Repository navigation
fix(metal): refuse to serve a pack whose per-layer embeddings cannot … - #334
Merged
Merged
Conversation
…be used On a Metal window, PLE off was a silent degradation: the startup line said "PLE OFF" and the model answered without its per-layer embeddings - fluent and wrong, with nothing failing. The table comes from the pack's PLE file and the block needs every ple_* tensor, so PLE off with those weights present means an incomplete or mismatched model folder. Fail the window's init with that reason instead, so the server reports it rather than serving a model that is not the one the pack describes. A pack without PLE tensors still runs: that is a different model, and `PleRun::ready()` returning false there is expected.
|
This pull request is labeled
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
…be used
On a Metal window, PLE off was a silent degradation: the startup line said "PLE OFF" and the model answered without its per-layer embeddings - fluent and wrong, with nothing failing. The table comes from the pack's PLE file and the block needs every ple_* tensor, so PLE off with those weights present means an incomplete or mismatched model folder.
Fail the window's init with that reason instead, so the server reports it rather than serving a model that is not the one the pack describes.
A pack without PLE tensors still runs: that is a different model, and
PleRun::ready()returning false there is expected.What & why
Fixes #
Area
edge0-infer/)edge0-infer/tools/assets/)edge0-infer/examples/)docs/)edge0-strata/How was it tested?
Checklist
ctest --preset cpu -L "L0|L1"passes locally inedge0-infer/(engine changes); CI also runs these and theL2tests not labeledlocal-onlylocal-onlyanswer suites (ctest --preset cpu -L local-only) on Apple silicon with i8mm (M2 or later), which CI cannot judgeedge0-infer/docs/testing.md)