Skip to content

Speedup ffn and gelu#15

Draft
certik wants to merge 3 commits into
mainfrom
gelu
Draft

Speedup ffn and gelu#15
certik wants to merge 3 commits into
mainfrom
gelu

Conversation

@certik

@certik certik commented Mar 8, 2023

Copy link
Copy Markdown
Owner

On my machine these changes speedup inference from 0.789s to 0.602s.

certik added 3 commits March 7, 2023 15:47
This provides about 4% speedup from 0.789 to 0.758s.
This provides about 20% speedup from 0.752s to 0.602s.
@certik

certik commented Mar 17, 2023

Copy link
Copy Markdown
Owner Author

With caching on, both main and this PR show 0.288s. With caching off, this PR is 0.543s, main is 0.716s.

@certik certik marked this pull request as draft March 17, 2023 15:20
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants