Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    language of instruction

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     signatures
    -0.08
    always
    -0.08
     Always
    -0.08
    ellipse
    -0.08
     zawsze
    -0.07
    legal
    -0.07
    telegram
    -0.07
     always
    -0.07
    odata
    -0.07
    Always
    -0.07
    POSITIVE LOGITS
     betrieben
    0.09
     kcal
    0.08
    要求
    0.08
    _except
    0.08
     requisito
    0.08
     exigir
    0.08
     voert
    0.08
     제외
    0.08
    リン
    0.08
     Voraussetzung
    0.08
    Act Density 0.013%

    No Known Activations