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

    Neuronpedia

    INDEX
    Explanations

    hope and suggestions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     while
    0.77
     (...
    0.71
     isn
    0.69
     (
    0.67
    ുന്നത്
    0.67
     (…)
    0.67
    while
    0.67
     sedangkan
    0.65
     when
    0.64
     aren
    0.64
    POSITIVE LOGITS
     semoga
    1.16
     hoffe
    1.15
     Espero
    1.15
     espero
    1.11
     conseille
    1.08
     nhé
    1.06
     надеюсь
    1.05
     consiglio
    1.04
    เพิ่มเติม
    1.03
    Hopefully
    1.03
    Act Density 0.127%

    No Known Activations