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

    Neuronpedia

    INDEX
    Explanations

    UI elements and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    1
    0.63
    8
    0.54
    acetic
    0.53
    3
    0.52
    gate
    0.48
    ंगामा
    0.48
    夹
    0.48
    5
    0.48
    curve
    0.47
    orragie
    0.47
    POSITIVE LOGITS
     таких
    0.54
    ्स
    0.49
     ませ
    0.49
     то
    0.48
    рных
    0.48
    表現
    0.48
     достовер
    0.48
    見
    0.47
     заклю
    0.47
    幅広い
    0.46
    Act Density 0.004%

    No Known Activations