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

    Neuronpedia

    INDEX
    Explanations

    sts:AssumeRole

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     scolded
    0.55
    ors
    0.53
    al
    0.53
    ators
    0.52
     flavored
    0.50
    ش
    0.50
     guessing
    0.48
     antics
    0.48
     nicknames
    0.48
    anians
    0.48
    POSITIVE LOGITS
    cumin
    0.54
     проведе
    0.52
    鏻
    0.52
     и
    0.49
    cima
    0.49
     અને
    0.48
    GPa
    0.48
     фокуси
    0.48
     эффективно
    0.47
     profissional
    0.46
    Act Density 0.000%

    No Known Activations