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

    Neuronpedia

    INDEX
    Explanations

    in its entirety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     userType
    -0.76
     było
    -0.76
     gjennom
    -0.75
     resultMap
    -0.75
     özellikleri
    -0.74
     otherwise
    -0.73
    otherwise
    -0.69
    thal
    -0.69
    と思ったら
    -0.69
     quedando
    -0.68
    POSITIVE LOGITS
     entirety
    4.03
     totality
    2.94
     its
    2.66
     totalidad
    2.25
     totalité
    1.97
     kokona
    1.88
     toto
    1.88
     full
    1.84
     fullness
    1.80
     entier
    1.69
    Act Density 0.043%

    No Known Activations