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

    Neuronpedia

    INDEX
    Explanations

    darkness shadows silence

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     praticamente
    0.53
     effettuare
    0.51
     extremadamente
    0.50
     extremely
    0.50
    非常に
    0.50
    めっちゃ
    0.49
     situazione
    0.49
     prácticamente
    0.48
     bzw
    0.47
     effectuer
    0.47
    POSITIVE LOGITS
     whispers
    0.68
     shadows
    0.63
     darkness
    0.60
     whispered
    0.59
     sombras
    0.54
     whispering
    0.53
     silence
    0.53
     Darkness
    0.53
     emptiness
    0.52
     sueños
    0.52
    Act Density 0.341%

    No Known Activations