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

    Neuronpedia

    INDEX
    Explanations

    Python PyQt/PySide QtWidgets import

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     at
    0.77
     were
    0.77
     Were
    0.75
    我们
    0.73
     for
    0.73
    ，
    0.71
    的
    0.71
     Marlon
    0.70
     [
    0.69
     seront
    0.67
    POSITIVE LOGITS
    ством
    0.80
    ва
    0.69
    ling
    0.67
    clipped
    0.67
    чет
    0.67
    して
    0.66
    straction
    0.66
    rases
    0.64
    стю
    0.63
    kelijk
    0.63
    Act Density 0.001%

    No Known Activations