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

    Neuronpedia

    INDEX
    Explanations

    programming code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Go
    0.80
    0.80
     Sol
    0.80
     tak
    0.78
     running
    0.78
     Mel
    0.77
     Cub
    0.77
     ________
    0.76
     Wol
    0.75
     Anglo
    0.75
    POSITIVE LOGITS
    ::
    1.20
    (''
    1.10
    <unused906>
    1.06
    ฺ
    1.04
     gemakkelijk
    1.03
    Utility
    1.03
    Util
    1.03
    Converter
    1.02
    뉸
    0.99
    <unused209>
    0.98
    Act Density 0.300%

    No Known Activations