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

    Neuronpedia

    INDEX
    Explanations

    describing states or explanations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     lieve
    0.43
     tzw
    0.35
    طيع
    0.35
     genauso
    0.34
     decays
    0.33
     بتوان
    0.32
    ونکہ
    0.32
     Theoretically
    0.31
     لگے
    0.31
     jett
    0.31
    POSITIVE LOGITS
    .”
    0.38
    И
    0.38
    ."
    0.37
    ↵↵
    0.37
     periódico
    0.36
    ！
    0.36
    सु
    0.35
    į
    0.35
    ↵↵↵
    0.35
     Archivado
    0.35
    Act Density 0.135%

    No Known Activations