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

    Neuronpedia

    INDEX
    Explanations

    . **Heading** "How it works"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     كيف
    0.47
     bagaimana
    0.45
    Какие
    0.44
    说明
    0.44
    很重要
    0.43
    Информация
    0.43
     explanations
    0.42
     יותר
    0.42
     объяс
    0.42
     કારણ
    0.41
    POSITIVE LOGITS
     this
    0.54
     einen
    0.50
     This
    0.48
     was
    0.46
     einem
    0.45
     Also
    0.44
     is
    0.43
     Meanwhile
    0.43
     Its
    0.43
     .
    0.43
    Act Density 0.076%

    No Known Activations