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

    Neuronpedia

    INDEX
    Explanations

    kindness and charm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     anore
    0.47
     lifeless
    0.42
    醜
    0.41
     selfish
    0.40
     primitive
    0.40
    🤮
    0.40
     stagnant
    0.38
     sécr
    0.38
     raped
    0.38
     adecuada
    0.38
    POSITIVE LOGITS
     kindness
    0.89
     wry
    0.88
     gentle
    0.88
     witty
    0.86
     charming
    0.85
     infectious
    0.84
     charm
    0.82
     humor
    0.76
     charisma
    0.76
     gent
    0.75
    Act Density 0.117%

    No Known Activations