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

    Neuronpedia

    INDEX
    Explanations

    more than imagined

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     MUST
    0.40
     преимущественно
    0.38
    ことが多い
    0.38
    する必要があります
    0.37
     cụ
    0.37
    MUST
    0.37
    Point
    0.36
     Must
    0.36
     보면은
    0.36
    されていない
    0.35
    POSITIVE LOGITS
     imagined
    0.86
     expected
    0.79
    想像
    0.77
    预期
    0.77
     erwartet
    0.75
     예상
    0.74
     anticipated
    0.74
     imagination
    0.74
    想象
    0.73
     imagines
    0.71
    Act Density 0.015%

    No Known Activations