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

    Neuronpedia

    INDEX
    Explanations

    shows purpose and effect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ..
    0.21
    \
    0.21
    $.
    0.20
    >.
    0.19
    !।
    0.19
    *.
    0.19
    。
    0.18
     Would
    0.18
    _.
    0.18
     skal
    0.17
    POSITIVE LOGITS
     ensures
    0.31
     позволяют
    0.30
     allows
    0.30
     позволяет
    0.28
     demonstrates
    0.28
     ensure
    0.28
     preclude
    0.28
     suggests
    0.27
     underscores
    0.26
     allow
    0.26
    Act Density 0.411%

    No Known Activations