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

    Neuronpedia

    INDEX
    Explanations

    used to describe functionality

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    5
    0.66
    4
    0.63
     Use
    0.63
    7
    0.61
     multipurpose
    0.57
     käyttö
    0.56
     използ
    0.56
     वापरा
    0.55
     overuse
    0.55
     usable
    0.54
    POSITIVE LOGITS
    ttt
    0.55
    ILIA
    0.54
    ت
    0.53
    ع
    0.53
     determining
    0.53
    தி
    0.49
    ет
    0.49
    ον
    0.48
    ܛ
    0.48
    ห
    0.47
    Act Density 0.050%

    No Known Activations