fix(test): widen hallucination detection tolerance (#809)#810
Merged
planetf1 merged 1 commit intogenerative-computing:mainfrom Apr 10, 2026
Merged
Conversation
…ing#809) Logprob-derived scores drift ~0.036 across runs due to inference non-determinism. Widen from abs=3e-2 to abs=5e-2 to absorb jitter while still catching real regressions.
Contributor
|
The PR description has been updated. Please fill out the template for your PR to be reviewed. |
jakelorocco
approved these changes
Apr 10, 2026
Contributor
jakelorocco
left a comment
jakelorocco
left a comment
There was a problem hiding this comment.
lgtm; I had previously widened it but I guess there's more variability than realized
Merged
via the queue into
generative-computing:main
with commit Apr 10, 2026
e0ffd3d
10 of 11 checks passed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Misc PR
Type of PR
Description
Widen
pytest.approxtolerance fromabs=3e-2→abs=5e-2intest_hallucination_detection. Scores are logprob-derived weighted expected values — inference non-determinism causes drift of ~0.036, breaching the old threshold while the categorical signal remains stable.Testing
^ on tests this now seems more consistent with the looser tolerance