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

    Neuronpedia

    INDEX
    Explanations

    plateau or stabilize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     डालें
    0.70
     وأ
    0.66
     öffnen
    0.66
     goût
    0.66
    ่วม
    0.66
     करणार
    0.65
     öpp
    0.65
    ussy
    0.64
     fühlen
    0.64
    撿
    0.64
    POSITIVE LOGITS
     plateau
    1.86
     plate
    1.55
     stabilize
    1.50
     Plateau
    1.48
    Plate
    1.43
     stabilized
    1.39
     stabilizes
    1.38
    plate
    1.37
     Plate
    1.33
     stabilise
    1.32
    Act Density 0.111%

    No Known Activations