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

    Neuronpedia

    INDEX
    Explanations

    years of, decades of, months of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     BUT
    0.47
    UK
    0.44
    com
    0.41
    的東西
    0.40
     twelve
    0.40
     SIX
    0.39
     YOU
    0.39
    তিন
    0.39
     WAY
    0.39
     אחד
    0.39
    POSITIVE LOGITS
     refrain
    0.48
     سایر
    0.47
     notificações
    0.45
     prohibitive
    0.44
    摇头
    0.44
     disbelief
    0.44
    馋
    0.44
     tantos
    0.43
     otros
    0.43
     stigmat
    0.43
    Act Density 0.002%

    No Known Activations