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

    Neuronpedia

    INDEX
    Explanations

    breakdowns and nuances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     macromolecules
    0.31
     It
    0.31
     have
    0.30
     et
    0.30
     And
    0.30
     and
    0.30
     parle
    0.29
     it
    0.29
     The
    0.28
     SLAs
    0.28
    POSITIVE LOGITS
     ngunit
    0.40
     براي
    0.39
     başlayalım
    0.38
    ့်
    0.37
     ambayo
    0.37
    ală
    0.36
     kanggo
    0.36
     nicknamed
    0.35
     jeśli
    0.35
    のは
    0.34
    Act Density 6.158%

    No Known Activations