compare host vs device-side chunk metadata computation#84
Draft
learning-chip wants to merge 14 commits intomainfrom
Draft
compare host vs device-side chunk metadata computation#84learning-chip wants to merge 14 commits intomainfrom
learning-chip wants to merge 14 commits intomainfrom
Conversation
learning-chip
commented
Apr 2, 2026
|
|
||
| ### Layout conventions | ||
|
|
||
| In general, the input to the `fast_inverse` kernels is a set of `D × D` sized triangular matrices. Depending on how these matrices are stored in memory, we might have `contiguous` layout, or the so-called `BSND` layout. The main input is a batch of sequences, and each sequence is then split in "chunks" of length `chunk_size`. This `chunk_size` is the same as the matrix size `D`. |
Collaborator
Author
There was a problem hiding this comment.
In FLA convention D is the hidden_dim, and chunk_size should probably be named C....
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
See file
metadata_overhead.mdfor summary. Using host CPU is not faster, thus not putting into #79