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

    Neuronpedia

    INDEX
    Explanations

    slice, Long, ambiguous, potential

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     makeshift
    1.15
     exh
    1.11
     hijacked
    1.11
     filles
    1.09
     Salt
    1.09
    goes
    1.07
    山の
    1.07
     Jugendliche
    1.06
    걀
    1.06
    ീയ
    1.05
    POSITIVE LOGITS
    是一家
    1.04
    šću
    1.04
     nhàng
    1.00
    ğından
    0.98
    𝒙
    0.95
    като
    0.94
    娀
    0.93
    രുന്ന
    0.93
    💅
    0.92
    😈
    0.91
    Act Density 0.000%

    No Known Activations