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

    Neuronpedia

    INDEX
    Explanations

    survival proportion, world speeds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    جاتا
    0.45
    మాల
    0.40
     francés
    0.39
    ArgsConstructor
    0.38
     açıkl
    0.37
     Stoner
    0.37
     bet
    0.37
    ガル
    0.37
     equalize
    0.37
    🎙
    0.37
    POSITIVE LOGITS
    <0xA7>
    0.39
    мый
    0.37
     Repe
    0.37
    ெ
    0.36
    мы
    0.36
     kunde
    0.36
    \{-
    0.36
     Puls
    0.35
     parameterType
    0.35
    찝
    0.35
    Act Density 0.001%

    No Known Activations