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

    Neuronpedia

    INDEX
    Explanations

    positive affirmations and well wishes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     determining
    0.85
     emphasizing
    0.82
    具有
    0.79
    determining
    0.76
     Determining
    0.74
     examined
    0.73
     qualitatively
    0.73
    完全に
    0.71
     detailed
    0.70
     providing
    0.70
    POSITIVE LOGITS
     goodies
    0.99
     galore
    0.95
     kitty
    0.83
     aiuto
    0.83
     yummy
    0.83
     baddies
    0.81
     hehe
    0.78
     bling
    0.73
     goodness
    0.73
     vibe
    0.71
    Act Density 0.062%

    No Known Activations