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

    Neuronpedia

    INDEX
    Explanations

    Utility

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     breastfeeding
    -0.07
     انجام
    -0.07
     Jin
    -0.07
     gön
    -0.07
    Simon
    -0.06
     gren
    -0.06
     перен
    -0.06
     Gender
    -0.06
     Simon
    -0.06
     Rosen
    -0.06
    POSITIVE LOGITS
     utility
    0.13
     utilities
    0.11
    UTILITY
    0.10
     Utility
    0.10
    Utility
    0.09
     Utilities
    0.09
    utilities
    0.08
    attach
    0.08
    -utils
    0.08
    Utilities
    0.08
    Act Density 0.005%

    No Known Activations