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

    Neuronpedia

    INDEX
    Explanations

    data types storage structures requirements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ot
    1.04
    .
    1.00
    رو
    0.98
    بد
    0.98
    با
    0.96
    رد
    0.92
    cial
    0.91
    م
    0.91
    ون
    0.89
    istä
    0.88
    POSITIVE LOGITS
    a
    2.44
    n
    1.62
     and
    1.52
    ي
    1.41
    ه
    1.39
    an
    1.33
    i
    1.33
    া
    1.32
    า
    1.32
    f
    1.29
    Act Density 0.201%

    No Known Activations