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

    Neuronpedia

    INDEX
    Explanations

    than humans

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.44
     tongs
    0.42
     تتم
    0.41
    Tow
    0.41
     зону
    0.40
     Nintendo
    0.39
    Gil
    0.39
    Kend
    0.38
    Glue
    0.38
     توا
    0.38
    POSITIVE LOGITS
    高
    0.51
    지
    0.48
     proč
    0.46
     puțin
    0.45
     মুর্
    0.44
     unlink
    0.44
    ವಾದ
    0.43
     확률
    0.42
    理
    0.42
    曳
    0.42
    Act Density 0.004%

    No Known Activations