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

    Neuronpedia

    INDEX
    Explanations

    bold, rebellious, assertive language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     peaceful
    0.74
    優しい
    0.73
     gentle
    0.73
     peacefully
    0.69
     douceur
    0.66
     Gentle
    0.64
    优雅
    0.64
    peace
    0.60
     gracefully
    0.59
     graceful
    0.57
    POSITIVE LOGITS
     aggressive
    1.04
     edgy
    1.04
     adrenaline
    0.93
     fiery
    0.88
     agres
    0.88
     rebellious
    0.88
     aggression
    0.86
     bold
    0.85
     agress
    0.85
     bolder
    0.84
    Act Density 1.090%

    No Known Activations