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

    Neuronpedia

    INDEX
    Explanations

    grammatical constructs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ault
    -0.28
    /preferences
    -0.26
    èĩªä¸»
    -0.24
    éĸ
    -0.24
     pals
    -0.24
     contracting
    -0.24
    atisfied
    -0.24
     dereg
    -0.24
    PERT
    -0.23
    uche
    -0.23
    POSITIVE LOGITS
    èĥŃ
    0.27
     backs
    0.26
    ·»
    0.26
    vik
    0.26
    ä¸Ŀä¸Ŀ
    0.25
    绺
    0.25
    æĸĩä¸Ń
    0.24
    sky
    0.24
     stepping
    0.24
    æĥķ
    0.24
    Act Density 0.032%

    No Known Activations