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

    Neuronpedia

    INDEX
    Explanations

    counts of five, six, or eighth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     некоторы
    0.43
     alguns
    0.39
    Settl
    0.38
    sparsebundle
    0.38
     некоторым
    0.38
    ljivo
    0.37
    kező
    0.37
    െല്ലാം
    0.37
    ొప్పి
    0.37
     diarrh
    0.37
    POSITIVE LOGITS
    -
    0.63
    0.47
    ↵↵
    0.45
     (
    0.43
    </h3>
    0.43
    р
    0.42
    </h2>
    0.40
    .
    0.39
    $
    0.39
    </b>
    0.38
    Act Density 0.099%

    No Known Activations