avereveard

4 ms·

1.
by avereveard 7d ago
btw that how mmlu score things to answer question instead of producing all the answer token they look at logprob of a b c d keys in 2020 making this technique old as dirt in nlp
2.
by avereveard 9d ago
Aws cto is basically opening every presentation with a everything breaks all the time slide, idk where you get that idea they oversell the cloud resilience
3.
by avereveard 11d ago
Opus specifically talks as if having a stroke. 4.6 was the last version that was pleasant to work with
4.
by avereveard 11d ago
Yep the only way out is hooks to forbid what can be detected by ast and second model to prune comments, flatten pyramids of fallback, and squash the test suite removing quirks maintaining wanted behaviors.
5.
by avereveard 12d ago
A tool aggregation layer made of code is a good way to save tokens Composition is an issue only as long as one keep demanding tool calls in json. If tools are goal predicates in prolog, it's easier.
6.
by avereveard 14d ago
Have your tests been ran on vpn with vpn cloaks?
7.
by avereveard 15d ago
Yes nobody is producing early 90s punk with the same intensity and dinosaurs never died, sorry nofx, so now I have my personal playlists.
8.
by avereveard 16d ago
Eh was ever confirmed they were under ZDR or not by them? Don't like to blame alleged victims but lack of a clear claim after these many days is not a good look. Was ai research allowed, under which guardrails, and what was the policy
9.
by avereveard 16d ago
No but non enterprise users will receive injected prompts. See how emany streaming now created a plus plus version while adding ads to what was the premium version. And the funny thing they just needed to increase base rates for users to st
10.
by avereveard 16d ago
This era reminded me of late 90s web culture, and it's heading the same way.
11.
by avereveard 23d ago
Eh we can detect opumism and gptisms our brain are very good at pattern recognition even if subconscious
12.
by avereveard 25d ago
Post edit hook that reject edit based on comment density, mine is at 5% you will also need to heed deny file edit in automode as the rascal will try that to preserve prose
13.
by avereveard 25d ago
There's layers, some of token selection is fingerprinting https://github.com/google-deepmind/synthid-text
14.
by avereveard 25d ago
You start with the obvious cases of sending you to a domain that serves malware, sure there will be instance of fraud that will slip trough, but lets not let best be the enemy of good
15.
by avereveard 1mo ago
not all regulation is bad regulation. micro packaging is waste, and waste is an issue in europe. the problem stem when you get conglomerate unfairly competing from deregulated countries, applying maximum legal optimization to skirt regulati
16.
by avereveard 1mo ago
Border control is border control, this is not a stop and frisk scenario.
17.
by avereveard 1mo ago
Don't forget also Screamers is happening
18.
by avereveard 2mo ago
No it is a vibecoded mess. Bet if you show the screen to a agent of your choice you will have yours in no time. The benefit is that if you give it self installing scripts you can just ask itself to add features Cgroup and nice it so you nev
19.
by avereveard 2mo ago
tried similar setups in the past, but messages is just not my jam, so built a html wrapper on top of claude code and open code that runs on a small-ish instance. looks like this https://i.imgur.com/lj9Fgco.png (screenshot a
20.
by avereveard 2mo ago
agree, technology doesn't change the basic mechanism of capitalism so I don't think this will be different, the only real question is how far means of production will be in term of capital intensity from upper class, which will de
21.
by avereveard 2mo ago
that was a big jump from what I said. and what I said was that research lives off grants so researcher can't really come back and cry about strings attached to grants. ofc the problem is faceted, as grant go to published authors with t
22.
by avereveard 2mo ago
Grants.They want the juicy grants and complain of the strings attached. Guess the single strongest predictor of paper in field per year?
23.
by avereveard 2mo ago
I've data where i cannot store metadata that i need to search semantically so i embed it on the fly at every search with static embedding and tokenizing was more than 99% of the cpu time. Granted that was due the naive implementation o
24.
by avereveard 2mo ago
Compute will become the means of production and we are not going to get a share of it because the world is hyper optimized for value extraction
25.
by avereveard 2mo ago
these materials are transparent to light it is not a property of the human vision system that light goes trough N2
26.
by avereveard 3mo ago
steering: You are the sole owner of the project implementation. User maintained documentation: - goals.md for the project overall goals - tech.md for guidance on how to build the project Agent maintained documentation, current state living
27.
by avereveard 3mo ago
Same. For some reason late opus model are very superficial doing ux work and so am using gpt for that, but backend is much better engineered by claude, gpt prefer to duplicate everything it needs on the spot causing class sprawl
28.
by avereveard 3mo ago
Very good on vision, really helped oneshotting complex ix thay I had so far to buukd piecemeal. Ended the 200$ plan weekly allowance in two days, so theres thay.
29.
by avereveard 3mo ago
Mybatis was a thing even back then... you still need a domain model after all
30.
by avereveard 3mo ago
never said purely functional, but there are only 4 data channels in each method (input, return, calling another method, setting a state) - and if you constrain your class design to pick only 2 for each method, your life is just a tad easier

More ›