Yeah 3.5 was pretty ass w bugs but could write basic code. 4o helped me sometimes with bugs and was definitely better, but would get caught in loops sometimes. This new o1 preview model seems pretty cracked all around though lol

Not sure entirely where this fits into this conversation, but one thing I’ve found really interesting that’s discussed in this Convo w Dr. K (I don’t have a timestamp sorry). Tai Chi has much more significant affect on all health perspectives than typical Western running/jogging/yoga etc.

And research papers can note this, but as soon as researchers start attempting to dig into the actual mechanical process behind why it has such a significant affect, their papers will be rejected because it dips too far into Woo/Spiritual territory despite not describing what the woo is, just acknowledging that “something” is there happening.

I think it’s interesting we can measure results and attempt to explain what we’re seeing but western research tends to be so tied to physical mechanisms it has almost started hindering our advancement.

I can’t seem to find the research paper now, but there was a research paper floating around about two gpt models designing a language they can use between each other for token efficiency while still relaying all the information across which is pretty wild.

Not sure if it was peer reviewed though.

which does support the idea that there is a limit to how good they can get.

I absolutely agree, im not necessarily one to say LLMs will become this incredible general intelligence level AIs. I’m really just disagreeing with people’s negative sentiment about them becoming worse / scams is not true at the moment.

I doesn’t prove it either: as I said, 2 data points aren’t enough to derive a curve

Yeah only reason I didn’t include more is because it’s a pain in the ass pulling together multiple research papers / results over the span of GPT 2, 3, 3.5, 4, 01 etc.

That’s definitely valid, but just because a tool is used for scam doesn’t inherently mean it’s a scam. I don’t call the cellphone a scam because most my calls are.

The jump from GPT-4o -> o1 (preview not full release) was a 20% cumulative knowledge jump. If that’s not an improvement in accuracy I’m not sure what is.

Compare the GPT increase from their V2 GPT4o model to their reasoning o1 preview model. The jumps from last years GPT 3.5 -> GPT 4 were also quite large. Secondly if you want to take OpenAI’s own research into account that’s in the second image.

Curious why your perspective is they’re are more of a scam when by all metrics they’ve only improved in accuracy?

Realistically this graph wouldn’t include Snapchat, as it’s less of a “social media” than the others. Most people nowadays use it strictly as a messaging platform.

This is actually the problem with mid sized communities. If there’s two political communities and their content is 70% overlapping then you’re obviously going to unsubscribe from one as your feed becomes cluttered. But then you’re missing that 30% of content from the other community.

Bans LGBT+ people

My fyp is full of bipoc LGBTQ+ people? Twitter on the other hand tries to feed me right wing news sources and disinformation all the time. There’s a spectrum here and TikTok is far on the other side.

Also to be frank the algorithm is by far the best out of any company with a “for you” algorithm. It blows pretty much anything else out of the water entirely.

One single god God, perhaps what makes a god a god? I do believe we form a network or web of connected consciousness. Who’s contained within this web and how interactions work on it is where it gets interesting.

I think the problem with IPFS is it’s terribly inefficient with large and or dynamic data storages, the exact problem bittorent tackles(large not dynamic).

IPFS does seem to provide great p2p functionality for small static files though, albeit pretty slow.

Your subscribed will be the same. All will differ as it grabs everything everyone on that instance is subscribed to.

Ex. If you’re on lemmyworld you’ll see everything there by default but if you swap to instance X and nobody in x is subbed to niche.community@lemmyworld then it won’t be brought to your All

I recently hired into a data analytics team

I work in Data Engineering and have spent most my time on analytics teams. They don’t have a SWE/CS background and generally because of that don’t follow any good programming practices. In my experience style guides are hard to get them to follow properly even if you set up SQLFluff for them., I can barely make them see the advantage of not committing directly to main (at least we’re using git). It’s very frustrating.

But honestly it’s not a huge deal, I swapped an esim in under 3 minutes last week

physical sim card trays

Seems like a backwards move IMO. honestly haven’t needed a tray in ~6 years

Being able to rip the battery out when the phone locks up

How often does your phone lock up that that’s necessary? I used to have lockups a lot more but android is damn stable now, haven’t had one in years.

Needing to make sure it’s actually off and can’t be remotely powered on

Do you disconnect your PC’s PSU?

When it’s 3 years old and the hardware is still well up to the task but the battery lasts 4 hours.

Fair, but if you’re getting 4 hours SOT after 3 years, it’s just not a great phone. My current Note 10+ still pumping out 10 hours SOT and I’m a HEAVY user.

Honestly in 2015 I’m totally on your side for this, but in the last few years I’ve never felt the need to hotswap batteries, and only slightly felt the need to replace it in general.

Just curious, what situations do you find yourself in relatively frequently that a hot swappable batter would be more convenient?

Nowadays w/ 15 SOT I don’t think I’ve actually needed one minus camping where I don’t really use my phone much anyways.

I’ve had an account for almost 10 years that I use at least every other day at work, and have seen plenty of questions I CAN answer but apparently don’t have the “reputation” to.

Honestly a really dumb system imo.