© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128733
    Prev
    Next
    INDEX
    Explanations

    The neuron detects mentions of assassination (and related attempted‐killing) events.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    assassination plots and targeted violence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ztő
    -1.01
     партнер
    -1.00
    ἵ
    -0.98
    Descriptions
    -0.91
     skrev
    -0.90
     poveznice
    -0.90
    wości
    -0.90
     behaviors
    -0.90
    onatos
    -0.88
     exhort
    -0.88
    POSITIVE LOGITS
    を狙
    1.16
    targets
    1.06
     waard
    1.02
     himself
    0.99
     veiligheid
    0.98
     assassination
    0.97
     security
    0.96
     beserta
    0.96
     risico
    0.95
     persoons
    0.94
    Activations Density 0.047%

    No Known Activations