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

    Neuronpedia

    INDEX
    Explanations

    here's why explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     پیر
    0.60
    ellis
    0.59
     Pops
    0.59
     Leon
    0.59
     league
    0.58
     fell
    0.58
     Glory
    0.57
     Pol
    0.57
     Falling
    0.57
     Clarks
    0.56
    POSITIVE LOGITS
    ba
    0.69
     জ্যোতি
    0.66
    inę
    0.61
    ätter
    0.60
    omét
    0.60
    Ba
    0.59
    定
    0.59
    ubuk
    0.59
    ausa
    0.59
    ッキ
    0.58
    Act Density 0.238%

    No Known Activations