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

    Neuronpedia

    INDEX
    Explanations

    push, pushed, pushing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    z
    1.29
    k
    1.26
    ،
    1.23
    j
    1.23
    2
    1.22
    ss
    1.13
    to
    1.13
     Increases
    1.13
    de
    1.11
    า
    1.11
    POSITIVE LOGITS
     pushed
    1.23
     combate
    1.22
     pouss
    1.19
     đẩy
    1.16
     unimagin
    1.15
    લ
    1.15
    ب
    1.13
     correcto
    1.12
     comprende
    1.11
     przede
    1.11
    Act Density 0.023%

    No Known Activations