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

    Neuronpedia

    INDEX
    Explanations

    waste, broken, silence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     local
    0.42
     political
    0.41
     quá
    0.39
    Parsing
    0.39
     Shortcuts
    0.37
     Melting
    0.37
     marca
    0.37
     LOCAL
    0.37
    fert
    0.37
     Sporting
    0.37
    POSITIVE LOGITS
    хам
    0.41
     ET
    0.39
     χρει
    0.39
     идет
    0.38
    Ⲱ
    0.38
     듣
    0.38
     棉
    0.38
    ühle
    0.37
    紡
    0.37
    愴
    0.37
    Act Density 0.000%

    No Known Activations