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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104779
    Prev
    Next
    INDEX
    Explanations

    acts of violence and crime

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

    The neuron strongly activates on explicit mentions of violent acts (e.g. rape, killing, assault, stabbing).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     to
    -2.38
     proporcionan
    -1.50
    as
    -1.50
    h
    -1.45
    o
    -1.41
     (
    -1.41
    p
    -1.38
    了下
    -1.35
    痙
    -1.34
       
    -1.30
    POSITIVE LOGITS
     sartén
    1.58
    ⡪
    1.50
     utensi
    1.47
     שונים
    1.43
     الاحتلال
    1.41
    
    1.34
     tebal
    1.30
    夘
    1.30
    สอง
    1.30
     campuran
    1.30
    Activations Density 0.044%

    No Known Activations