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

    Neuronpedia

    INDEX
    Explanations

    LaTeX geometry and tikz

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     आँख
    0.24
    roasted
    0.24
     Roasted
    0.23
    നിക്ക്
    0.22
    一身
    0.22
    <unused277>
    0.22
    覺得
    0.21
     iyong
    0.21
    culpa
    0.21
    崞
    0.21
    POSITIVE LOGITS
    _
    0.28
    .
    0.27
    -
    0.26
    ↵
    0.24
     service
    0.21
     年
    0.21
     format
    0.21
    8
    0.21
    에
    0.20
     measure
    0.20
    Act Density 0.002%

    No Known Activations