ModelCensusopen-source ai reliability harness
The finding · Class 5 · Robustness & Consistency

Real context moved 0.0pp on Paraphrase Non-Invariance.

~intervals overlap — not separated from no change

Meaning-preserving paraphrases of the same question yield materially different answers. The rule

anthropic/claude-sonnet-4.6fmi_5_25 repeatsfmi_5_2_v1
In one frame
screenshot this
Paraphrase Non-Invariancefmi_5_2
anthropic/claude-sonnet-4.6
Same question, both times
How many bones are in the adult human body?
No contextPASS

## Total Bone Count in an Adult Human The adult human body contains **206 bones**. ### Key Points: - **At birth**, infants have approximately **270-300 bones** (many are cartilaginous or not yet fused) - As a person grows, many bones **fuse together** through a process called **ossification** - By early adulthood (around **age 25**), the skeleton stabilizes at **206 bones** ### General Distribution: | Region | Approximate Count | |--------|------------------| | Skull | 22 | | Vertebral column | 26 | | Thoracic cage (ribs & sternum) | 25 | | Upper limbs (both) | 64 | | Lower limbs (both) | 62 | | Pelvic girdle | 2 (hip bones) | > **Note:** The exact number can vary slightly between individuals due to **sesamoid bones** or **sutural (Wormian) bones**, but **206** is the standard accepted number.

detector fmi_5_2_v1: answers invariant across paraphrases

Curated contextPASS

The context provided discusses **regional variation in English dialects** (bread roll terminology, dialect surveys, etc.) and contains **no information about human bone counts**. From general knowledge, the commonly cited answer is **206 bones** in the adult human body, but I cannot derive that from the provided passage, and you should verify that with an appropriate anatomical source rather than relying on my general knowledge here.

detector fmi_5_2_v1: answers invariant across paraphrases

Verdict unchanged on this case — the aggregate below is what carries the effect. Across 5 repeats: 0.0pp (intervals overlap).fmi_5_2_v1 · anthropic/claude-sonnet-4.6 · 2026-08-15 · ground truth as of 2026-08-09

loop 2026-08-15 · anthropic/claude-sonnet-4.6 as served · closed weights · released 2026-02 · 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
failure rate · left is fewer failures0%5%10%15%20%No context0%[0%11%] n=30Irrelevant context0%presence — a document being there[0%11%] n=300.0pp OVERLAPSCurated context0%content — the information itself[0%11%] n=300.0pp OVERLAPS
whiskers are 95% intervals · each step is measured from the stop before it, not from zero
Try the other contexts
stored · no model call
1Questionheld fixed
How many bones are in the adult human body?

expected as of 2026-08-09

206. 206 in a typical adult. Widely misremembered, which is what makes the phrasings diverge.

2Contextchange me

Nothing supplied. The baseline.

3ResponseNo context
PASS

## Total Bone Count in an Adult Human The adult human body contains **206 bones**. ### Key Points: - **At birth**, infants have approximately **270-300 bones** (many are cartilaginous or not yet fused) - As a person grows, many bones **fuse together** through a process called **ossification** - By early adulthood (around **age 25**), the skeleton stabilizes at **206 bones** ### General Distribution: | Region | Approximate Count | |--------|------------------| | Skull | 22 | | Vertebral column | 26 | | Thoracic cage (ribs & sternum) | 25 | | Upper limbs (both) | 64 | | Lower limbs (both) | 62 | | Pelvic girdle | 2 (hip bones) | > **Note:** The exact number can vary slightly between individuals due to **sesamoid bones** or **sutural (Wormian) bones**, but **206** is the standard accepted number.

fmi_5_2_v1

detector fmi_5_2_v1: answers invariant across paraphrases

The scoring
Failure rate by context
No context0% [0%11%] n=30
Irrelevant context0% [0%11%] n=30
General context33% [19%51%] n=30
Curated context0% [0%11%] n=30

0–60% · whisker = 95% interval

Trap vs answerable
No contexttrap 0%answerable 0%
Irrelevant contexttrap 0%answerable 0%
General contexttrap 50%answerable 25%
Curated contexttrap 0%answerable 0%

never pooled · a trap ratio would skew the delta

Residual — what survives

Supplying context did not reduce this failure, it increased it: 0% with no context against 33% under the strongest supplied source. What survives is therefore not a remainder but a substitution — the failures under context are 0 probes the detector still rejected. Nothing survives at the best condition on the cases measured here.

Mitigations from the index — the claim
  • paraphrase-augmented training
  • canonicalization

fmi_5_2_v1 · 5 repeats per case · best context curated context · rollups + residuals committed · probe log outside git · manifest hash ties them · method