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

    Neuronpedia

    INDEX
    Explanations

    superscripts and mathematical formulas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     EVER
    0.92
     hug
    0.88
     ESL
    0.82
     ผู้
    0.82
     typography
    0.81
     bookstore
    0.79
     Studio
    0.79
     bookstores
    0.78
     denied
    0.78
     Democratic
    0.76
    POSITIVE LOGITS
    k
    0.95
    i
    0.95
    n
    0.95
    ⺀
    0.94
    prime
    0.91
    *}
    0.88
    infty
    0.87
    j
    0.87
    degree
    0.84
    infinity
    0.80
    Act Density 0.116%

    No Known Activations