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

    Neuronpedia

    INDEX
    Explanations

    Asymptotes and holes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Nordic
    -0.07
     uppercase
    -0.07
     mastering
    -0.07
     nightlife
    -0.07
     operating
    -0.07
     envol
    -0.07
     footer
    -0.07
     archa
    -0.06
    oper
    -0.06
     operates
    -0.06
    POSITIVE LOGITS
     benign
    0.09
     harmless
    0.09
    —even
    0.08
     rağmen
    0.08
     aaye
    0.08
    pictured
    0.08
     desain
    0.08
     Redes
    0.08
     komp
    0.08
     wenigstens
    0.08
    Act Density 0.020%

    No Known Activations