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

    Neuronpedia

    INDEX
    Explanations

    limited scope or confinement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     livelli
    0.40
    شي
    0.40
     механиз
    0.40
     légende
    0.40
    Violation
    0.38
    Gen
    0.38
    Fellow
    0.37
    method
    0.37
    屽
    0.37
     Beats
    0.37
    POSITIVE LOGITS
     scope
    0.79
     Liability
    0.66
     confined
    0.63
     confines
    0.63
     liability
    0.62
     Scope
    0.61
     severely
    0.60
    liability
    0.55
    scope
    0.54
    Liability
    0.54
    Act Density 0.046%

    No Known Activations