Not financial advice.
Joined April 2013
- Tweets 117
- Following 6,213
- Followers 386
- Likes 2,058
19 Photos and videos
cryptotrader retweeted
6 Aug 2024
This incredible experiment demonstrated that a plant may have the ability to collapse quantum-probabilities towards the most likely outcome for its own survival, if a basic plant can do this...
What can we do as conscious, self aware entities?
367
1,702
7,082
755,222
cryptotrader retweeted
17 Apr 2024
My grandfather built motors that powered big machines. In a lab not so different from his workshop, my colleagues and I uncovered the assembly of the bacterial motor. Today, this discovery has been published in @NatureMicrobiol #MolecularNodes #Discovery nature.com/articles/s41564-0…
265
1,730
11,402
1,741,317
cryptotrader retweeted
2 Jun 2023
It’s the movie they really don’t want you to see: #WhatIsAWoman?
Watch the explosive documentary starring @MattWalshBlog FREE on Twitter for 24 hrs. x.com/realDailyWire/status/1…
28,697
154,399
437,497
194,209,499
cryptotrader retweeted
16 Mar 2023
Ever since I started using Midjourney, I had wanted to generate these images. It seems simple, but due to refractions and the difference in sizes between droplets, the results always came out strange. I did this today on #midjourneyv5.
#synthography #aicinema #aiartcommunity
1
3
34
3,081
cryptotrader retweeted
14 Feb 2023
Arriving in Egypt, Pythagoras tried to become a student of the temple priests. He applied again and again, but he was told that unless he goes through a particular training of fasting and breathing, he wasn't even allowed to enter the school.
7
10
25
15,246
cryptotrader retweeted
28 Feb 2023
Using tools like ControlNet, brands can transcend conventional boundaries of expression and explore new dimensions of creativity.
Here’s a cool experiment I did featuring the iconic Nike logo.
#ControlNet #Nike
49
426
4,652
648,430
cryptotrader retweeted
23 Feb 2023
ok, after many tests, I've managed to get my Additive Prompting structure working for interior design/architecture photography with 90% coherence to the prompt in Midjourney.
Examples & prompts in thread. Read them as you look at the images
🧵 Thread
ALT Editorial Style Photo, Off-Center, Mid-Century Modern, Living Room, Eames Lounge Chair, Wood, Leather, Steel, Graphic Wall Art, Bold Colors, Geometric Shapes, Design Within Reach, Track Lighting, Condo, Morning, Playful, Mid-Century Modern, Iconic, 4k --ar 16:9
154
486
4,486
1,399,616
cryptotrader retweeted
23 Jan 2023
Bitcoin: 6 on-chain metrics calling for a generational long-term buying opportunity
A thread 🧵
22
146
575
191,530
cryptotrader retweeted
13 Jan 2023
Language models are probably going to get a lot better in 2023, but not by getting bigger — by RL-style optimization toward specific tasks.
9
14
153
19,595
cryptotrader retweeted
12 Jan 2023
Delighted to share with you our latest paper. Here, we test the hypothesis that aging may be driven by DNA damage-induced changes to the epigenome. Mightbe due to a glitch in the software of the body that causes it to malfunction, which can be fixed with a reboot?👇 1/
287
1,675
7,061
1,522,953
cryptotrader retweeted
10 Jan 2023
3
44
413
26,606
imo this generation of AGI is data poor, per chinchilla. getting to the next stage will require massive synthetic data. in the case of copilot type models this will mean "self play", where the model generates code, interacts with a REPL, gets a stack trace, plugs that back in
39
14
431
The evolution of centralized AI ethics
v1: Don’t be evil
v2: We decide what is evil
v3: We decide you are evil
27 Aug 2022
every time i hear "AI ethicist" or some shit like that, I mentally replace it with "His Majesty's AI ethicist" or "the People's AI ethicist" to see how I really feel about it
25
103
605
cryptotrader retweeted
13 Aug 2022
AlphaGo didn’t beat us by studying our games, it beat us by studying vastly greater amounts of its own playing. The race is on to invent equivalent “self-play” for program synthesis, deduction, and other tasks currently limited by training data.
13 Aug 2022
"Language Models Can Teach Themselves to Program Better"
This paper changed my thinking about what future langauge models will be good at, mostly in a really concerning way. Let's start with some context: [1/11]
12
43
461
cryptotrader retweeted
3 May 2022
New jobs in the 21st century:
Model restart specialist
Hyperparameter psychic
Prompt engineer
Model janitor
Tensor shape mediator
Quantum state observer
Model footprint accountant
3 May 2022
The paper mentions 35 (!) manual restarts to train OPT-175B due to hardware failure (and 70 automatic restarts).
18
91
685
cryptotrader retweeted
18 Apr 2022
13
942
7,216
float i,s,e;for(vec3 p,q,d=vec3((FC.xy-.5*r)/r.y,-1);i <1e2;e=length(p)/s){p=q =i<50.?d*e:p-p 1e-4 e;p.z =7.;o =sqrt(e)/70.;p.xz*=rotate2D(t/8.);p=vec3(log(s=length(p)),atan(p.y,p.x),sin(t/4. p.z/s));s=1.;for(int j;j <6;)s*=e=PI/min(dot(p,p),.8),p=abs(p)*e-3.,p.y-=round(p.y);}
417
5,193
38,635
cryptotrader retweeted
7 Apr 2022
Projects like Hivemind (used by eg @huggingface) and BitTensor deserve a lot more attention!
github.com/learning-at-home/…
github.com/opentensor/BitTen…
11
48