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

    Neuronpedia

    INDEX
    Explanations

    examples to provide to them

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    чия
    0.48
    utch
    0.47
    訸
    0.47
    र्ष
    0.46
    േന
    0.46
     аген
    0.46
     व्
    0.46
     агент
    0.46
     पाता
    0.46
    ucing
    0.46
    POSITIVE LOGITS
     moderne
    0.49
     modernes
    0.43
     historiques
    0.41
     rage
    0.40
     historischen
    0.39
     historic
    0.38
     disguise
    0.38
     siglos
    0.38
     modernen
    0.38
     herbal
    0.38
    Act Density 0.008%

    No Known Activations