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

    Neuronpedia

    INDEX
    Explanations

    the followed by a modifier

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     canaliculato
    0.58
     커
    0.58
    .
    0.57
     rač
    0.56
    써
    0.56
     也
    0.56
     上
    0.55
     이유
    0.55
     ㅅ
    0.55
     對
    0.54
    POSITIVE LOGITS
     to
    0.88
     with
    0.82
     dengan
    0.78
     of
    0.67
    0
    0.67
     avec
    0.64
     be
    0.63
     и
    0.63
     в
    0.61
     by
    0.60
    Act Density 0.017%

    No Known Activations