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

    Neuronpedia

    INDEX
    Explanations

    Ch starts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    en
    0.90
    시
    0.85
    強い
    0.83
    En
    0.82
     av
    0.82
    жа
    0.80
     gir
    0.79
    න්ට
    0.79
    roje
    0.79
    nacht
    0.78
    POSITIVE LOGITS
     сад
    1.23
    ំពោះ
    1.12
    anooga
    1.11
    acter
    1.10
    munk
    1.05
     compris
    1.03
     Народ
    1.00
     resumen
    0.98
    persist
    0.96
     impresionante
    0.93
    Act Density 0.565%

    No Known Activations