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

    Neuronpedia

    INDEX
    Explanations

    kickstart, kickoff, kickass

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ir
    1.32
    $
    1.28
    ED
    1.26
    س
    1.26
    j
    1.24
    ES
    1.22
    2
    1.12
    !
    1.11
    ک
    1.10
    IC
    1.09
    POSITIVE LOGITS
    ير
    1.43
     on
    1.41
    ↵↵
    1.20
    to
    1.14
    出会
    1.13
     kicked
    1.07
     kicking
    1.04
    </h3>
    1.01
     kick
    1.00
     Kick
    1.00
    Act Density 0.003%

    No Known Activations