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

    Neuronpedia

    INDEX
    Explanations

    abstract qualities and contemplation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sadistic
    1.00
     egregious
    0.97
     ухуд
    0.96
     obviamente
    0.95
     일반적으로
    0.94
     agress
    0.93
     deleterious
    0.93
    இதனால்
    0.93
    📉
    0.92
     hypotension
    0.92
    POSITIVE LOGITS
     stillness
    1.25
     beautiful
    1.12
     perhaps
    1.11
     timeless
    1.10
     beautifully
    1.08
     Perhaps
    1.08
     humbly
    1.07
     life
    1.04
     treasures
    1.03
     heart
    1.02
    Act Density 0.247%

    No Known Activations