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

    Neuronpedia

    INDEX
    Explanations

    abandoned and irrelevant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     oike
    0.70
    特别
    0.66
    ave
    0.66
    semi
    0.65
     semi
    0.63
    explicit
    0.62
     chcete
    0.61
     not
    0.61
    ERK
    0.61
     বিশেষ
    0.60
    POSITIVE LOGITS
     irrelevant
    1.11
     anyway
    1.08
     anyways
    1.02
     insign
    0.99
     shrugged
    0.97
     unimportant
    0.92
    nonsense
    0.91
     Ignore
    0.90
     shrug
    0.90
     negligible
    0.89
    Act Density 0.358%

    No Known Activations