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

    Neuronpedia

    INDEX
    Explanations

    XYZ

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ared
    0.86
     broken
    0.85
    aren
    0.81
    蹉
    0.81
    損傷
    0.80
    eth
    0.78
    蹴
    0.77
    ayama
    0.77
    अधिकांश
    0.77
    owe
    0.77
    POSITIVE LOGITS
     POSSIBILITY
    0.74
     учун
    0.68
     leggi
    0.68
    Voici
    0.66
    HTree
    0.64
     pover
    0.63
    går
    0.63
    gelijk
    0.63
    υτό
    0.63
    dotto
    0.63
    Act Density 0.000%

    No Known Activations