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

    Neuronpedia

    INDEX
    Explanations

    arbitrary

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    檗
    -0.08
     Congo
    -0.07
     véhicule
    -0.07
    asyarakat
    -0.07
    隺
    -0.07
     coats
    -0.07
    螂
    -0.07
    sock
    -0.07
    isateur
    -0.06
    abyrinth
    -0.06
    POSITIVE LOGITS
     habit
    0.08
    Determin
    0.07
     access
    0.07
    Edition
    0.07
     drift
    0.07
    か
    0.07
     empty
    0.07
    icit
    0.07
     tailor
    0.06
    CHAPTER
    0.06
    Act Density 0.003%

    No Known Activations