Repository navigation
feat(gate1): add deterministic utility graders for WAVE results - #14
Merged
Merged
Conversation
Score locate/conflict/stale_dirty answers and compare D_SHADOW to max(B, C). CONTROL is recorded, not used for lift. WAVE5 against these graders is stop, not a broker authorization.
Deploying md-evals with
|
| Latest commit: |
c448294
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://2fe3c2b0.md-evals.pages.dev |
| Branch Preview URL: | https://feat-gate1-utility-grade.md-evals.pages.dev |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Evidence
Adds
python -m md_evals.gate1_utility_grade results.json. Deterministic string gold:src/base.ts/src/consumer.ts/src/entry.jsconflictplus disagreement languagefreshness: fresh/git_state: cleanLift: D_SHADOW mean rate vs max(B, C). CONTROL is not used for the hypothesis.
WAVE5 (
~/Escritorio/gate-1-utility-20260917-010004-WAVE5/results.json): scored_cells 24,hypothesis_liftfalse,utility_verdictstop. D mean 0.667 vs best single 0.833. Not a broker authorization. Billing unverified. Model answers are mostly agent-theater.Testing
Observed: EXIT=0, 5 passed.
PR #13 was merged with Tests 3.12/3.13 green; Cloudflare Pages had failed on that PR.