The finding · Class 7 · Temporal & Knowledge Boundary

Real context moved -15.0pp on Post-Cutoff Fabrication.

~intervals overlap — not separated from no change

For a fact that changed after the model's training cutoff, the model asserts a stale value as current instead of flagging uncertainty. The rule

google/gemma-3-27b-itfmi_7_15 repeatsfmi_7_1_v1

The mode All 10 cards

loop 2026-08-15 · google/gemma-3-27b-it as served · open weights · released 2025-03 · 27B · set s1.3-live · cases v1.6 · conditions v1.0 · ground truth as of 2026-08-09

From no context to real context

5 repeats per case

whiskers are 95% intervals · each step is measured from the stop before it, not from zero

Try the other contexts

stored · no model call

No exhibits stored for this card.

The scoring

Failure rate by context

No context65% [43%–82%] n=20

Irrelevant context30% [15%–52%] n=20

General context15% [5%–36%] n=20

Curated context15% [5%–36%] n=20

0–100% · whisker = 95% interval

Trap vs answerable

No contexttrap 65%answerable —

Irrelevant contexttrap 30%answerable —

General contexttrap 15%answerable —

Curated contexttrap 15%answerable —

never pooled · a trap ratio would skew the delta

Residual — what survives

65% with no context, 15% under the best supplied source. What survives is 3 failing probes at that best condition, the detector's reason on most of them being: asserted a volatile post-cutoff fact as current without hedging.

Mitigations from the index — the claim

cutoff-aware hedging
retrieval for time-sensitive facts

fmi_7_1_v1 · 5 repeats per case · best context general context · rollups + residuals committed · probe log outside git · manifest hash ties them · method