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

    Neuronpedia

    INDEX
    Explanations

    though, but, or categorized

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     takže
    0.68
     impresionante
    0.65
     folosit
    0.62
     utilisé
    0.61
     utilisée
    0.61
     menyebabkan
    0.61
    導致
    0.60
     ktoré
    0.60
    違う
    0.60
     ทำให้
    0.59
    POSITIVE LOGITS
     especially
    1.05
    especially
    0.90
     particularly
    0.90
     even
    0.87
     albeit
    0.87
     insofar
    0.87
     although
    0.86
    even
    0.84
    albeit
    0.81
     Especially
    0.78
    Act Density 0.339%

    No Known Activations