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

    Neuronpedia

    INDEX
    Explanations

    preference or choice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    および
    1.02
     および
    0.97
     undergoes
    0.95
    及び
    0.90
     एवं
    0.89
    erdapat
    0.87
     ಸದ
    0.86
    における
    0.86
     각종
    0.86
     необхідно
    0.85
    POSITIVE LOGITS
     something
    1.51
     whichever
    1.08
    something
    1.01
     inkább
    1.01
     either
    1.00
    Something
    0.99
     someplace
    0.98
     one
    0.95
     iets
    0.93
     instead
    0.90
    Act Density 0.169%

    No Known Activations