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

    Neuronpedia

    INDEX
    Explanations

    pre/post structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    奭
    0.41
     idols
    0.40
    onyms
    0.38
    vänd
    0.37
    Sil
    0.37
    roses
    0.37
     invasions
    0.37
    ouses
    0.37
    傀
    0.36
    TG
    0.36
    POSITIVE LOGITS
    етка
    0.47
     Rowe
    0.43
     Rockwell
    0.41
     Bhand
    0.40
     bap
    0.39
     Lips
    0.38
    Ⴊ
    0.38
    不在
    0.38
    чается
    0.38
     కాదు
    0.37
    Act Density 0.000%

    No Known Activations