robkop

5 ms·

1.
by robkop 1mo ago
One of the main goals at the moment is to keep thinking human legible. You can imagine how much harder it’d be to do root cause analysis on the recent OpenAI event if we couldn’t even tell what they’re thinking.
2.
by robkop 2mo ago
Claude design's prompt is trivial to verify. They bundle it in the frontend bundle and send it on every network request.
3.
by robkop 3mo ago
CoT legibility largely disappears which is quite concerning from a safety perspective
4.
by robkop 4mo ago
Just saying you’re not alone, very surprised by the reception given how brutally sloppified the OP is. Interesting problems space but I hope the author just gives dot points next time rather than bloating it and losing most of its meaning.
5.
by robkop 5mo ago
Could you please elaborate a bit more for my understanding? What in particular about this method breaks correct token boundaries? On my first read I read your comment as there are special tokens that require multiple tokens to emit, hence y
6.
by robkop 5mo ago
I use bedrock with 1M context every day. Not sure this is right
7.
by robkop 5mo ago
A lot of enterprises were doing that but now they hit the 150 user limit on Claude and are paying seat+api rates. Codex is still going strong but it’s hard to imagine they won’t do similar eventually. So now im honestly hearing a lot more f
8.
Show HN: Actual Claude Tokenizer(tokenizer.robkopel.me)
3 points by robkop 5mo ago|4 comments
9.
by robkop 5mo ago
There’s a lot of tradeoffs to play with, those inference ASICs may not carry the gradient but they are still optimised for larger batches and to run any model. They need enough memory for the weights, wide batch inference, and ideally lefto
10.
by robkop 5mo ago
You can ablate surprisingly large chunks of a model with near to no effect, you can try this easily - download an open weight model in torch. Obviously it’s not ideal but you could likely have single digit % of all weights affected and stil
11.
by robkop 5mo ago
I can’t speak for the states, but in AU I clearly see a massive displacement of undergrad and junior roles (only in AI exposed domains). I say this as both someone who works with many execs, hearing their musings, and someone who no longer
12.
by robkop 5mo ago
one of their highlights with mythos was it's ability to generate new puns I took a look and honestly they're the first AI puns that aren't bad Times are changing
13.
by robkop 7mo ago
We’ve got a long way to go in optimising our environments for these models. Our perception of a terminal is much closer to feeding a video into Gemini than reading a textbook of logs. But we don’t make that ax affordance at the moment. I wr
14.
by robkop 7mo ago
Rumours say you do something like: Download every github repo -> Classify if it could be used as an env, and what types -> Issues and PRs are great for coding rl envs -> If the software has a UI, awesome, UI env
15.
by robkop 7mo ago
I get this at least once a week. And then once you have to dig in and understand the full mental model it’s not really giving you any uplift anyway. I will say that doing this for enough months has made my ability to pick up the mental mode
16.
Show HN: You Are an Agent(youareanagent.app)
14 points by robkop 8mo ago|0 comments
17.
You Are an Agent – Try Being a Human LLM(youareanagent.app)
3 points by robkop 8mo ago|1 comments
18.
by robkop 8mo ago
I added a "Human" LLM provider to my local OpenCode a few months ago as a joke, and it turns-out acting as a LLM is quite painful. But it massively improve my agent harnesses dev skills. So I thought I wouldn't leave anyone o
19.
by robkop 8mo ago
It's a fair question - I think the fact that they hold abilities (read 200k tokens instantly, can clone themselves, ...) that we don't would suggest they will have quirks and differecnes. What downstream implication that will have
20.
Ax Not UX(robkopel.me)
3 points by robkop 8mo ago|2 comments
21.
by robkop 8mo ago
https://robkopel.me
22.
by robkop 9mo ago
Can you elaborate? I would have thought the main driver for the price of a service is the labor?
23.
by robkop 9mo ago
Does that cost to serve multiple stay the same when conventional sites are forced to shovel ai into each request? e.g. the new google search
24.
by robkop 9mo ago
I’ve heard too many rumors that much of that adoption is from copying ms i.e. bundling gemini into their office suite
25.
by robkop 9mo ago
Occam's Razor - this complexity arises from the human nature to try and build consistent abstractions over complex situations. It's exactly what we do in software too. To an outsider it's going to look nonsensical. I want to
26.
by robkop 9mo ago
For those curious about the "consistent principle of law" here - SCOTUS wrestled with nearly exactly this question in Free Speech Coalition v. Paxton earlier this year, and effectively emboldened more of these laws. Previously the
27.
by robkop 9mo ago
Interesting question - how much will end up as sound, or in the ever smaller tail of things like storing a bit in flash memory?
28.
by robkop 9mo ago
I remember you did some reverse engineering when they released agent, does it not feel quite similar to you? I know they didn’t dynamically scan for new skill folders but they did have mentions of the existing folders (slides, docs, …) in t
29.
by robkop 9mo ago
Hasn’t ChatGPT been supporting skills with a different name for several months now through “agent”? They gave it back then folders with instructions and executable files iirc
30.
by robkop 10mo ago
You could dual brand as vibe-npm, only install packages that are in your models training dataset

More ›