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

    Neuronpedia

    INDEX
    Explanations

    is a/an/the

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Các
    1.04
     които
    1.02
    They
    1.02
    これらの
    1.00
     этими
    1.00
     onları
    0.97
    các
    0.93
    Các
    0.93
    they
    0.92
     They
    0.91
    POSITIVE LOGITS
     meant
    1.75
     supposed
    1.67
     actually
    1.65
     going
    1.57
     used
    1.55
     called
    1.53
     gonna
    1.50
     considered
    1.41
     located
    1.40
     not
    1.37
    Act Density 0.733%

    No Known Activations