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

    Neuronpedia

    INDEX
    Explanations

    everywhere / frequently

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     importante
    0.96
    有一定的
    0.88
     gewisse
    0.87
    较大
    0.86
    较高
    0.80
     signifikan
    0.78
     Important
    0.77
    Significant
    0.76
     importanti
    0.74
    较高的
    0.74
    POSITIVE LOGITS
     everywhere
    2.24
     Everywhere
    1.96
     every
    1.87
     EVERY
    1.84
    every
    1.84
     endless
    1.79
     almost
    1.78
     ubiquitous
    1.76
    EVERY
    1.75
    Every
    1.67
    Act Density 0.326%

    No Known Activations