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

    Neuronpedia

    INDEX
    Explanations

    like or as analogies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     only
    0.50
     selfish
    0.49
     conditional
    0.45
     allowing
    0.44
     guarantee
    0.44
     ONLY
    0.44
     death
    0.43
     Death
    0.43
     Real
    0.42
    ਾਮ
    0.41
    POSITIVE LOGITS
     etcétera
    0.55
     düşünü
    0.53
    新冠
    0.51
     kült
    0.50
    是什麼
    0.50
     vorgestellt
    0.50
     étroites
    0.49
     hablaremos
    0.48
     ¿?
    0.48
    叭
    0.48
    Act Density 0.010%

    No Known Activations