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

    Neuronpedia

    INDEX
    Explanations

    code blocks and keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     \
    0.20
    ^{
    0.20
    \
    0.19
    $
    0.19
     TCP
    0.18
    I
    0.17
    #
    0.17
     för
    0.17
    0.17
     derivatives
    0.17
    POSITIVE LOGITS
     decoração
    0.21
     decorar
    0.20
     semangat
    0.20
    媢
    0.19
    ז
    0.19
     mercato
    0.19
    తున్న
    0.19
    manın
    0.19
     чувство
    0.19
    tuvo
    0.19
    Act Density 0.043%

    No Known Activations