chiisana

More than once I’ve heard the jokingly saying that ‘everything causes cancer in the state of California’ (regardless if they bore the warning label or not). I think while the intention may be good, the equivalent of notification fatigue is at play here and might not be delivering intended benefit/value.

If memory serves, 175B parameters is for the GPT3 model, not even the 3.5 model that caught the world by surprise; and they have not disclosed parameter space for GPT4, 4o, and o1 yet. If memory also serves, 3 was primarily English, and had only a relatively small set of words (I think 50K or something to that effect) it was considering as next token candidates. Now that it is able to work in multiple languages and multi modal, the parameter space must be much much larger.

The amount of things it can do now is incredible, but our perceived incremental improvements on LLM will probably slow down (due to the pace fitting to the predicted lines in log space)… until the next big thing (neural nets > expert systems > deep learning > LLM > ???). Such an exciting time we’re in!

Edit: found it. Roughly 50K tokens for input output embedding, in GPT3. 3Blue1Brown has a really good explanation here for anyone interested: https://youtu.be/wjZofJX0v4M

The models are not wrong. The models are nothing but a statistical model that’s really good at predicting the next word that is likely to follow base on prior information given. It doesn’t have understanding of the context of the words, just that statistically they’re likely to follow. As such, all LLM outputs are correct to their design.

The users’ assumption/expectation of the output being factual is what is wrong. Hallucination is a fancy word in attempt make the users not feel as upset when the output passage doesn’t match their assumption/expectation.

Another possibility: the console vendors are catching whiff of the whole gate keeper mess, and they don’t want to be at risk of being forced to open up their physical cartridge DRM mechanisms to allow third party cartridges for the relatively small EU market (compared to the rest of the world). Moving towards digital is much easier as result.

It is also clear as day that they’re testing the waters with Nintendo players who are generally deemed to be more casual, and lesser likely to push back compared to the more savvy other major consoles. Once this blows over, they will just move to digital everything across the board, citing successes and cost savings on the other platforms as basis for the move.

Stop addressing them as “normies” would be a great start.

Can’t speak for rest of the Fediverse as I’m not super active on microblogging anymore, but at least here on Lemmy, there is such a strong “in” culture and quirky skewed perception of the world, and often times come off as actively hostile against those that do not share the same quirky skewed world view. The anti-AI, anti-corporate, would rather shoot myself in the foot if it’s not FOSS, etc kind of views, with their own strong vocal proponents, comes off as unwelcoming. People are addicted to socials because of the positivity they can get, not the negative sentiments that’s often echo’ed.

Amongst those that doesn’t share the kind of view, you’d already be looking at an extreme small minority that might be willing to give the platform a try, but as long as the skewed perception of the world dominates the discussions, you can expect them to go back to main stream centralized platforms where they can get more main stream view points based discussions.

Wololo wololo

Technically, the words are adopted from Chinese (in this case both Traditional and Simplified are the same and have not diverged yet); but same meaning and reasoning, just different pronunciation.

Emojis used zero width joiner to combine multiple single code point emoji to a single combined emoji.

⛥ + ZWJ + ⬠ could form the combined character, and be rendered as desired.

It’s not threatening anyone… I don’t believe I’ve seen anywhere that the mods say or imply that. Also before anyone complain about singling people out, no, if I share anything from a non-reputable source, it’s going to get deleted, regardless of the subject. It’s about the quality of the source; the objective is to create a community sharing good trustworthy sources to improve the overall quality of content appearing on the community.

Again, you’ve been invited by the mods to repost from a more reputable source. If there aren’t any, then perhaps it is not !news worthy.

Looks like a case where poorly sourced article getting removed, with invitation to repost with a more reputable source… so do so with a better source. Or is the underlying article itself leaning too much towards propaganda that there is no more reputable source? and if that is the case, then is it really !news worthy?

Can you elaborate further? If the intention is to obscure the information by federating anonymous information outwards, then no third party instance owners can observe the true usernames for vote manipulation from the same user across instances. If more instances deploy this kind of poisonous behaviour across the fediverse, then it becomes untenable for instance owners and community moderator to protect themselves, which in turn hurts the fediverse as a whole.

Edit: if you mean the vote percentage thing, that’s utterly useless. Vote amplification works both ways, a bot user doesn’t have to vote just down or just up. So knowing the percentage of the anonymous user’s previous behavior doesn’t support identifying vote manipulation. An alleged abuser can easily create thousands of account and sample random % of account to mass drive sentiment without having them all appear with similar percentages.

I’ve seen posts being downvoted by user@instancea, user@instanceb, username@instancec etc. this will make tracking that kind of abuse much more difficult.

The code check needs to have some time value associated with it. I taped the link via app, which loaded it in an in-app browser. After getting by the code challenge, I have to back out from the in-app browser to receive the message. Upon returning, the site sends a new code invalidating the old one, which prevents me from logging in. Codes should be valid for X minutes, such that users can fumble around their apps.

As long as the hardware is relatively close to the target level defined by Nintendo, it really doesn’t matter if the chip was “already old”. The developers know what hardware to expect, if they want their game to run on that console, they’d be on the hook to optimize for it.

There’s also the problem that sadly Lemmy is filled with vocal users with skewed view of the world, and they tend to be extreme polarizing. The “if you’re not one of us, who firmly believes the world should work a certain way, and if you’re not willing to shoot yourself in the foot with a shotgun to prove it as a point, then you’re one of them; you should get the eff off of Lemmy and crawl back to Reddit” kind of way. They’re so scared of losing that pedestal that they’re going to go out of their way to alienate anyone who doesn’t drink their koolaid and push them off the platform so they can remain dominant. Sadly, these people also never really learned much of the real world, so those that are more experienced / educated gets pushed off the platform, and we end up with a bunch of weird superstonk culty kind of vibe everywhere.

I find myself more and more just make a comment and don’t look back. It’s quite literally futile and pointless trying to expect any discussion of any actual sustenance. You wonder why it’s just shitposting… well this is why.

It is probably best to think nothing on Lemmy is private. Any instance with at least one user subscribed to a community will receive updates (messages and votes) on the community. Instance admin can go into the database to see any private message between any user on that instance.

Their grievances with the Lemmy dev team is pretty universal. I can’t wait for Sublinks to become the mature release.

For projects, where they have their community presence also speaks to their ideology. Those projects’ communities chose to move off of Reddit, and be on Lemmy; those projects’ communities chose the instance they’re on.

One may plea ignorance in the early days of Lemmy, that they’re misguided by the instance description; but now a year later after all the drama, their decision to remain there will start to influence who will be able to interact with their community.

I have no sympathy for communities that chose to remain on CSAM infested instances that got defederated, and I will have no sympathy for project communities that continues to associate with ideologies by the ml admins.

Rather !announcements@lemmy.openmw.org, !general@lemmy.openmw.org, !bugs@lemmy.openmw.org, … etc.

So much more flexibility for organizations to build structured communities!

Cheesy or not, the game was incredibly fun to play with the little one. It is the first game my little one 100%’ed, and will always have a special place in my heart.