Skip to content

docs(efficacy): the control read at three judgings per tree - #1853

Merged
braboj merged 1 commit into
mainfrom
docs/control-three-judgings
Sep 25, 2026
Merged

braboj merged 1 commit into
mainfrom
docs/control-three-judgings

Conversation

@braboj

@braboj braboj commented Sep 25, 2026

Copy link
Copy Markdown
Collaborator

Part of #1767.

The gpt-6-astra top-up (judge.py --root C:/efficacy/control-six --repeat 3, 11 judgings, 0 failed) completed three judgings on each of the control's six trees. The control README gains a section with the means and the three scores per cell. Every primary falls on the tree built to damage it and rises on one built to improve it.

One existing sentence changed: the note that the top-up had stopped on the usage limit now points to the new section. Conformance counts this as an edit to an existing file under tests/. It is prose in a README, not a test.

No CHANGELOG entry: the benchmark is this repository's own.

Checks run: smoke 33/33; conformance 0 failed, its readings read.

🤖 Generated with Claude Code

The gpt-6-astra top-up completed three judgings on each of the six trees.
The control README records the means: every primary falls on the tree built
to damage it and rises on one built to improve it.

Part of #1767.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
@braboj braboj added P2 Medium — important but not blocking task Atomic implementable work labels Sep 25, 2026
@braboj
braboj merged commit 84c2571 into main Sep 25, 2026
1 check passed
@braboj
braboj deleted the docs/control-three-judgings branch September 25, 2026 09:16
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

P2 Medium — important but not blocking task Atomic implementable work

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant