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

    Neuronpedia

    INDEX
    Explanations

    trust atoms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    inidad
    0.47
    featureType
    0.38
    sensitivity
    0.37
     இல
    0.37
    تيب
    0.36
    आव
    0.36
     unsuccessfully
    0.35
    complexity
    0.35
    forbidden
    0.35
     interactivity
    0.35
    POSITIVE LOGITS
     trust
    2.33
     trusted
    2.19
     trustworthy
    2.16
     trusting
    2.13
    Trust
    2.09
     Trust
    2.08
    trust
    2.08
     trustworthiness
    2.02
     TRUST
    2.02
    信頼
    1.99
    Act Density 0.020%

    No Known Activations