Support larger hidden size in Attention Cuda kernel#7002
Merged
Conversation
gh-yewang
changed the title
gh-yewang
changed the title
gh-yewang
changed the title
tianleiwu
requested changes
Mar 15, 2021
tianleiwu
reviewed
Mar 15, 2021
tianleiwu
reviewed
Mar 15, 2021
tianleiwu
approved these changes
Mar 15, 2021
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
-Parity confirmed by comparing with CPU
-Perf test:
number of heads = 32
seq_len = 128
head_size | hidden_size | latency(ms)
32 | 1024 | 0.582
48 | 1536 | 0.802
64 | 2048 | 1.110
--------------------------------------(apply new kernels below)
65 | 2080 | 1.142
80 | 2560 | 1.332
96 | 3072 | 1.669
108 | 3456 | 1.930
128 | 4096 | 2.314
144 | 4608 | 2.697
160 | 5120 | 2.956