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

    Neuronpedia

    INDEX
    Explanations

    overriding original values

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     initialise
    0.54
     initialize
    0.45
     populate
    0.44
     initializes
    0.44
    اکي
    0.42
     desktops
    0.42
    栍
    0.42
     hydrate
    0.41
     establish
    0.41
     thaw
    0.41
    POSITIVE LOGITS
    ite
    0.48
    irt
    0.45
    ib
    0.44
     우리나라
    0.43
    iverr
    0.43
    Y
    0.41
    tem
    0.41
    ete
    0.41
    isting
    0.41
    ia
    0.40
    Act Density 0.005%

    No Known Activations