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

    Neuronpedia

    INDEX
    Explanations

    edible things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.91
    ти
    0.79
    is
    0.73
    те
    0.66
    ט
    0.66
    า
    0.64
    ‌
    0.63
     carbure
    0.62
     lathes
    0.61
    s
    0.59
    POSITIVE LOGITS
    し
    0.83
    ي
    0.77
    larda
    0.70
    يي
    0.69
     espí
    0.63
    WILL
    0.63
    মতী
    0.63
     bú
    0.63
    0.62
     bañ
    0.61
    Act Density 0.001%

    No Known Activations