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

    Neuronpedia

    INDEX
    Explanations

    python code placeholders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    '
    0.77
    .
    0.66
     DCE
    0.60
    తీ
    0.57
     Caucus
    0.57
     Sherlock
    0.55
     obvi
    0.55
    ский
    0.54
    سى
    0.54
     Président
    0.54
    POSITIVE LOGITS
    ه
    0.85
    ה
    0.77
    ۷
    0.72
    ną
    0.71
    Y
    0.71
    measures
    0.70
    T
    0.69
    Π
    0.65
    notes
    0.64
    ność
    0.64
    Act Density 0.001%

    No Known Activations