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

    Neuronpedia

    INDEX
    Explanations

    plate at / with

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     erhö
    0.48
     refleja
    0.46
     aforementioned
    0.45
     mük
    0.43
     revê
    0.43
     appunto
    0.42
     résult
    0.41
     démont
    0.41
     réel
    0.41
    bordered
    0.40
    POSITIVE LOGITS
     และ
    0.47
    后果
    0.47
     और
    0.46
     所有
    0.45
    ్ర
    0.44
    ור
    0.44
    不管
    0.44
     અને
    0.43
     ต้อง
    0.43
    琀
    0.43
    Act Density 0.008%

    No Known Activations