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

    Neuronpedia

    INDEX
    Explanations

    death, penalty, threats, core

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Wren
    1.00
    南京
    0.98
     broken
    0.97
    蓣
    0.94
    aec
    0.93
    ação
    0.90
     torn
    0.84
     massac
    0.84
     vomit
    0.84
    anity
    0.83
    POSITIVE LOGITS
     penalty
    1.12
    penalty
    1.05
    Penalty
    1.04
     Penalty
    1.03
     мозга
    1.02
    せる
    0.98
    posaż
    0.96
    bed
    0.92
    مة
    0.92
    缓
    0.91
    Act Density 0.089%

    No Known Activations