docs(EXPERIMENTS): head32 flip-rate CIs — heb repair overwhelming, khm a proper null - #124
Merged
Conversation
…m a proper null Paired bootstrap over per-pair position dumps (the new parity feature): heb-diac 10.985% -> 0.288% flips, delta -10.696pp, 95% CI [-11.661, -9.726]. khm-latn 2.325% -> 2.302%, CI [-0.173, +0.125] — statistically indistinct, exactly right for a family that was never broken. Probe-set note disclosed (full gate sets vs the original E1 golden subset).
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Completes TODO.training-work/05: paired bootstrap over the per-pair position dumps. heb-diac (the outlier family): 10.985 to 0.288 percent flips, delta -10.696pp, 95 percent CI [-11.661, -9.726] — the quantized-head repair is overwhelming. khm-latn (benign family): 2.325 to 2.302, CI [-0.173, +0.125] — statistically indistinct, which is exactly right where nothing was broken. Probe-set difference vs the original E1 table disclosed.