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

    Neuronpedia

    INDEX
    Explanations

    checksum, EXPAND, B-MIS, GCM, pp, nspawn, finishing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     and
    1.24
    _
    1.17
    </h2>
    1.05
    1.02
    🚺
    1.01
    🕣
    0.99
    0.99
    <0x0D>
    0.98
    0.98
    0.97
    POSITIVE LOGITS
    j
    1.61
    ar
    1.40
    ر
    1.40
    と思います
    1.39
    اد
    1.38
    ak
    1.34
    ou
    1.33
    ার
    1.30
    am
    1.28
    el
    1.27
    Act Density 0.183%

    No Known Activations