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

    Neuronpedia

    INDEX
    Explanations

    mathematical operations involving roots and logarithms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ik
    0.73
    Buenas
    0.71
    zne
    0.71
    ir
    0.68
    hon
    0.63
    re
    0.62
    Unter
    0.62
    Avoid
    0.61
    fw
    0.60
    aunque
    0.59
    POSITIVE LOGITS
     derivative
    0.49
     version
    0.49
     differentiator
    0.46
     component
    0.46
     pendant
    0.46
     debit
    0.46
     transformation
    0.45
     representation
    0.45
     gender
    0.44
    ↵↵
    0.43
    Act Density 0.618%

    No Known Activations