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

    Neuronpedia

    INDEX
    Explanations

    overly sentimental romantic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     negligently
    0.42
     negligent
    0.42
    懒
    0.41
     vibration
    0.41
    箓
    0.40
     Oxidation
    0.39
     relaxed
    0.39
     sterile
    0.39
     austerity
    0.39
     laziness
    0.39
    POSITIVE LOGITS
     sentimental
    1.10
     mush
    1.03
     gooey
    1.02
     romantic
    0.86
     cheesy
    0.84
     romant
    0.84
     swo
    0.82
     Romantic
    0.82
    Romantic
    0.82
     melodrama
    0.79
    Act Density 0.053%

    No Known Activations