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

    Neuronpedia

    INDEX
    Explanations

    automático machine learning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    0.55
    愆
    0.47
     \
    0.43
    胙
    0.43
    Fallback
    0.42
    [\
    0.42
    zechoslovak
    0.42
    initial
    0.42
    )$\
    0.42
    Estados
    0.42
    POSITIVE LOGITS
    ing
    0.59
    un
    0.59
    ５
    0.51
    ن
    0.48
     ৫
    0.48
    5
    0.47
    ↵
    0.46
    נ
    0.45
    ↵↵
    0.45
     היו
    0.45
    Act Density 0.002%

    No Known Activations