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

    Neuronpedia

    INDEX
    Explanations

    null safety

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     अफवाह
    0.47
     прямой
    0.44
     Linking
    0.41
     Vigil
    0.39
     linking
    0.39
     Chefs
    0.38
    elivery
    0.38
     Woman
    0.38
     chin
    0.38
     norme
    0.38
    POSITIVE LOGITS
    ポジション
    0.45
    ˆ‚
    0.39
    廂
    0.38
    新たな
    0.38
     refuted
    0.37
     vanwege
    0.37
    ธิบาย
    0.36
    ∝
    0.36
    ฏิ
    0.36
    കര
    0.35
    Act Density 0.001%

    No Known Activations