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

    Neuronpedia

    INDEX
    Explanations

    multilingual suffixes and inflections

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     není
    0.21
     eğer
    0.21
     आपण
    0.21
    资源
    0.20
     Importantly
    0.20
    Giving
    0.20
    если
    0.20
     बातों
    0.20
     ikke
    0.20
    資源
    0.20
    POSITIVE LOGITS
     통해
    0.24
    on
    0.22
     포함
    0.21
     (${
    0.20
     bevorzug
    0.19
     중심으로
    0.19
     ලබා
    0.19
     chiqar
    0.19
    ere
    0.19
     (`
    0.19
    Act Density 0.024%

    No Known Activations