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

    Neuronpedia

    INDEX
    Explanations

    identify the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    一个是
    0.40
    ㈤
    0.39
    ୍ୟ
    0.38
    valve
    0.38
     utter
    0.37
    rivate
    0.37
     averse
    0.37
     Asians
    0.36
     valve
    0.35
     palpable
    0.35
    POSITIVE LOGITS
     నియ
    0.40
    ذف
    0.40
    EMA
    0.37
    ckså
    0.37
     управ
    0.37
     सूट
    0.37
     تحصیل
    0.36
    চ্ছন্ন
    0.36
    ловек
    0.36
     Hats
    0.36
    Act Density 0.001%

    No Known Activations