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

    Neuronpedia

    INDEX
    Explanations

    aside from issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     therefore
    0.80
     cioè
    0.73
     VERY
    0.73
    즉
    0.72
    므로
    0.70
     quindi
    0.69
    라서
    0.68
    ってください
    0.65
     very
    0.64
     donc
    0.64
    POSITIVE LOGITS
    Despite
    1.80
    Alongside
    1.60
    Beyond
    1.58
     Despite
    1.51
     despite
    1.50
    Aside
    1.47
    While
    1.46
    Besides
    1.43
     Alongside
    1.39
    Amid
    1.39
    Act Density 0.103%

    No Known Activations