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

    Neuronpedia

    INDEX
    Explanations

    screw parts and actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ।
    0.84
     ਸਾ
    0.82
    '।
    0.82
    ۰
    0.82
    ।'
    0.78
    ❱
    0.77
    дам
    0.77
    $.}
    0.76
    amine
    0.74
    おり
    0.74
    POSITIVE LOGITS
     d
    0.95
    AY
    0.87
    Q
    0.82
     helix
    0.80
     on
    0.79
     rotation
    0.79
    V
    0.79
    us
    0.78
    Z
    0.77
     chiral
    0.77
    Act Density 0.014%

    No Known Activations