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

    Neuronpedia

    INDEX
    Explanations

    corruption

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    B
    1.44
     що
    1.28
    G
    1.24
    Point
    1.14
    F
    1.13
     а
    1.11
    ],
    1.10
    R
    1.10
    <0x95>
    1.09
    Brief
    1.06
    POSITIVE LOGITS
     conlle
    1.13
    jų
    1.13
    慗
    1.09
    Appuntamento
    1.07
     wreaths
    1.06
    邋
    1.05
     undue
    1.03
     lattices
    1.02
     glistening
    1.02
     glittering
    1.01
    Act Density 0.289%

    No Known Activations