fwip

8 ms·

151.
by fwip 2mo ago
They've had 20+ years to add one.
152.
by fwip 2mo ago
I believe Mojo has been primarily using AI-generated imagery of its mascot for several years now, which politically aligns with its pivot to being a language for AI development.
153.
by fwip 2mo ago
Customers are not perfectly-informed rational actors. The areas that companies are cutting costs on are precisely those that are most likely to escape the customer's notice - slightly smaller size, less durable, etc. It turns out that
154.
by fwip 2mo ago
I found public transit commuting to be quite peaceful most days. The rowdy children happened infrequently enough that they were often a nice change of pace, and the "drunks and thieves and undesirables" simply didn't happen w
155.
by fwip 2mo ago
Well, it's Science Alert - not exactly a bastion of level-headed reporting.
156.
by fwip 2mo ago
Gotcha, that makes sense. The LLM-written article said something about a "humanize" setting, so I thought it could have been something about that, so I spent a while looking for such a setting to disable.
157.
by fwip 2mo ago
What's the value of modeling these as petri nets instead of a regular sequencer? I was expecting some sort of branching, or at least multiple tokens per ring, but I can't figure out how to set that up. Also - is it just me, or is
158.
by fwip 2mo ago
You can also use a fixed seed for your prng. A hash of the input text (up to the current turn) should do.
159.
by fwip 2mo ago
In my experience, they had a lot of stuff working well in the first few years they rolled out the home voice assistants - Alexa, google home, etc. But for whatever reason, they've spent the last eight(?) years silently breaking things
160.
by fwip 2mo ago
What safety critical applications would be a good fit for LLMs?
161.
by fwip 2mo ago
It means the headline was vomited out by an LLM.
162.
by fwip 2mo ago
Simply knowing the quirk doesn't improve your skills, but the process of finding it often does. If Claude shortcuts that, then yeah, the problem might be solved, but you didn't skill up at all. Sometimes that's fine.
163.
by fwip 2mo ago
It definitely depends on the game. Some allow second (or third or fourth) players to play as "guests" (not signed in to any particular account), and some don't. Party games, or games designed primarily for co-op, are defini
164.
by fwip 2mo ago
You can play the DVD without your Xbox being online.
165.
by fwip 2mo ago
At least stock photos don't have the documented side effect of inducing nausea.
166.
by fwip 2mo ago
I mean, even the demo prompts on the ComfyUI page aren't adhered to by the model. From the first prompt, one of the four lines: > TRANSITION: a violent WHIP PAN off the rooftop that SMEARS the floating words away with it, motion-str
167.
by fwip 2mo ago
I wonder if the setup accidentally hit a cache at some layer.
168.
by fwip 2mo ago
And you'd have to be a pretty pig-headed exec to assume that the only experience an employee could possibly have with LLMs is positive.
169.
by fwip 3mo ago
And all of us have owned multiple keyboards that have lasted for years of daily heavy use without a single key failing. Even the $30 membranes you could buy at $local_retailer didn't fail in a month.
170.
by fwip 3mo ago
> That means zgm is early, but it is intentionally early in public. "Intentionally early" is one of those phrases that means they vibed this README out in an hour with AI, and now hope that they can create firmware for it in th
171.
by fwip 3mo ago
Oh, I understand what you mean. Thank you for explaining.
172.
by fwip 3mo ago
> This isn't because police are inherently horrible people, it is because they are normal humans. Well, you also have to remember that police aren't selected at random from the population. The kind of guy (or gal) who wants to
173.
by fwip 3mo ago
The big thing that's changed is that you're hearing about it. The second biggest thing is that we've given cops more and more access to automated systems that they can misuse.
174.
by fwip 3mo ago
> So either nobody cares about IP (at the very last in this context) and any AI company can just do whatever with data, or Chinese companies have to be held up to scrutiny. Anthropic is currently angling for a "IP restrictions for
175.
by fwip 3mo ago
Snarky answer: “It is difficult to get a man to understand something, when his salary depends on his not understanding it.” Realer answer: A combination of the above, plus group/bubble effect of all your coworkers saying the same thing
176.
by fwip 3mo ago
Anthropic is claiming that training an LLM to mimic another LLM is materially different and worse than slurping up stuff written by humans (even if that material is stolen). Basically, they want IP protection for Claude. This is a nakedly
177.
by fwip 3mo ago
I mean, if you want to count investors as revenue, that's not exactly standard economics.
178.
by fwip 3mo ago
No.
179.
by fwip 3mo ago
Or if you read the lines. "All sufficiently capable models, open and closed, should go through mandatory safety testing."
180.
by fwip 3mo ago
At time of writing, it was the only one on this thread that pointed out that the code was LLM-written.

More ›