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

    Neuronpedia

    INDEX
    Explanations

    wherein

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     démocr
    -0.99
     vectoriales
    -0.97
    resulting
    -0.97
     zelve
    -0.96
     normaux
    -0.96
     thereof
    -0.91
     resulting
    -0.90
     whereupon
    -0.90
     vectorielles
    -0.88
     wherein
    -0.88
    POSITIVE LOGITS
     
    0.64
     support
    0.63
    '
    0.62
     "
    0.61
     no
    0.61
    ​
    0.59
    \\
    0.58
    <
    0.57
    </em>
    0.56
     not
    0.56
    Act Density 0.049%

    No Known Activations