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

    Neuronpedia

    INDEX
    Explanations

    will precede a verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ка
    0.38
    ные
    0.34
    šky
    0.34
    да
    0.33
    мер
    0.32
    ні
    0.32
    ленні
    0.32
    és
    0.32
    мін
    0.32
     habíamos
    0.32
    POSITIVE LOGITS
    N
    0.41
    T
    0.31
    S
    0.30
    ur
    0.29
     be
    0.29
    ב
    0.28
    F
    0.28
    J
    0.28
    K
    0.27
    </h2>
    0.27
    Act Density 0.432%

    No Known Activations