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

    Neuronpedia

    INDEX
    Explanations

    citation markers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     by
    0.66
     îl
    0.62
    ă
    0.59
     aucun
    0.59
     variés
    0.58
     exib
    0.58
     consta
    0.58
     cheapest
    0.57
    ミス
    0.57
     acestea
    0.57
    POSITIVE LOGITS
    y
    0.82
    شور
    0.69
    })
    0.57
    u
    0.57
    }.
    0.57
    ার
    0.56
     وزار
    0.56
    (-
    0.55
    рке
    0.54
    }(
    0.54
    Act Density 0.001%

    No Known Activations