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

    Neuronpedia

    INDEX
    Explanations

    explain ... as if

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    d
    1.53
    ta
    1.47
    ்
    1.47
    actual
    1.46
    lens
    1.46
    drink
    1.43
     Toch
    1.43
    lined
    1.42
    dollar
    1.40
     गुजर
    1.38
    POSITIVE LOGITS
    ={\
    1.74
     orchestr
    1.66
    ال
    1.64
    oucester
    1.64
     particulars
    1.63
     prognosis
    1.63
     trigonometry
    1.61
    >/</
    1.60
    inology
    1.58
     ongoing
    1.57
    Act Density 0.001%

    No Known Activations