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

    Neuronpedia

    INDEX
    Explanations

    sequence after discord

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    artiste
    0.41
    no
    0.39
    بعاد
    0.38
    YNAM
    0.37
    SELF
    0.37
     জিনিসের
    0.37
     art
    0.37
    㸵
    0.37
     gjøre
    0.37
    padding
    0.37
    POSITIVE LOGITS
     as
    0.45
     TOT
    0.44
    Q
    0.44
     чолові
    0.43
     DOA
    0.43
    K
    0.42
    S
    0.42
     recklessly
    0.42
    $\
    0.42
    全体
    0.41
    Act Density 0.032%

    No Known Activations