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

    Neuronpedia

    INDEX
    Explanations

    key features: dynamic computational graph

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ies
    0.43
     stationery
    0.41
     nondegenerate
    0.40
     एकमात्र
    0.39
     incap
    0.39
     unleashed
    0.39
     daisies
    0.39
     heroine
    0.38
     Immobil
    0.38
     vacation
    0.37
    POSITIVE LOGITS
     удовле
    0.49
    必要的
    0.46
     nécessaire
    0.45
    하여
    0.45
    请求
    0.45
     клиента
    0.44
    гноз
    0.43
    맺
    0.43
    개를
    0.43
    의
    0.43
    Act Density 0.007%

    No Known Activations