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

    Neuronpedia

    INDEX
    Explanations

    whole, entire, all

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Every
    0.99
     Every
    0.90
    every
    0.76
    மையான
    0.71
    Included
    0.70
     every
    0.67
    removed
    0.67
    included
    0.67
     included
    0.66
    erson
    0.66
    POSITIVE LOGITS
     whole
    3.12
    whole
    2.67
    整个
    2.49
     tutto
    2.48
    Whole
    2.41
    整個
    2.41
     Whole
    2.40
     seluruh
    2.39
     tudo
    2.38
     entire
    2.33
    Act Density 0.157%

    No Known Activations