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

    Neuronpedia

    INDEX
    Explanations

    Data-Oriented Design

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    с
    0.54
    ك
    0.54
    ل
    0.52
    ප
    0.51
    ч
    0.48
    س
    0.47
    غ
    0.47
    أي
    0.47
    К
    0.47
    స్
    0.45
    POSITIVE LOGITS
     hyvin
    0.50
     国
    0.49
     ミニ
    0.47
     {}",
    0.46
     물질
    0.45
     ಮಾಡಿದ
    0.45
     vật
    0.45
     הח
    0.44
    した
    0.44
     évo
    0.44
    Act Density 0.001%

    No Known Activations