-
Notifications
You must be signed in to change notification settings - Fork 31
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add test that exports to MLIR a small sharded Llama model (#220)
Add test that exports to MLIR a small sharded Llama model The decode step requires exporting in non-strict torch mode due to pytorch/pytorch#135061 This export required to extend the registration functionality of our custom tensor types by provinding `flatten_with_keys_fn`. This is also required to bump the PyTorch version to >=2.4 for other export tests. The export to MLIR fails with TypeError: Unsupported torch type conversion for !torch.vtensor<[3,1,7],complex<f32>> Needs further debugging. Detailed error [here](https://github.com/user-attachments/files/17163454/sharded-llama-export-to-mlir-failure.txt).
- Loading branch information
Showing
3 changed files
with
207 additions
and
96 deletions.
There are no files selected for viewing
This file contains 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
This file contains 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
Oops, something went wrong.