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

    Neuronpedia

    INDEX
    Explanations

    crosses below

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused323>
    0.45
     moneda
    0.44
    ]=
    0.42
     cálculos
    0.42
    <unused549>
    0.42
     légère
    0.41
     quotidienne
    0.41
    <unused1074>
    0.41
     zau
    0.41
    jobSize
    0.41
    POSITIVE LOGITS
    def
    0.42
    ipple
    0.41
    abstract
    0.40
    willing
    0.39
    perm
    0.39
    raised
    0.39
    called
    0.38
    said
    0.38
    g
    0.38
    prod
    0.38
    Act Density 0.001%

    No Known Activations