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

    Neuronpedia

    INDEX
    Explanations

    on top

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ódigo
    0.73
    怎么
    0.68
    怎麼
    0.65
     escánd
    0.65
    Compute
    0.64
    rachten
    0.63
    compute
    0.63
     पंच
    0.62
     disertai
    0.62
     একত্রে
    0.62
    POSITIVE LOGITS
     top
    2.86
    top
    2.72
     atop
    2.38
    Top
    2.33
     Top
    2.25
    TOP
    2.15
     TOP
    2.11
     encima
    2.07
     tops
    1.99
     topped
    1.97
    Act Density 0.462%

    No Known Activations