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

    Neuronpedia

    INDEX
    Explanations

    improves efficiency and function

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     difference
    0.74
    原則
    0.73
     deems
    0.71
     যাবতীয়
    0.70
     necessity
    0.70
     opposites
    0.70
     niezb
    0.69
    Difference
    0.68
    事物
    0.68
     excessive
    0.68
    POSITIVE LOGITS
     efficiency
    1.85
     efficiencies
    1.81
     Efficiency
    1.70
    efficiency
    1.66
    efficiencies
    1.65
    Efficiency
    1.65
     transparency
    1.62
    效率
    1.54
     eficiencia
    1.49
     eficiência
    1.45
    Act Density 0.269%

    No Known Activations