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

    Neuronpedia

    INDEX
    Explanations

    harming others or self

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chased
    0.43
     inning
    0.38
    بوط
    0.37
    licher
    0.36
     p
    0.36
    諸
    0.36
    izante
    0.35
     spanning
    0.35
     hiding
    0.35
    scanf
    0.34
    POSITIVE LOGITS
    whom
    0.54
     whom
    0.52
    उनकी
    0.44
     мнению
    0.43
     INCLUDING
    0.42
    आपके
    0.41
     आपके
    0.40
     उनकी
    0.39
     včetně
    0.39
     emociones
    0.38
    Act Density 0.024%

    No Known Activations