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

    Neuronpedia

    INDEX
    Explanations

    measuring specific factors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    да
    0.63
    Б
    0.58
    на
    0.55
    М
    0.53
    ד
    0.52
    ETER
    0.50
    О
    0.50
    ста
    0.49
    Ти
    0.48
    Ли
    0.48
    POSITIVE LOGITS
     før
    0.44
     initial
    0.42
     قبل
    0.42
     sebelum
    0.42
     ingin
    0.41
     potentiel
    0.40
     أثناء
    0.40
     කිරීමට
    0.40
     πριν
    0.40
     trước
    0.39
    Act Density 0.008%

    No Known Activations