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

    Neuronpedia

    INDEX
    Explanations

    sex slurs and explicit terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    部
    0.39
    范围
    0.37
     значительно
    0.37
    部の
    0.36
    部に
    0.36
    читать
    0.35
    frist
    0.35
    ⏲
    0.35
    澈
    0.34
    বে
    0.34
    POSITIVE LOGITS
     vomit
    0.39
    vom
    0.38
     Collins
    0.36
     squad
    0.36
    もう
    0.36
     sow
    0.36
     dreamy
    0.35
     következő
    0.35
     vomiting
    0.35
    wedge
    0.35
    Act Density 0.272%

    No Known Activations