Skip to content

docs(autotrain): continuous-openui-local screening cycles 8-10 closeout - #1509

Merged
Tyler-R-Kendrick merged 3 commits into
mainfrom
claude/great-dirac-rx8wy9
Aug 8, 2026
Merged

docs(autotrain): continuous-openui-local screening cycles 8-10 closeout#1509
Tyler-R-Kendrick merged 3 commits into
mainfrom
claude/great-dirac-rx8wy9

Conversation

@Tyler-R-Kendrick

Copy link
Copy Markdown
Owner

Summary

Second /autotrain continuous-loop iteration of this scheduled run (follow-up
to #1480, which merged). Restarted the designated branch from the latest
main per the merged-PR restart rule and ran 3 more supervised screening
cycles of continuous-openui-local (wf_smoke_v2 fixture train data, 20
steps, honest ship gates on):

  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c{8,9,10}-results.{md,json}
    — one JSON + markdown pair per cycle (documenting-experiment-results).
  • docs/MODEL_CARD.md / README.md — per-cycle checkpoint-provenance
    notes (fixture/scratch, not a ship promotion), each heading suffixed with
    its campaign id per the MD024 convention landed in docs(autotrain): continuous-openui-local screening cycles 1-5 closeout #1480.
  • src/slm_training/resources/versions.json — matching no-bump: history
    entries for harness.experiments.slm228_spectral_disposition.

Honesty status — all 3 cycles are non-positive

Cycle 8's component-edge arm didn't execute this pass (control-only fixture
screen); cycle 9 was a frozen replay of the same arm pair that completed and
still failed honest fixture ship gates; cycle 10 (component-edge vs
control) also completed and failed honest ship gates. Per autotrain's
continuous-mode law, non-positive cycles get local docs commits only — no
gh stack layer
, so this PR is plain documentation/provenance housekeeping,
not a claimed model or harness win. No checkpoint here is being promoted or
shipped.

Cycle diagnostics feed the next continuous iteration's component-plan arm
(queued in NextRunPriorityV1).

Test plan

  • scripts.verify_version_stamps --check passes on every commit in this PR
  • Pre-commit hook test shards (tests/test_scripts/test_verify_checkpoint_references.py,
    tests/test_versioning/*) pass on every commit
  • Each cycle's honest ship-gate scoreboard is captured verbatim in its
    docs/design/*-results.json
  • No training/eval claims beyond fixture-scale screening — do not
    interpret any of these 3 cycles as ship-ready

Generated by Claude Code

@vercel

vercel Bot commented Aug 8, 2026

Copy link
Copy Markdown

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated (UTC)
slm-training Ready Ready Preview Aug 8, 2026 3:19pm

Request Review

@coderabbitai

coderabbitai Bot commented Aug 8, 2026

Copy link
Copy Markdown

Warning

Review limit reached

@Tyler-R-Kendrick, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 39 minutes

You've used all free OSS reviews for now. Wait for the free limit to reset to keep reviewing this public repository.

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews.

How do review limits work?

CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability.

For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window.

Please refer docs for additional details.

Review details
⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 833f1e7a-f6a3-4845-bb3d-568347629259

📥 Commits

Reviewing files that changed from the base of the PR and between 6678d8f and 4853c41.

📒 Files selected for processing (9)
  • README.md
  • docs/MODEL_CARD.md
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c10-results.json
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c10-results.md
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c8-results.json
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c8-results.md
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c9-results.json
  • docs/design/continuous-loop-20260808-continuous-openui-local-8c0b60dd-c9-results.md
  • src/slm_training/resources/versions.json

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands.

claude and others added 3 commits August 8, 2026 10:18
Cycle continuous-loop-20260808-continuous-openui-local-8c0b60dd-c8:
control-only fixture screen (component-edge arm not executed this
pass); non-positive, still fails honest ship gates. Local docs commit
only, no stacked PR. Also proactively suffixes the c8 "Continuous
autotrain note" heading with its campaign id to keep README/MODEL_CARD
MD024-clean per the convention landed in PR #1480.

Appends the matching no-bump version-stamp history entry for
harness.experiments.slm228_spectral_disposition.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QBCuFmvGEyYJsjKGCccbNQ
…0260808-continuous-openui-local-8c0b60dd-c9 closeout
Cycle continuous-loop-20260808-continuous-openui-local-8c0b60dd-c10
(component-edge vs control, wf_smoke_v2 fixture): non-positive fixture
screen, still fails honest ship gates. Local docs commit only, no
stacked PR. Appends the matching no-bump version-stamp history entry
for harness.experiments.slm228_spectral_disposition.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01QBCuFmvGEyYJsjKGCccbNQ
@Tyler-R-Kendrick
Tyler-R-Kendrick force-pushed the claude/great-dirac-rx8wy9 branch from 61da386 to 4853c41 Compare August 8, 2026 15:18
@Tyler-R-Kendrick
Tyler-R-Kendrick merged commit dd174d9 into main Aug 8, 2026
1 of 3 checks passed
@Tyler-R-Kendrick
Tyler-R-Kendrick deleted the claude/great-dirac-rx8wy9 branch August 8, 2026 15:19
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants