Sir John de Mindeville

human mind cartographer making inaccurate maps of where algorithms end and brains begin

[Sir John de Mindeville @Hanzimmer]

made a more thorough version of this for people with longer attention spans.

37 questions instead of 10 and added likert scales for more psychometric reliability.

i like the original but it feels a bit flat imo

have fun

https://t.co/kFlBoV9Ivi

[Michael Millerman @millerman]

A new diagnostic quiz

10 questions for you to think about.

This is probably more fun for people who are relatively new to philosophy. Try it out and see!

I am having a lot of fun creating these tools, quizzes, and resources.

Enjoy!

https://t.co/XsoQ8IwmN5

[Sir John de Mindeville @Hanzimmer]

if musk was a country, he’d be in the top 20 economies of the world

[Sir John de Mindeville @Hanzimmer]

YES

[Joshua Kushner @JoshuaKushner]

in markets, to go long on something is to bet it grows more valuable over time. much of the conversation today is short on humans, wagering that ai makes people redundant. we believe the opposite is true for the industries @ThriveHoldings operates in.

we are long humans.

https://t.co/oqygDJKZGO

[Sir John de Mindeville @Hanzimmer]

@biiordache this is distributionmaxxing at its peak

[Sir John de Mindeville @Hanzimmer]

don’t cut a rut in the road

[Sir John de Mindeville @Hanzimmer]

@yacineMTB 😂😂😂

[Sir John de Mindeville @Hanzimmer]

nvm the surface area maxxers, this makes a lot of sense

[🕊️ @lichthauch]

The most frightening person i ever witnessed was a man who only wanted one thing. everybody around him had a hundred plans, a hundred projects they were going to start next month, and he had one, just one. and he never talked about it much, and that was the part that unnerved you, because when someone talks a lot about what they want, you can relax. talking is letting the pressure out, so it's harmless. but he was airtight.

Everything in him was moving toward one point, in silence. within a few years he was in a place no one else could reach, and everybody who had watched him stood there scattered and confused, with their mouths full of plans that never solidified into anything. i learned from watching him that the mouth is a leak. every time you say what you are going to do, you lose a piece of the will that would have done it

[Sir John de Mindeville @Hanzimmer]

@iamgingertrash 12 Against the Gods

[Sir John de Mindeville @Hanzimmer]

is anyone from tpot in bali?

[Sir John de Mindeville @Hanzimmer]

@WatcherGuru how is american airlines late to this? air france already has it

[Sir John de Mindeville @Hanzimmer]

never leaving this app

[Delicious Tacos @Delicious_Tacos]

“It’s my honor to announce that the true pope is working with us from Avignon”

[Sir John de Mindeville @Hanzimmer]

music is the final frontier. in 10 years from now we will stare into the abyss and yell and the echo will throw us back to the sky to remind us what it feels to be human

[Sir John de Mindeville @Hanzimmer]

@iruletheworldmo what's this slack channel you mention?

[Sir John de Mindeville @Hanzimmer]

attention really is all you need

[Alexander Whedon @alex_whedon]

Introducing SubQ - a major breakthrough in LLM intelligence.

It is the first model built on a fully sub-quadratic sparse-attention architecture (SSA),

And the first frontier model with a 12 million token context window which is:

- 52x faster than FlashAttention at 1MM tokens
- Less than 5% the cost of Opus

Transformer-based LLMs waste compute by processing every possible relationship between words (standard attention).

Only a small fraction actually matter.

@subquadratic finds and focuses only on the ones that do.

That's nearly 1,000x less compute and a new way for LLMs to scale.

[Sir John de Mindeville @Hanzimmer]

daily reminder to buy compute

[Sir John de Mindeville @Hanzimmer]

just saw karpathy apologizing for vague posting on stage

[Sir John de Mindeville @Hanzimmer]

@seelffff wtf is this slop? no mention of https://t.co/iuU7Mn5xh5

[Sir John de Mindeville @Hanzimmer]

@wordgrammer it’s pretty bad for adding local models

[Sir John de Mindeville @Hanzimmer]

@0xSero can't you use your local rig to run the full model with engram offloading?

[Sir John de Mindeville @Hanzimmer]

@human_named_gui and he knows how to pick the best architecure such that complexity stays localized and hidden in the system from day one