Fold hourly 0348 HIGH - #63
Merged
Merged
Conversation
Catalog open-cricket BYOM (wire-compat ≠ logit-equiv), jev-decision-arena (confidence ≠ P(correct)), typesafe-ai-test 8,400 calls, werr JevBench 81.65 *theirs*, and OpenJev-Kit (training not complete). Densify jev-voice-browser §82 and is-malicious (does not sandbox). notes.md §139 / composition 633–648 / findings batch #121. Does not bump 0.5.0. Co-authored-by: Basit Mustafa <24601@users.noreply.github.com>
meijustory123/OpenJev-Kit is a rename of meijustory123/openjev (GitHub id 1379187719), not a distinct namesake. Keep the lock vs Zefan-Cai/Open-Jev. Gate densify original IDs for jev-voice-browser §82 and is-malicious §17. Co-authored-by: Basit Mustafa <24601@users.noreply.github.com>
|
ADV_PASS. Adversarial review + testing of #63 (HEAD Fixed on this branch before ready: Tests (
IDs unique vs |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Fresh PR off latest
mainafter merged #62 (hourly 0248 /notes.md§138). HARD RULE: do not reopen or amend PR #23–#62. Does not bump 0.5.0. Skip Archer (Hub still HTTP 401).invented_signal: false.IDs
notes.md§139 / composition 633–648 / findings batch #121 / archive2026-09-21T10.What this fold catalogs
92 novel HIGH items (0 REVISIT). Third-party benches stay theirs.
PRIMARY
JonathanHHenson/open-cricketBYOM Qwen2.5-1.5B. Wire-compat ≠ logit-equiv. Qwen2.5 ≠ Archer. replica ≠ TypeSafe.virtualman333/jev-decision-arenaGreedy 0.90 vs Oracle 0.82 theirs. Random conf 0.00 still 20.5% theirs. confidence ≠ P(correct). seed 42 n=1 is not Harbor.dopeCape/typesafe-ai-test8,400 calls $0.39 theirs. Noul 0.7 true 44% theirs. ≥0.9 conf 91.7% AG News theirs.pCwOrM/werrJevBench 81.65 theirs not Harbor. WindTunnel 49/49 theirs not Harbor. 0-byte Mandelbrot is not a replica.meijustory123/OpenJev-Kittraining not complete. Qwen3.5-0.8B ≠ Archer. OpenJev-Kit ISmeijustory123/openjev(same GitHub id 1379187719). ≠ Zefan-Cai/Open-Jev.Densify (not sibling first sightings)
moritzkremb/jev-voice-browseron §82: HEAD 054db0f3→198a0764395a, README fa033303→816309fc22e6. context is the conversation so far.luantak/is-malicious: A clean report is not proof. does not sandbox.Also
Checks
uniqueness_gate.pyok (0843…0248+0348)evaluate_decisions.py --self-testokrevisit_fingerprints.py --self-testok