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

    Neuronpedia

    INDEX
    Explanations

    conditional execution and system operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    is
    0.51
    สาร
    0.49
    ead
    0.48
    Search
    0.47
    ia
    0.46
    يا
    0.46
    على
    0.45
    ata
    0.45
    i
    0.45
    ية
    0.44
    POSITIVE LOGITS
     penalties
    0.50
     introns
    0.47
     cakes
    0.46
     algún
    0.46
     límites
    0.45
     prácticas
    0.45
     cancelación
    0.44
     electrolytes
    0.44
     políticas
    0.43
     रखेंगे
    0.43
    Act Density 0.003%

    No Known Activations