pandeiro

Use it all the time. Occasional hiccups.

Vulture Capitalists, obviously

There's no hot reload in Android (kotlin) dev? Wow

Get out of this thread and never come back.

I'm using Meta-Llama-3-8B-Instruct.

BTW it's a great project and I see it's gotten massive adoption on Github, with plenty of Issues to follow. Curse of success!

Thanks again.

EDIT: Created an Issue for it, probably a more appropriate place to file this ;)

https://github.com/simonw/llm/issues/561

Your tool is really cool, thanks for making it and sharing.

I notice that a lot of times, the multi-line output I get from my prompts is truncated or the model just aborts or something. The exit code is still 0 but it seems like something went wrong.

I remember that, too, and switched to IE for the next X years

Can't believe unfettered greed and self-interest would ruin something like this.

Another update:

https://www.bleepingcomputer.com/news/security/cdk-global-ou...

Reasonable approach and I applaud your effort to maintain the site's purpose while also not ignoring these issues (that do have some relation to the tech industry, as we've seen). Thank you.

One man's politically biased is another's politically astute.

The die hard masking advocate is your example? Wow

Especially when they're factchecking for partisan narrative compliance

lol just go along with it, we need to label them something, and "non-partisan" just doesn't give anyone a self-righteous dopamine hit

University of California

"With this recommendation, the UC Vaccine policy now requires all staff and students to get the bivalent booster, when eligible"

[1] https://campusready.ucdavis.edu/covid-vaccination

Re: 'the level of discussion'

What you posted was condescending and demonstrative of a large blindspot as to the degree of adverse event surveillance actually being carried out.

Hope that helps you achieve the tone of discussion you're aiming for, in the future.

Nuts is that on the basis of the flimsiest of evidence, your ascribing an efficacy in disease prevention to an intervention (wearing masks) that was widely understood to do no such thing prior to 2020.

Furthermore, respiratory disease avoidance on a society-wide scale has never and almost certainly will never be achieved. The assumption that it is an attainable goal is truly nuts.

Why not wait until COVID hysteria is over? Seems like the right call.

COVID IFR for your probable age group: 0.05%

Wonder of the number and frequency of these 'surprises' has anything to do with them being trialed and authorized in 3-6 months rather than the standard 6-10 years.

Yeah how did I know the first actual concrete example would involve COVID? jfc

Holy crap reading the replies here I'm really at a loss to imagine what an "UNSAFE and effective" vaccine would look like.

"Yeah Bob, got the flu shot yesterday and I can't hear, can't sleep, and can barely walk." WTF?

Yeah yeah yeah, downvote away.

Not surprised this degenerated immediately into a mass ranting session about pandemic science with pretty much the entire catalog of cognitive biases and logical fallacies on display.

I guess that's interesting to some, from a sociological standpoint, but for me, there's more than enough of this on literally every other social media platform.

I'm interested in a discussion about the actual dynamic of spotify vs youtube dissemination and whether the claims made in the article are valid. Because the entire premise is backed by the "secondary metric" of how many Twitter followers a guest's account grew by -- this seems pretty ripe for confounding variables, like the appeal of the guest, auxiliary appearances elsewhere, the news cycle at the time overlapping with the guest's subject matter, and other things.

Curious what others think of this.

The State is always benevolent and we should always believe and do everything they say. End of story.

You "literally have evidence" that what doesn't happen in humans, ADE in general? Seems pretty easy to disprove; there's a lot of research based on the phenomenon, eg:

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7211844/

If you're talking about ADE with COVID specifically, I agree there's no evidence, but there's the obvious principle that absence of evidence doesn't equate to "evidence this doesn't happen", which, if you have it, please link?

Side-note: definitely helps your argument to throw around the term "anti-vaxxers", though. You could probably just remove all the other words in your comment and keep just that, to the same dog-whistling effect?

Not a dumb question at all. Not a domain expert but an enthusiast so I'll take a crack at an answer:

Maximal fire suppression strategies have been the most prevalent approach to avoiding property damage and loss of life caused by forest fires for most of the past century, despite the known harm to ecological systems that rely on periodic burning, because nature is difficult to control/predict and prescribed burns (an alternative or supplemental approach) are time and labor intensive.

So you're saying 'net net', you think the COVID mortality is undercounted? If I had to bet something valuable, I'd take the other side. That's just what seems most probable to me, based on the evidence I cited.

Another plausible explanation to excess mortality differences is that lockdowns and other measures caused people to engage in behaviors that increased mortality, from putting off necessary medical visits to exacerbated mental health due to social isolation or financial stress.

Personally my guess is that COVID was, yes, responsible for the vast majority of those deaths.

But I also do think that those cases of overcounting are very likely NOT the only, isolated ones. The dynamics exposed (how hospitalizations were counted, the basic 'with' vs 'of' question), almost certainly apply more broadly.

So, apologies for attempting nuance, but I think COVID can at once be a) a massive pandemic (and tragic for millions), while also being b) somewhat overblown due to unrigorous metrics that in turn feed more sensationalism than otherwise warranted.

Aside from the distraction of the author's nuttery, the issue is probably the most salient one for the American middle class over the next century.

Either very strong measures are imposed to limit / de-incentivize real estate speculation and hoarding, or the vast, vast majority of the population is entirely excluded from any kind of property ownership.

And while simplistic 'conspiracy' thinking definitely is of a much lower value than rigorous analysis, there is IMO merit to at least acknowledging that the politically powerful class might overlap substantially with the rentier (property ownership) class and thus resist any such reform taking place. It shouldn't be out of bounds to speculate on the motives of powerful actors, even if it is obviously not the same as making evidence-based claims.

Agreed, I use euromomo to look at total mortality across Europe and it's difficult to deny that something was killing lots more people than usual last spring and winter (northern hemisphere).

However it can also be the case that COVID deaths (and hospitalizations as well) may be over-reported. We have recent reports that seem to be pointing strongly to this possibility.

[1]: Alameda County revising COVID deaths downward by 25% - https://oaklandside.org/2021/06/04/alameda-countys-new-covid...

[2]: Research in CA suggests COVID hospitalization in children overcounted by 40% - https://nymag.com/intelligencer/2021/05/study-number-of-kids...