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

    Neuronpedia

    INDEX
    Explanations

    initial or first steps

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    efined
    0.43
     Needless
    0.41
     caffe
    0.41
     dined
    0.38
     Plenty
    0.37
     sheltering
    0.37
     cleansed
    0.36
     sultry
    0.36
     caffeine
    0.35
    <h4>
    0.35
    POSITIVE LOGITS
     initiale
    0.52
     перший
    0.48
    最初に
    0.47
    最初の
    0.46
    для
    0.45
    Digite
    0.44
    してから
    0.43
    כי
    0.42
     тест
    0.42
     ਟ
    0.42
    Act Density 0.006%

    No Known Activations