The article starts with a pretty weak simile and is structured in a way that reminds me of llm output. Made me stop reading pretty quickly. I’m wary of essays that take a genuinely complicated organizational problem and…
what the brain needs is Default Mode Network, not more stimuli
Seemed interesting until I read that Phosh pulls in GNOME - gnome-settings, gnome-session etc. Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery…
Decaf is typically not 100% free of caffeine, and coffee contains other minor stimulants as well
> There are now Asian models coming , optimized focused on cybersecurity defense at a high level, so I suspect this will be a relatively moot point soon. Source?
OpenAI, the non profit organization, is going to become a publically traded profit maximizing corporation
Your work experience does not represent the current state of the art in car manufacturing automation
The hard parts are not based on OSS
EBITDA is typically used to evaluate profitability.
Russia is the aggressor. What do you propose?
Satellites have multi kW solar panels
What a nothingburger
But it's a 26 (Tahoe) bug. Earlier OS versions unaffected
Your cloud act is making American cloud vendors lose customers in droves
Lesser of two evils etc. How about south america as a counterexample?
If it ever comes to that, we'll pick China, thanks
Australia is pretty close to the equator
The quoted writing is AI slop, and OP is reacting to the fact that they did not write even the introductory text themselves (or at least bother to edit out clear AI/slop indicators)
Why store uncompressed layers? Surely decompression is fast enough that there's little gain in storing the decompressed layers?
It takes like an hour. So that's a quick mvp then
But if it stores the uncompressed layers, why store the compressed ones too? Why both at the same time?
Is your experience with this new quantization approach from Intel? Otherwise your comment is a bit offtopic at best, misleading at worst.
Well, you see, they just can't find a checkbox for ipv6 support in the IIS GUI on their ingress servers.
It's only open because nobody who's interested in this model would send their data to openai to be stripped of PII. If they thought otherwise, it would be closed-weights and API-only for "safety" reasons
Well let's talk again when the problems have been solved, then. Until then, manually curated skills and documentation will beat this
The article starts with a pretty weak simile and is structured in a way that reminds me of llm output. Made me stop reading pretty quickly. I’m wary of essays that take a genuinely complicated organizational problem and…
what the brain needs is Default Mode Network, not more stimuli
Seemed interesting until I read that Phosh pulls in GNOME - gnome-settings, gnome-session etc. Seems like a very strange bundle to bring in for an extremely power constrained device, where every % of increased battery…
Decaf is typically not 100% free of caffeine, and coffee contains other minor stimulants as well
> There are now Asian models coming , optimized focused on cybersecurity defense at a high level, so I suspect this will be a relatively moot point soon. Source?
OpenAI, the non profit organization, is going to become a publically traded profit maximizing corporation
Your work experience does not represent the current state of the art in car manufacturing automation
The hard parts are not based on OSS
EBITDA is typically used to evaluate profitability.
Russia is the aggressor. What do you propose?
Satellites have multi kW solar panels
What a nothingburger
But it's a 26 (Tahoe) bug. Earlier OS versions unaffected
Your cloud act is making American cloud vendors lose customers in droves
Lesser of two evils etc. How about south america as a counterexample?
If it ever comes to that, we'll pick China, thanks
Australia is pretty close to the equator
The quoted writing is AI slop, and OP is reacting to the fact that they did not write even the introductory text themselves (or at least bother to edit out clear AI/slop indicators)
Why store uncompressed layers? Surely decompression is fast enough that there's little gain in storing the decompressed layers?
It takes like an hour. So that's a quick mvp then
But if it stores the uncompressed layers, why store the compressed ones too? Why both at the same time?
Is your experience with this new quantization approach from Intel? Otherwise your comment is a bit offtopic at best, misleading at worst.
Well, you see, they just can't find a checkbox for ipv6 support in the IIS GUI on their ingress servers.
It's only open because nobody who's interested in this model would send their data to openai to be stripped of PII. If they thought otherwise, it would be closed-weights and API-only for "safety" reasons
Well let's talk again when the problems have been solved, then. Until then, manually curated skills and documentation will beat this