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

    Neuronpedia

    INDEX
    Explanations

    word parts or suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    做
    0.36
    ounds
    0.35
    أت
    0.34
    菸
    0.34
     أنت
    0.33
    時候
    0.32
    で
    0.32
     looks
    0.32
     tío
    0.32
     peeps
    0.32
    POSITIVE LOGITS
     delle
    0.42
    )}$
    0.39
    izioni
    0.38
    ుల
    0.35
     della
    0.34
    之事
    0.34
     }}\
    0.34
     commendable
    0.34
    nykh
    0.34
    㣩
    0.33
    Act Density 0.003%

    No Known Activations