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

    Neuronpedia

    INDEX
    Explanations

    innovative designs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     apparently
    0.44
    ).
    0.41
     goodies
    0.40
     anti
    0.39
     designed
    0.39
     Kn
    0.39
    PostBody
    0.39
     presumably
    0.39
     Gilmore
    0.39
     throws
    0.38
    POSITIVE LOGITS
     جديده
    0.43
    現在
    0.42
     ยัง
    0.41
    覺
    0.40
    樹
    0.40
     ještě
    0.39
     еще
    0.39
     ešte
    0.39
    ቃ
    0.38
     ඉක්
    0.38
    Act Density 0.002%

    No Known Activations