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

    Neuronpedia

    INDEX
    Explanations

    Kh/خ/ข/X followed by various scripts and suffixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kh
    0.55
     Kh
    0.54
    Kh
    0.52
    kh
    0.48
    кха
    0.45
    Khud
    0.44
    kha
    0.43
     Kha
    0.43
     bou
    0.41
     kha
    0.41
    POSITIVE LOGITS
    Mir
    0.44
    mir
    0.43
     Mir
    0.42
    azoline
    0.41
    رد
    0.40
     Kel
    0.40
    rol
    0.38
    rta
    0.38
    İL
    0.38
    <0xB0>
    0.37
    Act Density 0.001%

    No Known Activations