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

    Neuronpedia

    INDEX
    Explanations

    mass-energy equivalence E=mc²

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    cutting
    0.57
     escribir
    0.57
    ഗത
    0.56
     writing
    0.55
    lekt
    0.55
    lesen
    0.55
     Orts
    0.55
    厂
    0.54
     PHONE
    0.53
     dobr
    0.52
    POSITIVE LOGITS
    ج
    0.48
    enn
    0.46
    ну
    0.46
    狸
    0.45
    amia
    0.45
     rounds
    0.45
     bouillon
    0.45
    eduanya
    0.44
     ceinture
    0.44
     }
    0.44
    Act Density 0.018%

    No Known Activations