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

    Neuronpedia

    INDEX
    Explanations

    greetings followed by names or titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dvara
    0.59
     bidirectional
    0.54
     interrelated
    0.52
     hypothalamic
    0.52
     nontrivial
    0.51
    akati
    0.51
     vatth
    0.51
     idiosyncratic
    0.51
     ሌሎች
    0.51
    搞
    0.51
    POSITIVE LOGITS
     dear
    1.63
     sir
    1.54
    Dear
    1.34
     dearest
    1.34
    dear
    1.34
    sir
    1.29
     monsieur
    1.27
     جناب
    1.26
     Dear
    1.25
     Sir
    1.23
    Act Density 0.377%

    No Known Activations