Kofi

837 posts

Kofi banner
Kofi

Kofi

@k3baah

optimist

Brighton, England Katılım Ocak 2012
1.7K Takip Edilen432 Takipçiler
Michael Edward Johnson
Michael Edward Johnson@johnsonmxe·
This is one reason I think longevity research needs a science of latches — huge parts of our nervous system’s innate dynamic range get latched when we’re young, never to reopen. And a surprising amount of age-related “loss of neuroplasticity” may simply be gradual accumulation of latches Some latches are doing important things, but carefully reopening e.g. the least useful 0.1% every week could lead to e.g. a reversion to the radical learning rates of youth, the youthful feeling of magical possibility, & in general having a ‘resonant’ and ‘fresh-feeling’ nervous system Anecdotally, this sort of reversion can happen from breakthroughs on the meditative path (which I expect to be latch-opening events!) I think of this as longevity research’s “skin folds problem” — after someone loses a lot of weight, they sometimes need to get surgery to make their skin taut again — it doesn’t happen on its own. Likewise, even if a future science of longevity can roll back someone’s biochemical age, their total amount of latches will more closely track their experiential age. A full rollback of phenomenological age will require a proper science of latches
English
9
12
224
23.4K
Nora Belrose
Nora Belrose@norabelrose·
If deep learning can predict weather better than an explicit physics simulation, does that mean that deep learning is more "fundamental" than physics? Or that nothing is fundamental?
Google DeepMind@GoogleDeepMind

Today in @Nature, we’re presenting GenCast: our new AI weather model which gives us the probabilities of different weather conditions up to 15 days ahead with state-of-the-art accuracy. ☁️⚡ Here’s how the technology works. 🧵goo.gle/49trAOv

English
386
163
3.3K
641.1K
Kofi
Kofi@k3baah·
@samwhoo which model is this?
English
1
0
0
220
Sam Rose
Sam Rose@samwhoo·
I've been slowly working voice to text more into my workflow to reduce the strain on my hands. If you've not tried voice to text for a little while, I'd recommend giving it a go in 2024. The Whisper models are incredibly good. I use superwhisper.com on Mac. Sorry the text is a bit small at the end.
English
5
1
35
2.8K
Kofi
Kofi@k3baah·
@shaoruu @cursor_ai - not having to @ every file that could be relevant - knowing if context has been exceeded for a given composer session (ie if certain files/snippets are being dropped)
English
0
0
0
40
ian
ian@shaoruu·
what do you want the most in @cursor_ai composer? making sure i don't miss out on great ideas in this next release
English
413
31
972
228.8K
Jimmy Apples 🍎/acc
Jimmy Apples 🍎/acc@apples_jimmy·
About to watch this for the first time, hope it lives up to expectations
Jimmy Apples 🍎/acc tweet media
English
107
13
592
58K
Dwarkesh Patel
Dwarkesh Patel@dwarkesh_sp·
Huge props to my editors Conor & Aaron for animating this character so compellingly They (& Conor especially) have worked tirelessly & ingeniously to produce what it takes entire video game studios to make Btw, before I hired them, Conor was a farmer in Argentina, & Ishan (my amazing shorts editor) was a maths student in Sri Lanka A privilege to work with these 3!
Dwarkesh Patel@dwarkesh_sp

The @gwern interview. 0:00:00 – Anonymity 0:01:09 – Automating Steve Jobs 0:04:38 – Isaac Newton's theory of progress 0:06:36 – Grand theory of intelligence 0:10:39 – Seeing scaling early 0:21:04 – AGI Timelines 0:22:54 – What to do in remaining 3 years until AGI 0:26:29 – Influencing the shoggoth with writing 0:30:50 – Human vs artificial intelligence 0:33:52 – Rabbit holes 0:38:48 – Hearing impairment 0:43:00 – Wikipedia editing 0:47:43 – Gwern dot net 0:50:20 – Counterfactual careers 0:54:30 – Borges & literature 1:01:32 – Gwern's process 1:19:17 - Gwern's finances 1:25:05 - Random

English
11
15
489
45.4K
Kofi
Kofi@k3baah·
@m_ashcroft @Karaminder @adhdjesse i basically got stuck thinking I had to do low priority stuff first to free up mental space to work on high priority stuff but that led to overwhelm by low priority tasks (they don’t stop) and guilt for not doing high prio reversing that did the trick for me
English
0
0
8
201
Kofi
Kofi@k3baah·
@m_ashcroft @Karaminder @adhdjesse but the usual time management heuristics are actually pretty effective. for me its - scheduling my day by the hour (very generously) the previous night - next action = highest priority (whenever I remember) i also made a customised todo list on obsidian which helped
English
1
0
5
194
Michael Ashcroft
Michael Ashcroft@m_ashcroft·
1) I get antsy, unmotivated and distractible when things are too calm 2) I get calm, motivated and focussed when things are chaotic 2 is a helpful strategy that has served me well… but 1 just seems really dumb because i end up turning ‘1 situations’ into ‘2 situations’
English
4
2
96
5K
Abhi Sivasailam
Abhi Sivasailam@_abhisivasailam·
I understand if people are skeptical, but I really have replaced the overwhelming majority of last-mile data analysts. And it was *easy*. What isn't at all easy: replacing all the telemetry, pipeline glue work, data modeling, etc., that makes last-mile analytics possible...
English
5
2
38
5.3K
Luke
Luke@luke_b_91·
Luke tweet mediaLuke tweet media
ZXX
8
15
296
18.2K
Kofi retweetledi
Elon Musk
Elon Musk@elonmusk·
ZXX
34.7K
80K
680.5K
148.1M
jason liu
jason liu@jxnlco·
Starting to freelance? A great proposal is the first step - present options to get paid more and show you understand their needs, while adding urgency. This template landed me my first $10k monthly contract, my first $60k prepaid, and even equity in 3 startups. Want to charge more and move to value-based pricing? Reply 'proposal' or retweet and I'll DM you my proposal template
English
364
54
286
71.4K