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

    Neuronpedia

    INDEX
    Explanations

    trust and untrustworthiness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ֡
    0.42
     céré
    0.38
    ခန်း
    0.38
     bumper
    0.37
    aloo
    0.36
    ລາ
    0.35
     appass
    0.35
    ೇಳ
    0.35
    镶
    0.34
     illumination
    0.34
    POSITIVE LOGITS
     trust
    4.28
    trust
    3.88
    Trust
    3.84
     Trust
    3.81
     TRUST
    3.67
    信任
    3.66
     trusts
    3.53
     trusting
    3.44
    TRUST
    3.36
     trusted
    3.34
    Act Density 0.162%

    No Known Activations