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

    Neuronpedia

    INDEX
    Explanations

    traditional methods and concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    つ
    0.76
    </h2>
    0.67
     szé
    0.65
     of
    0.63
    })}{\
    0.62
     dată
    0.62
     höch
    0.62
     normalize
    0.61
     familiarize
    0.60
     ungefähr
    0.59
    POSITIVE LOGITS
    us
    1.09
    ت
    1.07
    et
    1.03
    b
    1.01
    ang
    0.95
    ir
    0.90
    al
    0.89
    ing
    0.88
    و
    0.87
    m
    0.86
    Act Density 0.001%

    No Known Activations