Dark Hacker News
new
|
best
|
ask
|
show
|
jobs
karinemellata | Dark Hacker News
user:
karinemellata
created:
July 7, 2020
karma:
59
submissions
comments
1.
1 points
by
karinemellata
304 days ago
|
discuss
2.
304 days ago
|
discuss
3.
Alignment is not free: How model upgrades can silence your confidence signals
(variance.co)
121 points
by
karinemellata
1 year ago
|
67 comments
4.
We used sparse autoencoders to explain LLM moderation flags of violent threats
(variance.co)
6 points
by
karinemellata
1 year ago
|
0 comments
5.
3 years ago
|
discuss