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

    Neuronpedia

    INDEX
    Explanations

    represents value types

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     +
    1.23
     -
    1.20
     if
    1.09
        
    1.08
    5
    1.07
    +
    1.06
     with
    1.05
     at
    1.05
    _
    1.04
     the
    1.02
    POSITIVE LOGITS
     inherently
    1.34
     proizvoda
    1.31
     predicates
    1.30
     predicate
    1.25
     collezione
    1.25
     benutzer
    1.18
     voidaan
    1.17
     accessori
    1.17
    Zawodnik
    1.14
     linguaggio
    1.13
    Act Density 1.039%

    No Known Activations