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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     week
    -0.08
    Weak
    -0.07
    弦
    -0.07
    弧
    -0.07
    ailable
    -0.07
    apg
    -0.07
    _short
    -0.07
    =i
    -0.07
    night
    -0.07
     &,
    -0.07
    POSITIVE LOGITS
     Sek
    0.07
    חבר
    0.07
    어요
    0.06
     rescued
    0.06
    但仍
    0.06
    と思いました
    0.06
     conceded
    0.06
    قدر
    0.06
     deutschen
    0.06
    ونة
    0.06
    Act Density 0.368%

    No Known Activations