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

    Neuronpedia

    INDEX
    Explanations

    length of collections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     "
    -1.32
    u
    -1.17
    各种
    -1.17
    EdgeInsets
    -1.15
    iric
    -1.13
    还有
    -1.12
     تضيف
    -1.11
    istice
    -1.10
     zrobić
    -1.10
     for
    -1.10
    POSITIVE LOGITS
     of
    2.70
     it
    2.25
     the
    1.77
     its
    1.63
     our
    1.63
     you
    1.50
     this
    1.48
     your
    1.41
    謇
    1.35
     my
    1.26
    Act Density 0.002%

    No Known Activations