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

    Neuronpedia

    INDEX
    Explanations

    proper names and prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     да
    0.59
     а
    0.57
     е
    0.54
     с
    0.50
     у
    0.50
     по
    0.50
     \\
    0.50
     это
    0.49
     або
    0.49
     се
    0.48
    POSITIVE LOGITS
     Chiefs
    0.58
     Baseball
    0.57
    ihanna
    0.57
     Scorsese
    0.53
     County
    0.53
     Pharmaceuticals
    0.51
     Yankees
    0.51
     Seahawks
    0.50
    ufthansa
    0.50
     Seattle
    0.49
    Act Density 0.057%

    No Known Activations