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

    Neuronpedia

    INDEX
    Explanations

    parsing structured text and attributes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    mouth
    0.47
    िका
    0.44
    ﻱ
    0.43
    Πα
    0.41
    مس
    0.41
    στ
    0.41
    ك
    0.40
    ひ
    0.40
    階
    0.40
    िख
    0.39
    POSITIVE LOGITS
    ity
    0.50
    azgo
    0.50
    aço
    0.48
     raport
    0.47
     شرکت
    0.47
    ermak
    0.46
    öst
    0.46
     예측
    0.46
    >//
    0.45
     توز
    0.45
    Act Density 0.000%

    No Known Activations