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

    Neuronpedia

    INDEX
    Explanations

    abilities that heal or protect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ору
    0.46
     архе
    0.46
    mathb
    0.43
     섬
    0.43
     अपराधी
    0.40
     Adjusted
    0.40
    листи
    0.39
     thước
    0.38
     чувстви
    0.38
    мор
    0.38
    POSITIVE LOGITS
     buffs
    0.80
     potions
    0.80
     buff
    0.76
     restorative
    0.69
    Buff
    0.68
     healing
    0.68
     regeneration
    0.68
     curative
    0.66
     protective
    0.66
     regenerative
    0.66
    Act Density 0.018%

    No Known Activations