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

    Neuronpedia

    INDEX
    Explanations

    danger and safety risks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     څنګه
    0.40
     Login
    0.40
     beyaz
    0.40
     CRUD
    0.39
     सियासी
    0.39
     login
    0.39
     cleansed
    0.39
     ظهور
    0.39
     fief
    0.38
     polinom
    0.38
    POSITIVE LOGITS
     safety
    2.11
    Safety
    1.98
     Safety
    1.95
     dangerous
    1.95
     unsafe
    1.93
     опас
    1.91
    危険
    1.91
    safety
    1.91
    危险
    1.88
     SAFETY
    1.85
    Act Density 0.045%

    No Known Activations