© 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. 34-GEMMASCOPE-RES-131K
    7. 77884
    Prev
    Next
    INDEX
    Explanations

    serial killer

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on mentions of notorious killers or serial-murder figures (e.g. “serial,” “killer,” Ted Bundy, Juan Covington).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Diplomacy
    -0.86
     fraudulent
    -0.85
     gangster
    -0.84
    τες
    -0.82
     diplomacy
    -0.81
    êng
    -0.81
    欺
    -0.80
     vueltas
    -0.79
     rempli
    -0.78
     Dicapai
    -0.78
    POSITIVE LOGITS
     serial
    1.96
    Serial
    1.60
     Serial
    1.52
    serial
    1.48
     spree
    1.43
     victims
    1.38
    SERIAL
    1.31
     profiler
    1.17
     killings
    1.15
     SERIAL
    1.15
    Activations Density 0.035%

    No Known Activations