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

    Neuronpedia

    INDEX
    Explanations

    patching model, buttons, goat, without needing, angles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     espèce
    0.44
     הנ
    0.42
    لبية
    0.42
     विषय
    0.39
     hä
    0.38
    ニア
    0.38
    hala
    0.38
    窿
    0.37
     அறிய
    0.37
    dana
    0.37
    POSITIVE LOGITS
     reduction
    0.41
     padlock
    0.39
     ys
    0.38
    ap
    0.38
     lock
    0.38
    cot
    0.37
    टकों
    0.36
     plac
    0.36
    YT
    0.36
     लगना
    0.36
    Act Density 0.000%

    No Known Activations