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

    Neuronpedia

    INDEX
    Explanations

    focusing and analyzing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.48
     Iodine
    0.46
     supplementation
    0.46
    缓冲区
    0.45
    ící
    0.44
    .}
    0.43
     iodine
    0.43
     بٹ
    0.42
    um
    0.41
    ONE
    0.41
    POSITIVE LOGITS
    ために
    0.50
    共に
    0.45
    Brush
    0.44
     всіх
    0.44
    すべての
    0.43
    この
    0.42
    多くの
    0.42
    あなたの
    0.42
     ഇപ്പോൾ
    0.42
    Enroll
    0.42
    Act Density 0.006%

    No Known Activations