ModelCensusopen-source ai reliability harness
Failure Mode Index · fmi_4_4

Dimensional-Analysis Failure.

Reasoning & Calculationinstrumented · tier 1 · status: stable

The model produces a quantity with inconsistent or wrong units, or mixes unit systems, yielding a dimensionally invalid result.

The question that exposes it

A car travels 150 km in 2 hours. Give its average speed — with units.

Probes: Whether the result carries correct, consistent units.

Trap: Models mix systems or drop dimensions, reporting a speed in the wrong unit.

Signature

Parse the answer's units and run dimensional analysis against the question's expected dimension; failure = unit mismatch or impossible dimension.

Example

Reports a speed in "kg/s"; adds meters to seconds.

Severity axes
prevalence
medium
harm
high
stealth
medium
Engineering guidance · class 4
  • Do arithmetic in code. Hand the model a calculator and ground the answer in what it returns.
  • Assert units and dimensions separately from the value.
  • Ask for method and answer as separate fields, then check they agree.
  • Do not ask a model to verify its own output in the same turn — it confirms far more readily than it catches.
Mitigations
  • explicit unit tracking
  • dimensional checkers
Measurement

One tab per model, newest first. Each card is that model against this mode under every condition on a single day. The mitigations above are the claim; the residual on each card is what actually survived the intervention.

fmi_4_4_v1 · 5 repeats · served 2026-09-10

loop 2026-09-10 · x-ai/grok-4.6 as served · closed weights · released 2026-01 · set s1.4-live · cases v1.7 · conditions v1.0 · ground truth as of 2026-08-09

ConditionFailure rate95% intervalnTrapAnswerable
No context
bare · The model alone, with no supplied context.
0%0%–10%n=50 · eff 360% (n=25)0% (n=25)
Irrelevant context
null · Length-matched irrelevant context. The control.
0%0%–10%n=50 · eff 360% (n=25)0% (n=25)
General context
recipe_a · Broad retrieval — the first passage the source returns, unreviewed.
0%0%–10%n=50 · eff 360% (n=25)0% (n=25)
Curated context
recipe_b · Hand-checked retrieval — the same passage plus the provenance chain behind it.
0%0%–10%n=50 · eff 360% (n=25)0% (n=25)
Presence effect · no context → irrelevant context
0.0pp

intervals overlap — not separated

What a document being there did, before any of its content. Should be small.

Content effect · irrelevant → best real context
0.0pp

intervals overlap — not separated

What the information itself did. This is the publishable claim.

Residual — what survives

Unchanged by context: 0% with none, 0% with the strongest supplied source. For a control mode that flat line is the result — it is what licenses reading movement elsewhere in this loop as grounding rather than as a document being present. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim
  • explicit unit tracking
  • dimensional checkers

replay this card →