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

    Neuronpedia

    INDEX
    Explanations

    uses of the words "and", "or", "will", "would", "should", and "to", possibly indicative of logical combinations, cause/effect, or actions taken.

    oai_token-act-pair · gemini-2.0-flash

    and/or

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     geweest
    -0.82
     đã
    -0.79
    Went
    -0.78
    Has
    -0.77
     InputDecoration
    -0.77
    选择了
    -0.77
     Has
    -0.75
     Exactos
    -0.74
    Took
    -0.73
     désolés
    -0.72
    POSITIVE LOGITS
     will
    1.04
     would
    0.88
     might
    0.81
     could
    0.78
     may
    0.73
     be
    0.69
    will
    0.63
     serão
    0.63
     serán
    0.62
     sẽ
    0.61
    Act Density 1.345%

    No Known Activations