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

    Neuronpedia

    INDEX
    Explanations

    action followed by consequence or description

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    薷
    0.49
     фон
    0.49
     Punto
    0.47
    狨
    0.46
    Ан
    0.46
    移民
    0.45
     PublicKey
    0.44
     און
    0.44
    ゼロ
    0.44
     Andrei
    0.44
    POSITIVE LOGITS
    嵩
    0.45
     بیش
    0.44
    ক্ষা
    0.43
    ovol
    0.42
    rought
    0.42
     FSC
    0.41
    opis
    0.41
    avio
    0.41
    og
    0.41
    accurate
    0.40
    Act Density 0.002%

    No Known Activations