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

    Neuronpedia

    INDEX
    Explanations

    let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     wherein
    0.43
    θυ
    0.43
     removing
    0.39
     hereby
    0.39
     Figs
    0.38
     Using
    0.38
     Coulomb
    0.37
     pairwise
    0.37
     permettant
    0.37
    ၍
    0.37
    POSITIVE LOGITS
     довольно
    0.63
     काफी
    0.55
     unusually
    0.55
     väldigt
    0.55
     ரொம்ப
    0.55
     очень
    0.52
     굉장히
    0.52
     gerçekten
    0.52
     elég
    0.52
     complicated
    0.51
    Act Density 0.059%

    No Known Activations