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

    Neuronpedia

    INDEX
    Explanations

    chatting and creating

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ना
    0.45
    गा
    0.45
    ần
    0.44
    лку
    0.43
    STTS
    0.43
    anakk
    0.42
    ل
    0.42
    }&=
    0.42
    일본
    0.42
     દા
    0.42
    POSITIVE LOGITS
     where
    0.56
     confided
    0.53
     you
    0.52
     explicitly
    0.51
     aesthetically
    0.51
     configuring
    0.50
     according
    0.50
     entrusted
    0.50
     deploying
    0.49
     savoir
    0.48
    Act Density 0.001%

    No Known Activations