anthropic/claude-sonnet-5.5 · Model identity and integrity report

Family

anthropic

V3 sub-model reference

Available ✓

Checks

知識截止 · 能力探針 · 拒絕模板 · 拒絕溫度參數

Family-fingerprint coverage

60 weighted rules · 8 feature categories

Concrete reference answers, refusal templates, and individual signal weights are not published because disclosure would help spoofing endpoints evade detection.

315 runs · 114 hosts · 258 anomalies

Family-fingerprint consistency

  anthropic      270  ██████████████████████░░░   86% (expected)
  unknown         42  ███░░░░░░░░░░░░░░░░░░░░░░   13% (suspicious)
  openai           3  ░░░░░░░░░░░░░░░░░░░░░░░░░    1% (suspicious)

Observed V3 sub-model mapping

  → abstain                                172 (55%) — low confidence
  → unknown                                 53 (17%) ⚠ mismatch
  → anthropic/claude-opus-4.8               47 (15%) ⚠ mismatch sim=0.81
  → anthropic/claude-haiku-4.5              26 ( 8%) ⚠ mismatch sim=0.79
  → anthropic/claude-sonnet-4.6              9 ( 3%) ⚠ mismatch sim=0.79
  → anthropic/claude-sonnet-5.5              4 ( 1%) ✓ correct sim=0.90
  → anthropic/claude-opus-4.7                2 ( 1%) ⚠ mismatch sim=0.83
  → openai/gpt-5.6-sol                       1 ( 0%) ⚠ mismatch sim=1.00
  → anthropic/claude-sonnet-4.5              1 ( 0%) ⚠ mismatch sim=0.83

Capability score distribution (auxiliary)

  0-49     █████████████████████████  315
  50-69    ░░░░░░░░░░░░░░░░░░░░░░░░░    0
  70-89    ░░░░░░░░░░░░░░░░░░░░░░░░░    0
  90-100   ░░░░░░░░░░░░░░░░░░░░░░░░░    0

BazaarLink Probe combines a Bayesian V2 family classifier across nine fingerprint categories with V3 sub-model references to verify endpoints claiming anthropic/claude-sonnet-5.5. Low family consistency, a different V3 match, or divergence from selfClaim is flagged as anomalous. BazaarLink Probe home.