INDEX
    Explanations

    Desirability

    This neuron activates on erotic or sexually arousing vocabulary.

    New Auto-Interp
    Negative Logits
     szy
    -0.07
     gönder
    -0.07
     مالی
    -0.06
     подаль
    -0.06
    _ed
    -0.06
    Idx
    -0.06
    queueReusable
    -0.06
     oud
    -0.06
     pcl
    -0.06
    -0.06
    POSITIVE LOGITS
     fertilizer
    0.08
     delight
    0.07
    -awaited
    0.07
    ivative
    0.07
     goodness
    0.07
     Lawrence
    0.07
     dividends
    0.06
    =length
    0.06
    due
    0.06
     Combo
    0.06
    Act Density 0.017%

    No Known Activations