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

    Neuronpedia

    INDEX
    Explanations

    Okay, "Okay"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     seen
    0.80
     نقصان
    0.78
     temporary
    0.76
     خواهد
    0.75
    些
    0.74
     簡単
    0.74
     undirected
    0.74
     사용
    0.72
     compatible
    0.72
     indicates
    0.72
    POSITIVE LOGITS
    Oh
    1.91
     Oh
    1.77
     oh
    1.76
    Ah
    1.55
    oh
    1.42
    Honestly
    1.40
     gosh
    1.37
    Wow
    1.35
     OH
    1.32
    Absolutely
    1.31
    Act Density 0.904%

    No Known Activations