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

    Neuronpedia

    INDEX
    Explanations

    Def

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     stunted
    0.59
     Cessna
    0.58
     grunn
    0.54
    ショッピング
    0.52
    ВИ
    0.50
     কাশ
    0.50
     fah
    0.49
    FOL
    0.48
     infallible
    0.48
     Pelican
    0.47
    POSITIVE LOGITS
    Def
    1.16
     Def
    1.16
    def
    1.09
     def
    1.02
     Defin
    0.95
     definit
    0.93
    initions
    0.93
    inition
    0.88
     déf
    0.86
    iciencies
    0.86
    Act Density 0.028%

    No Known Activations