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

    Neuronpedia

    INDEX
    Explanations

    fire suppression

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    OutOfBounds
    0.66
     Lights
    0.62
    柵
    0.61
    rectomy
    0.60
     disqualified
    0.59
     বিশ্র
    0.59
     Pauline
    0.59
    Pil
    0.59
     Nails
    0.58
    โต
    0.58
    POSITIVE LOGITS
     extin
    1.42
     extinguish
    1.39
     put
    1.34
     extinguishing
    1.33
     extingu
    1.31
     suppression
    1.29
     Ext
    1.29
     extinguished
    1.27
     Suppression
    1.26
     extinction
    1.19
    Act Density 0.015%

    No Known Activations