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

    Neuronpedia

    INDEX
    Explanations

    allegedly, reportedly, supposedly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Usually
    0.43
    annel
    0.43
    probabilities
    0.38
    Refer
    0.38
    Generally
    0.38
     абсолютно
    0.38
    Ill
    0.37
    Prob
    0.37
    Signed
    0.37
    一批
    0.37
    POSITIVE LOGITS
     allegedly
    1.83
     reportedly
    1.63
     supposedly
    1.56
     якобы
    1.54
     purportedly
    1.48
     according
    1.42
     menurut
    1.40
    据
    1.36
     volgens
    1.35
    据说
    1.32
    Act Density 0.022%

    No Known Activations