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

    Neuronpedia

    INDEX
    Explanations

    dialogue structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    urin
    0.27
    ’
    0.26
    Q
    0.26
    var
    0.25
    name
    0.25
    G
    0.25
    ume
    0.25
     Place
    0.24
    ref
    0.24
    oldo
    0.24
    POSITIVE LOGITS
     evils
    0.31
     棉
    0.30
    ⺠
    0.29
    🙎
    0.29
     injustices
    0.28
     Shit
    0.28
     unapolog
    0.28
    喟
    0.28
    ंपरिक
    0.28
    雱
    0.27
    Act Density 0.003%

    No Known Activations