maizeq

6 ms·

1.
by maizeq 1y ago
I would encourage everyone to check out the other two major priorities of this movement. Of particular importance I think is the removal of recommender systems and opaque algorithms from feeds, which can and have been powerful instigators o
2.
by maizeq 2y ago
1. This doesn’t indicate language is necessary for thought but rather that language is useful for refining thought. If anything it shows that some form of thought exists before it is articulated into language. I would assume inverse cases a
3.
by maizeq 3y ago
Great guide. Really surprised air cooling is enough. My single 4090 kicks out enough heat that it’s a miniature heater for my room. Also, only 512Mb RAM??
4.
by maizeq 3y ago
This is not about open source AI, and the people who are saying it is don’t seem to understand the point Anthropic are making here. The point here is that malicious hidden behaviour encoded during pre-training seems to be very resistant to
5.
by maizeq 3y ago
How much JS is being shipped in that case though, and if it’s loaded async and not being used would it ever even impact page paint time?
6.
by maizeq 3y ago
This news is not about vertical integration or competitive advantage but of potential conflicts of interest. Meta/Amazon/Alphabet chips are made in house or from acquisitions (presumably with the CEO having no financial stake in t
7.
by maizeq 3y ago
What’s your favourite MCMC method/technique atm? I know this can be quite domain dependent, but curious. From my experience, it can seem more of an art than a science at times.
8.
by maizeq 3y ago
I’ll bite. I’ll stick to heuristics and intuitions since much of the heavy lifting to better quantify these risks has and continues to be handled by others already, as the other comments have mentioned. Here are some seemingly obvious intui
9.
by maizeq 3y ago
Any link to this? I actually haven't seen any reported results on less than 2 bits.
10.
by maizeq 3y ago
https://www.theverge.com/2023/11/18/23967199/breaking-openai... https://www.theverge.com/2023/11/18/23967199/breaking-openai... “The OpenAI board is in discussions wi
11.
by maizeq 3y ago
The reporting on this in the last two days has been bizarre and of such shoddy quality, particularly from The Verge. So many articles with no real sources saying the board was desperate for him to come back, or reneged on their decision, or
12.
by maizeq 3y ago
I love that he called this out. In my experience, e/acc people love using this inane reasoning
13.
by maizeq 3y ago
I work in a AI research lab for a large tech company and my observations are completely opposed to yours. Almost every researcher I have spoken believes that real risk exists, to some degree or other. Recent surveys of people in industry ha