Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Select op decomposes to unsupported ops #671

Open
rpavlovicTT opened this issue Nov 12, 2024 · 1 comment
Open

Select op decomposes to unsupported ops #671

rpavlovicTT opened this issue Nov 12, 2024 · 1 comment
Assignees
@rpavlovicTT
Labels
Ops Support new op in tt-forge and tt-mlir
Milestone
[Training] Llama ...

Comments

@rpavlovicTT
Copy link
Contributor

rpavlovicTT commented Nov 12, 2024 •
edited
Loading

Running llama backward pass generates select op from concat op from forward pass. Concat op is then lowered to select op in autograd pass. Next, in post-autograd stage, select op is decomposed in sequence of few ops consisting of pad_tile, narrow, sparse_matmul which are not supported in TT-MLIR. This is legacy code anyway.

Solution for select op can be to lower it to ttir.slice op instead.

Image
@rpavlovicTT rpavlovicTT added the Ops Support new op in tt-forge and tt-mlir label Nov 12, 2024
@rpavlovicTT rpavlovicTT added this to the [Training] Llama bwd pass milestone Nov 12, 2024
@rpavlovicTT rpavlovicTT self-assigned this Nov 12, 2024
@rpavlovicTT
Copy link
Contributor Author

tenstorrent/tt-mlir#1179

@ashokkumarkannan1 ashokkumarkannan1 mentioned this issue Nov 26, 2024
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@rpavlovicTT rpavlovicTT

Labels
Ops Support new op in tt-forge and tt-mlir
Projects
None yet
Milestone
[Training] Llama bwd pass
Development

No branches or pull requests
1 participant
@rpavlovicTT