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

    Neuronpedia

    INDEX
    Explanations

    altered by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    utiliser
    0.59
    Constant
    0.58
    കൊണ്ട്
    0.58
    Weiter
    0.55
     bizim
    0.54
     കൊണ്ട്
    0.54
    మ్
    0.53
    Record
    0.53
    tså
    0.53
    ಂದರೆ
    0.53
    POSITIVE LOGITS
     before
    2.27
     after
    2.21
     shortly
    1.91
     BEFORE
    1.76
    before
    1.74
     afterward
    1.73
     AFTER
    1.72
     setelah
    1.72
     sebelum
    1.72
     afterwards
    1.70
    Act Density 0.481%

    No Known Activations