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

    Neuronpedia

    INDEX
    Explanations

    workbook! validation are

    np_acts-logits-general · gemini-2.5-flash-lite

    code context identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    .
    0.74
    ется
    0.68
    نا
    0.64
    ник
    0.63
    ä
    0.61
     .
    0.60
    లు
    0.60
    \
    0.60
    ния
    0.59
    aker
    0.59
    POSITIVE LOGITS
    x
    0.69
    ᱐
    0.68
    ᱗
    0.66
    provides
    0.65
    ၀၀
    0.65
    excludes
    0.65
    and
    0.63
    ת
    0.63
     eléctricas
    0.62
    ០
    0.62
    Act Density 7.341%

    No Known Activations