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

    Neuronpedia

    INDEX
    Explanations

    weak or no rights bypass

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     best
    0.73
     never
    0.72
    Al
    0.71
     still
    0.68
     C
    0.68
     similar
    0.66
     to
    0.66
     S
    0.65
     offered
    0.64
    De
    0.64
    POSITIVE LOGITS
     ಸರಿಯ
    0.93
    kter
    0.92
    <unused933>
    0.92
     தனிநாயகம்
    0.92
    tał
    0.91
    edLeft
    0.91
    <unused800>
    0.91
    InitFlag
    0.89
    <unused1312>
    0.89
    <unused1567>
    0.89
    Act Density 1.265%

    No Known Activations