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

    Neuronpedia

    INDEX
    Explanations

    building character

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    optimal
    0.46
    行为
    0.45
    Optimal
    0.45
     pillar
    0.45
     Behavior
    0.44
     behavior
    0.43
     mathematically
    0.43
    behavior
    0.43
     pillars
    0.42
    Behavior
    0.42
    POSITIVE LOGITS
     character
    0.96
    character
    0.95
     vernacular
    0.92
     karakter
    0.87
     carácter
    0.82
     Character
    0.80
     caractère
    0.77
    Character
    0.77
     харак
    0.77
     carattere
    0.76
    Act Density 0.008%

    No Known Activations