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

    Neuronpedia

    INDEX
    Explanations

    caveats and emphasis

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    就没有
    0.45
     the
    0.43
     возникает
    0.40
    什么是
    0.39
    对于
    0.39
    这个时候
    0.38
    把
    0.38
    读取
    0.37
    oniazid
    0.37
     when
    0.36
    POSITIVE LOGITS
     caveats
    0.59
     énfasis
    0.57
     emphasis
    0.55
     caveat
    0.53
     intentions
    0.53
     proviso
    0.51
     учетом
    0.51
     vengeance
    0.51
     finalidad
    0.50
     connotations
    0.50
    Act Density 0.226%

    No Known Activations