Enable concatenation of unpadded prompts and a few other optimizations #1178
Conversation
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
sixiang-google
requested review from
gobbleturk,
khatwanimohit,
bvandermoon,
vipannalla and
RissyRan
as code owners
sixiang-google
force-pushed
the
sixiang-prefill-packing
branch
2 times, most recently
from
5ad72e7 to
653be9c
Compare
| .compile() | ||
| ) | ||
| min_num_prompts = max_length // length | ||
| max_num_prompts = max_length // (length // 2) |
There was a problem hiding this comment.
Did not understand this part?
There was a problem hiding this comment.
Save the compiled prefill_insert_batch function based on (padded_length, num_prompts). For example, if we have 12 sequences of length 80 (padded to 128), we now concatenate them into a sequence of length 960 and would invoke compile(128, 12)
sixiang-google
force-pushed
the
sixiang-prefill-packing
branch
from
653be9c to
0394b0d
Compare
sixiang-google
force-pushed
the
sixiang-prefill-packing
branch
from
0394b0d to
e35f3cd
Compare
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Outdated suggestions cannot be applied.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.
Description
-Based on #1162 and #1168, allow unpadded seqs to attend to each other.
-Enable overloading under len of 1024.
-Sorted data for less slot idle.
-Reconfigured garbage collection.
Tests
Please describe how you tested this change, and include any instructions and/or
commands to reproduce.
Checklist
Before submitting this PR, please make sure (put X in square brackets):