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

    Neuronpedia

    INDEX
    Explanations

    acronyms in parentheses

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     that
    1.24
     on
    1.20
      
    1.19
     (
    1.18
     That
    1.18
     *
    1.12
     que
    1.11
     من
    1.08
     get
    1.07
     to
    1.07
    POSITIVE LOGITS
    including
    1.53
    /)
    1.50
    நபியே
    1.50
    англ
    1.41
    excluding
    1.36
    siehe
    1.36
    %)
    1.32
    粡
    1.31
    ;;)
    1.28
    및
    1.27
    Act Density 1.001%

    No Known Activations