalise by surface area, which makes defects rate comparable across processes. At least to a first order approximation, keeping all other parameters constant, bigger chips are more likely to have a defect because they have larger surface areas.
If a process has a defect rate of x, the probability of a die having a defect is A*x. The limit is then the defect rate for a die with the surface area of a transistor, you cannot really get any better than this. But that’s not very useful if you want to extrapolate what the defect rate would be on an improved or cheaper process.
Even though there are so many packed buttons and links it seems to like a vastly superior design compared to modern websites. I don't know if it's me being used to this or if we went wrong somewhere.
Funny, I entirely disagree. On one hand, the design itself is heartwarming because it's old internet. Also, once you navigate into the topics, things are neatly organized, and there I agree with you. But the content on the main page itself lacks a common thread, no idea where i'm supposed to read or what belongs together.
I'd say there is a pretty common thread for it, but man is it a mishmash of interests, which really makes it charming.
It's certainly not focused. Sure this site is about RF, but the owner decided they'd throw in basically all their interests into one website. I mean, it has crossword puzzles and a jokes page :D
This is the site to send someone to who wonders "What was the internet like before Web 2.0".
I just browsed it on mobile. Each article preview is in a little box, just scroll around to it. It works fine for me.
The alternative modern mobile view leaves very little on the page, and forces you to click back and fwd, and back button functionality is usually broken. I think it's easier to scroll
I've been using Reolink for years and been very satisfied with it.
The only quip is the default UI isn't very good. When changing that reaches the top of my priority list, I'll switch it since they don't force you into a walled garden. Plan is to run it through frigate into HomeAssistant and use a UI from them. I've never used frigate before though so it'll be a learning process if plug and play solutions aren't already available
But I still don't know how people work with codex when it keeps resetting the context so often, I mean it's surprisingly good at making notes to itself and can follow a long task, but if there are multiple instructions it sometimes forgets some of them.
reply