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

    Neuronpedia

    INDEX
    Explanations

    Scaffold followed by open parenthesis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.59
    =\$
    0.57
    .=
    0.56
    ianSpace
    0.56
     saddhim
    0.56
    ittees
    0.55
    رة
    0.54
    ,=
    0.52
     आये
    0.51
    ائیں
    0.48
    POSITIVE LOGITS
     to
    0.54
    ка
    0.54
     jadwal
    0.51
     प्रयोग
    0.49
     transcript
    0.48
     palatable
    0.47
    生活中
    0.47
     was
    0.47
     निशाना
    0.47
    Was
    0.47
    Act Density 0.000%

    No Known Activations