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

    Neuronpedia

    INDEX
    Explanations

    require("express")

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ing
    1.52
    lz
    1.25
    lords
    1.24
    etään
    1.20
    lığı
    1.18
    ة
    1.18
    larda
    1.16
    yg
    1.16
    lé
    1.16
    lw
    1.15
    POSITIVE LOGITS
    ди
    1.46
    "${
    1.41
    "।
    1.38
    не
    1.35
     dictum
    1.34
     hearsay
    1.32
     Danville
    1.32
    ний
    1.31
     midsole
    1.29
     tangy
    1.28
    Act Density 0.001%

    No Known Activations