iterateoften

6 ms·

31.
by iterateoften 4mo ago
Yeah, I always wonder if they do some type of obfuscation and transformation on the private data and find a way to backdoor the info without technically using it directly.
32.
by iterateoften 4mo ago
I noticed since 5.5 GPT has been adding "lattice" to a lot of things. Not sure if it is the new Gremlins.
33.
by iterateoften 4mo ago
More like its a regulated space and it makes basic sense to have regulations
34.
by iterateoften 4mo ago
“The model mimicked the output of the training data” is a less impressive press release.
35.
by iterateoften 4mo ago
Same thing with extraterrestrials. One side is confidently shouting maybe aliens exist and visited earth. The other side calmly explains every example brought up about aliens visiting is easily explained by something more simple. The “ali
36.
by iterateoften 4mo ago
Aren’t all the reasons people use mobile because of its capabilities? Like the fact that it is in fact mobile?
37.
by iterateoften 4mo ago
In my experience, investors push hard to go in office when fundraising
38.
by iterateoften 4mo ago
How much dilution? Who’s getting the value?
39.
by iterateoften 5mo ago
Just a question, but on these threads it’s nonstop talk about how dangerous the paper is like using it for one dnd game will give you cancer but we don’t blink twice at cashiers handling it 8hrs a day?
40.
by iterateoften 5mo ago
Going from Bambu to a self printed 3d printer I don't even think counts as the same category of devices. Bambu is concentrated on making a plug and play device that just works.
41.
by iterateoften 5mo ago
People love defending Anthropics shortcomings… “Mythos isn’t supposed to be that good at security, because actually Anthropic was referring more about running llms than mythos specifically” “The opus model is worse because they have no comp
42.
by iterateoften 5mo ago
It’s easier to judge an objective output like number of words than subjective like quality. Same as lines of code, etc.
43.
by iterateoften 5mo ago
focus groups are like the sobriety tests on the side of the road. Its just performance and the conclusion was made before it even started.
44.
by iterateoften 5mo ago
People did and still do the same with electricity and magnets. Both incredibly useful despite it.
45.
by iterateoften 5mo ago
This is funny because it’s a silly topic, but I think it shows something extremely seriously wrong with llms. The goblins stand out because it’s obvious. Think of all the other crazy biases latent in every interaction that we don’t notice b
46.
by iterateoften 5mo ago
The next loop
47.
by iterateoften 5mo ago
Lol, I respect karpathy a lot, but this is such an obvious in your face idea that it is laughable to put someone’s name on it. What’s next “karpathy investing” where ai in a loop builds a portfolio?
48.
by iterateoften 5mo ago
Im left delighted to find out something new, but left wanting to know how to use it. Like if im 75% on the green transition, how do i use this information.
49.
by iterateoften 5mo ago
Yeah seems like nonsense advise. Have a code word that was never recorded? I don’t see how that would tote y anything. Like the point of these systems is they can say stuff you never said convincingly
50.
by iterateoften 5mo ago
Zen meditation for an hour staring at a wall is a marathon that at the end results in a semi-psychedelic state for me. Exercising and sitting b meditating are two related but seriously different things. Which is why there are many other typ
51.
by iterateoften 5mo ago
Unless you know exactly why paper trading sims are so hard to backrest in practice, it’s silly to make arguments on why your paper trading sim works. It’s insanely easy to make a trading algo profitable on historical data.
52.
by iterateoften 5mo ago
You don’t need for every item. Just low frequency high profit goods
53.
by iterateoften 5mo ago
It’s also that agents and ML reach local maximima unless external feedback is given. So your wiki will reach a state and get stuck there.
54.
by iterateoften 6mo ago
Cultural influence is another benefit. China is securing its sphere of influence as well as keeping us ai in check.
55.
by iterateoften 6mo ago
Beware. I had Claude code with opus building boards and using spice simulations. It completely hallucinated the capabilities of the board and made some pretty crazy claims like I had just stumbled onto the secret hardware billion dollar pro
56.
by iterateoften 6mo ago
No it’s because Anthropic can’t message anything to its customers without lying.
57.
by iterateoften 6mo ago
It’s the official communication that sucks. It’s one thing for the product to be a black box if you can trust the company. But time and time again Boris lies and gaslights about what’s broken, a bug or intentional.
58.
by iterateoften 6mo ago
The high cost of labor has nothing to do with reusable parts of the washing machine. If it was all standardized, you could do a lot of repairs yourself.
59.
by iterateoften 6mo ago
I see an objective difference in speed between Team account and my personal max 20 and a subjective difference in quality. On my personal account projects get 90% and then get stuck on a bug and then I have to hand the project over to codex
60.
by iterateoften 6mo ago
The definitions of what is messy or clean will change will llms… But there will always be a spectrum of structures that are better for the llm to code with, and coding with less optimal patterns will have negative feedback effects as the lo

More ›