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

    Neuronpedia

    INDEX
    Explanations

    support resources and hotlines

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Gir
    0.86
     reliable
    0.78
     Bor
    0.77
     embank
    0.76
     {{{
    0.76
     Does
    0.75
     trag
    0.75
     brave
    0.74
     قط
    0.74
    juris
    0.74
    POSITIVE LOGITS
    inside
    0.71
    czenia
    0.68
    defines
    0.67
    ionista
    0.66
    väl
    0.66
    forma
    0.66
     realizzazione
    0.65
    Dentro
    0.65
    value
    0.65
     szen
    0.65
    Act Density 0.056%

    No Known Activations