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

    Neuronpedia

    INDEX
    Explanations

    advanced capabilities and techniques

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    of
    1.40
    ون
    1.27
    that
    1.23
     a
    1.21
    1.13
    th
    1.07
    ود
    1.04
    ας
    0.99
    1
    0.98
    vä
    0.97
    POSITIVE LOGITS
    u
    1.48
    r
    1.38
    z
    1.31
    ر
    1.30
     avanç
    1.27
    y
    1.23
     advancements
    1.16
    quela
    1.12
    i
    1.08
     subpar
    1.05
    Act Density 0.034%

    No Known Activations