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

    Neuronpedia

    INDEX
    Explanations

    community,

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     duck
    0.42
     Duck
    0.42
     tang
    0.38
     건
    0.38
    orum
    0.38
     ducks
    0.38
    Alk
    0.38
    duck
    0.37
     Ducks
    0.37
     Album
    0.37
    POSITIVE LOGITS
     Fujita
    0.42
     المثلث
    0.39
    DEVICE
    0.39
    াদে
    0.38
    Trimethyl
    0.37
    デバイス
    0.37
    談
    0.37
     مقدس
    0.37
    㴓
    0.37
     ऊपरी
    0.35
    Act Density 0.005%

    No Known Activations