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

    Neuronpedia

    INDEX
    Explanations

    digestive, damages, cause, aligns with intent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hydroxide
    1.23
    পূর্বক
    1.22
     quarantined
    1.16
    挝
    1.16
     (|
    1.14
     multilevel
    1.13
    apiVersion
    1.10
     genders
    1.09
     Pending
    1.09
     erotic
    1.08
    POSITIVE LOGITS
    ве
    1.62
    ي
    1.56
    i
    1.48
    ০
    1.38
     dolayı
    1.31
    л
    1.30
    ু
    1.30
    е
    1.29
    이
    1.28
    り
    1.28
    Act Density 0.002%

    No Known Activations