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

    Neuronpedia

    INDEX
    Explanations

    place, interaction, something

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     whose
    0.50
    **
    0.48
    当你
    0.44
     and
    0.44
       
    0.41
     yang
    0.39
     cuya
    0.39
     that
    0.38
     meng
    0.38
     both
    0.37
    POSITIVE LOGITS
     obwohl
    0.50
     oppure
    0.48
     এছাড়া
    0.47
    蜼
    0.45
    ؛
    0.44
     😂😂
    0.42
    '،
    0.41
    었고
    0.41
     tiež
    0.40
    выше
    0.40
    Act Density 0.005%

    No Known Activations