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

    Neuronpedia

    INDEX
    Explanations

    what is believed or thought

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     付き
    0.45
    🫢
    0.43
     NOTHING
    0.41
    attachments
    0.40
    Adj
    0.40
    ҳ
    0.40
     ВС
    0.39
     تطبيق
    0.39
    partite
    0.38
    covers
    0.38
    POSITIVE LOGITS
    ίναι
    0.41
    是如何
    0.39
    лец
    0.39
     attention
    0.38
    인의
    0.38
    心中的
    0.38
     medium
    0.38
     defining
    0.38
    士
    0.36
     is
    0.36
    Act Density 0.011%

    No Known Activations