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

    Neuronpedia

    INDEX
    Explanations

    options or ways to proceed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     offensive
    0.55
     fü
    0.46
     körper
    0.46
     uden
    0.45
    น
    0.45
     damaged
    0.45
     Offensive
    0.45
     mám
    0.45
     conformance
    0.43
     jestem
    0.43
    POSITIVE LOGITS
    ickets
    0.48
    undi
    0.47
    umbai
    0.47
    ilie
    0.46
    iden
    0.46
    ate
    0.46
    에서
    0.45
    ossal
    0.45
     سای
    0.44
    ales
    0.44
    Act Density 0.001%

    No Known Activations