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

    Neuronpedia

    INDEX
    Explanations

    withhold and withstand

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    vature
    0.40
    rafpunkte
    0.38
    സിന്
    0.38
     käytt
    0.38
    ClassLoader
    0.38
    겉
    0.38
     renderEncoder
    0.38
    szön
    0.38
    𒉣
    0.38
    Baj
    0.37
    POSITIVE LOGITS
     withholding
    2.59
     withhold
    2.50
     withheld
    2.44
     withstand
    1.86
     withstood
    1.62
     With
    1.29
    With
    1.25
     Withers
    1.23
    WITH
    1.21
    Wit
    1.18
    Act Density 0.038%

    No Known Activations