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

    Neuronpedia

    INDEX
    Explanations

    how things are done

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    。
    1.51
    ।
    1.44
    ։
    1.42
     :).
    1.33
    ے۔
    1.29
    $.
    1.25
    ។
    1.25
    。（
    1.22
    。...
    1.22
    。【
    1.21
    POSITIVE LOGITS
     solely
    0.76
     عندما
    0.73
     כאשר
    0.72
     when
    0.71
     relating
    0.69
     relying
    0.66
     involving
    0.65
     melewati
    0.63
     using
    0.63
     containing
    0.62
    Act Density 0.670%

    No Known Activations