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

    Neuronpedia

    INDEX
    Explanations

    tightness or control

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ﻃ
    0.58
    <0x0D>
    0.55
    tabla
    0.52
    悟
    0.51
    sthe
    0.50
     metade
    0.49
    忔
    0.49
    di
    0.48
    tableau
    0.47
     illiterate
    0.47
    POSITIVE LOGITS
     tighter
    1.26
     tight
    1.22
    Tight
    1.13
     Tight
    1.10
     tightness
    1.09
    tight
    1.07
     tightened
    1.02
     tightly
    1.01
     tighten
    0.98
    緊
    0.98
    Act Density 0.053%

    No Known Activations