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

    Neuronpedia

    INDEX
    Explanations

    maintain balance and order

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    reward
    0.79
     Merci
    0.76
     destiny
    0.75
     offres
    0.74
     possibilités
    0.74
     métall
    0.74
     releg
    0.73
     remer
    0.73
     rewarded
    0.73
     moveTo
    0.72
    POSITIVE LOGITS
     balance
    1.01
    balance
    0.87
     composure
    0.87
    Balance
    0.86
     homeostasis
    0.84
     equilibrio
    0.83
     equilibrium
    0.82
     Balance
    0.81
     equilib
    0.79
    平衡
    0.79
    Act Density 0.149%

    No Known Activations