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

    Neuronpedia

    INDEX
    Explanations

    refusal to assist

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Columns
    0.45
     attribution
    0.42
     columns
    0.42
     ColumnName
    0.41
     column
    0.38
     rights
    0.37
    trait
    0.37
     #
    0.37
     assimilation
    0.37
     integration
    0.36
    POSITIVE LOGITS
    >′
    0.42
    Rew
    0.39
     المستقيم
    0.38
    баев
    0.38
    स्करी
    0.38
    Negative
    0.38
    Sorry
    0.38
    бари
    0.37
    फेस
    0.37
    Android
    0.37
    Act Density 0.003%

    No Known Activations