Widen data/original: 22 more model-written passages (14 Train, 14 Validation); recall test pinned to live items #44

Merged
weksa merged 3 commits from weksa-lowering/corpus-widen into main 2026-11-10 18:16:48 +00:00
Member

Cut cut-corpus-widen.r1 (reports h1, h2), Soul verdicts s1 and s2 (no blockers). Test stays 0: ModelWritten is never in Test (split.rs). Round orig-5 over this corpus: 13/14 scored per split (smoke; briefed-subagent lowerer).

🤖 Generated with Claude Code

Cut cut-corpus-widen.r1 (reports h1, h2), Soul verdicts s1 and s2 (no blockers). Test stays 0: ModelWritten is never in Test (split.rs). Round orig-5 over this corpus: 13/14 scored per split (smoke; briefed-subagent lowerer). 🤖 Generated with [Claude Code](https://claude.com/claude-code)
Items original:7..28, written for the non-memorised target; groups chosen so
the V1 split derives Train or Validation. Test is not filled: a model-written
item of a test group is in no split (split.rs), so none can be held out.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
weksa merged commit da6bc7aa14 into main 2026-11-10 18:16:48 +00:00
Sign in to join this conversation.
No reviewers
No labels
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
GameCult/Weksa!44
No description provided.