throwaway314155

8 ms·

91.
by throwaway314155 9mo ago
> Most doctors are just people who had a strong ability to pass tests and finish medical school. This is a tautology. “Most doctors are just (doing lots of work there) people who had the ability to meet the prerequisites for…becoming a d
92.
by throwaway314155 9mo ago
> Doctors are usually just normal people that had the means, memory, drive and endurance to get through an exclusive education that will guarantee them a life in relative affluence and maximized adoration. Is that all?! Also, everyone is
93.
by throwaway314155 9mo ago
Or as those in the Valley put it "Adderall is all you need"
94.
by throwaway314155 9mo ago
So what if they are?
95.
by throwaway314155 9mo ago
Oh come on.
96.
by throwaway314155 9mo ago
And nothing of value would be lost.
97.
by throwaway314155 10mo ago
> told disgruntled iPhone 4 users that they were holding their phones wrong That was never proven. Although their PR response was atrocious.
98.
by throwaway314155 10mo ago
How is that relevant?
99.
by throwaway314155 10mo ago
Playing two Steam games simultaneously typically results in a logout from the second (offending) PC.
100.
by throwaway314155 10mo ago
> Microsoft licensing rules mean that you are not allowed to hide Windows. What do you mean? As in - manufacturers can't create an overlay similar to Steam Deck's "Gaming Mode"?
101.
by throwaway314155 10mo ago
My first job did IRL code reviews with at least two senior devs in the loop. It was both devastating and extremely helpful.
102.
by throwaway314155 10mo ago
Ah fair enough.
103.
by throwaway314155 10mo ago
Yeah, I'm calling bullshit as well. The OP responds but doesn't seem to acknowledge that --dangerously-skip-permissions is a thing.
104.
by throwaway314155 10mo ago
Do skills get access to the current context or are they a blank slate?
105.
by throwaway314155 10mo ago
> which presumably hasn't done a fresh pre-training over the web What makes you think that? > Did they figure out how to do more incremental knowledge updates somehow? It's simple. You take the existing model and continue pr
106.
by throwaway314155 10mo ago
GPT 4o was an MoE model as well.
107.
by throwaway314155 10mo ago
> it is really hard to look at for me. What were you expecting?
108.
by throwaway314155 10mo ago
Somehow I suspect Claude Code (in an interactive session with trial, error, probing, critiquing, perusing, and all the other benefits you get) would do better. This example seems to assume Claude can do things in "one shot" (even
109.
by throwaway314155 10mo ago
Not arguing with any of that, just stating plainly that there are practical reasons for time limits and one of the many reasons is that tests are done supervised and thus must have _some_ sort of time limit. Everything else is you projectin
110.
by throwaway314155 10mo ago
> Why? Teachers have lives, including needing to eat and sleep.
111.
by throwaway314155 10mo ago
Realistically there has to be _some_ time limit. No one is going to sit in a room for 10 hours while you finish your test.
112.
by throwaway314155 10mo ago
It has no idea what it's own knowledge cutoff is.
113.
by throwaway314155 10mo ago
What?
114.
by throwaway314155 10mo ago
I'm confused. This just seems like feel-good bullshit advice that only works for people in extremely good circumstances. There's a false equivalence between - “All it takes is for one to work out.” and the following: - "You d
115.
by throwaway314155 11mo ago
Silence says it all.
116.
by throwaway314155 11mo ago
I’d love to follow your work. Got a GitHub?
117.
by throwaway314155 11mo ago
What?
118.
by throwaway314155 11mo ago
How is an image generation model serving the market of...developers? I mean I know we all focus on these models and get excited about what they can do. But why would we pay for them for more than a few tests?
119.
by throwaway314155 11mo ago
> as opposed to the weird choice to use CLIP and T5 in the original FLUX This method was used in tons of image generation models. Not saying it's superior or even a good idea, but it definitely wasn't "weird".
120.
by throwaway314155 11mo ago
There aren’t any YOLO models for captioning and the other models aren’t robust enough to make for good embedding models.

More ›