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

    Neuronpedia

    INDEX
    Explanations

    faster, database, generation, made

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     perceive
    0.76
     tzv
    0.73
     redeem
    0.72
     آس
    0.70
     खोजने
    0.70
     unsatisfactory
    0.68
     Ekim
    0.68
    ﺔ
    0.68
     uncover
    0.66
     ghost
    0.65
    POSITIVE LOGITS
    وتی
    0.82
    Build
    0.80
    ته
    0.79
    costs
    0.79
    ngth
    0.77
    ज
    0.76
    ZX
    0.75
    COP
    0.74
    ти
    0.73
     условия
    0.73
    Act Density 0.001%

    No Known Activations