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

    Neuronpedia

    INDEX
    Explanations

    pronouns and self-references

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ذلك
    0.67
     reducing
    0.66
     it
    0.65
     while
    0.65
     While
    0.63
     disrupting
    0.63
    적으로
    0.62
     sha
    0.62
     am
    0.61
     USA
    0.59
    POSITIVE LOGITS
    zelf
    0.86
    Примечания
    0.81
    Personally
    0.73
    缀
    0.72
    <unused369>
    0.72
     جميعا
    0.71
    self
    0.69
    所有人
    0.69
    倆
    0.69
    <unused59>
    0.68
    Act Density 0.245%

    No Known Activations