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

    Neuronpedia

    INDEX
    Explanations

    half and compound terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ９
    0.47
    ７
    0.42
    ５
    0.42
    อุ
    0.41
    ulis
    0.39
    Bueno
    0.38
    upuncture
    0.38
    ４
    0.38
    lihat
    0.37
    ulfill
    0.37
    POSITIVE LOGITS
     half
    0.83
     Half
    0.82
     dozen
    0.75
    half
    0.74
     दर्जन
    0.73
    半
    0.73
    Half
    0.73
     HALF
    0.71
     نصف
    0.70
     半
    0.64
    Act Density 0.023%

    No Known Activations