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

    Neuronpedia

    INDEX
    Explanations

    describing importance or key attributes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     also
    0.68
    asi
    0.66
     preferable
    0.64
     immobile
    0.64
     mük
    0.63
     neither
    0.63
     十分
    0.62
     acetylation
    0.62
    aded
    0.61
    чна
    0.61
    POSITIVE LOGITS
    perhaps
    1.14
     pleasures
    1.12
     Perhaps
    1.03
    maybe
    1.01
    Perhaps
    1.01
     Maybe
    1.01
    Maybe
    0.97
     happenings
    0.97
     maybe
    0.95
     Faculties
    0.93
    Act Density 0.072%

    No Known Activations