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

    Neuronpedia

    INDEX
    Explanations

    explanation of rules

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ون
    0.63
    ر
    0.54
    ap
    0.50
     архіви
    0.48
    ين
    0.44
    ים
    0.43
    𝕕
    0.43
    yatiti
    0.43
    rometry
    0.42
     કારણે
    0.42
    POSITIVE LOGITS
     on
    0.56
     is
    0.54
     of
    0.45
     are
    0.44
     sont
    0.43
     jika
    0.43
     with
    0.42
    A
    0.42
    是
    0.40
     (
    0.40
    Act Density 1.198%

    No Known Activations