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

    Neuronpedia

    INDEX
    Explanations

    transforms or limits or plays or returns or supports or drops

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ли
    1.21
    о
    1.19
    lo
    1.18
    ε
    1.14
    і
    1.14
    ت
    1.10
    ted
    1.09
    ுள்ள
    1.03
    torn
    1.03
    ten
    1.01
    POSITIVE LOGITS
    heets
    1.56
    heet
    1.25
    creens
    1.24
    ysteem
    1.22
    hift
    1.20
    criptors
    1.15
    essions
    1.10
    ources
    1.10
    喈
    1.10
     esophagus
    1.08
    Act Density 0.143%

    No Known Activations