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

    Neuronpedia

    INDEX
    Explanations

    multiple languages and specific linguistic structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     owner
    0.60
     cari
    0.53
     dot
    0.51
     disrupt
    0.49
     renal
    0.49
     enter
    0.49
     prä
    0.48
     Dot
    0.48
     hydroxy
    0.48
     loin
    0.47
    POSITIVE LOGITS
     радика
    0.48
    とされる
    0.46
    たす
    0.45
    お
    0.44
    ராத
    0.44
    徘
    0.44
    бира
    0.43
    ાવો
    0.43
    है
    0.43
    いた
    0.43
    Act Density 0.000%

    No Known Activations