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

    Neuronpedia

    INDEX
    Explanations

    here

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    er
    1.72
    ㅎㅎ
    1.59
    ي
    1.59
    ुणा
    1.52
    骧
    1.52
     Ironically
    1.52
    е
    1.48
     Einen
    1.48
    υ
    1.48
    ۔
    1.45
    POSITIVE LOGITS
    il
    1.72
    на
    1.60
    یم
    1.57
    дование
    1.55
    é
    1.53
    cps
    1.52
    años
    1.48
    }}^{*
    1.48
    の一部
    1.45
    crs
    1.45
    Act Density 0.255%

    No Known Activations