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

    Neuronpedia

    INDEX
    Explanations

    distinguishing categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ▛
    0.46
     短
    0.45
    计
    0.44
    ة
    0.44
    一股
    0.43
    极
    0.42
    elim
    0.42
    itated
    0.41
    赋予
    0.40
    友
    0.40
    POSITIVE LOGITS
     distinguish
    0.47
     tachy
    0.46
     involution
    0.45
     hypertrophy
    0.44
    चाकी
    0.44
     disting
    0.44
     perceive
    0.44
    その後
    0.44
     উচ্চারণ
    0.43
     inhomogeneous
    0.43
    Act Density 0.002%

    No Known Activations