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

    Neuronpedia

    INDEX
    Explanations

    what it would be like

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dominio
    1.32
     marche
    1.13
     decide
    1.12
     segura
    1.11
     domínio
    1.09
     sociedad
    1.08
     其实
    1.08
     formado
    1.07
     assunto
    1.06
     relazione
    1.06
    POSITIVE LOGITS
    etown
    1.33
    諧
    1.08
    '،
    0.96
    呎
    0.91
    ти
    0.91
     यी
    0.91
     Dayton
    0.90
    𝘰
    0.89
    "\
    0.89
    ró
    0.88
    Act Density 0.000%

    No Known Activations