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

    Neuronpedia

    INDEX
    Explanations

    explains why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    न्वयन
    0.45
     otherwise
    0.43
     unscrupulous
    0.43
    യില്ല
    0.41
    ნის
    0.41
     όχι
    0.40
     Selatan
    0.40
     keenly
    0.40
    謦
    0.39
    ukam
    0.38
    POSITIVE LOGITS
     tué
    0.46
     marqués
    0.44
    alon
    0.43
    ؛
    0.43
     guerr
    0.43
     và
    0.41
    a
    0.41
    上記
    0.40
     blancs
    0.40
     وإ
    0.40
    Act Density 0.014%

    No Known Activations