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

    Neuronpedia

    INDEX
    Explanations

    escape or escapism

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Coming
    0.46
     pochod
    0.45
     Desde
    0.43
    Coming
    0.43
    कडून
    0.42
    ذج
    0.41
     omitted
    0.40
     Accurate
    0.40
     Stationary
    0.40
     Directly
    0.39
    POSITIVE LOGITS
    逃
    1.06
     escape
    1.05
     escaping
    1.05
     khỏi
    1.02
    escape
    0.98
     escapes
    0.96
     Escape
    0.91
    escap
    0.91
    Escape
    0.91
     escaped
    0.90
    Act Density 0.023%

    No Known Activations