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

    Neuronpedia

    INDEX
    Explanations

    technical documentation and lists

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     três
    0.53
     exclusivement
    0.50
     trzech
    0.50
     cidade
    0.50
    Tenemos
    0.47
    سے
    0.47
     ಎರಡು
    0.47
    deux
    0.46
     quatro
    0.46
    கடந்த
    0.46
    POSITIVE LOGITS
    /
    0.47
    자기
    0.45
     lengthening
    0.45
     paperwork
    0.44
     “
    0.44
    脾
    0.44
    (
    0.42
     product
    0.40
     dig
    0.40
     자기
    0.40
    Act Density 0.001%

    No Known Activations