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

    Neuronpedia

    INDEX
    Explanations

    making things accessible globally

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    0.98
    ادي
    0.88
     as
    0.82
    ુદ્ધ
    0.79
    s
    0.79
     integrante
    0.76
     jeopard
    0.75
    O
    0.73
     that
    0.71
    াই
    0.70
    POSITIVE LOGITS
     accessible
    1.11
     Accessible
    1.10
     Accessibility
    1.08
    Accessibility
    1.08
     accessibility
    1.07
    м
    1.06
    il
    1.04
    accessible
    1.02
    Accessible
    0.97
    ە
    0.95
    Act Density 0.010%

    No Known Activations