Experiment ·
ORACLE-REAL
Does forcing the fact chunk to passage 1 help on real items?
From the record
- What was tested
- ORACLE-REAL
- Result
- U 0.427 vs 0.412
- Verdict
- closed: rule failed, CI excluded the hoped-for gain, a step-0 gate failed, or the mechanism was falsified
- Detail
- KILL (reading binds)
- Set
- 30 keyed v0.1 (dev)
- Judge
- Claude Haiku
- Pre-registered
- yes (790fe1d0)
Linked facts · 1
- ORACLE-REAL: Does forcing the fact chunk to passage 1 help on real items?ORACLE-REAL: Does forcing the fact chunk to passage 1 help on real items?U 0.427 vs 0.412Set30 keyed v0.1 (dev)StatusdevJudgeClaude Haikunnot recordedSeedsnot recorded95 % CI+0.015 (−0.029..+0.059)Date2026-10-21Caveatcriterion frozen: yes (790fe1d0); arm verdict: KILL (reading binds)Sourcenot in the public repositories yetOpen the data file