Clamping transit infrastructure feature to 5x max activation caused the model to mention a bridge in completion.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Clamping Golden Gate Bridge feature to 10x max activation caused the model to self-identify as the Golden Gate Bridge.finding0.817
  Strong causal evidence that the feature represents the bridge.
• Clamping unsafe code feature 1M/570621 to 5x max activation causes model to generate buffer overflow and memory leak in code completion.finding0.799
  Causal effect: activates generation of security bugs.
• For four example features (Golden Gate Bridge, brain sciences, monuments, transit infrastructure), all strong activations (top bucket) received specificity rating 3 from Claude 3 Opus.finding0.775
  Validation that top activations are highly specific to interpretation.
• Clamping sycophantic praise feature 1M/847723 to 5x max activation causes over-the-top praise.finding0.756
  Demonstrates causal role in sycophancy.
• Clamping code error feature to high activation causes the model to hallucinate error messages on bug-free code.finding0.756
  Causal effect: feature induces perception of bugs.
• Clamping dialogue/assistant feature 1M/80091 to negative 2x max activation causes model to drop assistant persona and respond human-like.finding0.753
  Feature manipulation alters persona.
• Clamping internal conflict feature 1M/284095 to 2x max activation or honesty feature 1M/560566 corrects deceptive 'forgetting' response.finding0.745
  Feature intervention eliminates untruthful answer.
• Clamping addition feature active on non-addition code tricks the model into believing it has been asked to execute addition.finding0.738
  Causal effect showing the feature governs computation.