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

    Neuronpedia

    INDEX
    Explanations

    terminal emulator code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AL
    0.57
     وي
    0.57
     andro
    0.56
     durch
    0.56
     Dub
    0.56
     Ald
    0.55
    ।
    0.55
     في
    0.55
     DE
    0.55
     Sv
    0.55
    POSITIVE LOGITS
    anlı
    0.55
    ofile
    0.54
     جهانی
    0.52
    immediate
    0.52
    incremental
    0.50
    immersive
    0.48
    ۰
    0.48
     continuité
    0.48
    compressed
    0.47
    λλη
    0.47
    Act Density 0.000%

    No Known Activations