Skip to content

llama : bump max layers from 256 to 512 - #8530

Merged
ggerganov merged 2 commits into
masterfrom
gg/bump-max-layers
Jul 19, 2024
Merged

llama : bump max layers from 256 to 512#8530
ggerganov merged 2 commits into
masterfrom
gg/bump-max-layers

Conversation

@ggerganov

Copy link
Copy Markdown
Member

fix #8528

@ggerganov
ggerganov force-pushed the gg/bump-max-layers branch from fac0355 to b268edf Compare July 17, 2024 07:04
@slaren

slaren commented Jul 17, 2024

Copy link
Copy Markdown
Member

I would suggest moving the assert to the case where the key is actually an array only. That way it wouldn't affect other models. The assert should also be replaced with an exception instead to avoid crashing on invalid models. Ideally it should be a dynamic container instead of a fixed array, but that requires more changes.

@ggerganov

Copy link
Copy Markdown
Member Author

I would suggest moving the assert to the case where the key is actually an array only. That way it wouldn't affect other models.

Wouldn't this loop potentially overflow if we check just array keys?

https://github.com/ggerganov/llama.cpp/blob/b268edf87c3bb996feb88e2d8b2c85b38bff6f36/src/llama.cpp#L4061-L4063

@slaren

slaren commented Jul 17, 2024

Copy link
Copy Markdown
Member

Yes, I assumed that it would only set the first element in that case. From what I can tell, only OpenELM ever uses more than the first element.

@ggerganov
ggerganov requested a review from slaren July 17, 2024 07:50
@ggerganov
ggerganov merged commit d197545 into master Jul 19, 2024
@ggerganov
ggerganov deleted the gg/bump-max-layers branch July 19, 2024 13:50
arthw pushed a commit to arthw/llama.cpp that referenced this pull request Jul 27, 2024
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
Seunghhon pushed a commit to Seunghhon/llama.cpp that referenced this pull request Apr 26, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
ljubomirj pushed a commit to ljubomirj/llama.cpp that referenced this pull request May 6, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
my-other-github-account pushed a commit to my-other-github-account/llama.cpp that referenced this pull request May 15, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
my-other-github-account pushed a commit to my-other-github-account/llama.cpp that referenced this pull request May 15, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
phibya pushed a commit to ziee-ai/llama.cpp that referenced this pull request May 29, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
AlexiAlp pushed a commit to minghaop/llama.cpp that referenced this pull request Jun 2, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
AlexiAlp pushed a commit to minghaop/llama.cpp that referenced this pull request Jun 2, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
fukuro-kun pushed a commit to fukuro-kun/fukuro-llama-cpp-turboquant that referenced this pull request Jul 5, 2026
* llama : bump max layers from 256 to 512

* llama : replace asserts with exceptions
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.

Bug: Can't quantize 405B Mega merge

2 participants