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

    Neuronpedia

    INDEX
    Explanations

    violence and hate speech

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    =|
    0.41
     recordings
    0.39
    聞いた
    0.38
     committee
    0.38
    attie
    0.37
     allocate
    0.37
     Correspondence
    0.37
    aha
    0.36
     electrical
    0.36
     meals
    0.36
    POSITIVE LOGITS
     demost
    0.47
    Nuestro
    0.44
    インド
    0.43
     dimost
    0.43
    Чи
    0.43
     demostrar
    0.42
     interrom
    0.42
    イチ
    0.42
    リ
    0.42
    rendre
    0.42
    Act Density 13.922%

    No Known Activations