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

    Neuronpedia

    INDEX
    Explanations

    perfectly followed by positive descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Below
    0.45
    %
    0.41
    Dec
    0.40
    Sub
    0.40
    West
    0.39
    Bast
    0.38
    char
    0.37
    care
    0.37
    Equality
    0.37
    Since
    0.37
    POSITIVE LOGITS
    <unused2121>
    0.48
     kesehatan
    0.46
     Един
    0.45
     zuen
    0.44
     opportunità
    0.44
     كانت
    0.44
    <unused1794>
    0.43
    <unused2127>
    0.43
    न्हा
    0.43
     وكانت
    0.43
    Act Density 0.004%

    No Known Activations