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

    Neuronpedia

    INDEX
    Explanations

    temporal adverbs

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Jusqu
    -0.65
     Houſe
    -0.64
     Theſe
    -0.62
     [](
    -0.59
     houſe
    -0.59
     Anſ
    -0.57
     greateſt
    -0.56
     centrifug
    -0.55
     Beſ
    -0.55
    !")
    -0.54
    POSITIVE LOGITS
     he
    0.85
     I
    0.67
     it
    0.65
    AsUp
    0.64
     his
    0.63
    ,
    0.63
     they
    0.60
    rungsseite
    0.56
     we
    0.56
     she
    0.56
    Act Density 0.031%

    No Known Activations