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

    Neuronpedia

    INDEX
    Explanations

    local control

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ৬০
    0.43
     まし
    0.43
     horsepower
    0.40
     nee
    0.39
    ぶん
    0.39
    cooling
    0.38
    inflation
    0.38
     shag
    0.38
     volat
    0.37
    eburger
    0.37
    POSITIVE LOGITS
     similarly
    0.43
    ję
    0.42
     roues
    0.41
     as
    0.41
     terletak
    0.41
     Wrath
    0.41
     übrigens
    0.40
     proprio
    0.39
     vehículo
    0.39
    esseur
    0.39
    Act Density 0.002%

    No Known Activations