Fix unpassable 53_casual_attention performance test (input range -100..100 -> -1..1) - #333
Open
kunal-mansukhani wants to merge 5 commits into
Open
kunal-mansukhani wants to merge 5 commits into
kunal-mansukhani wants to merge 5 commits into
Conversation
Updated the range for generating Q, K, and V tensors in performance test.
kunal-mansukhani
requested review from
ishaan-arya and
shxjames
as code owners
October 2, 2026 01:13
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fixes #332.
The performance test drew Q, K, V from U(-100, 100) with d=128 while atol = rtol = 1e-5, so logits have std ~3333 and fp32 reassociation alone puts correct kernels ~0.046 off the reference. This changes the perf-test range to U(-1, 1), matching 112_attention_with_sinks (same M=5000, d=128, same tolerance), and updates the matching constraint in challenge.html.