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

    Neuronpedia

    INDEX
    Explanations

    break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zusätz
    0.66
    additional
    0.65
    他の
    0.63
    Additional
    0.63
     знак
    0.60
    商品は
    0.60
    appears
    0.58
    )]=
    0.58
     其他
    0.58
    任何
    0.58
    POSITIVE LOGITS
     discuss
    1.77
     explore
    1.72
    探讨
    1.67
     Discuss
    1.60
     discussing
    1.56
    Discuss
    1.55
     membahas
    1.55
     examine
    1.52
     поговорим
    1.51
     рассмотрим
    1.49
    Act Density 0.745%

    No Known Activations