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

    Neuronpedia

    INDEX
    Explanations

    numbered list items

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Origins
    0.66
    ₂+
    0.63
     auxilia
    0.63
    zsche
    0.63
     benefits
    0.63
     benefici
    0.63
    观测
    0.62
     beneficial
    0.61
     amén
    0.61
     aiding
    0.61
    POSITIVE LOGITS
     NEVER
    1.37
     PLEASE
    1.36
     DO
    1.33
     NO
    1.32
     MOST
    1.29
     BEFORE
    1.28
     STOP
    1.27
     ALWAYS
    1.24
     DON
    1.23
     WE
    1.22
    Act Density 0.372%

    No Known Activations