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

    Neuronpedia

    INDEX
    Explanations

    coercion, abuse, or manipulation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Che
    0.42
     Via
    0.41
     প্রেমিক
    0.39
    |$.
    0.38
     Fra
    0.38
     Canary
    0.38
     bilgis
    0.38
    |(
    0.37
     azide
    0.37
    iov
    0.37
    POSITIVE LOGITS
    Personal
    0.46
    Pat
    0.43
    Mental
    0.43
     personal
    0.42
    パー
    0.41
    🍵
    0.39
    tear
    0.39
     personalization
    0.39
    ナス
    0.39
    녹
    0.38
    Act Density 0.000%

    No Known Activations