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

    Neuronpedia

    INDEX
    Explanations

    offensive word detection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    하거나
    0.46
    하면서
    0.46
    여
    0.41
     монта
    0.40
     있다는
    0.40
    하지
    0.40
    Reached
    0.39
    kadot
    0.39
    타
    0.39
    하지만
    0.39
    POSITIVE LOGITS
     Peralta
    0.45
     ethnic
    0.44
     redesigned
    0.43
     ajustes
    0.43
     patrols
    0.43
     peine
    0.43
     revamped
    0.42
     surveillance
    0.41
     zan
    0.41
     Ethnic
    0.41
    Act Density 0.001%

    No Known Activations