Hacker Newsnew | past | comments | ask | show | jobs | submit | AndrewKemendo's commentslogin

This is the only logical explanation for Robert Scoble’s popularity

I’m exactly there with you with my 13 mini and I did realize that it’s not gonna last as we really get into the era of local LLMs

Running deepseek 6B on the Private LLM app on the iPhone 13 basically set my phone on fire


> Running deepseek 6B on the Private LLM app on the iPhone 13 basically set my phone on fire

Hey, I’m the author of Private LLM. I hope you’re joking about the phone catching fire. Btw, there’s no DeepSeek 6B model, you’re likely talking about the DeepSeek Distill 7B model.


Yes thank you for the correct correction

The last model I tried to run was Dolphin 3B but even Zephyr1.6B kills the phone pretty fast even with a full battery

Even Dolphin completely killed my phone so yeah I can’t really use the Private LLM app for anything heavier than the default 1.6B


Yeah, I agree. Really hard to fit anything larger on the 4GB of RAM on the iPhone 13 of which, only about (depending on what iOS version you're on) 2.1-2.5GB is usable by apps.

Seems like a collapse of the alcohol market would be a benefit on the order of removing leaded gas

But don’t let my biological logic stand in the way of cultural madness - if a coherent society in your view requires a poison in order to facilitate then that society is probably not worth keeping going


Are you living in a separate system that’s not associated with capitalism?

If so I’d love to know what that is so I can go there.


Exactly this

Why train a junior human when I can train a junior robot that I can copy-paste that can’t quit and has infinite margins?


How’s it any different than a diff log?

Better question: how is it in any way similar?

If you read the history of both and assuming that there’s good comments and documentation, it shows you the reasoning that went into the decision-making

That’s correct

I’ve spent most of the last 20 years doing reinforcement learning and it is exceptionally simple conceptually

The challenge is data acquisition and the right types of process frameworks


This is where I’m at

I stopped trying to recruit cofounders because I don’t need them anymore

I can do everything I need to do by myself plus tools at a pace no set of humans can achieve


Clearly it didn’t matter

Capital always wins because there’s an infinite line of psychopaths at the ready to screw everybody over for slightly less money than the previous person did


His own employees helped reinstate him, not the capital class which was actually the board that fired him.

The board that fired him wasn’t really “the capital class” in the traditional sense. It was a nonprofit board with an unusual governance structure specifically designed to limit investor controlling. Ilya and Helen were acting on safety/governance concerns, arguably against the interests of capital (Microsoft, VCs).

Like literally he’s doing right now the thing that would not have been done had Ilya and the other board members retained their positions


Not psychopaths. I recall it being rank and file who were concerned about their options or whatever. Greed is fundamental human behavior.

“There was a time when nobody trusted either aircraft nor elevators. Today people have pure unquestioned faith in both. Existential faith in fact, they test their faith with their lives. You may chuckle and laugh but that's simply because you are ignorant of the systems that keep you alive and safe”

https://kemendo.com/Faith.html


Elevators are suspended in a way that holds brakes open, if all of the multiply-redundant cabling snaps, the breaks activate. There's an airbag equivalent at the bottom of the shaft, too.

I don't really have a point I just think the typical elevator braking failsafe is so genius in its simplicity that I got excited to share.


" Today people have pure unquestioned faith in both"

Not true at all. We accept the risks to obtain benefits but we also know having an accident in the air or in elevators is highly unlikely given what we know; so therefore its perfectly rational behaviour.


Nonsense

that would assume that your average person has any concept of the relative statistics and has a sense of making decisions based on statistics

People make decisions based on what other people around them are doing

this is well known in safety engineering in architecture and civil engineering which is why you have standards for egress doors because left of their own devices humans will follow crowds to their own death

https://en.wikipedia.org/wiki/Crowd_collapses_and_crushes

https://www.sciencedaily.com/releases/2008/05/080512172901.h...


One does not need to know of relative statistics to know that a) you dont see planes randomly dropping out of the sky on a regular basis b) people enjoy flying to hot destinations and are willing to accept the small chance the flight may not be risk-free - people are aware of this when they experience some level of turbulence when flying.

Finally, Ive seen plenty of your posts on here. You write with a particular tone. Who are you? A nobody who's spent a lot of time posting crap on here.


Attacking the person rather than their argument only serves to make your argument look weaker.

I agree. This looks rather childish.

I’m curious if you can actually describe the tone

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: