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

    Neuronpedia

    INDEX
    Explanations

    Times New Roman

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    LookAndFeelInfo
    0.70
    BetMap
    0.68
     目前
    0.64
    能
    0.62
    ז
    0.62
     جوړونک
    0.61
     atthakath
    0.61
    نة
    0.60
    ר
    0.60
    اتي
    0.60
    POSITIVE LOGITS
     —
    0.81
     I
    0.57
    u
    0.55
    LA
    0.55
     *
    0.54
     |
    0.54
     הוא
    0.53
     되어
    0.53
     subtraction
    0.52
     gearbox
    0.52
    Act Density 0.001%

    No Known Activations