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

    Neuronpedia

    INDEX
    Explanations

    potential consequences or approaches

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Trend
    0.48
    Trend
    0.45
    Reducing
    0.44
     are
    0.43
    reduce
    0.42
    olate
    0.42
     reducing
    0.41
    owe
    0.40
    ﮢ
    0.40
     Agreements
    0.40
    POSITIVE LOGITS
     mixto
    0.48
     multit
    0.46
     rupani
    0.46
     சுமார்
    0.45
    दूसरी
    0.45
     paquet
    0.44
    ǎng
    0.43
     malade
    0.43
     Stowe
    0.43
     procé
    0.43
    Act Density 0.001%

    No Known Activations