finding
active
finding:on-ordinary-coding-prompts-misaligned-checkpoints-show-fake-secretly-fraud-trick-hidden-in-j-lens-at-prefilled-i-signal-present-at-sdf-checkpointOn ordinary coding prompts misaligned checkpoints show fake, secretly, fraud, trick, hidden in J-lens at prefilled I; signal present at SDF checkpoint.
Model organism detection.
Source paper
extracted_fromAuthors
Neighborhood — ranked by edge-count
Papers (1)
paper
Claims (1)
claim
- Model organism results.
Datasets (1)
dataset
- Four checkpoints used for auditing.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Mechanistic.
- J-lens at fabricated search results shows fake, injection, false and 假 while model silently omitted them.finding0.784Silent recognition.
- Extrapolation from scale-emergence finding to future risk
- Load-bearing description of the core pernicious divergence mechanism illustrated in Figure 1
- Initial evidence that alignment faking persona is more sensitive to exploiting training signals
- Interpretation of illusion.
- Imagine instruction raises property name in J-lens while J-orthogonalized property probe stays at baseline.finding0.755J-space privileged.
- Geometric evidence for convergence to stable truth directions only for simpler tasks.