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

    Neuronpedia

    INDEX
    Explanations

    safe words, GO, retro

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ؤ
    0.42
     seat
    0.42
    支給
    0.41
     treasurer
    0.40
    டித்த
    0.40
    विण्यात
    0.40
    coil
    0.39
     windscreen
    0.39
    Size
    0.38
     procesu
    0.38
    POSITIVE LOGITS
    んに
    0.47
     тут
    0.46
     такими
    0.46
     ато
    0.46
     απο
    0.46
     дизайна
    0.45
     ре
    0.45
     οικο
    0.45
     traduce
    0.45
     அபி
    0.44
    Act Density 0.004%

    No Known Activations