Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
zachglabman
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
1.
▲
takeaways from 28 factory tours in 14 days
(twitter.com)
4 points
by
zachglabman
1y ago
|
0 comments
2.
▲
by
zachglabman
2y ago
If a transformer had infinite data and parameters, I'm sure it could simulate human reasoning to a high degree. Humans don't work that way, so we may need to create a more general definition for artificial reasoning
3.
▲
by
zachglabman
2y ago
LLMs are wrong for most things imo. LLMs are great conversational assistants, but there is very little linguistic rigor to them, if any. They have almost no generalization ability, and anecdotally they fall for the same syntactic pitfalls t
4.
▲
Deepfakes, Zero Emissions, Alpaca
(practicum.substack.com)
1 points
by
zachglabman
6y ago
|
0 comments
5.
▲
Conscious Consumerism, Toilet Paper, Words
(practicum.substack.com)
1 points
by
zachglabman
6y ago
|
0 comments
6.
▲
Gap Years, Future of Education, Future of Work
(medium.com)
1 points
by
zachglabman
6y ago
|
0 comments