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

    Neuronpedia

    INDEX
    Explanations

    software license terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     desempen
    -1.02
     isKindOfClass
    -1.00
     funcionam
    -0.99
     poderão
    -0.97
     freundlichen
    -0.94
     Vorschlag
    -0.93
     Eindruck
    -0.91
     especially
    -0.90
     terão
    -0.89
    ：……
    -0.88
    POSITIVE LOGITS
     or
    1.30
    ")
    1.08
    "),
    1.03
     美丽
    0.92
     tengah
    0.92
    \},
    0.90
    ",
    0.90
    textEdit
    0.90
     商店
    0.88
    реш
    0.86
    Act Density 0.106%

    No Known Activations