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

    Neuronpedia

    INDEX
    Explanations

    intensive invite cells what by bottle weak why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     abras
    0.47
    وبة
    0.40
     solvation
    0.40
     accelerator
    0.38
    ™,
    0.38
     fixture
    0.38
     cyclopent
    0.38
     inches
    0.38
     advis
    0.38
    ,
    0.38
    POSITIVE LOGITS
    ו
    0.53
    i
    0.49
    恂
    0.48
    י
    0.48
    ።
    0.48
     года
    0.47
    ți
    0.46
    u
    0.46
    ู
    0.46
    Maine
    0.45
    Act Density 0.008%

    No Known Activations