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

    Neuronpedia

    INDEX
    Explanations

    third and fifth concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    1.41
    ي
    1.36
    ان
    1.31
    л
    1.23
    ле
    1.22
    ות
    1.19
     ganzen
    1.18
    两
    1.16
     medicamentos
    1.09
    8
    1.09
    POSITIVE LOGITS
     revital
    1.23
    the
    1.18
    <0x80>
    1.13
    ang
    1.13
    ot
    1.11
     third
    1.11
    third
    1.09
    s
    1.09
    ab
    1.06
     fifth
    1.05
    Act Density 0.011%

    No Known Activations