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

    Neuronpedia

    INDEX
    Explanations

    is a or are

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     interacted
    0.63
     ——
    0.62
     -
    0.61
     —
    0.61
     cannot
    0.60
     என்னும்
    0.59
     baffle
    0.58
     can
    0.58
    。
    0.58
     Which
    0.57
    POSITIVE LOGITS
     الحكوم
    0.66
    Rez
    0.63
    ñar
    0.63
    ándole
    0.62
     Kays
    0.62
    +)$
    0.62
    matics
    0.61
    Res
    0.61
    substack
    0.60
     anecdotal
    0.60
    Act Density 0.007%

    No Known Activations