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

    Neuronpedia

    INDEX
    Explanations

    former and latter

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     상당히
    0.88
     любых
    0.88
    Very
    0.86
     정말
    0.85
    一些
    0.83
     farklı
    0.82
     äußerst
    0.82
     qualsiasi
    0.82
    বিভিন্ন
    0.82
     jeszcze
    0.81
    POSITIVE LOGITS
     latter
    3.99
    后者
    2.94
    前者
    2.79
     former
    2.25
    former
    1.86
     Former
    1.79
     letz
    1.70
    Former
    1.58
     Latter
    1.46
     later
    1.41
    Act Density 0.395%

    No Known Activations