Skip to Main Content
Dataflow matrix machines (by Anhinga anhinga) (
dmm
) wrote
2023-10-29 05:51 pm (UTC)
no subject
~29:00 it seems that most computations only go through a couple of layers (residual stream gives it the freedom to do this).
(So the bulk of computations are probably shallow, with a bit of "true deepness" sprinkled on top of it.)
(
27 comments
)
Post a comment in response:
From:
Anonymous
This account has disabled anonymous posting.
OpenID
Identity URL:
Log in?
Dreamwidth account
Account name
Password
Log in?
If you don't have an account you can
create one now
.
Subject
HTML doesn't work in the subject.
Formatting type
Casual HTML
Markdown
Raw HTML
Rich Text Editor
Message
Log in
Account name:
Password:
Remember me
Other options:
Forget your password?
Log in with OpenID?
Close
menu
Log in
Create
Create Account
Display Preferences
Explore
Interests
Directory Search
Site and Journal Search
Latest Things
Random Journal
Random Community
FAQ
Shop
Buy Dreamwidth Services
Gift a Random User
DW Merchandise
Interest
Region
Site and Account
FAQ
Email
no subject
(So the bulk of computations are probably shallow, with a bit of "true deepness" sprinkled on top of it.)